Euradwaste '08 - EU Bookshop - Europa

Interested in European research?
Research* eu is our monthly magazine keeping you in touch with main developments (results, programmes,
events, etc.).
It is available in English, French, German and Spanish. A free sample copy or free subscription can be obtained
from:
European Commission
Directorate-General for Research
Communication Unit
B-1049 Brussels
Fax (32-2) 29-58220
E-mail: research-eu@ec.europa.eu
Internet: http://ec.europa.eu/research/research-eu
Cover:
Background: Mont Terri Underground Rock Laboratory
© Cornet Photoshopping GmbH, Zürich
Graphic:
courtesy of ANDRA, France
Courtesy of ONDRAF/NIRAS, Belgium
EUROPEAN COMMISSION
Directorate-General for Research
Directorate J — Energy (Euratom)
Unit J.2— Fission
European Commission
Office CDMA 01/61
B-1049 Brussels
Tel. (32-2) 29-61 670
Fax (32-2) 29-54 991
E-mail: Christophe.Davies@ec.europa.eu

EUROPEAN COMMISSION
Euradwaste '08
Seventh European Commission Conference on the Management
and Disposal of Radioactive Waste
Community Policy and
Research & Training Activities
Edited by C. Davies
Directorate-General for Research
2009 Euratom EUR 24040

LEGAL NOTICE
Neither the European Commission nor any person acting on behalf of the Commission is responsible for the use
which might be made of the following information.
The views expressed in this publication are the sole responsibility of the author and do not necessarily reflect the
views of the European Commission.
A great deal of additional information on the European Union is available on the Internet.
It can be accessed through the Europa server (http://europa.eu).
Cataloguing data can be found at the end of this publication.
Luxembourg: Publications Office of the European Union, 2009
ISBN 978-92-79-13105-9
doi: 10.2777/46864
© European Communities, 2009
Reproduction is authorised provided the source is acknowledged.
Printed in France
PRINTED ON WHITE CHLORINE-FREE PAPER
EUROPE DIRECT is a service to help you find answers
to your questions about the European Union
Freephone number (*):
00 800 6 7 8 9 10 11
(*) Certain mobile telephone operators do not allow access to 00 800 numbers
or these calls may be billed

FOREWORD
The Euratom Treaty celebrated its 50 th anniversary in 2007. Promotion of research and spreading of scientific
and technical knowledge for the peaceful use of nuclear energy have been and remain a core task of the
Treaty. Since the mid-70s, the European Commission (EC) has supported R&D on the management of radioactive
waste as part of Euratom multi-annual programmes, the results being reported at successive
Euradwaste conferences. This was the 7 th conference in the series and not only was a showcase event for
R&D performed in the 6 th Euratom Framework Programme (FP6, 2002-2006) but also presented more general
socio-political issues and related EU strategy in the field.
Energy is now at the top of the political agenda, with nuclear power an important element in the debate. A
number of initiatives have recently been launched that contribute to the Community's evolving strategy in
this field. The European Nuclear Energy Forum (ENEF) is facilitating a broad stakeholder dialogue and
analysis of key issues. On the regulatory side, a High Level Group on Nuclear Safety and Waste Management
(since renamed ENSREG) has been established, and the EC has published a new proposal for a Directive
on nuclear safety. In the area of low-carbon technologies, the Community's Strategic Energy Technology
Plan specifically mentions nuclear energy and the need for waste management solutions to be developed
over the next 10 years. All this is against a backdrop of only slow progress towards implementation of geological
disposal in most (but not all) Member States' national programmes. These and related issues were the
focus of day 1 of Euradwaste'08.
In the area of European research in general, new funding instruments were introduced in FP6 to tackle fundamental
concerns such as fragmentation, lack of critical mass and coordination, and general underinvestment.
A number of Integrated Projects and one Network of Excellence were subsequently launched in the
waste area within Euratom: EUROPART and EUROTRANS in P&T; NF-PRO, FUNMIG, and PAMINA in
near-field/far-field processes and performance assessment in geological disposal; ESDRED on engineering
studies and repository designs; and ACTINET-6 on fundamental actinide sciences. These and other Euratom
projects have helped redefine the state-of-the-art in their respective areas, and have also had a major integrative
effect on the sector as a whole. They were used as focal points for the Euradwaste'08 technical sessions
in days 2 and 3, which also turned an eye to the future by soliciting views on key remaining research and
how Euratom FP7 and later Community programmes can contribute. These sessions were complemented by
technical visits to either the French URL (underground research lab) at Bure or the HADES URL and PRA-
CLAY experimental gallery at the Belgium nuclear research centre, Mol.
All conference sessions included invited presentations and panel discussions. Summaries and proceedings
are available at http://cordis.europa.eu/fp7/euratom-fission/euradwaste2008_en.html. These include summaries
for a less technical audience prepared by an independent journalist.
The conference attracted some 270 participants from 30 countries. Publicity and media involvement was a
key preoccupation. A media briefing on the status of research in geological disposal was held two weeks
before the conference at the Bure URL. Twenty journalists from eleven EU Member States took part, and
several articles were subsequently published in major newspapers or magazines. As a result of this interest,
local French public TV broadcast interviews recorded at the conference or during the technical visit to the
Bure URL on 23 October.
The EC would like to express its gratitude to all those who contributed to making Euradwaste'08 a resounding
success, in particular the chairs, rapporteurs, panel members, speakers, ANDRA for co-organising and
hosting several communication activities and the complementary session on developments in repository
technologies held at Bure on 23-24 October, and both ANDRA and EURIDICE for hosting the technical visits.
S. Webster O. Quintana Trías
Head of Unit Fission, DG Research Director Energy (Euratom), DG Research
iii

TABLE OF CONTENTS
FOREWORD iii
CONFERENCE SUMMARY
Working together to make geological disposal a reality 1
Scientific and technical presentations 7
KEYNOTES BY THE EUROPEAN COMMISSION
Mr Peter Faross, Director, DG Energy and Transport, Directorate H "Nuclear
Energy" 15
Mr Octavi Quintana Trías, Director, DG Research, Directorate J "Energy"
(Euratom) 23
SOCIO-POLITICAL AND STRATEGIC ISSUES
General introduction and objectives 27
Session I Current situation of geological disposal in the EU
Introduction and objectives 29
“Radioactive waste management: Where do we stand?” 31
Mrs Marie-Claude Dupuis, ANDRA (FR)
Panel discussion 41
Session II Economical factors governing geological disposal programmes
Introduction and objectives 45
“Assessment of financial provisions for nuclear waste management
– long-term perspective from a Finnish viewpoint” 47
Mr Eero Patrakka, Posiva Oy (FI), et al.
Panel discussion 57
v

Session III Co-operation in geological disposal
Introduction and objectives 61
“Cooperation in the development of geological disposal concepts –
benefits and challenges” 63
Ms Monica Hammarström, SKB (SE), et al.
Panel discussion 69
Session IV Communication of risk and uncertainties
Introduction and objectives 73
“Communicating the safety of radioactive waste disposal – the
perspective of a person responsible for science and technology
within an implementing organisation” 75
Mr Piet Zuidema, NAGRA (CH)
Panel discussion 87
COMMUNITY RESEARCH in RADIOACTIVE WASTE MANAGEMENT – Partitioning
and transmutation and geological disposal: 6 th Euratom Framework Programme for nuclear
research and training activities (2002-2006)
Introductory keynote:
“The Euratom research and training programme in radioactive
waste management” 91
Mr Simon Webster, Head of Unit ‘Fission’, EC, DG Research
Session V Partitioning and Transmutation and its impact on geological disposal
Introduction and objectives 103
“An overview of partitioning activities in Europe” 105
Mr Jean-Paul Glatz, EC, JRC-ITU (DE)
“Overview of activities in Europe exploring options for transmutation”
113
Mr Dankward Struwe, FZK-IRS (DE), et al.
“Impact of partitioning and transmutation on nuclear waste management
and the associated geological repositories” 125
Mr Enrique M. González Romero, CIEMAT (ES)
vi

“Impact of advanced fuel cycle scenarios on geological disposal” 141
Mr Jan Marivoet, SCK•CEN (BE), et al.
Panel discussion 153
Sessions VI to VIII Geological Disposal
Introduction and objectives 157
Session VI Near-field processes
“Advances in integrating European research on the near-field system”
161
Mr Alain Sneyers, SCK•CEN (BE), et al.
“Challenges of assessing long-term performance of nuclear waste
matrices in repository near-field environments – insights from the
NF-PRO and MICADO projects” 173
Mr Karel Lemmens, SCK•CEN (BE), et al.
“Key processes affecting the chemical evolution of the engineered
barrier system” 183
Mr David Savage, Quintessa Ltd (UK), et al.
“Impact of thermo-hydro-mechanical processes on repository performance”
193
Mr Patrik Sellin, SKB (SE), et al.
“Disturbed and damaged zones around underground openings –
effects induced by construction and thermal loading” 203
Mr Peter Blümling, NAGRA (CH), et al.
“Near-field processes – the challenge of integration into performance
assessment” 213
Mr Lawrence H. Johnson, NAGRA (CH), et al.
Panel discussion 223
Session VII Repository technologies, actinides and far-field migration processes
“ESDRED – an integrated European project focused on technology
development” 229
Mr Wolf K. Seidler, ANDRA (FR), et al.
vii

“Achievements of the ESDRED project in Buffer Construction
Technology” 239
Mr Chris De Bock, ONDRAF/NIRAS (BE), et al.
“New transport and emplacement technologies for vitrified waste
and spent fuel canisters” 259
Mr Wilhelm Bollingerfehr, DBE Technology GmbH (DE), et al.
“Emplacement of heavy canisters into horizontal disposal drifts
using fluid (air/water) cushion technology” 269
Mr Stig Pettersson, SKB (SE) et al.
“Application of low pH concrete in the construction and the operation
of underground repositories – ESDRED Module 4” 279
Mr José Luis García Siñeriz, AITEMIN (ES), et al.
“ACTINET – a network of excellence in actinide sciences” 291
Mr Thomas Fanghänel, EC, JRC-ITU (DE), et al.
“IP FUNMIG: the FP6 far-field project” 299
Mr Gunnar Buckau, FZK-INE (DE), et al.
“Radionuclide migration in clay-rich host formations: process understanding,
integration and upscaling for safety-case use (FUN-
MIG RTDC 3 + 1 & 2)” 309
Mr Scott Altmann, ANDRA (FR), et al.
“Laboratory and in situ investigations on radionuclide migration
in crystalline host rock” 327
Ms Tiziana Missana, CIEMAT (ES), et al.
“Investigation of far-field processes in sedimentary formations at
a natural analogue site – Ruprechtov” 343
Mr Ulrich Noseck, GRS (DE), et al.
“Radionuclide migration in the far-field: The use of research results
in safety cases” 353
Mr Bernhard Schwyn, NAGRA (CH), et al.
Panel discussion on far-field processes 361
Session VIII Performance assessment studies – coordination of RD&D for
waste disposal
“Performance assessment methodologies in application to guide
the development of the safety case” 369
Mr Jörg Mönig, GRS (DE)
“The treatment of uncertainty in PA and the safety case” 377
Mr Daniel Galson (Galson Sciences Co.)
viii

“Sensitivity analysis techniques for the performance assessment of
a radioactive waste repository” 387
Mr Ricardo Bolado-Lavín, EC, JRC-IE Petten (NL), et al.
“Proposed European technology platform for the co-ordination of
RD&D for geological disposal” 399
Mr Alan Hooper, NDA-RWMD (UK), et al.
Panel discussion on Coordination of RD&D for waste disposal in
Europe 409
Posters based on projects performed as part of Euratom FP6, ISTC and supported by the EC-
DG Energy and Transport (TREN)
Topic: Partitioning and transmutation
Development of the methods for immobilisation of long-lived nuclear
waste in carbon matrices for storage and transmutation
M. Abdulakhatov, V.G. Khlopin Radium Institute (RU), et al.
Topic: Near-field processes
TIMODAZ – Thermal Impact on the Damaged Zone around a Radioactive
Waste Disposal in Clay Host Rocks
Xiangling Li, EURIDICE (BE)
TIMODAZ – Lining stability under thermal load
Jaroslav Pacovsky, CTU Prague (CZ), et al.
TIMODAZ – Characterisation of Rock Mass Crack Damage Using Ultrasonic
Surveys
Juan M. Reyes-Montes, ASC (UK), et al.
TIMODAZ – Modelling the Excavated Damage Zone around an underground
gallery - Coupling mechanical, thermal and hydraulical aspects
Robert Charlier, Univ. Liège (BE), et al.
TIMODAZ – Large-scale heater experiments in boom clay
Jan Verstricht, EIG EURIDICE (BE), et al.
ix
415
423
429
435
441
447

NF-PRO – Concrete degradation and its influence on the geochemical
conditions at the concrete/bentonite interface under repository conditions
A. Escribano, CIEMAT (ES) et al.
NF-PRO – Influence of THM-GCh behaviour of the bentonite barrier on
the corrosion processes of the carbon steel canister
E Torres, CIEMAT (ES) et al.
NF-PRO – Experimental and modelling studies of the THM behaviour of
the clay barrier
María Victoria Villar, CIEMAT (ES) et al.
NF-PRO – Mechanical and permeability properties of highly precompacted
granular salt bricks
Klaus Salzer, IfG Germany (DE), et al.
NF-PRO – Impact of Bedding Planes to EDZ Evolution and the Coupled
HM Properties of Opalinus Clay
Till Popp, IfG Germany (DE), et al.
NF-PRO – Studies on long-term stability of spent fuel
Vincenzo V Rondinella, EC JRC, Karlsruhe (DE), et al.
THERESA – Evaluation and improvement of numerical THM modelling
capabilities for rock salt repositories
K. Wieczorek, GRS mbH (DE), et al.
Topic: Far-field processes
FUNMIG – Research on well-defined processes
Pascal Reiller, CEA Saclay (FR)
FUNMIG – Real system analyses of PA relevant processes in sediments:
The Ruprechtov natural analogue site
Vaclava Havlova, NRI Rež (CZ), et al.
Topic: Training fellowships
SMARAGD: The Study of Mineral Alterations of Clay Barriers Used for
Radwaste Storage and its Geological Disposal
Miroslav Honty, SCK-CEN (BE), et al.
x
453
459
465
471
477
483
489
497
503
511

Topic: Support actions
SAPIERR-II – Shared, regional repositories, developing a practical implementation
strategy
Ewoud Verhoef, COVRA (NL) et al.
Topic: EC-DG TREN
Improving financing schemes for nuclear decommissioning and radioactive
waste management in European member states and on EU level
Wolfgang Irrek, Wuppertal Institut für Klima, Umwelt, Energie (DE)
LIST OF PARTICIPANTS 533
xi
519
527

xii

CONFERENCE SUMMARIES

Working together
to make geological disposal a reality
General summary
The Euradwaste ’08 conference in Luxembourg on 20-22 October 2008 brought together researchers,
radioactive waste management organisations, policy-makers, regulators, engineers and educators
to discuss the underground disposal of spent nuclear fuel and long-lived high-level radioactive
waste as well as the impact of advanced fuel cycles (partitioning and transmutation) on deep geological
repositories. Three days of presentations, poster sessions, panel discussions and questionand-answer
sessions focused on the full spectrum of issues facing implementation of this waste
management option.
The first day of the conference dealt with the strategic, economic and socio-political aspects of geological
disposal, and presentations were universally followed by lively discussions. As the strategy
and needs of each country vary so widely, finding common ground to some of the issues on a European
level proved to be a challenging task.
Finding solutions now
In the opening address, Dr Quintana Trias, Director at the European Commission’s DG Research,
said, ‘Significant quantities of high-level radioactive waste already exist in interim surface storage,
and it is inconceivable that these accumulations remain in this situation indefinitely. Sooner or
later, society must implement a permanent long-term management solution that respects high levels
of safety and adequately protects the public and the environment, both now and in the future.’
In his keynote address, Mr Peter Faross, Director at the Commission’s DG Energy and Transport,
summarised the issue: ‘Following 30 years of research, geological disposal now represents a passively
safe and sustainable option for the long-term management of nuclear waste … and there are
solutions as regards all technical, financial and social aspects.’ The consensus among the scientific
community that geological disposal is the only option capable of fulfilling the long-term safety requirements
was made clear in subsequent presentations and discussions.
Several speakers underlined the importance of moving ahead sooner rather than later with geological
disposal. Addressing the question of how partitioning and transmutation, which aim to reduce
the amount and toxicity of the waste, effect the implementation timeline of geological disposal, Mrs
Marie-Claude Dupuis of ANDRA in France commented, ‘There will always be waste and it has to
go somewhere.’
Mr Faross referred to the recently published Eurobarometer report on attitudes towards radioactive
waste, which noted that approximately 93 % of EU citizens believe that finding a solution for managing
radioactive waste should not be left to future generations. On the same theme, Dr Quintana
Trias remarked, ‘It is the responsibility of the present generation to implement a solution, since we
have benefited from the electricity produced by today’s nuclear power plants.’
Mr Faross concluded: ‘It is time to implement this solution… Wait-and-see approaches putting burdens
on future generations and certainly waste export to countries outside the EU should not be
1

supported. The Commission believes that the eventual implementation of deep geological disposal
is an essential condition for the continued use and possible expansion of nuclear power.’
Dr Hans Forsström of the IAEA opened the first session by saying that incredible technological advances
have been made in recent decades in the development of geological repositories. ‘We know
how to do it and we know how to show that it works. Our challenge is confidence-building in the
public and in our research. It must be improved,’ he said.
‘Who fails to plan plans to fail’
Time was one of the recurring themes of the conference. ‘Time is the only thing that will make radioactivity
entirely harmless,’ said Dr Hans D. K. Codée of COVRA in the Netherlands, ‘and time
is what we must manage.’
Mrs Dupuis of ANDRA noted that while it is crucial to make solutions available right away, providing
safety demonstrations are a major scientific challenge because they are done on ‘space- and
timescales that are way above the field of experimental capabilities’.
Communicating the unavoidably long-term nature of this kind of research is a real challenge. For
example, simply building an underground laboratory takes decades, while observing the movement
of a radioactive nuclide through a millimetre of rock might take as many years. These are also serious
considerations when it comes to continuity of funding support. Representatives of the European
Commission emphasised their commitment to enabling this continuity, as the disposal of nuclear
waste continues to be high on the political agenda.
Follow the money
The session on economical factors governing geological disposal programmes revealed that determining
the ultimate cost of a geological repository is a challenge that would puzzle the most ambitious
economist. Nuclear power offers high electricity yield and low impact on the carbon cycle;
however, each country is responsible for managing its own nuclear waste, and building a geological
repository costs as much as building a nuclear power plant. This can be a problem for countries
with few reactors. Importantly, the cost of operating such a facility for the required 100 or so years
before closing it up far exceeds the building costs. As facilities must be built and operated on such
long timescales, the exact costs are very difficult to pin down, and the question of financing becomes
exceedingly complex.
Dr Eero Patrakka of Posiva in Finland presented the case of his country’s costing and financing of a
geological repository. He explained, ‘The funds for radioactive waste management must be collected
in advance and they must be available when the waste management operations are carried
out’. However, while determining how much to put aside relies on sound cost assessment, costs
themselves (such as the price of copper, the cost of labour, or the effects of continuous research) are
ever changing.
‘Footing the bill’
Dr Patrakka of Posiva showed how seemingly small changes can have serious financial impacts on
the planning of geological repositories. For example, adding one tonne of spent fuel will add approximately
EUR 0.5 million to the overall costs; adding one year to operating time has an impact
of EUR 10 million; and a change in the price of copper of EUR 1 per kg will have an impact of
2

oughly EUR 35 million. ‘It is impossible to foresee what will happen globally during the operation
of a repository over the next 100 years or so,’ he concluded, adding, ‘We need these cost calculations.
We have to demonstrate that the solutions are economically feasible. But it’s not cheap,
and the funds must be collected in advance.’
During the panel discussion that followed, Mr Jean-Paul Minon of ONDRAF/NIRAS in Belgium
said, ‘We have to foot the bill at the end of the day, and we need to think about what mechanisms
will ensure that the bill can be paid.’
Dr Codée of COVRA promoted a multinational approach to building and operating geological repositories,
explaining that the cost of a repository is economically not feasible with a very small
nuclear programme; ‘you could, alternatively, share the repository with others and share the cost.
It’s not easy, but it’s a way, and it should be a European way.’ Dr Codée added, ‘Fuel-making is an
international business. We buy electricity from other countries. Why should the disposal part be
non-international?’
Working together
As one might imagine, the question of how nations and industry manage the finances of the money
put aside for nuclear waste disposal was a matter of some debate. In some countries, such as France
and the Netherlands, the process is open, while in others, such as Belgium, that information is not
made available to the public. Because of the wide differences between countries in how finances
are managed, the question of shared-cost repositories remained open.
There is clearly widespread cooperation in research, both within Europe and worldwide. Mrs Dupuis
of ANDRA remarked, ‘Our area is one where we are not in competition with one another. We
are all moving forward together. Progress in one country will help the others.’
Several representatives of countries with smaller nuclear programmes agreed that implementing
plans for geological disposal would be helped by seeing a working example in a country with a larger
programme, such as Sweden or Finland.
This is no small feat. Countries that have put a lot of resources into basic research and development
are open to sharing what they’ve learned with other countries and allowing countries that haven’t
yet started to learn from their mistakes. However, close cooperation during the phase leading up to
actual implementation is difficult because, as Dr Patrakka of Posiva said, ‘programmes are at various
stages, their scopes and volumes are different, technical solutions vary, implementation is organised
differently and funding schemes are different’.
In the case of Slovakia, their spent-fuel strategy is based on their relationship with the former Soviet
Union, which makes implementation of geological repositories difficult for legislative reasons.
Slovenia faces a different set of challenges. Dr Irena Mele of ARAO in Slovenia explained, ‘we
have one plant, and we own half of it; Croatia owns the other half. How can we train the critical
mass, especially in terms of human resources? … International cooperation is essential for the viability
of our programme.’
Dr Piet Zuidema of NAGRA in Switzerland said, ‘Failure is the real cost issue. If you can save
money with cooperation you should do it, but not if it’s going to negatively impact the quality.’
3

Ms Monika Hammarström of SKB in Sweden presented the case of fruitful collaboration between
Sweden and Finland. ‘Cooperation enhances political and social acceptance,’ she said. ‘The work
between Finland and Sweden strengthened relations and improved knowledge in all stakeholders
involved.’ She cautioned, though, that the shift in focus from research and development to industrialisation
will change the context of cooperation. ‘International contacts and cooperation make
sense only in the context of a sufficiently extensive own programme,’ she said, ‘You need a strong
will to find solutions.’
Mr Minon of ONDRAF/NIRAS spoke about the need for harmonising the regulatory framework in
the EU: ‘Not the details, which are important in implementation, but consensus on the principal values
of security and safety.’ Dr Forsström of the IAEA added, ‘There is a clear need for political and
industrial commitment. It’s important that the changes stick, and that things can move forward.’
Mr McCombie of the Arius association in Switzerland described the work of the SAPIERR-I and -
II projects, which looked into the possibility of establishing regional repositories, while Mr Mathieson
of NDA in the UK presented the activities of a complementary project, CATT, that explored
the scope for technology transfer between Member States under the assumption that each country
has its national repository. The issue of setting up an international entity to make further steps towards
the possibility of shared facilities between interested countries after the SAPIERR projects
will be discussed at a first meeting to be held on 28 January 2009, in Brussels.
Building trust through communication and political commitment
Communication was really seen as the key issue, both between the scientific community and industry
and between regulators and the public. According to the Eurobarometer report mentioned
above, a significant number of people would change their minds about nuclear energy if they felt
there was a safe way to dispose of the waste, but at the same time most people think that it is not
possible to safely dispose of the waste.
Dr Zuidema of NAGRA presented the challenges of communicating about the safety of radioactive
waste disposal. He said that while there has been tremendous progress in communication techniques,
especially in terms of visualisation, ‘the wealth of information available and its complexity
may distract from the real issues and confuse the non-specialist.’
The reality, he said, is that nuclear energy generates relatively small volumes of waste, that there
are significant financial resources available for its management, and that there is enough time for
careful planning and implementation of geological repositories. But from a social point of view,
managing radioactive waste is different from other projects; the responsibility of the implementers,
policy-makers and regulators is to take the time to create trust and confidence.
Creating and maintaining trust, Dr Zuidema said, relies on providing adequate legislation, presenting
understandable overall goals, demonstrating visible progress within reasonable timescales, and
involving all interest groups. Dr Jordi Bruno of Amphos 21 in Spain summed it up nicely, saying
‘First, build credibility. Second, put the proper authorities in place. Third, have independent monitoring.’
4

Meeting face-to-face
Dr Daniel Galson of Galson Sciences in the UK agreed that strong and independent monitoring was
essential to building confidence. Also, he said, ‘We should talk to the public about concern-driven
risk management. Small group discussions are better than larger community group meetings or
written communication because trust is built up with an individual.’
Mr Kaj Nilsson of the Oskarshamn municipality in Sweden (a community that hosts several nuclear
facilities) emphasised the importance of respecting the people who live where repositories are
planned. ‘You must meet people person-to-person so they can see you are reliable,’ he said, adding,
‘Regulators must come out of their offices and meet people where these projects are planned.’
Perhaps one of the most poignant observations was provided by Ms Nadia Zeleznik of ARAO in
Slovenia, who reminded the audience that ‘fear is a crucial factor to the understanding of acceptability.’
It is important to keep in mind the strong psychological factors at play, she said, when
working to develop credibility and trust.
Mr Minon of ONDRAF/NIRAS acknowledged that ‘safety is a psychological concept’, and cautioned
against tailoring messages for different audiences. ‘You have to develop a message you can
take to everyone,’ he said. ‘You can say things in a simple way that is still honest.’
When to start?
So if disposing of high-level radioactive waste in geological repositories is safe and economically
feasible, why are there none in operation in Europe? Certainly public opinion plays a major role.
But additionally, the science of managing nuclear waste is constantly improving, and for policymakers
this can make it hard to resist the urge to put things off. Mr Mats Sjöborg, a Swedish policy-maker,
commented that politicians face a ‘new computer’ dilemma: each new development potentially
improves proposed designs, and these projects involve such long time frames and such astronomical
sums of money that it’s difficult to decide when it’s time to begin.
A new technology platform
From several discussions that were taking place throughout the conference, it was clear that the
community is ready to move on from more basic research and development towards implementation.
This means a shift in focus from science to engineering. To that end, a new European technology
platform (ETP) on the implementation of geological disposal of radioactive waste has been
initiated, led by SKB (Sweden) and Posiva (Finland). Dr Alan Hooper of NDA in the UK presented
outcomes of the Sixth Euratom Framework Programme (FP6) CARD project, aimed at assessing the
feasibility of this ETP. A step-by-step implementation plan has been outlined where the next step is
the drafting of a vision document that will provide the basis for organisations to commit to participation.
The vision document will be finalised early next year. The platform’s launch is planned for
the second half of 2009.
In a separate meeting about the technology platform, Mr Webster of DG Research said that the
Commission supports the effort, saying that this area of research is one with the most unifying basic
vision – that of implementation of safe geological disposal. He expressed hopes that it will follow
the success of the Sustainable Nuclear Energy Technology Platform, dealing mainly with new reactor
systems, which was launched in September 2007.
5

Dr Bruno of Amphos 21 urged the swift implementation of the platform, saying, ‘Please, can everyone
get on board as soon as possible [implying the research community alongside the radioactive
waste management organisations]. The sooner we are ‘we’, the better off we will be.’
6

Scientific and technical presentations
The second part of the Euradwaste ’08 conference was dedicated to discussing the scientific and
technical aspects of partitioning and transmutation, which aim to reduce the amount and toxicity of
radioactive waste, the near- and far-field issues that impact the development of geological repositories,
engineering studies, and aspects such as overall performance and safety assessment of these
repositories.
Approximately 270 scientists, engineers, politicians and regulators, and specialists in converging
areas had a rare opportunity to hear about the state of play in the various disciplines related to radioactive
waste management. Results from myriad FP6 (Sixth Framework Programme) projects
were presented and future directions for projects funded under Euratom in FP7 were discussed.
Because of the large spread of interest areas, and the limited time, the information imparted was for
the most part quite general and easy to follow, and dialogue was encouraged. Occasionally time
allowed projects to be reported in greater depth. Many of those who had attended the last Euradwaste
conference in 2004 felt the panel discussions added a lot to their understanding of the issues
and allowed for a meaningful level of dialogue.
Partitioning and transmutation
The technical part of the conference began with presentations on partitioning (chemical separation)
and transmutation (radionuclide conversion) of the most radiotoxic long-lived radionuclides, which
are carried out on spent fuel before disposal. Mr Ved Bhatnagar of the European Commission’s DG
Research opened the session with a broad overview of these processes and followed up with a lively
panel discussion on the effect of partitioning and transmutation (P&T) research on implementation
of geological disposal.
P&T research seeks to reduce the long-term radiotoxicity of nuclear waste to a few hundred years,
rather than the tens of thousands of years of the initial spent fuel. This would serve two purposes:
making the disposal of radioactive waste safer, in case of inadvertent human intrusion in a repository,
and reducing the size of the repositories, if the heat-bearing radionuclides are removed. The
research, as several presenters at the conference noted, does not seek to replace geological disposal
as a waste management option, but to optimise it. Whether or not P&T is effective, geological repositories
will be needed for the final product.
‘Dialogue between P&T and geological disposal communities is a goal of the European Commission,’
said Mr Bhatnagar. ‘It is also important to remember that P&T is necessary for developing
sustainability in nuclear energy.’
Mr Enrique González Romero of CIEMAT in Spain explained that ‘spent fuel is a complex material
with many kinds of radioactive isotopes with largely different characteristics. After partitioning,
the idea is that each component will be used or treated separately.’
Mr Jean-Paul Glatz of the EC Joint Research Centre’s Institute for Transuranium Elements based in
Germany gave an overview of research efforts in partitioning, which seeks to separate fission products
and actinides that share very similar properties with one another. ‘We are still doing basic
7

measurements of actinides to determine their properties and behaviour,’ he said, noting that advantages
had been found to using aluminium in pyro-processing of metallic fuels.
Future efforts in this field, Mr Glatz said, require the installation of large-scale facilities for reprocessing
demonstrations. Mr Glatz spoke about the contributions of FP7 efforts such as the projects
ACTINET-I3 (currently being negotiated and a follow-up to an existing network of excellence) and
ACSEPT, which must all be seen in the context of the Sustainable Nuclear Energy Technology
Platform.
Dr Dankward Struwe of Forschungszentrum Karlsruhe (FZK) in Germany presented the state of
transmutation activities in Europe. ‘As a result of the European research effort around the EURO-
TRANS project,’ he said, ‘first answers to the most relevant open questions of new design solutions
of plants meeting requests for optimal transmutation will be obtained. Now, the main interest is to
demonstrate the feasibility of this system using our current knowledge base.’
Mr González Romero explained how studies of transmutation, which seek to reduce long-lived radioactive
isotopes to short-lived or even stable ones, ‘highlight the importance of detailed planning
to achieve the best results of advanced facility utilisation’. The main difficulty, he said, is that the
cost of technological development and facility deployment is very high, and it can be difficult to
justify because it won’t necessarily be used for very long. As more is learned, new facilities will
eventually need to be designed and deployed.
P&T and geological disposal
The sciences of partitioning and transmutation are largely in harmony with the goals of geological
disposal, and the presentations showed that much has been achieved in recent years. Specialists in
the two areas pointed to politics as having played a role in inciting rivalry between the communities,
and stated unequivocally that research and development efforts in P&T should remain ongoing
but should not delay implementation of geological disposal.
‘We are not two scientific communities that are in confrontation (although we are chasing the same
funding),’ said Dr Jordi Bruno of Amphos 21 in Spain, ‘but politicians are using P&T to drag their
feet, and they will take a lifetime to make up their minds.’ Mr Philippe Lalieux of
ONDRAF/NIRAS in Belgium added, ‘It is important to balance the role of P&T and to avoid delaying
disposal-related decisions.’
Mr Gianluca Benamati, a member of the Italian Parliament, shared his perspective on communicating
the benefits of P&T in the management of nuclear waste. ‘Information about nuclear waste
management is most trusted when it comes from independent sources, notably scientists (41 %) and
NGOs (38 %),’ he said, referring to results of the recent Eurobarometer survey. ‘In Italy, the public
would see geological disposal differently if we had treatment beforehand to decrease the radioactivity.’
Mr González Romero of CIEMAT emphasised the need to consult the public when developing
waste-management policies that incorporate P&T, saying, ‘We are working here to develop technological
options. But each country has to consider its own constraints to deployment, including policy.
Public acceptance requires the participation of all stakeholders in decision-making.’
Mr Bernard Boullis of CEA in Saclay, France, said, ‘The back-end of the fuel cycle must be considered
soon enough to avoid possibly huge stockpiles of spent fuel.’
8

‘P&T offers a number of very interesting perspectives for waste management of future fuel cycles,’
said Mr Jan Marivoet of SCK-CEN in Belgium. ‘The interaction between materials engineers and
waste management organisations is highly desirable.’
Geological disposal
The European Commission’s framework programmes encourage implementation-oriented research
in the area of geological disposal of radioactive waste. Since 1990 there have been many EU projects
focussed on the ‘near field’ and related basic processes, especially the behaviour of engineered
barrier systems.
The NF-PRO integrated project, funded under FP6, sought to integrate European research on the
near-field processes, including the waste matrix, the canister, the man-made barriers and the surrounding
host rock. Mr Lawrence Johnson of Nagra in Switzerland reviewed this and other integrated
projects, noting achievements and areas for further study. NF-PRO, he said, ‘improved process
understanding, enhanced collaboration between specialists across national programmes, improved
communication between performance assessment specialists and those seeking deeper process
understanding, and helped identify key areas for future R&D’.
He stressed, however, that joint efforts between individual national programmes and the larger research
community are necessary for a complete synthesis of the issues. ‘A future project that attempts
to provide a synthesis of all information into improved models would be valuable. It should
consider as wide a spectrum of studies as possible, including continuing results from ongoing largescale
experiments.’
Dr David Savage of Quintessa in the UK also spoke about NF-PRO and engineered barrier systems.
In terms of modelling and integration of the data, Dr Savage said that ‘work so far has barely
touched the surface’ on experimental data that could be modelled. ‘We have a whole raft of data
now to test our models against,’ he explained.
‘I want to emphasise that we need to look at the engineered barrier system evolution on different
physical and temporal scales. The data from the lab and in situ experiments need to be placed in
context with those from natural systems. It is unrewarding to have separate programmes for experimental
and natural analogue studies,’ Mr Savage concluded.
Mr Karel Lemmens of SCK-CEN in Belgium presented some insights from the NF-PRO project
and of the MICADO project, which focused on the modelling of spent fuel dissolution. His presentation
on the performance of the waste matrix clearly imparted some valuable insights to specialists
in other scientific areas.
The far field
Mr Gunnar Buckau of Forschungszentrum Karlsruhe (FZK) in Germany remarked that in terms of
far-field studies, ‘There are no “show-stoppers”, or generic issues in the pipeline.’ One thing that
might be needed, he said, is to determine long-term changes in far-field properties that result from
anthropogenic influences or natural cycles such as deep permafrost, salination or melt-water intrusion.
It would be wise to determine the impact of such changes on the concerned minerals and retention
processes.
9

Dr Peter Blümling of Nagra in Switzerland imparted some insights from the FUNMIG project, describing
the disturbed and damaged zones around underground openings caused by construction and
thermal loading. ‘Alterations, fractions, stress, redistribution, shear and spalling are problems
common for all rock types,’ he explained, adding, ‘If you go in deep mines you see similar processes.’
The visualisations of Dr Blümling’s presentation certainly provided some food for thought. Tunnelling
causes fracture propagation, heating can increase swelling, and fissures apertures change
with the seasons. This is where the value of ‘plastic’ clays as a host environment comes into play:
the self-sealing and self-healing properties have been well studied and are well-known processes.
The same processes are not apparent in crystalline rock and are more complicated in industrial clay
or shales; the implication was that these are areas where future research efforts might do well to focus.
Dr Tiziana Missana of CIEMAT in Spain spoke about migration of radionuclides in crystalline host
rocks, concluding, ‘Realistic conditions are very important to assess the real role of colloids in radionuclide
transport. This is important for future experiments.’ ‘Studies in realistic, dynamic conditions
are needed,’ she added, noting that evaluation of overlooked processes such as microbiological
aspects is also important.
From basic research to repository engineering
Mr Wolf Seidler of ANDRA in France presented the achievements of the ESDRED project, which
aimed at demonstrating, at an industrial scale, the technical feasibility of developing, manufacturing
and testing equipments and components necessary for the construction, operation and closure of a
deep geological repository, for which there are no industrial equivalents anywhere. He explained
that all the demonstrators had met or exceeded the design specifications. Making a parallel with
Henry Ford’s first industrial car more than one hundred years ago, which paved the way to uninterrupted
improvement in car mass production, he added that ‘technical solutions for the emplacement
of waste packages, the backfilling and the sealing of cells and drifts are now at hand’.
The safety case
An enormous amount of data has clearly come out of FP6 research programmes, in all areas of nuclear
waste management. The big challenge is to put it all together, and to create multiple models
on different scales that can be used to validate conclusions.
To that end, the development of a safety case is one of the main goals of the industry at large. ‘A
safety case is a book,’ explained Dr Jörg Mönig of GRS in Germany. ‘Everyone who starts later
can benefit from previous programmes because of the publication of safety cases. A safety case
uses multiple scales to show that a case is reasonable.’
Speaking from the perspective of performance assessment, Mr Johnson of Nagra said, ‘If you look
at it all, you’ve got a whole lot of information but there is no good way to really put it all together.
Future synthesis studies will be helpful for integrating results. A larger effort to bring the story together,
to explain some of these observations, would be helpful.’
The application of knowledge generated by the FP6 projects in the safety case necessitates digesting
the large amount of data, tools and knowledge generated. Because of the sheer amount of data and
10

the high stakes involved, there is a danger that incorporating the results might become overcomplicated
and mask the real results.
‘If you compile everything with everything,’ cautioned Mr Johnson, ‘then you’re going for complexity
for complexity’s sake. Not every uncertainty is critical.’
‘I do believe very strongly that models should be fit to purpose,’ Dr Savage of Quintessa commented.
‘We can become obsessed by fitting the data from very short experiments. Short-term
data can’t automatically be extrapolated to long-term models.’
Mr Johnson concluded by saying, ‘We need multiple approaches to evaluate uncertainty. We all
understand the limitations of using individual models. If you understand your system well, you can
design around uncertainties. Joint efforts are needed to determine the priority areas with greatest
impact on the safety case.’
Daniel Galson of Galson Sciences in the UK spoke about the PAMINA integrated project and the
treatment of uncertainty in performance assessment. Commenting on the role of the regulator, he
said, ‘Most regulators want to match the level of scientific understanding and knowledge of the developer,
and this allows meaningful reviews of research, development and demonstration programmes,
safety cases and licence applications.’ He urged close dialogue between regulators and
developers, saying, ‘Dialogue must be controlled and documented and not lead to a compromise of
a regulator’s freedom to make decisions.’
Dr Jörg Mönig of GRS in Germany explained that the PAMINA project hopes to deliver a European
handbook on the state of the art of methods and tools needed for assessing the safety geological
repositories. ‘This involves so much paper you can’t imagine,’ he said. ‘Safety functions are
strongly similar on a national scale, but there are a lot of differences, even in terminology, in the
details.’
Education and basic research
Thomas Fanghänel of the EC JRC in Karlsruhe, Germany, presented some of the achievements in
basic actinide research in the ACTINET network of excellence. ‘There has been a decline in actinide
sciences in universities over the past decades because of lack of facilities demands,’ he said,
explaining that the network’s goal was to address this drop-off and get actinide sciences back into
the curricula of European universities.
The project encouraged big laboratories to open their doors to researchers, allowing them to use
their facilities, because universities don’t usually have the right set-up. ‘This is really not a simple
thing to do,’ he said. ‘We approached actinide laboratories with specialised competences, tools,
dedicated beamlines for actinide research and experienced scientists, and succeeded in gaining between
six and seven person-years at facilities.’ This effort significantly helped to bring young scientists
into the field.
On-the-job training was seen as crucial to the development of all the sciences involved, and many
speakers commented on the value of working on several sites and of continuing to train, no matter
what the area of expertise.
11

Future directions
There seemed to be consensus amongst the participants that future studies should focus on on-site
experimentation at potential repository sites.
Mr Patrik Sellin of SKB in Sweden expressed questions that remain for the study of geological disposal:
‘How can we upscale results of short-term studies to long-term projections? How can we be
sure we’re extrapolating results correctly to project to hundreds of thousands of years? Which questions
do we address for which periods of time?’
Laboratory studies can be very limiting. For example, the act of crushing rock can change its properties,
so studying the migration of radionuclides on a micro-scale can be expected to generate far
different results from studying the same phenomenon in larger pieces of rock or in the rock on-site.
Similarly, because it is extremely difficult to reproduce water-flow conditions in a laboratory, work
in this area must be done on-site.
‘Advanced geochemical modelling on sites helps generate understanding we can’t obtain in a laboratory,’
said Dr Missana of CIEMAT. ‘It is important for making projections into the future.’
‘We need to have several layers, several models because the system is so complex,’ added
Dr Altmann of ANDRA. ‘We need to use models that are appropriate for each scale.’
‘Data from lab and in situ experiments need to be placed in context with those from natural systems,’
said Dr Savage of Quintessa. ‘It is unrewarding to have separate programmes for experimental
and analogue studies.’ Mr Lemmens of SCK-CEN added, ‘Natural analogues give better stability
than we see in our labs. Natural analogues need to be understood better to explain this gap.’
Mr Buckau of FZK spoke about the need to invest resources in confidence-building, a sentiment
shared by Mr Wolf K. Seidler of ANDRA in France. ‘I am encouraged that there will be a new
technology platform,’ said Mr Seidler. ‘A complete cycle demonstration would do more for confidence-building
than anything I can think of. If the whole process was well documented and videos
well publicised, I think you could do a lot of good.’
Dr Simon Löw of ETH in Switzerland remarked, ‘One thing we’ve failed to do is conduct projects
with adequately long attention spans. Some of these questions need to be looked at for 20 years for
it to be useful. We need studies on a longer timescale.’
Mr Johnson of Nagra concluded, ‘It’s going to take decades to have a ready repository. I favour
moving ahead because we have a solid basis in fundamental research. In recent years we’ve started
spinning our wheels a bit because we can’t do our work on sites. If we were all doing this work on
site we’d learn our lessons along the way. Moving to the next step would make the quality of the
science better.’
12

KEYNOTES BY THE EUROPEAN COMMISSION
13

1. Introduction
Keynote Address
Peter Faross
European Commission
Director, DG Energy and Transport, Directorate H "Nuclear Energy"
Welcome to Luxembourg – a place with a very interesting and exciting history.
In the past, Luxembourg was an invincible fortress. Today, it is - beyond others - the seat of
the nuclear "pole" of the Commission. Some say it is still a fortress…
Long time spans appear to be also a typical feature in waste management.
– Hundred thousands of years, that is the period the safety of geological repositories
has to be demonstrated. That looks like a long time, but we also know this is nothing
compared to the duration of geological processes.
– Six years – that is the time spent already on the debate on how to urge MS to take decisions
and to develop long-term solutions, in particular for High Level Waste and
Spent Fuel. Six years are a long time span as well when comparing it to the few days
it took to come to EU rulemaking in the current bank crisis. A short period when realising
that even a well progressing repository project still requires decades from the
principle decisions to the final implementation.
Is this a reason to relax? No, the contrary is true.
Nuclear Energy is high up on national agendas again after a period of stagnation. But the
waste issue still is perceived as an unsolved issue in quite a number of countries.
Given the long times required for the implementation of geological disposal, wait-and-see
approaches are risky and are close to gambling. Who guarantees the availability of human
and financial resources over extended time spans? Who ensures the necessary political stability?
Who gives young people confidence that their children and grandchildren are not
confronted with a major waste problem without having the knowledge, the money and the
required political commitment? Will they accuse us of not having taken and implemented
decisions in time?
Who fails to plan, plans to fail!
2. The situation
Traditionally, the first day of the Euradwaste conference is the "political day", focussing on
policy, strategy and socio-economic issues.
So I would like to offer to you a snapshot of how the Commission sees the situation today,
what initiatives have been taken in the political area and what conclusions could be drawn.
15

This year saw the publication of two important Commission reports in the area of Radioactive
Waste Management:
– the 6 th Situation Report
– the 4 th Special Eurobarometer on radioactive waste
The 6 th Situation report has good and bad messages:
– First good message: there is clear progress in the disposal of short-lived low and intermediate
level waste. Today 7 of the 16 EU Member States already operate repositories
for this waste type. By 2020 disposal solutions are expected to be in place in
almost all EU Member States.
– Another good message: Finland, Sweden and France appear to progress well towards
the implementation of deep geological repositories for high level waste and spent
fuel. Operational repositories are expected up to 2025. Germany and Belgium will
possibly follow by 2040. In addition, Germany will most likely also have a deep repository
for non-heat emitting short and long-lived waste already by the end of 2013.
– Third good message: The responsibility for waste management is clearly assigned in
all the Member States
– The bad message: most of the remaining countries appear to be much less advanced
when talking about high level waste and spent fuel.
This situation has to be put in contrast to what our "clients", the citizens, feel and expect
from us as confirmed by the 2008 Eurobarometer on radioactive waste
– Citizens are now split 50-50 in their opinions about the use of nuclear energy, which
is a significant increase in favour of nuclear compared to earlier polls of this kind.
But radioactive waste management is still regarded as a major stumbling block. 4 out
of 10 of those opposed to nuclear would change their mind if there where safe and
permanent solutions in place for the management of high level waste
– Moreover, an overwhelming majority of more than 90 percent of the citizens feel that
this issue should not be left for future generations, and that each MS should have in
place a Radioactive Waste Management plan with fixed deadlines.
– But, here we have a bad message as well: citizens feel not well informed about radioactive
waste, and they are not convinced yet that geological disposal is really a safe
solution.
– In this situation, citizens clearly expect the EU to develop consistent and harmonised
methodologies and to monitor national practices and programmes.
– Do you see the parallel to the recent bank crisis? I do.
You know it better than me: for 30 years, research has been undertaken in geological disposal.
Under the Research Framework Programmes, the European Commission has invested
around 200M€ alone since 1994 to support Member States' research in the area of radioac-
16

tive waste management, a considerable part of this amount being allocated to geological
disposal of high level waste.
Today, experts agree that it has been sufficiently demonstrated that geological disposal
represents the safest and most sustainable option for the long-term management of highlevel
waste.
So why the implementation of solutions is so difficult in so many MS and even worldwide?
Of course, the Commission is well aware that geological disposal implies quite a number of
technical, financial and societal challenges and needs time for its implementation.
But, we in Europe are the leaders in the area of nuclear technology. We have mastered all
aspects of the nuclear fuel cycle and market our equipment and services throughout the
world. We should demonstrate that we also can successfully manage the radioactive waste
produced.
I hope you agree that a "mañana"-type wait-and-see approach is not an appropriate response,
given the mentioned decades required for the implementation.
3. The challenges
So, given that the technology is available, what are the real obstacles? Financing and cost?
Public acceptance? Lack of political commitment?
Lets cover Financing and Cost first.
Here we have to distinguish two issues,
– the provision of sufficient and timely funding – by the industry - for the realisation of
geological disposal and
– the total cost which create a challenge in particular for Member States with small nuclear
programmes.
Small programmes: yes, there are high fixed costs caused by research, site investigation,
underground laboratory and construction. Yes, MS with small nuclear programmes would
have to bear similar cost as those with big programmes.
But: the case of Finland demonstrates that even countries with relatively small programmes
can afford their own national repository if the process starts in time.
Cooperation with others is an important tool to minimise cost. Its potentials are proven by
the cooperation between Finland and Sweden and were in addition analysed in the CATT
project 1 under the 6 th Framework Program.
1 Coordinated Action for Transfer of Technology
17

What about multinational solutions? Theoretically, they have much to commend in terms of
economy of scale. As part of the 6 th Euratom Framework Programme, the Commission
funded the SAPIERR 2 project, which looked at the feasibility of such solutions in Europe.
But: first of all it is clear that a country must be willing to host such multinational centre
which requires political and social acceptance.
And equally important, in no way should the hope for multinational solutions be used as argument
for a wait-and-see policy. Instead, each Member State should actively seek for solutions
on its own territory which actually does not exclude offering a repository at a later
point in time to others, if the local population has build up trust and accepts the widening of
the scope.
It is also a clear Commission view that proposals from non-EU states for disposal of waste
and spent fuel should not be encouraged for technical, economical and also safety and security
reasons. This is in particular true when the potential receiving state has not put in place
the same technical, political and societal requirements and conditions as given at EU level.
At the end of the day, the real issue and the key to the problem is the timely collection of
funds covering the cost for radioactive waste management including decommissioning and
disposal. This being in place, the perceived "insurmountable barrier" disappears as the costs
of disposal are still manageable when compared to the total cost of electricity production.
The European Parliament and the Commission are putting a high importance on this very
issue. In 2006, the Commission has released a Recommendation covering form, transparency,
adequacy and auditing of funds earmarked for decommissioning and waste disposal
and is closely following-up the progress since then. Today we see that, although provisions
for a decommissioning and waste management fund are in place in the majority of the
Member States, the adequacy of the funds is still a matter of concern in most EU MS.
Timely funding is the key. Else, we would have to tell nuclear newcomers that unfortunately
they cannot proceed as their programme is not big enough to afford the management
of their wastes...
What about Public Acceptance?
The last Eurobarometer on radioactive waste has revealed that only a minority, namely 43%
of the EU citizens, believes that deep underground disposal represents the most appropriate
solution for the long-term management of highly radioactive waste. In addition, 51% fear
most the possible effects on environment and health if a repository would be built near their
home.
This lack of public acceptance and the resulting political hesitance is possibly the main reason
for the lack of progress in the EU Member States.
In fact, quite a number of repository projects failed at an early stage as they were build only
on pure scientific and technical evidence without properly involving the local population in
2 Support Action: Pilot Initiative for European Regional Repository
18

the decision making process. The sequence "decide – announce – defend" was therefore
mostly followed by "abandon".
The Waste Eurobarometer indeed revealed that almost 60% of the EU citizens want to be
directly consulted and to participate in the decision making process if a repository would be
built near their home.
That is why the implementation of geological disposal requires modern governance concepts,
building on a step-wise approach and early involvement of local stakeholders.
Such modern governance concepts have laid the foundation of the Finnish and Swedish approach.
Finally, let's address the issue of Political Commitment.
All so far successful programs have demonstrated one important issue: at the very start of
the Radioactive Waste Management process there must strong political commitment and
clear political decisions.
This should be followed by the transposition of the political decisions into a legal, regulatory
and organisational framework and all other steps required realising a repository.
So we are talking a top-down approach. What we see in practice is quite often the contrary.
Organisations created in principle to identify solutions work in a political vacuum, which I
imagine must be a frustrating task.
This has to change. Politics has to respond to the expectations expressed by the citizens, as
seen in the Eurobarometer poll.
4. EU initiatives
What is presently done at EU level?
Well, here we see at present three important initiatives which complement each other:
First of all, the European Nuclear Energy Forum, short "ENEF", was established last
year by the Commission following a European Summit dedicated to an EU Low Carbon
Energy Strategy. The Forum is jointly hosted by the Czech Republic and Slovakia and aims
at a broad discussion among all relevant stakeholders on the opportunities and risks of nuclear
energy.
Three ENEF working groups discuss opportunities, transparency and of course also the
risks of nuclear energy.
It goes without saying that Radioactive Waste Management forms an important topic here
and a dedicated subgroup is at present developing a 'Roadmap to successful implementation
of geological disposal in the European Union, in particular highlighting the essential elements
for the implementation of geological disposal as well as recommendations for national
roadmaps and actions at EU level.
19

Another important development is that the European Council endorsed in March 2007 a
Commission proposal to set up an EU High Level Group on Nuclear Safety and Waste
Management.
The mandate of the group is to progressively develop a common understanding in these
fields and, eventually, to suggest additional European rules. This work has to take into account
a set of actions formulated in Council Conclusions in May 2007.
Triggering progress in Radioactive Waste Management is forming an important objective
and again, a dedicated Subgroup has been created to cover a number of issues, including
obstacles and drivers, the better use of the Joint Convention process, guidelines for national
programmes as well as the use of best practices and peer reviews.
Finally let's talk about Research and a very recent initiative to create a Technological Platform
on the IMPLEMENTATION of Geological Disposal of Radioactive Waste.
Following a feasibility study, an interim Executive Group under Swedish leadership is right
now developing a Vision Document describing the possible scope and objectives of this
Platform.
The group arrived at the view that after decades of extensive research, development and
technical demonstration the time is ripe now to focus on the implementation of geological
disposal without further delay. It is no surprise that the draft vision suggests to cover three
key topics: confidence building, establishment of national waste management plans as well
as facilitating and maintaining access to existing competence and technology.
The Technological Platform is scheduled to be launched during 2009.
We very much hope and expect that these three initiatives will lead to significant progress in
Radioactive Waste Management.
5. Conclusions
In summary, I would like to offer the following conclusions for your consideration.
Following 30 years of research, geological disposal now represents a passively safe and
sustainable option for the long-term management of high-level waste, also proven by natural
analogues.
European citizens attach to this issue a high importance.
There are solutions as regards all technical, financial and societal aspects.
– It is therefore time to implement this solution, independent from medium- to longterm
national energy policies. Wait-and-see approaches putting burdens on future
generations and certainly waste export to countries outside the EU should not be supported.
– The Commission believes that the eventual implementation of deep geological disposal
is an essential condition for the continued use and possible expansion of nuclear
power.
20

All initiatives leading to encouraging and facilitating progress towards identification and
operation of waste repositories are therefore highly welcome.
The Commission also expects that the High Level Group will address without delay the action
imposed on them by the Council Conclusions of 8 May 2007:
– to clearly identify the need for "developing strategies for the safe management of all
types of spent fuel and radioactive waste" and
– to "urge EU Member State to establish and keep updated a national programme for
the safe management of radioactive waste and spent fuel that includes all radioactive
waste under its jurisdiction and covers all stages of management."
The Commission also has high expectations in the driving forces of ENEF and the guidance
it will provide.
The same applies to the planned new Technological Platform on the Implementation of
Geological Disposal of Radioactive Waste.
This first day of the Euradwaste conference will address many of issues I mentioned before,
the situation, the obstacles and drivers, the question of economics, and finally the important
issue of public trust.
We trust that this day will form a further important piece in the puzzle towards final solutions.
In any case I hope that you will experience today interesting and fruitful debates.
I am sure, this will not be the last time we discuss these issues.
But I have a dream: at the Euradwaste 2012 the Commission comes up with a Keynote
Speech, which has got nothing but good messages.
Geological processes last long, but even they lead to changes from time to time…
I thank you for your attention.
21

Ladies and Gentlemen,
Keynote Address
Octavi Quintana Trías
European Commission
Director, DG Research, Directorate J "Energy" (Euratom)
As Director of Energy-Euratom at the European Commission's Research Directorate-General, I
would like to welcome you here today to the 7 th Euradwaste conference. As a representative of DG-
Research, I will focus my speech on more scientific and technical issues. My colleague from the
Directorate-General for Energy & Transport has a view more from the political and strategic perspective,
particularly in relation to the continued use of nuclear energy. Clearly these different perspectives
are strongly interlinked and together form a coherent Community policy on the subject of
management of radioactive waste.
The legal basis for all research in the field of applied nuclear science and technology is the Euratom
Treaty, one of the founding Treaties of European integration, and which celebrated its 50 th anniversary
last year. Already 50 years ago, this Treaty foresaw the importance of carrying out research at
the Community level on key issues of interest to Member States. Over the intervening years, the
Euratom Framework Programmes have demonstrated that this collaboration can be extremely effective,
improving our understanding of the science involved, developing a common European view on
the technical issues, maximising the Community added value, and thereby ensuring protection of
the public and the environment in a field with important cross-border implications.
Over the years, the focus of the Euratom Framework Programmes in the field of radioactive waste
management has shifted from fundamental research on basic phenomena in the early days during
the 1980 and 90s, to more applied R&D in the later programmes. Throughout this period, important
Community support has been provided to the research efforts in Member States in all areas of radioactive
waste management, in particular geological disposal.
In 2007 the European Commission launched the 7 th Euratom Framework Programme. This was
unanimously adopted by all European Union Member States. Management of radioactive waste remains
a key thematic priority. This includes both "partitioning and transmutation" as well as continued
research on the ultimate disposal of high-level and long-lived waste in geological rock formations.
In particular, the programme clearly stresses the importance of "implementation-oriented"
R&D on geological disposal.
During the sixth Euratom Framework Programme, 2002 – 2006, some 90 million euros was committed
to research on radioactive waste. Many of the projects are still in progress. In particular, a
small number of large "integrated projects" were launched covering the principal fields of research.
These will be presented in detail tomorrow and on Wednesday, so I will not mention anything now,
except to say that all of them are redefining the state of the art in their respective areas. In the area
of geological disposal, the Commission believes this will enable Member States to push forward
confidently towards eventual implementation of actual disposal systems in those countries where
the socio-political climate is favourable. In partitioning and transmutation, important progress has
been made on understanding the chemical separation and nuclear transformation processes and in-
23

tegrating them as part of advanced fuel cycles. The integrating nature of all these projects is also
helping to fundamentally restructure the way research is being conducted in these fields in Europe.
Looking further into the past, some 100 million euros were devoted to research on radioactive waste
management over the 4 th and 5 th Framework Programmes, split approximately 2 to 1 in favour of
geological disposal. A significant part of this funding went on large-scale demonstration experiments
carried out in underground research laboratories. These demonstrations had to run for many
years. Preparation and decommissioning of the apparatus also took years. The construction of the
underground research laboratories themselves takes even longer and can face the same delays and
opposition as an actual waste disposal facility. Similarly, research on partitioning and transmutation
is also very long term, involving investigations into very complex chemical and nuclear processes.
This underlines the unavoidable long-term nature of research and the need for continuity in the
funding support – something that has been provided by the Community programme. … If we want
to get the science right we must take the time to do it properly!
This advancement in science depends on a close interplay between theory, experiment, demonstration
and reproducibility of results. Throughout this process there is an important principle of peer
review, expert analysis and interpretation. In a complex multi-disciplinary field it is crucial to allow
this scientific consensus time to become established.
In geological disposal for example this process has resulted in widespread agreement within the scientific
community regarding not only the feasibility of long-term confinement of radioactive waste
in deep and stable rock formations, but also the fact that it is the only safe option. Significantly, this
scientific community extends beyond those directly involved in the research effort itself. Many national
geological societies and other academic scientific bodies have also published favourable
opinion papers.
Of all the issues raised in this scientific debate on geological disposal, one stands out as the most
intractable. How can we be assured of safety in the very long-term? Here it is important to appreciate
the limitations of science – research can never provide absolute certainty, nor demonstrate that
risks are zero. However, research does allow us to understand and model the processes involved.
For example, the study of analogues, both natural and man-made, allows an understanding of similar
processes that have occurred in the past.
Significant quantities of high-level radioactive waste already exist in interim surface storage, and it
is inconceivable that these accumulations remain in this situation indefinitely. Sooner or later, society
must implement a permanent long-term management solution that respects high levels of safety
and adequately protects the public and the environment both now and in the future. Though developments
in "partitioning and transmutation" may have significant impacts by reducing long-term
waste toxicity and overall volumes for disposal, it is clear that there will always be a need for geological
disposal of the "ultimate" wastes. The scientific consensus is that geological disposal is the
only option capable of fulfilling the long-term safety requirements, and most national waste management
strategies now recognise this fact. These strategies must also recognise that it is the responsibility
of the present generation to implement a solution, since we have benefited from the
electricity produced by today's nuclear power plants.
However, to implement waste management solutions requires both political will and public acceptance,
certainly in regions surrounding potential sites for geological repositories. In this process,
science must provide a neutral frame of reference in which to present technical issues. The research
conducted must be beyond reproach. It must be thorough, detailed and capable of supporting robust
arguments. Throughout the Euratom programmes, these have been the guiding principles.
24

The Euratom programme also includes more strategic projects looking at the transfer of technology
between larger and smaller waste management agencies, and whether countries could share waste
management facilities rather than each having to construct the full range of installations. This is
particularly important for Member States with small nuclear programmes, or with unfavourable geology.
The Euratom programme also provides support to basic actinide science, which is important
not only for research on geological disposal, but also in "partitioning and transmutation" and the
fuel cycle in general.
Questions of a technical nature still remain to be answered in all areas radioactive waste management,
and an integrated European research effort is the best guarantee that these can be addressed
both effectively and efficiently. This effort should be clearly focussed on the key identified outstanding
issues and solidly based on the wealth of accumulated scientific knowledge, to which must
be added the results of on-going research in the 6 th and 7 th Framework Programmes. In the case of
geological disposal, the best people to drive this process forward are the national waste management
agencies, since they are ultimately responsible for the implementation of disposal options in
the respective Member States. Technical Safety Organisations and the major research institutes are
also key players, and provide additional expertise ensuring that research and the interpretation of
the results are robust and reflect the state of the art.
Last year saw the launch of the "Sustainable Nuclear Energy Technology Platform". This was a pivotal
moment in R&D in the nuclear sector in Europe, bringing together a broad range of stakeholders
in the nuclear research and industrial sectors around a common vision for future research in
the field of nuclear installation safety and advanced nuclear technology, including also the research
in the field of partitioning and transmutation. The Technology Platform is now finalising its draft
"Strategic Research Agenda" for presentation at the General Assembly next month.
The Commission recognises the importance of establishing a similar "Technology Platform" in the
specific field of geological disposal. Not only will this enhance integration of all research players
around a shared vision of geological disposal, it will also enable a much more effective and targeted
use of Euratom funding. The Commission acknowledges the efforts of the Swedish and Finnish
waste management agencies who, with the support of a drafting team made up of experts from a
number of countries, are currently finalising the first draft of the all-important Vision Document.
Indeed, this document will be presented this evening to a meeting of all interested R&D stakeholders,
and represents the first important steps towards the creation of a Technology Platform in
this field.
With such a structure in place, the Euratom Programme can continue to provide invaluable support
to national programmes in their endeavours to implement safe, timely and cost-effective geological
facilities for the disposal of high-level radioactive waste and in the management of radioactive
waste in general.
Finally, I would to thank all those in my service in Brussels, and in DG-Energy and Transport and
the Conference service here in Luxembourg, for their hard work and dedication in organising
Euradwaste'08.
I wish you all a constructive and stimulating conference over the next 3 days.
Thank you.
25

General introduction and objectives
SOCIO-POLITICAL AND STRATEGIC ISSUES
The first day of the 2008 Euradwaste conference is devoted to national and European policy in the
area radioactive waste management (RWM). A series of sessions involving keynote speakers, panellists
and debates with participants will help to gain a better insight into success stories, obstacles
and means to overcome them.
At both national and EU level, energy is high on today’s political agenda. The “20-20-20” targets
agreed by political leaders at the European Spring Council Summit in 2007 will be crucial in the
development of a low-carbon economy in Europe. The role of nuclear energy in fighting climate
change, reducing dependence on foreign energy imports, and maintaining competitiveness has been
clearly recognised, but so has the need to maintain a high level of nuclear safety and to continue to
make progress in the management of high-level radioactive waste and spent fuel. That is why the
European Council requested the creation of a high-level group on nuclear safety and waste management
and a European nuclear energy forum, both of which have since been established by the
European Commission and both of which treat radioactive waste management as a key issue with
important repercussions for the present and future use of nuclear energy.
While the management of low- and intermediate-level short-lived waste is today a mature industrial
practice, the situation regarding the most hazardous waste category, namely high-level long-lived
waste, is characterised by ongoing R&D programmes in several Member States. A few of these are
quite advanced, but at the same time many are stalled and “wait-and-see” approaches were adopted
in other Member States. After some 30 years of research it is now internationally accepted by the
scientific and technical community that geological disposal is the only appropriate and feasible solution
for long-term management of such waste, though the number of active national programmes
implementing this solution can be counted on the fingers of one hand, and even in these cases actual
disposal operations will not begin until 2020 at the earliest. The reasons for such low take-up are
essentially socio-political, related to public and political acceptance, though logistical and economic
reasons such as the relative ease of temporary storage and the cost penalty of being one of the first
also undoubtedly play a role. Nevertheless, successful implementation of such programmes in the
leading countries will boost confidence and uptake in others.
The four sessions of the first day of the conference are dedicated to a closer look at all such issues:
why some countries are well on their way to constructing geological repositories and why others lag
behind; whether economics plays an important role and if there are solutions to reducing financial
burdens; how risks and uncertainties can be communicated in ways which build trust; and what role
the European Union should play.
Each session has been designed to provide a broad insight into the topic addressed and is introduced
by a keynote speech, followed by a panel debate involving representatives from a number of European
Member States at various stages of implementation and invited experts with technical, regulatory,
and policy backgrounds.
27

SESSION I: Current situation of geological disposal in the EU
Chairman: Dr Hans Forsström, IAEA Vienna
Introduction and objectives
The objective of this session is to set the scene regarding progress towards implementation of sustainable
waste management solutions in the EU Member States. While geological disposal is internationally
recognised as the best option for the final management of high-level and long-lived radioactive
waste, no repository has yet been implemented anywhere in the EU. Nevertheless, Member
States such as France, Sweden, and Finland (the latter having a relatively small nuclear programme)
are well advanced in the development of a definitive solution for such waste and have established
appropriate policies, milestones and endpoints. Conversely, other countries lag far behind
and are yet to establish even a clear political strategy for dealing with this issue.
In line with the competences accorded under the treaties, the European Union can have an important
role to play. This has been bolstered by public support for action at EU level coming from the
results of Eurobarometer surveys. It has led the European Commission, supported by Council conclusions,
to request that each EU Member State establish and keep updated a national programme
for the safe management of all radioactive waste produced in the country.
The session will paint a picture of the current European situation, consider the progress towards
geological disposal made so far in the different countries, and identify success stories and potential
problems and obstacles (policy, economics, public acceptance, etc.) in implementing both strategies
and solutions.
29

Radioactive waste management: Where do we stand?
Marie-Claude Dupuis
Chairperson, OECD/NEA Radioactive Waste Management Committee
Chief Executive Officer, French National Radioactive Waste Management Agency (ANDRA,
France)
Summary
Since protecting human beings and the environment is the ultimate goal of radioactive waste
management, stabilisation and conditioning techniques have been developed to process the
waste before disposing of the resulting packages under safe conditions and over compatible
timescales with the radioactive half-lives of the radionuclides contained in the waste. Industrial
solutions are already available for certain waste categories or are being developed for the
most active or long-lived residues with implementation prospects by the end of the next decade.
In the meantime, scientific, as well as technical and political challenges are progressively
resolved thanks to the combined efforts of all stakeholders involved and to the contribution
and drive of the European Commission for Research.
1. Introduction
All radioactive waste generated by the nuclear power industry must be stabilised and disposed of
under safe conditions in order to protect human beings and the environment. Although industrial
solutions have been developed several decades ago for operating waste, which are mostly shortlived,
it is still impossible to dispose of the residues resulting from the burnup of uranium and consisting
of either spent fuel or vitrified waste depending of the cycle options. Until the end of the
1980s, repeated attempts were made in various countries mostly in response to an increasing demand
of information and involvement in the decision-making processes, but all failed. Over the last
decade, a very large number of advances were achieved, not only about governance in the radioactive
waste management sector, but also about newly acquired knowledge. Not only various alternatives
to final disposal, such as transmutation, were investigated, but increasingly sophisticated options
requiring a more and more precise knowledge about the evolution of a radioactive waste repository
were also modelled and simulated. In parallel, the required technological procedures for
handling high-level packages weighing several tonnes have been undertaken, especially in the
framework of the VI th European Project that is now coming to an end.
The purpose of this paper is to provide an overview of the situation at stake, including the major
pending challenges and the proposed responses, notably with the support of the European Commission
(EC).
2. Managing low-level short-lived radioactive waste in Europe
Radioactive waste was produced from the very discovery of radioactivity. The first mining of such
materials inevitably gave rise to residues, although famous researchers were in no position to gauge
the impact of radioactivity upon human health and the quality of the environment. The mining sector,
at first, soon followed by the industry specialising in radiological applications, were both un-
31

aware of its potential effects. Residues were simply left unattended in various backyards that
formed several “uncontrolled” landfills or dumps, some of which have even been rediscovered over
the last few years. Those so-called “orphan sites” date back to the discovery of radioactivity by
Henri Becquerel in 1896 until the Second World War, and include the laboratory of Pierre Curie
and later of his wife, Marie. However, such legacy from the early years of the nuclear age is far
from being the most hazardous.
The development of military and power applications, together with the matching array of research
infrastructures, triggered the production of large quantities of waste that were initially entombed in
trenches or dumped into the sea. Those first disposal operations, which may still be qualified as
“entombment”, were conducted either on NPP sites or in their close neighbourhood, as in the case
of the disposal facility located near the village of Drigg, beside the Sellafield Complex in the
United Kingdom or of the Centre de la Manche Disposal Facility near the La Hague Site, in France.
Similar operations were performed on many NPP sites, especially in Eastern European countries,
such as Ignalina, in Lithuania, or Dukovany, in the Czech Republic. All such pioneer operations
underwent or are still undergoing remediation and consolidation work on their installations in order
to protect human beings and their environment. Sometimes, the waste is retrieved and reconditioned
before being disposed of in dedicated structures, a solution that allows waste packages to
meet the overall safety functions of the disposal facility over compatible timescales with the activity
of the radionuclides contained in the waste. Figure 1 shows a trench from which waste packages
have been retrieved with a view to disposing of them under permanent safety conditions.
Figure 1: Photograph of an open-ground disposal trench, after waste removal
Sea dumping stopped once the London Convention on the Prevention of Marine Pollution by
Dumping of Wastes and Other Matter was adopted in November 1972 and entered into force in
1975 [1] with a view to prohibiting the immersion of land residues, including radioactive waste. It
should be noted, however, that the direct release of materials into the sea through ducts is not considered
as immersion.
32

The intense development in nuclear power generation, especially as a result of the first petroleum
crisis, led to a considerable stream of radioactive waste. In practice, only operating waste resulting
from the routine operation of NPPs, including used filters and resins, residues from the treatment of
chemical effluents containing radionuclides and all common maintenance items (rags, papers, small
tools, etc.) are considered as low-level short-lived radioactive waste. Their generally recognised
threshold corresponds to the half-life of caesium 137, which is slightly over 30 years. Stabilisation
and solidification processes have been developed and implemented on most nuclear-power generating
sites in order to compact solid waste into packages, to embed ion-exchange resins into an epoxy
matrix, to produce cemented waste or to bituminise sludges. Special facilities were built for the
disposal of such waste. Hence, safety is ensured by the design itself from the very beginning of the
waste-disposal operations thanks to the cumulated effects of the respective characteristics of package
stability, their protection guaranteed by structures and facilities, as well as the control of potential
discharges by the properties of the sites, namely with regard to hydrogeology.
Rather than multiplying the number of disposal facilities and the associated safety demonstrations,
many countries have preferred a centralised approach. Power generators who dispose directly of
their own waste will have focused their investigations on the reactor sites themselves. Such is the
case of the underground repositories located in Forsmark, in Sweden, and Olkiluoto, in Finland. In
the most favourable cases, site characteristics were specified. In combination with relatively simple
structures, the overall safety functions of the repository may be fulfilled, as in the case of the
French and Spanish surface disposal facilities, located at the Centre de l’Aube and El Cabril, respectively,
as shown in Figure 2.
Figure 2: (left) The Centre de l’Aube Disposal Facility, France,
and (right) the El Cabril Disposal Facility, Spain
Since both facilities have a national vocation, they also receive short-lived radioactive waste resulting
from medical, research and industrial activities. The volume of such residues remains small and
represents slightly over 3% of all low-level short-lived waste produced in France.
A special note should also be made about new projects under construction, notably a surface disposal
facility on the Mol-Dessel Site, in Belgium, and the shallow disposal facility at Bátaapáti, in
Hungary. In both cases, the sites were selected after holding a consultation process in the vicinity
of nuclear facilities.
33

3. Intermediate-level long-lived and high-level radioactive waste
No disposal facility for intermediate-level long-lived waste and for high-level waste of the nuclear
power cycle is yet available today. The option selected by most States is disposal within a deep
geological formation, the guiding principles of which are included in the collective opinion published
by the Radioactive Waste Management Committee of the OECD Nuclear Energy Agency
(NEA) [2].
All intermediate-level long-lived and high-level waste result directly from the fission of uranium.
Consequently, they are found in spent fuel or in reprocessing residues. Some countries, like
Finland and Sweden, rely on an open-cycle management according to which spent fuel is considered
as waste, whereas countries, like France, the United Kingdom and the Federation of Russia
applying a closed-cycle management reprocess their spent fuel in order to extract recoverable materials,
such as uranium and plutonium. Until now, the United States has been operating an opencycle
system, but investigations have resumed in the framework of the Global Nuclear Energy Partnership
(GNEP) [3] that involves the potential reprocessing of spent fuel. Many countries have
adopted the GNEP approach proposed by the United States. Beside the advantage of retrieving and
recycling recoverable materials, reprocessing reduces significantly the waste volumes intended for
disposal, thus ensuring considerable space saving, especially in countries with an impressive nuclear
power fleet. It should be noted that the fraction of residual waste after reprocessing represents
only 5% in mass, which equals to about 0.05 t per tonne of spent fuel. Consequently, the quantities
of materials intended for disposal are also being reduced, thus making the disposal option less attractive
to future generations in terms of metal resources the materials might hold. However, since
the geological formations are particularly suitable for radionuclide immobilisation due to their low
water circulation rate and their reducing character (anoxic), reprocessing does not offer determining
differences with regard to the safety performance of the repository, as far as the retention capability
is sufficient.
Investigations carried out in several countries since the early 1980s provided a wealth of information
and focused on the study of disposal concepts in different types of geological formations. The
principles that were identified at the time are still used as reference material. From their inception,
they were supported by international organisations, which not only organised exchange meetings
and shared information and feedback, but also shared the costs of complex and costly programmes.
The exchange of information between waste-producing countries has promoted their respective
evolutions. Initially, support was mostly provided on scientific and technical issues. Such was the
case of the joint experiments led by the NEA at the old Stripa iron mine, in Sweden, and the many
projects launched by the EC at the then-called “methodological” laboratories in a clay formation
located at Mol, in Belgium (Figure 3), in the Asse salt formation, in Germany, and in the Grimsel
granite formation, in Switzerland. The first attempts at selecting disposal sites were not very fruitful,
often due to the lack of convincing arguments, and because of poorly-structured political approaches.
Well-known failures were recorded in different European countries, including France
and Great Britain.
34

Figure 3: Schematic diagram of the Mol Underground Laboratory, Belgium
In the mid-1990s, a second generation of underground laboratories was developed, notably in a
granite formation at Äspö, Sweden, and in a clay formation at Mont Terri, Switzerland. During the
same period, Germany prepared the structures for the implementation of a repository for its waste
and spent fuel in a salt dome at Gorleben, but in 2000 the government voted a moratorium that is
still valid in order to suspend its activities.
The lessons learnt from the successive failures in Scandinavian countries and in France have led
gradually to integrate the decision-making process in a structured approach involving political leaders
at both the national and local levels. It is within that context that a specific underground laboratory
was implemented as early as 2000 in a Callovo-Oxfordian argillite formation, at Bure. It is
also within that new context that the United Kingdom’s Committee on Radioactive Waste Management
(CoRWM) was entrusted with the mission to study governance in more detail concerning
radioactive waste management in that country in a rather similar fashion to the reassessment of the
same issue in Canada under the responsibility of the Nuclear Waste Management Organization
(NWMO).
A large number of socio-political studies were initiated, with an increasing support from the EC.
Concurrently, the NEA set in place the Forum on Stakeholder Confidence (FSC), which also reflects
the evolution of the concerns at stake.
Today, three projects are well under way in Europe (i.e., Finland, Sweden and France). Scientific
and technical advances have supported political decisions on the way to the successful implementation
of deep geological repositories.
Feasibility has already been demonstrated in various geological media, such as salt, granite or clay.
Comparable performance and safety levels have been achieved thanks to facilities that are adapted
to the characteristics of their host formation and to the specificities of each site. In every case, the
combination of the characteristics and performance levels of waste packages, structures and sites
has allowed to fulfil all required safety functions over the short and long terms.
4. Major scientific challenges
The major challenge consists in providing safety demonstrations over quite uncommon space and
time scales. Such task requires developing not only a suitable methodology (currently called “the
safety case”), but also the overall knowledge necessary to support them.
35

According to a clearly-defined and demonstrable physical approach to quality, the reliability of the
safety assessment of a long-term repository relies on our capability to describe the different phenomena
occurring over time. In certain cases, those phenomena may be coupled, as in the case of
expectable thermomechanical effects after the emplacement of highly-exothermal packages. As for
processes likely to occur over the very long term (i.e., mostly chemical processes and radionuclide
migration), a sound physical control reduces the propagation of uncertainties relating to an insufficient
description or model. The required parameters to describe that process must also be known
with an appropriate level of precision, and the associated uncertainties must be submitted to essential
sensitivity tests in order to ensure that the repository meets its operational and performance targets.
In order to provide a sound description of the repository’s determining phenomena, I would
like to point out the methodology developed and implemented by Andra, which consists of a phenomenological
analysis of repository situations and a description of the phenomena tracking down
the evolution of the repository at each space and time field. From the previous example, we may
retain the thermal effects relating to the presence of exothermal packages that will need to be taken
into account throughout the heat-release period of the packages. In the case of vitrified waste, heat
releases decrease after a few centuries and cease to determine the behaviour of disposal structures
over time. Inversely, any phenomenon involving radionuclide migration may only be taken into
account once waste containers are corroded. That type of extremely rigorous and systematic analysis
conducted over the entire repository avoids excessive couplings between the thermal, mechanical,
hydraulic and chemical characteristics of the repository, which are not always necessarily described
properly. The reader may refer to Andra’s Dossier 2005 for complementary information on
the selected approach for the phenomenological analysis of repository situations and a sound perspective
view of the phenomena in relation to the space and time dimensions to be taken into account
[4].
Keeping in line with the demonstration capability and, consequently, with the confidence that our
researchers may have in their models, the goal will be to seek the simplest designs possible. The
purpose is not only to describe and model easy geometries, but also to favour certain operating conditions.
With reference once again to the Callovo-Oxfordian argillites mentioned in the Dossier
2005, the temperature is limited to 90°C on the walls of disposal structures, with a view to staying
within a describable operating field without having to go through more complex diphasic behaviours.
By opting for simplicity, not only does modelling become easier, but the demonstration as
well.
However, some inevitable fields of physics remain pending and insufficiently described, as in the
case of gas production and behaviour throughout the lifetime of the repository. All organisations in
charge of 0-waste management have agreed to raise the issue among the priorities for the
7 th Framework Project for Research and Development (FPRD), and the FORGE Project should provide
responses to fulfil the ambitions of the safety demonstration for waste repositories.
Physical modelling is therefore a determining factor. However, we are confronted with space and
time scales that are way above the field of experience. All investigations conducted in many surface
laboratories and all experiments carried out in underground laboratories will only cover several
decades at best, whereas safety demonstrations extend over timescales that are much longer by
three to five orders of magnitude. It is therefore necessary to deepen the analysis over several tens
or hundreds of millennia. Once again, the quality of the analysis will rely on the quality of the
simulation means and of the digital approaches. Qualified methods must be available in order to
change both space and time scales efficiently, to analyse the impact of parameter uncertainties and
to study the nature of the various effects throughout the lifetime of the repository. Generally speaking,
it will be possible to analyse the impact of a leaking package or of a borehole intrusion within
the repository. Today, thanks to the growing efforts in modelling and in digital simulations for
more than 10 years, we may considerer the “black-box era” to be obsolete and welcome rather increasingly
clear and convincing descriptions. The numerous projects supported by the EC will have
36

contributed to enhance knowledge about processes and phenomena, such as the mechanisms regulating
the chemical composition of deep waters (ARCHIMEDE Project), to reduce uncertainties on
data (CHEMVAL Project on thermodynamic data and the NEA Thermochemical Database Project),
to carry out an intercomparison of modelling approaches (DECOVALEX for hydromechanical couplings).
More recently, the Projects on the Near-Field Phenomena (NF-PRO), on Fundamental
Processes of Radionuclide Migration (FUNMIG) and on Performance Assessment Methodologies
in Application to Guide the Development of the Safety Case (PAMINA) made it possible in the
framework of the 6 th FPRD to reassess our current knowledge, to generate stabilised data and to
propose integration modalities for those data in the safety assessments.
5. Major technological challenges
From the technological standpoint, the major challenges consist in being able to take into account
simultaneously nuclear-safety requirements during the operating stage and the constraints of the
mining environment. Agencies have already achieved decisive advances in the field with the emergence
of the first industrial complexes. The most remarkable achievement so far is undoubtedly the
encapsulation plant for which the Swedish Nuclear Fuel and Waste Management Company (SKB)
has submitted an application in 2008 with a view to using that copper friction-welding process.
During the 6 th FPRD, significant efforts were made throughout Europe in the framework of the Project
on Engineering Studies and Demonstration of Repository Designs (ESDRED) led by Andra.
Several handling demonstrators for waste packages were built, including an emplacement robot for
vitrified waste packages in horizontal cavities (Figure 4), as well as the moving of charges up to
80 t for the horizontal transfer of spent-fuel packages. Those demonstrators are on display at the
Technological Centre, in Saudron, which is located very close to the Meuse/Haute-Marne Underground
Research Laboratory, on the outskirts of Bure.
Figure 4: General layout for the horizontal emplacement of a waste canister
There are still many development and implementation needs to fulfil, especially in order to carry
out operations with automated and often-sophisticated devices in order to minimise human interventions
and, hence, ensure maximum safety to the operating staff. With regard to design and construction,
several options remain open for the different projects, notably for the choice of access to
underground installations, either down a ramp or vertical shafts. The overall arguments are analysed
in relation to their specific environments and limitations in order to make timely choices.
Requirements are also becoming clearer concerning the phase following the emplacement of disposal
packages in underground cavities. The first challenge is the reversibility of the repository and
the retrievability of the waste packages. That issue is particularly important in France and is being
addressed by various activities at the international scale. Following the preliminary reflections
made in the late 1990s and the European project on that topic, a new step was launched by an NEA
37

working group [5]. Its main objective is to identify reversibility limits and mostly the modalities to
assess them. In parallel and consistently with the group’s activities, Andra has proposed the principle
of a reversibility scale allowing, first of all, the different teams to exchange opinions on the basis
of a common terminology [6].
Reversibility also opens up the issue about repository monitoring and environmental follow-up. As
a matter of fact, the ultimate decision to retrieve waste packages shall rely on objective arguments
and data that are only obtainable if a suitable monitoring and follow-up programme is implemented
around disposal facilities. The purpose of the MoDeRn Project, which has been launched in the
framework of the 7 th FPRD, is to provide the first response elements and to identify which requirements
need to be met regarding technological developments in order to find corresponding responses.
Lastly, it is impossible to speak about waste disposal over the long term without mentioning conservation
of information and of the repository memory during the ultimate reversibility phase and
over the very long term. The issue is already well assimilated over a few centuries on disposal sites
for short-lived waste, but will need to find appropriate responses over the longer term.
6. Societal challenges
Decisions on controversial issues such as those relating to nuclear energy are prepared and taken
within a complex social and political context that may involve contradictory interests at times or
lead to political compromises, but always with a concern to protect human health and the environment.
The opposition and failures encountered in various countries since the 1980s have led to consider
more open and transparent governance modes in order to associate the different stakeholders in the
reflections, the procedures and the preparation of relevant reports. In Europe, Scandinavian countries,
such as Finland and Sweden, were the first to be faced with determined opponents who were
either purely and simply opposed to the disposal of high-level radioactive waste, thus hoping to
block the generation of nuclear power, or advocating decision-making processes combining a better
participation of stakeholders and enlightening more objectively the nature of the risks in the vicinity
of such facilities. Investigations finally focused on nuclear-power generating sites with the support
of local authorities and populations. On the other hand, the proximity of a nuclear facility such as
Sellafield, was not sufficient to convince populations and authorities to pursue investigations for a
deep geological repository in the United Kingdom. Once again, the claim for more transparent decision-making
processes involving a better participation of all stakeholders was emphasised, as the
CoWRM’s conclusions and recommendations clearly illustrate [7].
The approach that was finally adopted in Belgium for the disposal of low-level and intermediatelevel
short-lived waste [8] will have shown ultimately that the participation of local communities
and of the different stakeholders, even in the selection of certain design options, had led to a generally
accepted project. The integration of the radioactive waste disposal project into a regional development
plan has undoubtedly been a determining factor.
Germany experienced an extremely fast development of its facilities, but was faced with a demand
as explicit as in other countries. A moratorium is still being imposed on the Gorleben Site for political
reasons reflecting the same request as in other countries for an open and transparent decisionmaking
process after having reviewed all other disposal alternatives in the Gorleben salt dome. In
the case of non-exothermal waste, the Konrad was finally licensed following a procedure that extended
over more than 20 years.
In France, following the failure of site implementations for investigation purposes, the Law of
30 December 1991 prescribed a new deadline for Parliament to discuss the issue after 15 years of
investigations. In 2005, the results were made public and the government asked the National
Commission on Public Debate to organise a public inquiry on radioactive waste management. For
38

the first time in French history, a public debate was set up in accordance with a national policy and
not on a specific project. The debate took place from September 2005 to January 2006 and involved
close to 600,000 documents being distributed to 3,000 participants who attended 13 meetings
throughout the country. It should be pointed out, however, that the debate remained a discussion
among pro-nuclear supporters and anti-nuclear protesters most of the time, although it also
provided an opportunity to refine and clarify expectations and challenges.
Following the public inquiry and considering the research results and their assessment, Parliament
adopted on 28 June 2006 the Planning Act Concerning the Sustainable Development of Radioactive
Materials and Waste [9]. The new legislation encompasses all radioactive waste, irrespective of
their activity level, and prescribes, among other requirements, specific procedures and deadlines
with a view to commissioning by 2025 a new deep geological repository for high-level and intermediate-level
long-lived waste. With reference to the geological disposal, the Planning Act renews
the assessment system of Andra’s studies and investigations by the National Review Board (Commission
nationale d’évaluation – CNE) and by the Parliamentary Office for the Assessment of Scientific
and Technological Options (Office parlementaire d’évaluation des choix scientifiques et
technologiques – OPECST) and the assessment of its future licence application by the French Nuclear
Safety Authority (Autorité de sûreté nucléaire – ASN). In addition, a new act will set forth
the prerogatives of the Local Information and Follow-up Committee (Commission locale
d’information et de suivi – CLIS), responsible for public information and consultation, and prescribe
a public debate and inquiry before any licence may be issued.
Situations are always very complex, and any transfer of experience remains quite limited not only
through the different Community projects, but also over time. Political and cultural conditions at
the local, regional and national levels are determining in all cases. However, the exchange of approaches
and practices develops a favourable frame of mind for listening and sharing for which the
NEA’s FSC plays a very useful role for all partners involved.
7. Conclusions
With regard to the different challenges at stake, considerable advances have been achieved thanks
to the respective efforts, support and advice of international organisations. The EC, through its successive
FPRDs, has been contributing very largely for more than 20 years to the different scientific
disciplines that have been mobilised about the issue of geological disposal. The experiments conducted
in the first generation of so-called “generic” underground laboratories were useful to prepare
investigation and measurement methods that were subsequently implemented in underground laboratories
on site. The NEA’s efforts also contributed to further advances, first about ethical issues,
then on the structuring of safety cases. Today, the NEA’s efforts are most valuable in the field of
governance and decision-making relating to radioactive waste management. Lastly, special emphasis
should be given to the key role played by the International Atomic Energy Agency (IAEA), especially
in the development of reference documents.
8. Acknowledgements
May all those who have contributed for many years to the significant advances of radioactive waste
disposal projects be congratulated and thanked here for their continuing efforts in research and development
and for their determination to convince those who have to make relevant decisions.
39

References
[1] International Maritime Organization, Convention on the Prevention of Marine Pollution by
Dumping of Wastes and Other Matter, [London Convention 1972], 1972 and 1996 Protocol
Thereto.
[2] Technical Appraisal of the Current Situation in the Field of Radioactive Waste Management.
A Collective Opinion by the Radioactive Waste Management Committee, OECD/NEA, Paris,
1985.
[3] Joint Statement on the Global Nuclear Energy Partnership and Nuclear Energy Cooperation,
Washington, D.C., May 21, 2007, www.gnep.energy.gov/pdfs/GNEP_Joint_Statement.pdf.
[4] F. Plas, P. Landais, The Phenomenological Analysis of Repository Situations (PARS) – Application
Within the Dossier 2005 Argile (Meuse/Haute-Marne Site), in Safety Cases for
Deep Geological Disposal of Radioactive Waste: Where Do We Stand? – Symposium Proceedings
Paris, France, 23-25 January 2007
[5] Reversibility and Retrievability in Geologic Disposal of Radioactive Waste,
NEA/RWM(2007)7, 08 March 2007
[6] Proposal for an International Reversibility-Retrievability Scale, Jean-Noël Dumont, Marie-
Claude Dupuis, Jean-Michel Hoorelbeke, Thibaud Labalette, Gérald Ouzounian, 2008 International
High-Level Radioactive Waste Management Conference, Sept. 7-11, 2008, Las Vegas,
Nevada
[7] Committee on Radioactive Waste Management, Managing our Radioactive Waste Safely,
CoRWM’s recommendations to Government, CoRWM Doc. 700, July 2006
[8] La Mise en dépôt final, sur le territoire belge, des déchets radioactifs de faible et moyenne
activité et de courte durée de vie, Rapport de clôture de l’ONDRAF relatif à la période 1985-
2006, invitant le Gouvernement fédéral à décider de la suite à donner au programme de dépôt,
NIRON 2006-02F, mai 2006
[9] Consolidated version of the Planning Act No. 2006-739 of 28 June 2006 Concerning the Sustainable
Management of Radioactive Materials and Waste
(www.andra.fr/publication/produit/loi-VA-12122006.pdf) .
40

PANEL DISCUSSION
Summary of the Panel Discussion Concluding
Session I: Current Situation of Geological Disposal in the EU
Panel members:
Hans Forsström (Chair), IAEA Vienna
Gheorghe Negut (Rapporteur), ANDRAD, Romania
Michael Äbersold, SFOE Switzerland
Alvaro Rodríguez Beceiro, ENRESA, Spain
Vit�zslav Duda, RAWRA, Czech Republic
Esko Ruokola, STUK, Finland
This report summarizes the outcome of session I on the current situation of geological disposal in
the EU. The session included a keynote speech by Mrs Marie-Claude Dupuis (CEO, ANDRA,
France) as chairwoman of the OECD/NEA Radioactive Waste Management Committee (RWMC)
on the global situation of geological disposal projects in the EU. This was followed by a panel discussion
based on a set of questions.
The questions put forward for discussion were:
� What is the RWM situation in the EU Member States?
� What are the success stories?
� What are the obstacles for development of a solution? Policy? Public acceptance? Economics?
� Is it possible to overcome wait-and-see approaches?
� Long term storage: does it create or solve problems?
� Are multinational solutions envisaged and/or desirable in the EU?
� Who or what are the potential drivers?
� Do we need binding EU legislation or can we count on voluntary measures?
Apart from the presentation of the history and current status of geological disposal in the EU (see
keynote paper), M.C. Dupuis insisted on the fact that intermediate storage cannot be considered as
long-term solutions for high level radioactive waste and spent fuel. As well transmutation process
will in any case produce high level waste. In conclusion, deep geological disposal remains the preferred
option for management of high level radioactive waste and spent fuel.
Implementation of geological disposal solutions requires parliamentary commitment on regulation
and principles, as well as it necessitates social, technical and scientific cooperation.
The panellists then presented their statements before the word was given to the floor.
41

Michael Äbersold
Switzerland is in the process of developing geological disposal and among others one of the important
concerns is the idea of trans-border transparency. According to the Swiss legislation, the project,
for which a political strategy is in place for the next 20 years, needs local and national acceptance
by the means of referendums. In the same way the development of potential new NPPs requires
to go through a referendum process.
As part of the Swiss programme, the costs of disposal are anticipated by the establishment of budget
procedures and social policies will be implemented for those communities which will be involved
in the disposal project.
Alvaro Rodríguez Beceiro
The attitude of Spain with regards to geological disposal is not positive. The siting process has
been interrupted in the mid 90's and so far the situation is focused on an interim storage during 50
to 60 years and geological disposal is not treated as a priority.
According to the actors involved in the Spanish geological disposal project, the programme suffers
from a lack of political commitment as well as a weak public acceptance. Beside, the problem is
that given Spain’s public mentalities, history and other specificities, public acceptance is still a matter
of future wish.
Moreover, generally speaking, the nuclear industry is rather in favour of extended storage solutions
than progressing in the direction of geological disposal. Spain rather considers that a national project
for deep geological repository is at the moment unrealistic.
Vit�zslav Duda
In the Czech Republic, nuclear energy receives a favourable public acceptance and apart from the
fact that a national law prohibits import of radioactive waste, some politicians would be in favour of
accepting and managing foreign radioactive waste.
There is no state policy on reprocessing and the decision is left to CEZ, which does not perceive it
as being economically interesting. However, the question remains open.
As part of the geological disposal programme, site selection is scheduled for 2015, and repository
construction should start after 2050.
Due to the fact that the Czech Republic is a small country with a high population density, the transfer
of information on the nuclear programme is not optimized.
Under the Atomic Energy Act 2002, CEZ the nuclear plants operator is required to put aside funds
for waste disposal, lodging these with the Czech National Bank.
Gheorghe Negut
In Romania, the first unit of Cernavoda NPP has been operated since 1996, the second unit started
operation in 2007. Plans exist for construction of Unit 3 and 4 in Cervanoda and a new NPP.
Polls show increasing acceptance of nuclear energy and disposal as the solution for radioactive
waste management.
The main and major challenge in Romania with regard to waste management is the construction of
a near surface facility for LLW in Saligny near Cernavoda NPP to be commissioned in 2014.
According to the Government, the national strategy for geological disposal is to develop a repository
to be operated in 2055, a faraway horizon.
42

With the support of IAEA preliminary siting investigations have been performed in green schists in
Dobrogea, a sparsely populated area near Cernavoda.
Esko Ruokola
Finland was presented as a success story with regard to political commitment on geological disposal,
with a decision taken in 1983.
In the mid 90's with the establishment of a legislation prohibiting import and/or export of radioactive
waste, export of radioactive waste to Russia was stopped, reinforcing the option to go for deep
geological disposal.
As confirmed by the series of Eurobarometers on nuclear safety and on radioactive waste (the last
one issued in June 2008), a large majority of Finns is favourable, both at national and local levels,
to nuclear energy and consider that geological disposal is a safe solution to implement for the management
of their spent fuel. Moreover most Finns, who have been deeply involved in the nuclear
programme early in its development, are confident in their policy makers, regulators and operators.
43

SESSION II: Economical factors governing geological disposal programmes
Chairman: Mr Thibaud Labalette, ANDRA, France
Introduction and objectives
In the EU there is a large range in the size of national nuclear power programmes and associated
RWM programmes, resulting in major differences regarding the availability of financial and human
resources as well as influencing political decision-making. The development and implementation of
the geological disposal option is often considered as expensive or even financially insurmountable,
in particular by Member States with very small nuclear programmes, in which the option is made
conditional on increased national nuclear capacity and hence higher revenues, or the development
of shared facilities.
Consequently, the objective of this session is to discuss the economic factors governing geological
disposal programmes, such as the cost of disposal compared to the overall cost of a nuclear programme,
the distribution of costs over the different phases of development of a disposal solution as
well as the principles, methods and timing for collecting waste management funds, including foreseen
contingencies. What mistakes have been made in the past that have led to the perception that
geological disposal is unaffordable, and how can these impasses be avoided in the case of future
initiatives?
The session will concentrate on discussing principles, parameters and relations rather than the detailed
cost of disposal. It could also provide some input to the following session on cooperation,
knowledge transfer and/or the potential sharing of facilities, especially for countries with small nuclear
programmes.
45

Assessment of Financial Provisions for Nuclear Waste Management
Long-Term Perspective from Finnish Viewpoint
Summary
Eero Patrakka, Jussi Palmu, Kimmo Lehto
Posiva Oy, Finland
The financial provisions for radioactive waste management must be secured in advance and
collected in phase with electricity production due to the long lead times in waste management.
Financing of waste management necessitates reliable assessment of waste management costs.
Sufficiently accurate information of waste management costs is needed also for the evaluation
of the economics of nuclear power, i.e. when new nuclear power plant projects are considered.
The scopes, implementations and progresses of waste management programmes in different
countries are very different. Continuous research and development produces modified
solutions and, consequently, changes cost estimates. Experiences from mechanisms related to
the functioning of global markets and changes in industrial structures show that caution must
be followed when making long-term predictions.
According to the Finnish nuclear legislation the waste producers are responsible for all activities
related to waste management including financial provisions. The provisions are collected
in an external fund managed by the government. As to spent fuel management, direct geological
disposal is stipulated in the nuclear energy act. The provisions collected in the nuclear
waste management fund cover in principle liabilities for the waste generated so far in the operating
NPP units. The cost estimate for spent fuel management of Olkiluoto and Loviisa
NPP's is presented and the impact of various technical and economical parameters discussed.
1. Introduction
Compared to other means of power generation, nuclear power has the disadvantage of producing
very long-lived radioactive waste. Although the waste volumes are relatively minor, radioactive
waste management requires special methods and facilities that will be in operation for decades,
even centuries. This fact is also reflected in the public attitudes towards nuclear power. In the special
Eurobarometer study on radioactive waste in 2008 44% of EU citizens were in favour of nuclear
energy, however, if those against nuclear felt that the issue of radioactive waste were solved,
four out of ten would change their mind.
The solution of radioactive waste management is imperative for expanded use of nuclear energy in
power generation. Concerning low and short-lived intermediate level waste, industrially developed
repositories are operational in most EU member states. Although there are no repositories for high
level waste in operation, there exists wide technical acceptance that deep geological disposal of
high level waste is the best available solution from a safety point of view.
47

As in all industrial activities, "Polluter pays" principle is valid for nuclear power, too. Considering
the long lead times in radioactive waste management, the financial provisions for waste management
must be secured in advance and collected in phase with electricity production. The present
generations that take the advantage of nuclear electricity shall have the burden to bear the waste
management costs and not push them forward to future generations.
The financing of waste management necessitates reliable assessment of waste management costs.
The cost calculations are, however, not too straightforward especially when related to high level
waste management. Prediction of future costs over a time span of one hundred years is very challenging
while bearing in mind that the technology is continuously developing. In all activities related
to nuclear power "Safety first" principle is mandatory. The application of this in financial provisions
implies that conservatism is included in cost calculations.
It should be noted that sufficiently accurate information of waste management costs is needed also
for the evaluation of the economics of nuclear power, i.e. when new nuclear power plant projects
are considered. Such calculations can utilise the same data base although they might be based on
different premises and boundary conditions.
The purpose of this presentation is to discuss the costs of spent fuel management, first using the
Finnish situation as an example and then trying to generalize our experiences and conclusions into
an international framework as far as possible. Some long-term perspective is available on the basis
of the past evolution of our cost estimates but an extrapolation into the future is doubtful at its best.
2. Basis for calculation of waste management costs in Finland
2.1 Legal framework for assessing financial liabilities
Preparations for nuclear waste management were commenced in Finland already in the 1970s when
the power plants were still under construction. In 1983, the Government confirmed a target schedule
for spent fuel management, in which the construction of the final disposal facility was scheduled
for the 2010s and the start of final disposal for 2020.
The Nuclear Energy Act regulates the implementation of nuclear waste management in Finland.
According to the Act, the operators of the NPPs bear the responsibility for nuclear waste until final
disposal has taken place. All nuclear waste generated in Finland, including spent fuel, must be handled,
stored and permanently disposed of in Finland.
The financial side of final disposal is also covered by legislation. The assets required for the management
of wastes produced in nuclear power plants are collected in advance from the waste producers
and transferred to the State Nuclear Waste Management Fund. The two nuclear power companies,
Teollisuuden Voima Oy (TVO) and Fortum Power and Heat Oy (Fortum), are responsible
for the safe management of the their waste and for all associated expenses. TVO and Fortum established
a joint company, Posiva Oy, in 1995 to implement the disposal programme for spent fuel
from their NPPs, whilst other nuclear wastes are handled and disposed of by the power companies
themselves.
The State Nuclear Waste Management Fund is a reserve for future costs. The Fund was introduced
in the Nuclear Energy Act of 1987 and is operative since 1988. It is not included in the budget of
the state, but is an external fund controlled by the Ministry of Employment and the Economy
(TEM). The Finnish Fund fulfils the two globally accepted principles for such funds: the funds are
48

collected in the cost of the nuclear electricity production giving rise to the wastes and the collected
funds should be available when the related waste management operations are carried out. The nuclear
operators are entitled to borrow back, at the market interest, 75% of the capital against full securities.
The State has the right to borrow the remaining 25% at the same interest rate.
It is worth emphasizing that the Fund does not pay for the waste management activities but only
keeps in safe the money corresponding to the costs of the remaining measures. Theoretically, all the
funds have been returned to the operators when they have carried out all the necessary waste management
operations. As the costs of dismantling and decommissioning immediately turn to remaining
waste management costs when a nuclear facility is made critical for the first time, there is a
transition period of 25 years to collect these costs that constitute a considerable portion of the total
waste management costs.
A second and more universal financial provision system was introduced with the adoption of the
International Financial Reporting Standard (IFRS) as basis for the financial reporting of Finnish
public companies in 2005. IFRS requires that the waste management liabilities are accounted for in
the balance sheet. Funds for future waste management activities have to be collected by the end of
the operational lifetime of NPPs.
2.2 Practical prerequisites and principles of assessment
The implementation of spent fuel management for Olkiluoto and Loviisa NPP units is based the
Decision-in-Principle (DiP) issued by the Finnish Government in December 2000 and ratified by
the Parliament in May 2001. According to this decision, Posiva will locate its repository in crystalline
bedrock at Olkiluoto and the disposal would be based on the so called KBS-3 concept (Fig. 1).
Based on the target schedule, final disposal of spent fuel will commence in 2020.
Host rock
Backfill
KBS-3V
Bentonite
Canister
49
KBS-3H
Host rock
Bentonite
Canister
Figure 1: Schematic illustration of KBS-3 disposal concept with two alternatives
The cost estimates for financial provisions and liabilities are based on the latest detailed waste management
plans and schedules. These are revised at three years intervals; latest revision was made in
2006.

The financial provisions according to the Nuclear Energy Act have to cover all the future costs of
the management of the existing wastes. This means that the actual waste management plans and
schedules have to be adjusted to correspond to the smaller amount of waste accumulated so far. In
the estimates all costs must be presented at present day cost level and no discounting is allowed.
The fund calculations are revised annually.
The cost estimates for the liability calculations according to IFRS requirements are based directly
on the actual waste management plans and schedules during the total expected lifetime of the operating
units. The costs are divided into two categories: costs independent on the quantity of spent
fuel and spent fuel costs. The former costs are treated as investment costs and the latter ones as fuel
costs in the calculations. The costs are presented as discounted cash flows. IFRS estimates are revised
quarterly, if necessary.
In the beginning TVO and Fortum carried out all the work related to the calculations of financial
provisions. Since the foundation of Posiva, this activity is carried out by Posiva for its owners. The
calculation of IFRS cash flows is made by Posiva but their further modification for the reporting
starting with discounting is performed by the utilities.
The estimated liability of nuclear waste management activities and related cumulative cash flow for
the expected lifetime of the operating Olkiluoto and Loviisa NPPs and Olkiluoto 3 unit is depicted
in Figure 2. With the given assumptions, a payback from the Fund will begin about 2030 and the
Fund has been emptied at the time when all waste management obligations have been carried out.
ro
u
E
n
ilio
M
6000
5000
4000
3000
2000
1000
0
50
Liability
Cash flow
Figure 2: Funding liability and cumulative cash flow of nuclear waste management for Olkiluoto
and Loviisa NPPs (future costs in December 2007 level)
3. Cost estimate for final disposal of spent fuel in Finland
3.1 Basis and scope of cost estimate
Cost estimates for spent fuel management are made for two BWR units at Olkiluoto (Olkiluoto
1&2, 2 x 860 MWe, operated by TVO), two PWR units at Loviisa (Loviisa 1&2, 2 x 488 MWe, op-
Total

erated by Fortum) and one PWR unit (Olkiluoto 3, 1600 MWe) under construction at Olkiluoto by
TVO.
Posiva published an updated preliminary design of the spent fuel disposal facility at Olkiluoto site
at the end of 2006. The reference for the design work and technical development is a KBS-3 type
disposal concept. The design available uses the best knowledge on the geological features of
Olkiluoto, current understanding of the technical implementation of the disposal concept and accounts
for the infrastructure of Olkiluoto island. Based on this preliminary design phase 2, Posiva
updated the cost estimate for spent fuel disposal in 2007 [1].
The disposal facility consists of an above ground facility and an underground repository. The most
important building of the above ground facility is the encapsulation plant. The spent fuel that has
been transported from Loviisa and Olkiluoto sites after it has been stored there in interim stores for
several decades is received in the encapsulation plant and the fuel is packed into disposal canisters.
The underground repository consists of disposal tunnels and deposition holes deep in the bedrock,
in which the encapsulated fuel is emplaced, and of required underground auxiliary facilities and access
connections.
Currently Posiva is aiming at submitting the application for the construction license for the disposal
facility in 2012. The operation of the facility is planned to be commissioned in 2020 after the operation
license has been granted. The disposal of spent fuel from the operating units (Olkiluoto 1&2
and Loviisa 1&2) is scheduled for 2020 to 2070 while the disposal of Olkiluoto 3 fuel is estimated
to take place between 2070 and 2110.
3.2 Summary of results
Based on the information provided by TVO and Fortum the predicted accumulated quantities of
spent fuel to be disposed of are given in Table 1. As shown in the table, Olkiluoto units are supposed
to operate for 60 years and Loviisa units 50 years. The total accumulated quantity of spent
fuel over this period is calculated to about 5500 tU.
Table 1: Predicted accumulated fuel quantities at Olkiluoto (OL) and Loviisa (LO) units
OL1 OL2 OL3 LO1 LO2
Design operating life (years) 60 60 60 50 50
Number of fuel assemblies 7,224 7,288 3,728 3,993 4,380
Average burnup of all assemblies 38.0 38.6 45.5 37.0 38.1
(MWd/kgU)
In tons of uranium (tU) 1,270 1,263 1,980 492 526
Number of canisters 1,210 932 698
The corresponding total cost estimate is about 3 Billion Euro (December 2006) and broken down in
Table 2. No discounting has been applied to the future costs. Neither interests nor value added taxes
are included. A general contingency of 20% has been added to all costs. The calculated value of 3
Billion Euro for the spent fuel disposal of a fleet of 4300 MWe of nuclear power may be compared
to the published investment cost of 3 Billion Euro for Olkiluoto 3 unit.
The cost estimate covers the construction and operating costs of the rock characterisation facility,
ONKALO, above ground and underground investment costs including the relevant infrastructure,
above ground and underground operating costs, spent fuel transports from the interim stores to the
51

disposal facility and decommissioning costs of the repository. The following costs are not included:
R&D costs, nuclear regulatory costs, land lease/acquisition and real estate taxes. By definition, the
costs related to interim storage of spent fuel are excluded.
Table 2: Total estimated costs (Million Euro) of spent fuel disposal from Olkiluoto and Loviisa
units (December 2006)
Construction 630
above ground facilities 150
repository (incl. ONKALO) 480
Operation 2,140
encapsulation plant 1,120
canisters 530
repository 470
transports 20
Decommissioning and sealing 240
dismantling & waste management 10
repository closure and sealing 230
Total 3,010
3.3 Impact of parameters
The results of cost estimate calculations for spent fuel are dependent on input parameters, some of
which are based on the technical boundary conditions and some on economical and financial assumptions.
The impact of parameters must be considered separately for the encapsulation process
and underground disposal process.
The crucial technical boundary conditions are obvious: quantity and burnup of spent fuel. The impact
of quantity is straightforward as far as the capacity of encapsulation plant and underground repository
is sufficient. There might be a case where a stepwise increase in a process becomes necessary,
whereupon a corresponding stepwise rise in the costs will occur.
Increasing burnup of spent fuel will increase the residual heat, which means either longer cooling
times or larger disposal rock volumes. In practice, an optimisation will be made between these two
factors. Prolongation of cooling times is feasible as long as the NPPs are operating but the situation
changes once all units have been shut down.
Scheduling of disposal activities is not only an economic issue, as it is dependent on the progress of
the whole waste management programme, including policy decisions. A right timing of disposal
activities facilitates an efficient use of encapsulation capacity. Many other factors may be considered,
such as optimisation of resources in the whole disposal chain.
There are many other technical parameters that affect the disposal costs: detailed design of the disposal
concept, canister manufacturing process, rock excavation methods, tunnel backfilling process.
In our case, there are two alternatives for the KBS-3 concept – vertical or horizontal location of the
disposal holes (Fig. 1) – which possibly differ in costs. We consider at least three manufacturing
processes for the copper canister: pierce and draw, extrusion, forging. Their unit costs are different
but the final selection will be based on a combination of criteria. Our reference excavation method
is drill and blast, but if for some reason another method must be used cost increase is to be ex-
52

pected. As to tunnel backfilling, there are many factors to be considered: backfill material, fabrication
of backfill blocks, installation of backfill.
We have now proceeded to an outline planning stage of the technical implementation of our repository.
It is absolutely necessary to compile a data base of the relevant cost drivers and estimate
the implementation costs of different alternatives simultaneously with design process. Also it is important
to collect and analyse the realised costs. In our case, we have the unique possibility to receive
cost information during the ongoing construction of ONKALO facility.
Economical and financial parameters are typically outside our control. These include the price of
raw materials, inflation, interest rate and return on investment.
The essential raw material in our concept is copper. Latest observations demonstrate that its price is
fluctuating rapidly and substantially. In these circumstances it is very difficult to predict the price of
copper even at the moment when the disposal starts, not to mention the whole operating period of
100 years.
The impact of inflation, interest rate and return on investment varies depending on the funding
scheme. In our case, the return on investment of the State Fund is bound to market interest rate. In
practice this means that profit of the Fund is dependent on the real interest rate.
The impact of parameter variations in the range of the total costs of 3,000 Million Euro is illustrated
in Table 3 for our concept and prerequisites. Relatively minor changes in parameters may cause a
change of the order of 100 Million Euro in the total costs.
Table 3: Impact of variation of different parameters in the total estimated costs (December 2007)
Parameter Change in parameter Impact
Amount of spent fuel 1 tU ~ 0.5 million Euro
Burnup of spent fuel 5 MWd/KgU 7–8 years cooling time
Operating time of disposal 1 year ~ 10 million Euro
facility
Price of copper 1 euro/kg ~ 35 million Euro
Real interest rate 1 %-point ~ 20 million Euro
3.4 Evolution of cost estimates
Spent fuel disposal costs have been estimated since the very beginning of the Finnish spent fuel
management programme. First calculations were made in the early 1980s, and they have been updated
regularly. There have been notable changes in the cost estimates based on the quantity of
spent fuel, timing of disposal and development in disposal technology.
The first estimate was based on the disposal of spent fuel from two Olkiluoto units. The next ones
considered two Olkiluoto and two Loviisa units with relatively short design operating life times (40
years). The newest calculations include Olkiluoto 3 and extended life times (50-60 years). The new
unit has an important effect in the operating time of the disposal facility, as the disposal of its spent
fuel can start more that 40 years later than the disposal of fuel from the old units.
53

The development of disposal technology has had and will have a twofold impact in the cost estimates.
In some cases improvements in technology have reduced the costs. An example of this is the
manufacturing process of copper canister. In other cases new research results or more stringent requirements
have led to more expensive solutions. This was the case when the tunnel backfill
method was changed. Although we already are at an advanced stage of designing the disposal process,
changes in cost estimates are still probable.
Table 4 is a summary of the evolution of cost estimates for our spent fuel disposal. The impact of
different factors cannot be separated as they overlap each other. The most important reason for increased
total costs is the introduction of the new unit in 2003. Olkiluoto 3 had an impact through
the increased amount of spent fuel and the prolonged operating time of disposal facility. The latter
was the major reason for the rise in the unit costs due to a stretched operation of disposal facility.
The first cost estimate made in 1980 may be interpreted as a kind of exercise that was based on generic
cost information of KBS-3 concept without detailed data of all cost components. The estimate
of 1994 was the first calculation including Loviisa fuel for which no technical plans had yet been
made. Since then the changes in unit costs have been moderate when excluding the introduction of
Olkiluoto 3 discussed above.
Table 4: Evolution of cost estimates for spent fuel disposal (December 2007)
4. International situation
Estimation Cost estimate Euro / kgU Euro cent /
year [Million Euro]
kWh
1980 650 540 0.27
1994 815 340 0.10
1999 1,020 400 0.12
2000 1,030 400 0.12
2003 3,000 530 0.15
2006 3,140 570 0.16
Most national high level waste management programmes are still at a preparatory stage. This means
that the technical plans drafted in such programmes are at a generic level and efforts for site selection
are still pending. The estimation of related costs cannot rely on detailed technical and site information
but have to be based on conceptual studies. Cost estimates, of course, will become more
accurate with developing requirements and technical plans. In our experience the costs tend to rise
when the development of technology advances and the implementation plans get more detailed.
The international development in high level waste management is characterised by continuous research
that produces new findings which often lead to new requirements to be considered in design.
The combination of developing requirements and the long lead times needed in the development of
technical solutions makes it difficult to fix the final design. This, however, is necessary in order to
produce reliable long-term cost information.
Although the waste management programmes and solutions are national there is an interest in every
country to compare costs for the purpose of benchmarking. This is, however, very difficult for several
reasons: the programmes are at various stages, their scopes and related waste volumes are dif-
54

ferent, the technological solutions vary and, finally, little information is available. In addition, the
implementation is organised in different ways and the funding schemes differ.
Some information of nuclear waste management costs has been published lately. A rough interpretation
of this data is the following:
- In Finland the costs of final disposal of 5,500 tU spent fuel is 3 billion in 2006 Euro [1].
- In USA the total cost of Yucca Mountain repository for 109,300 tU is $90 billion (about 60
billion Euro) [2].
- In Sweden the cost estimate for final disposal of 9,100 tU spent fuel is 35 billion SEK (about
3.5 billion Euro in 2007) [3].
- In UK the estimated disposal cost for an amount of 16,400 tU high level waste and spent fuel
is 12.2 billion £ in 2008 (15 billion Euro) [4].
- In Spain the total cost estimate for disposal of 6,800 tU spent fuel and high level waste is
about 3 billion in 2006 Euro [5].
It is essential to observe that the figures given above for the final disposal of spent fuel or high level
waste are not comparable as such. With all the limitations inherent in any comparison, the costs
seem to be in the order of 0.5 to 1.0 billion Euro per 1000 tU to be disposed of. Related to the number
of NPP units served by the disposal facility, the figures vary between 0.3 to 0.6 billion Euro per
unit. We only can comment the costs calculated for Finland, which are at the high end of the comparison:
they can be explained by the small number of NPP units and the very low utilisation rate of
disposal operations, which is a deliberate choice and not a result of cost optimisation.
5. Reflections and conclusions
Estimation of costs is indispensable for assessing the financial provisions needed for radioactive
waste management. It is also unavoidable to demonstrate that there exists a solution for waste management
that is economically feasible. The funds for radioactive waste management must be collected
in advance and they must be available when the waste management operations are carried
out.
Cost estimates for final disposal of spent fuel from Olkiluoto and Loviisa NPPs have been made
regularly since 1980. They have become more accurate in pace with the development in technology
and implementation plans. The developments have a twofold impact in cost estimates. In some
cases improvements in technology have reduced costs while in other cases new research results or
more stringent requirements have led to more expensive solutions.
The results of cost calculations are dependent on input parameters, some of which are based on
technical boundary conditions and some on economical and financial assumptions. The crucial
technical boundary conditions are obvious: quantity and burnup of spent fuel. Scheduling of disposal
activities is not only an economic issue, as it is dependent on the progress of the whole waste
management programme, including policy decisions. A data base of the relevant cost drivers must
be collected and the implementation costs of different alternatives must be estimated simultaneously
with design process.
The international development in high level waste management is characterised by continuous research
that produces new findings which often lead to new requirements. The combination of developing
requirements and the long lead times needed in the development of technical solutions
makes it difficult to fix the final design. This in turn necessitates repeated and revised cost calcula-
55

tions. This is a real dilemma: reliable and predictable cost calculations are required while at the
same time cost estimates are subject to changes due to modified technical plans.
Radioactive waste management is not a stand-alone industry. The technology to be used there is dependent
on other industrial structures and their developments. It is impossible to foresee what will
happen globally during the operation of a repository over the next 100 years or so. Issues like availability
of raw materials, their prices and the existence of subcontractors can only be mentioned but
no justified long-term predictions can be made. In addition, high level waste management schemes
may encounter fundamental changes due to changes in the operational environment. This is suggested
by recent international developments in the back-end of nuclear fuel cycle.
References
[1] Palmu J. (2008). Summary of the Cost Estimate for Spent Nuclear Fuel Disposal of the
Olkiluoto (OL1-3) and Loviisa (LO1-2) Nuclear Power Plants. Memo POS-003662. Posiva
Oy. 5 p.
[2] DOE/OCRWM (2008). Analysis of the Total System Life Cycle Cost of the Civilian Radioactive
Waste Management Program, Fiscal Year 2007. DOE/RW-0591. U.S. Department of Energy,
Office of Civilian Radioactive Waste Management. 34 p + app.
[3] SKB (2007). Plan 2007, Kostnader för kärnkraftens radioaktiva restproducter. Svensk
Kärnbränslehantering AB. 56 p.
[4] NDA (2008). Annual Report & Accounts 2007/08. Nuclear Decommissioning Authority. 187
p.
[5] MITYC (2006). Sixth General Radioactive Waste Plan. Ministerio de Industria, Turismo y
Comercio. 249 p.
56

PANEL DISCUSSION
Summary of the Panel Discussion concluding
Session II: Economical factors governing geological disposal programmes
Panel members:
Thibaud Labalette (Chair), ANDRA, France
Milena Christoskova (Rapporteur), DPRAO, Bulgaria
Hans D. K. Codée, COVRA, The Netherlands
Vit�zslav Duda, RAWRA, Czech Republic
Jean-Paul Minon, ONDRAF/NIRAS, Belgium
Ján Timul'ák, DECOM A.S., Slovakia
This report summarizes the outcome of session II on the economical factors governing geological
disposal programmes. The session included a keynote speech by Dr Eero Patrakka (CEO, Posiva
Oy, Finland) on the "Assessment of financial provisions for nuclear waste management – long-term
perspective from a Finnish viewpoint". This was followed by a panel discussion based on a set of
questions.
The questions put forward for discussion were:
� What are the main economic parameters governing RWM and disposal in the EU?
� What are the main funding schemes and principles of financing waste disposal?
� What are the main elements to consider in the overall estimation of RWM costs?
� What are the influencing factors for costs variation?
� What is the influence of the volume of waste on the disposal costs?
� How can new decisions on nuclear energy (new NPP…) influence costs and funding schemes
and how are they taken into account?
The panellists presented their statements before the word was given to the floor.
Hans D. K. Codée
COVRA is a pioneer company in long-term management. If we make a simple calculation based on
nuclear power units installed in the Netherlands, 30 years of operational period and energy production
the cost of repository is unbearable for the country. Even if the operational period is doubled
and new NPPs are constructed, the cost cannot be afforded in the country.
The only opportunity is to share the repository and to share the costs. It is not easy but it is the
European way. The fact the money will be kept for 100 years has to be accepted and the money has
to be managed in a proper way. The volume of waste produced in France in one year is equal to the
volume of waste produced in the Netherlands in course of 100 years. The operating cost depends on
57

the shipment. The Netherlands produce 6-7 canisters per year and one shipment of 25 canisters is
carried out once every 4 years.
Since time is the crucial factor for making radioactive waste harmless 100 years is a suitable time to
keep the material in storage and to provide the financing for the deep geological repository. Price
paid by the polluter nowadays for the waste generated covers all the expenses including collection,
100 years storage and disposal.
Vit�zslav Duda
The Czeck republic has a big nuclear sector and can afford to cover the expenses for the deep geological
disposal. The bank system is reliable and there should not be problems in financing these
activities.
Jean-Paul Minon
The economical factors in geological disposal programme are heavily influenced by four factors:
1. Volume of waste;
2. Time schedule (construction, investment, calculation of money –we always need more than
expected);
3. Regulatory framework (the high level group is dealing with harmonization of regulatory
framework in the EU). During the design process agreement on security and safety must be
achieved;
4. Technical and economical aspects and environmental policy having in mind the life time of
these projects. For example the monitoring period for surface disposal is 300 years. For the
same period of 300 years the agricultural economy has become information economy.
Someone has to pay the bill for deep geological disposal. In Belgium a draft bill is developed including
the risks what will happen if there is no money. Mechanisms for securing the money with a
kind of insurance mechanism have to be prepared, in order to ensure that whatever happens, the bill
will finally be paid.
Ján Timul'ák
Slovakia has established a body responsible for RAW management funds. In fact, three bodies dealing
with radioactive waste exist in Slovakia:
� One dealing with RAW processing;
� National nuclear fund dealing with the money;
� DECOM – responsible for strategic management.
In 1994 a new act on the safe use of nuclear energy has been adopted. A national nuclear fund has
been established in 2006.
A national strategy on safe management of spent fuel and radioactive waste has been developed.
The technical and financial aspects of RAW and SF management are concerned. Three approaches
for SF management are considered:
� Direct disposal;
� Shipping to the country of origin and external disposal;
� Regional geological disposal facility (SAPIERR).
58

Discussion:
As a response to a question on the existence of examples of research to decrease the costs of geological
disposal programmes instead of increase, E. Patrakka indicated that if the design basis of the
disposal facility is not fixed as in the Finnish example, R&D activities keep carrying out and the
last scientific achievements can be implemented. In general, scientific and legislative developments
need money added J-P. Minon.
The opportunities for using or not using discounted costs have been discussed. Experience from
COVRA is that it is not possible to use discounted costs. POSIVA Oy example considers both calculations.
The way to regularly re-evaluate the estimated cost of radioactive waste management has been discussed.
In France the cost is evaluated every three years under the control of the administrative authority.
In the Netherlands, the costs are reviewed every five years; for the waste generated in the
past, a revision of prices is not foreseen. The status of the cost evaluation is different among the
countries (public information, commercial data). On the question on how the money accumulated in
the funds is managed – for example by investment banks, shares, real estate, E. Patrakka answered
that in Finland it is a state fund and POSIVA Oy is not responsible for the money management. J-P
Minon indicated that in Belgium long-term funds are established by a Royal decree and money are
used to the state profits. The surveillance committee takes care to approve the solidity of money
investment. They are used by the economic network and saved by economic cycle. In France, the
provisions of the nuclear operators are controlled by the administrative authority.
In relation with the example of copper canister thickness which decreased twice during the past
years, there is a concern on losing the flexibility by creating a picture that cannot change if the price
decreases. On this issue E. Patrakka indicated that the raw material prices are forecast, the real cost
is unknown.
H. Codee insisted that the only option for the small countries to cope with the costs is to try to find
a way to share the cost. If 25 or 27 repositories in Europe are constructed they will not work. Multilateral
repository will be a really working facility. The principle “Together in Europe” has to be
applied.
The cost estimation has to include total cost of siting, R&D activities, and design.
The small countries will benefit if a market of disposal activities is established in Europe and if an
opportunity for transport of spent fuel and vitrified waste is created commented V. Duda.
The Finnish example and the panel discussion illustrate the importance of economical aspects
within radioactive waste management. The cost evaluation process is not simple since it has to address
long periods of time and uncertainties. Many countries have already implemented dedicated
funds or assets to prepare the realization of long-term waste management solutions.
59

SESSION III: Co-operation in geological disposal
Chairman: Mr Jean-Paul Minon, ONDRAF/NIRAS, Belgium
Introduction and objectives
While multilateral cooperation in geological disposal at the stage of R&D is already widespread in
Europe, and has to a large extent been fostered by Euratom framework programmes over the years,
this must now be capitalised upon through true joint strategic planning of implementation-oriented
research activities. Cooperation between programmes faced by similar challenges, such as in
Finland and Sweden, or between countries favouring a specific host rock type, has also proved to be
effective and can be the basis for closer ties in the future.
In fact, exchange of information, transfer of technology and know-how, joint construction works or
even shared facilities at all stages of RWM cooperation could be efficient means to overcome the
financial and human resource challenges addressed in Sessions I and II. The FP6 projects CATT
and SAPIERR, dealing with technology transfer for the realisation of national solutions and the feasibility
of shared solutions, have demonstrated the interest in and the need for a debate on the concept
of common initiatives. This debate needs to be conducted very carefully in the case of the very
sensitive issue of multinational repositories.
Consequently, the objective of the session is to establish an overview of the actual situation in the
EU regarding cooperation in all aspects of RWM, including their benefits, challenges and drawbacks.
61

Cooperation in the development of geological disposal concepts –
Benefits and challenges
Monica Hammarström, SKB, Juhani Vira, Posiva
Research and development supporting the development of a final disposal concept for
spent nuclear fuel
This presentation will give an example of how research and development performed in international
cooperation in the field of spent nuclear fuel have supported the development of a
concept for final disposal of spent nuclear fuel.
1. Nuclear Power in Sweden and Finland
Both Sweden and Finland have been operating nuclear power reactors since the 1970s. The Swedish
nuclear power programme consists of 12 reactors. Two of the reactors, situated at Barsebäck
were shut down in 1999 and 2005 due to political decisions. The reactors in operation (9,000 MWe
net total) generated around 70 billion kWh in 2007 which corresponds to almost half of the country’s
electricity. An extensive uprate programme is ongoing for the Swedish reactors to compensate
for the loss of the production from the Barsebäck reactors.
The Finnish nuclear programme consists of 4 reactors at two sites. The two reactors situated at
Olkiluoto are operated by Teollisuuden Voima Oy (TVO) and the two at the Loviisa site are operated
by Fortum Power & Heat Oy (Fortum). A 5 th reactor is being constructed at the Olkiluoto site.
The plans are to start the operation of that reactor in 2011. The existing reactors (2,696 MWe net
total) generated 22.5 billion kWh net in 2007, which equals to about one quarter of the country's
electricity.
In March 2007 TVO and Fortum announced that they were each about to commence environmental
impact assessments (EIA) for new nuclear power units at the Olkiluoto and Loviisa sites respectively.
This would clear the way for either company to seek government approval for a new unit,
though no investment decision had been made. TVO's EIA for Olkiluoto-4 was submitted to the
government in February 2008, for a 1,000-1,800 MWe PWR or BWR unit. Fortum's plans for an
EIA on a 1,000-1,800 MWe unit at Loviisa were submitted in June 2007.
In June 2007 a new consortium of industrial and energy companies announced plans to establish a
joint venture company - Fennovoima Oy - to construct a new nuclear power plant in Finland. The
group consisted initially of stainless steel producer Outokumpu, mining and melting company Boliden,
energy utilities Rauman Energia and Katterno Group, and electricity supplier E.On Suomi
(the Finnish subsidiary of Germany-based E.On) which is leading the project. Then the ownership
base expanded from five to over 60 as electricity consumers sought to insure against future energy
cost blowouts.
2. The waste issue became a political issue
The nuclear waste issue became already in the 1970’s a political question in both countries. In
Sweden a special Government Committee on Radioactive Waste was set up. The committee, with
63

members also from the industry, recommended interim storage of spent nuclear fuel, a final repository
for LILW, a transportation system, necessary arrangements for reprocessing of the spent fuel
and also for direct disposal of spent fuel in a deep geological repository. It was also recommended
that the waste producer should bear all costs… The work by the committee resulted in the passing
of a new law, stipulating that new nuclear power reactors could not be put into operation unless the
owner was able to show that the waste problem was solved in a safe way. The nuclear power industry
in Sweden decided to give highest priority to the waste problem in order to meet the requirements
in the law. A jointly owned company, which today is SKB, was established to fulfil the requirements
in the new legislation. Planning of new waste management facilities started and the interim
storage facility was taken into operation in 1985 and the final repository for LILW in 1988.
The law stipulated that the owner of the reactor had to show how and where a completely safe storage
could be provided for either the high level reprocessing waste or the spent nuclear fuel. “The
storage facility must be arranged in such a way that the waste or the spent nuclear fuel is isolated as
long a time as is required for the activity to diminish to harmless level.” “These requirements implied
that measures should be taken which during all phases of the handling of the spent nuclear
fuel, can ensure that there will be no damage to the ecological system.
The waste management programme in Finland started in the early 1980’s. The accident at Three
Mile Island led to increased concerns about the use of nuclear power and also of waste issues. The
general plans for nuclear high level waste management in Finland were based on future reprocessing
of the spent fuel. TVO was at that time discussing about reprocessing contracts with BNFL and
Cogema. IVO (later Fortum Power and Heat) was still sending the spent fuel back to the Soviet
Union. The high prices for reprocessing together with low uranium prices lead to second thoughts
and TVO withdrew from negotiations regarding reprocessing contracts.
In 1983 a decision was taken by the Finnish Government that TVO should either seek for possibilities
to send their spent fuel abroad permanently or start a systematic for direct geological disposal
of spent fuel in Finland. Future milestones for a disposal programme were defined. A site for a final
repository for spent fuel was to be selected in 2000, the repository construction should start in
2010’s and the start of disposal of spent nuclear fuel operations in 2020. Reacting to various national
and international developments in 1994 an amendment was made to the Nuclear Energy Act
that prohibited both the exports and imports of any nuclear waste from or to Finland. In practice
the amendment meant that all the nuclear waste arising should be disposed of in Finnish bedrock or
soil.
3. Start of research and development work
Comprehensive and intensive research work started in Sweden as a result of the new Swedish legislation.
A “crash program” was started and the very first international discussions on experiments in
an abandoned iron mine, Stripa, in central Sweden started. The international cooperation in the
Stripa mine started as a joint project between Sweden and the United States. An early American
proposal about basic research together with a more down to the earth Swedish research program
was the start of a very fruitful period which set the standards and ambitions for international cooperation
in the area of research on geological disposal of spent nuclear fuel. Around 1980 an important
step was taken when OECD/NEA took the initiative to organize a broader international participation
in the Stripa Project. This initiative resulted, in a very short time, in the creation of a network
of the most excellent researchers, experts and laboratories around the world. Both Finnish
nuclear power companies, TVO and IVO joined the Stripa project during the 1970s.
64

During the 1980’s it was realized that in order to fully understand the environment down to the repository
depth of a deep geological repository it would be necessary to perform all steps from investigations
in boreholes to the excavation of tunnels and deposition galleries and holes. The plans
to start the construction of an underground laboratory, the Äspö Hard Rock laboratory, in an virgin
rock environment thus was started. Experiences and knowledge gained in the Stripa project was
transferred to the planning of the various activities at Äspö as well as the international cooperation
and the network of specialists.
A consequence of amendment to the Finnish Nuclear Energy Act, made in 1994, meant that also
Fortum had to start planning the disposal of their spent fuel in Finland instead of sending it back to
Russia. After negotiations between TVO and Fortum, a joint company, Posiva Oy, was established
to take care of the disposal of spent fuel from the Olkiluoto and Loviisa reactors. In practice Posiva
took over the programme started by TVO but modified the plans for larger amounts of spent fuel.
The time schedule for preparations for geological disposal remained the same as before. The participation
by TVO in the international cooperation at the Äspö laboratory was transferred to Posiva.
4. Benefits from international cooperation
The model for international networking and cooperation has, ever since the Stripa period, been a
key component in the development of the Swedish wastewaste management programme and the
KBS-3 concept. The Stripa project enhanced the level of international cooperation on geological
disposal and in situ research. It served as a signal to many countries to launch their own field work.
The cooperation between SKB, TVO and IVO and later on with Posiva was strengthened by bilateral
information exchange agreements signed in 1987. This information exchange was performed
in regular meetings on various organisational levels and contacts between experts in various areas.
Joint development of a canister concept (”ACP” Canister) was started as a result of this information
exchange and later on cooperation started at the Äspö Hard Rock Laboratory.
SKB’s progress in encapsulation and repository technology and Posiva’s progress in the siting process
during the 1990’s of course further increased the incentives to enhance the cooperation. It was
also recognized that since SKB’s and Posivas’s objectives and programs for final disposal of spent
nuclear fuel are very similar both companies would gain considerable benefits both in economical
terms, including sharing of resources but also to maintain and enhance both political and public acceptance.
At the end of the 1990’s the cooperation increased to cover in principle all areas of the
implementation of the KBS-3 concept. A Joint Steering Group was created to manage the cooperation
and to develop future joint strategies. A five-year agreement on extensive technical cooperation
between SKB and Posiva was signed in 2001. During this 5-year period the cooperation comprised
in total 80 joint projects corresponding to a total cost of 30 million Euros. The agreement
was renewed in 2006 for another five years.
The participation in activities at SKB’s laboratory facilities; the underground laboratory at Äspö,
the Bentonite laboratory and the Canister laboratory, are strategic components in Posiva’sPosiva’s
spent fuel programme. These facilities give the possibilities to studying generic problems in crystalline
rock, developing methods and techniques and for demonstration activities. These laboratories
attract many organisations and specialists around the world. An extensive international cooperation
programme has been in operation at Äspö since the laboratory was constructed. This participation
by international experts ensures high scientific and technical quality of the research and
development.
65

In 2003, Posiva started the construction of the ONKALO, an underground site characterisation facility,
for the purpose of final characterisation of the bedrock on the site for the final repository.
The facility will also give possibilities to test final disposal technology in actual deep underground
conditions. The experiences that will be gathered during the construction will be important complements
to SKB’s planning for the future construction of the final repository in Sweden. It will
give us the opportunities to compare results from experiments and tests that have been performed in
the Äspö laboratory. It will give opportunity to verify methods and quality control programmes and
to demonstrate the whole KBS-3 disposal sequence in full scale at a real site.
5. Siting of facilities
The siting programmes for the final repositories both in Finland and Sweden have been performed
in parallel and also integrated to research and development activities. According to the Swedish
legislation reactor owners had to ensure safe management of the spent fuel and also to present a
R&D program for final disposal of the fuel. SKB published the KBS-3 report in 1983, 25 years
ago, covering these requirements. The R&D program should also cover the siting of a repository.
The report was approved in 1984 by the Swedish government and gave green light for the start-up
of new reactors.
The Finnish strategy at that time was to adopt the KBS-3 concept and focus resources on the siting
of the final repository to be able to select one site in 2000 which was required by the Government
guidelines. The site selection process started in Finland in 1983 and ended in 1999 when a site at
Olkiluoto in the community of Eurajoki was selected. The corresponding site selection process
started in 1992 in Sweden, when SKB in its RD&D Programme 1992 described the stepwise process
for siting and construction of the final repository for spent fuel. SKB started site characterization
at two sites in 2002 and one of these sites will be selected during 2009.
SKB and Posiva have had comprehensive information exchange and collaboration in all aspects in
their siting activities. This collaboration has not only supported investigations and the technical
work performed by SKB and Posiva but has also strengthened relations between all stakeholders
involved and increased our knowledge in all aspects of nuclear waste management.
It is our belief that the joint research, joint development work and the sharing of information in all
steps all the way to the point when both countries repositories are taken into operation will support
our work in ensuring that quality and safety requirements are met. We also believe that sharing and
comparing work and results will help us to reach the necessary level of confidence in all steps of
repository development.
6. Challenges
Of course challenges will appear during this kind of cooperation which comprises issues of such
diversity. Some are dealing with cultural differences, some are technical and some are of organizational
character. The differences in legislations and regulations of course have impact on the cooperation.
The timescales of SKB and Posiva have the same target date for start of operation of repositories.
This does not necessary mean that the delivery of results before this date are the same. These differences
mainly come from differences in regulations in each country and how the license process
has been set up. In Finland a two-step licensing process is followed: first an application is made to
the government on the construction license, after construction a separate application is made to the
66

government on the operating license. This means that the information requirements at the stage of
construction license are somewhat different from Sweden where the main licensing process takes
place before the construction of the facility. The differences in the licensing process mean different
priorities and different interests in various RD&D tasks. Cooperation has to be seen in long-term
perspective but this is not always easy when time schedules are tight and when resources have to be
used in the own programme. The cooperation and the transfer of knowledge will not be efficient if
the differences in staffing and the access to expert resources are imbalanced. Different cultures
both organizational and on personal levels put strains on the practical cooperation and should not be
neglected.
Both SKB and Posiva have entered the licensing process and have very focused programmes with
the aim to have repositories for spent nuclear fuel in operation in 2020. We are both in the stage of
shifting from research oriented to implementation oriented. These steps and changes also have to
be considered and will influence on the scope and the form of our cooperation.
Conclusion
Seen from Posivas perspective, with a relatively small nuclear programme, at least historically, it is
possible to have a programme for geological disposal of spent fuel without excessive costs. The
cooperation with SKB and also with other international organisations has been important for the
success in Finland. The language barrier to Sweden has not been a problem. Both SKB and Posiva
have support both from the political side and from its owners to solve the problem “at home” and
“now”.
It is our experience that international cooperation make sense only in the context of a sufficiently
extensive own programme, otherwise information transfer and application is difficult to realise.
When stakes increase the fairness of cooperation also becomes important. The ties created by close
cooperation need to be acknowledged and accepted as well as differences in the national contexts
and conditions.
I would like to take the opportunity to inform all participants during this conference that stakeholders
from a number of countries in Europe foresee continued cooperation in the field of implementation
of geological disposal. Such cooperation we believe can be boosted through what is
called “Technology Platforms”. It is my pleasure also to inform you that SKB and Posiva have
been entrusted with the task to lead the initial steps towards such a platform. More information on
this task will be given later during this conference. You will also get background information in the
presentation on the CARD–project later during this conference. The CARD-project looked into the
feasibility and the interests of establishing such a platform.
On behalf of my Finnish colleagues and SKB I would like to end this presentation by recommending;
fruitful cooperation is reached with open and flexible minds, the strong will to find solutions
and targeted programmes.
Thank you!
67

PANEL DISCUSSION
Summary of the Panel Discussion Concluding
Session III: Cooperation in geological disposal
Panel members:
Jean-Paul Minon (Chair), ONDRAF/NIRAS, Belgium
Ewoud Verhoef (Rapporteur), COVRA, The Netherlands
Charles McCombie, ARIUS, Switzerland
John Mathieson, NDA, United Kingdom
Irena Mele, ARAO, Slovenia
Piet Zuidema, NAGRA, Switzerland
This report summarizes the outcome of session III on�cooperation in geological disposal. The session
included a keynote speech by Ms Monica Hammarström, SKB, Sweden, on "Cooperation in
developing geological disposal concepts – benefits and challenges". This was followed by a panel
discussion based on a set of questions.
The questions put forward for discussion were:
� What is the situation in the EU with regards to multi-national solutions (overview)?
� What is their potential regarding overcoming the perceived financial burden for small programmes?
� Is technology transfer feasible? What is transferable and what are the limits of such transfers?
What is the added value (national / EU)?
� How can collaboration in research be enhanced in the future? How can research cooperation be
optimised between national programmes of different sizes, speeds and states of advancement?
Can this bring mutual benefit for both small and large programmes? Can underground laboratories,
and other research facilities, be better utilised or even shared between research programmes?
� In particular, how can countries collaborate effectively in RWM training programmes?
� Are shared facilities realistic in the EU?
� What benefits, challenges and risks are seen today for the implementation of shared facilities?
What are the regulatory implications of such solutions?
Introduction by the chairman
This session deals with cooperation in geological disposal in general and multinational repositories
in particular. There is already quite some cooperation in the field geological disposal. Many organisations
and forums are in place to facilitate cooperation, think of the IAEA, the OECD-NEA
and the EC. A first step towards cooperation is the collective commitment to safely manage radioactive
waste, for example by signing the Joint Convention. The EC funded research in the Euratom
Framework Programmes can be seen as a second step. In addition to the Framework Programmes,
the EC has established different forums to facilitate cooperation: ENEF (stakeholders), HLG (regulators)
and CARD proposing a Technology platform for end-users (WMOs) and research organisa-
69

tions. EDRAM is another example of cooperation between WMOs from EU member states (with
active geological disposal programmes and Underground Research Laboratories) and the USA,
Canada and Japan.
The panellists then presented their statements before the word was given to the floor.
C. McCombie and J. Mathieson reported on two specific EC funded projects, SAPIERR I and II and
CATT respectively, that were launched as part of the 6th Euratom Framework Programme to explore
how Member States with relatively small amounts of nuclear waste can implement long-term
solutions through collaboration.
Charles McCombie: SAPIERR
Shared repositories have been investigated over the past 3 years in the SAPIERR projects. The objective
of SAPIERR I (Support Action: Pilot Acton for European Regional Repositories) was to explore
the feasibility of regional European solutions for deep geological disposal, mainly in terms
strategy and legal aspects. The project was coordinated by DECOM of the Slovak Republic. The
present SAPIERR II project (Strategic Action Plan for Implementation of Regional European Repositories)
examines in more detail specific organisational, legal, societal, economic, safety and security
issues that directly influence the practicability and acceptability of such facilities. The project
is coordinated by COVRA, the Dutch WMO. The goal of the project was to propose a practical
implementation strategy and organisational structures to create a formalised, structured organisation
for implementing shared EU radioactive waste storage and disposal activities. The tasks proposed
in the project are designed to meet this goal:
1. Preparation of a management study on the legal and business options for establishing a
European Development Organisation (EDO) leading to one or more proposed frameworks
(options) for such an organisation;
2. A study on the legal liability issues of international waste transfer within Europe. Even in
national disposal programmes, the issues associated with long-term transfer of liabilities are
complex;
3. A study of the potential economic advantages but also benefits of European regional storage
facilities and repositories. The economic implications are analyzed at local, national and regional
level;
4. Outline examination of the safety and security impacts of implementing one or two regional
stores or repositories relative to a number of national facilities;
5. A review of public and political attitudes in Europe towards the concept of shared regional
repositories;
6. Development of practical implementation strategy. Should we set up an EDO? This part of
the project is more related to political/strategical aspects, than research. Therefore, policymakers
from some twenty different European countries have been invited to participate in
the follow-on working group and defined terms of reference, organisational form, programme
etc. of the EDO.
The countries that have confirmed participation in the working group to date are: Bulgaria, Czech
Republic, Estonia, Italy, Latvia, Lithuania, Netherlands, Romania, Slovakia, Slovenia and Spain.
The first meeting of the working group will be held together with the closing seminar of the SAPI-
ERR II project 20 th of January 2009 in Brussels.
70

John Mathieson: CATT
CATT stands for cooperation and technology transfer on long-term radioactive waste management
for Member States with small nuclear programmes. The overall objective of CATT is to investigate
the feasibility of Member States with small nuclear programmes (referred to as Technology Acquirers)
implementing long-term radioactive waste management solutions within their national borders,
through collaboration on technology transfer with those Member States with advanced disposal
concepts (referred to Technology Owners). Members of the consortium CATT are the national
waste management organisations of the UK (Nirex), Sweden (SKB), Germany (DBE), Lithuania
(RATA), Bulgaria (DPRAO) and Slovenia (ARAO), together with Joint Research Centre in The
Netherlands and the IAEA.
Although CATT assumes that each nuclear Member State has its own repository whereas SAPIERR
assumes that there will be both national and also shared repositories in Europe, the two actions are
complementary. The focus of CATT is on HLW and SF, rather than LILW for which solutions already
exist in many Members States. It looked at commercial, policy and legal aspects of technology
transfer. Technologies of long-term storage, and encapsulation in particular were taken as
cases to examine the feasibility of technology transfer. Different scenarios were investigated ranging
from using a foreign technology to sharing an encapsulation facility – even a mobile encapsulation
facility was examined – work out collaboration models. The idea of a technology platform
emerged as a result, which was investigated in another 6 th Euratom Framework Programme project,
CARD.
Irena Mele
She presented Slovenia as an example of a small nuclear programme seeking cooperation. The
main problem for Slovenia is to maintain sufficient critical mass for a geological disposal programme,
in terms of financial and human resources. International cooperation is vital to accumulate
and maintain sufficient expertise, but also to attract young people. ARAO employs 20 persons
only. This means that when the staffs are involved in the development of the LILW repository, it is
very difficult to also upkeep the HLW programme. In Slovenia there is no industry to (financially)
support the programme, and only limited R&D capacity, therefore, ARAO participate in six FP6
projects (and will participate in FP7 as well) to be able to further develop the programme and work
out cost estimates.
Piet Zuidema
He provided a personal view on cooperation. The multinational solution may be attractive, but is
prohibited by law or policy in many countries. The multinational solution is possible in Switzerland
under certain conditions, but does not play a role in the current waste management strategy.
Financial costs should not be an important argument for multinational solutions in small programmes.
Energy production from nuclear in small programmes even including disposal is less expensive
than energy production from renewables. Technology transfer is only feasible with fully
developed technologies and can then be left to the market. More cooperation in science may not be
necessary, as results can often be found in the open literature. International cooperation on training
is considered difficult and national training on the job should be preferred. R&D can be done internationally
to share resources and costs. It however results in R&D compromises, shared operational
power and stumble upon different expectation of and opinions on the programme.
Mr Zuidema also suggested further cooperation between regulators, to harmonize their rules and
messages, between policy-makers to synchronize approaches, processes, norms etc. Finally he emphasized
the role of different international organisations in cooperation.
71

Discussion:
The chairman opened the debate by stating that sharing facilities is an idea that must be discussed
and presented the EDRAM position on cooperation in geological disposal:
1. Each country responsible for their waste;
2. Each country has the right to prohibit import/export of radioactive waste;
3. Successful implementation of national programmes within 10-20 years is top priority;
4. Countries can share repositories as long as they comply with international safety standards;
5. Same ethical considerations should apply to national and multinational repositories;
6. International cooperation plays an important role in R&D and can promote progress.
From the discussions and comments made by the panellists, the main elements are summarized below:
If successful implementation of national programmes within 10-20 years is top priority,
should we dedicate all EC resources to advanced programmes (F, SE, FI)?
Every country is responsible for its own waste. Therefore, cooperating with other countries
to help them find solutions for their waste without working on its national solution (i.e. waiting
for a multinational solution), is an irresponsible approach. In addition, the proposed
waste directive applies to everybody; EDRAM identified the multinational option as a reason
not to include tight time schedules in the waste directive.
Multinational approaches should not make the same mistakes national programmes made
before: don’t go too fast, start with the support of the public, which is not yet the case in
many countries. Examine social support at local level at various locations before progressing
Multinational solutions are necessary for countries like Slovenia. For instance, the price to
encapsulate waste represents a major part of the overall cost of disposal. Some countries
cannot afford the price of such encapsulation plant as well as the canister fabrication plant
which would double the cost of disposal. As it is done for the front-end, international (industrial)
services must be considered for the back-end.
There appears to be a strong opposition to multinational solutions especially from countries
currently developing their own national programmes. This debate has become over-heated.
Both national and multinational solutions can be envisaged and studied. After all the whole
fuel cycle is international, why make the disposal of radioactive waste an exception.
There is an important difference between other steps in the fuel cycle and waste disposal.
Other fuel cycle products are geographically widespread, while waste is placed at a specific
location.
Wrap-up by the chairman
Sharing facilities is an idea that must be discussed. We have to investigate the basic reasons for
multinational solutions. At the same time we have to realize that some countries have difficulties
working on national programmes and are afraid that the multinational solutions may set them back
20 years. There are contrasting perspectives on the idea, but we need to watch the temperature in
the debate and take it back to a sane level.
72

SESSION IV: Communication of risk and uncertainties
Chairman: Dr Claudio Pescatore, OECD/NEA, France
Introduction and objectives
One of the specificities and challenges of geological disposal is the extremely long time frames
over which the safety of the facility has to be demonstrated, involving a number of assumptions
with inherent uncertainties and the impossibility of assuring zero risk.
Public acceptance of geological disposal is the key to success. However, it is highly dependent on
developing trust in the implementer and the authorities, leading to reassurance that the repository
will provide the required level of protection. Given the complex nature of the subject, this requires
transparent decision-making processes and the ability to communicate information on risks and uncertainties
in an open and objective manner to the different stakeholders, including the public in
general. This issue of risk communication has been a topic of a number of framework-programme
projects in the past and in particular is currently being investigated in two FP6 projects, ARGONA
and PAMINA. The former is looking at the broad area of risk governance and communication,
whereas the latter is a technical project on performance assessment but with a mandate to communicate
and disseminate to a wider audience.
The objective of the session is therefore to debate means and approaches to manage the communication
of risk and uncertainty in order to provide convincing arguments why, despite these risks and
uncertainties, a sufficient level of safety can be maintained even over very long timescales. The session
will therefore not cover technical aspects, but might address possible needs for technical and
legislative harmonisation as a potential precondition to achieve trust.
73

Communicating the safety of radioactive waste disposal – the perspective
of a person responsible for science and technology within an implementing
organisation
Piet Zuidema, Nagra, Switzerland
Communicating the safety of radioactive waste disposal is a major issue. For the scientists involved,
this communication has different target groups (different groups within the implementer organisation,
safety authorities, licensing bodies, decision-makers, public, etc.). Most of the work by a typical
scientist is directed at internal clients (people within the organisation, scientific community directly
related to project work, etc.) and at the authorities concerned with reviews and licensing. For
that purpose the scientist has today a whole suite of highly complex communication tools (3-D
visualisation, etc.) that are extensively used. However, these tools are often not the key to success
in communicating to the public. For the public, waste management is an issue of concern that often
also involves emotional components (fear, mistrust, etc.) and these are not necessarily best addressed
with the tools the scientist normally works with. Therefore, communication with the public
is very challenging for many scientists; being knowledgeable on a topic does not necessarily make a
scientist a good communicator.
The reasons for concern by the public are often related to radioactivity and the long time scales involved
in assessing the safety of disposal facilities, both issues being outside the 'area of experience'
of most members of the public. To bring these areas closer, there are limitations on what can be
done, because the public normally has neither the time to listen to, nor the background to understand,
all the detailed scientific arguments.
In communicating with the public, besides transferring the key arguments for safety at the highest
level, other elements are equally important to achieve confidence. These elements are related to
trust and credibility and include: (i) organisational issues (the role and responsibility of the different
groups, the decision-making process, the interaction – perceived or real – between the different
stakeholders); (ii) behavioural issues (openness to new findings (adaptive management), ability to
listen, open interaction with science ('we are devoted to good scientific practise'), …); (iii) issues
related to the design of the planned facilities (possibilities for monitoring, reversibility and retrievability,
etc.). This, however, also requires to respect the different roles in communication; some areas
are the main responsibility of the implementer, other that of the regulator, and again other that of
the policy maker ('the message and the messenger').
The observation of communication mechanisms shows that high quality information products are
not sufficient to generate trust. Indeed, communication with the public has to take place on a personal
basis. Scientists should be perceived as 'normal human beings' with a high sense of responsibility.
In their communication with the public, the scientists should be open and also honestly address
the challenges and difficulties of the tasks involved without confusing the public with too
many details; this also includes addressing remaining uncertainties.
With regard to safety, it may be helpful for the public to put the scientific messages in perspective
and connect them with issues familiar to the audience and to demonstrate that solutions exist, e.g.
by referring to successful projects and activity fields also in other countries, or to work performed
in underground laboratories and on natural analogues. Finally, the issue of 'how good is good
75

enough' and 'how safe is safe enough' can be important in the interactions with the public. Here, international
collaboration and especially better harmonisation between the different countries (e.g.
on what is considered to be safe (regulations) and on how to demonstrate safety) is an important
issue where room for improvement still exists.
76

Communicating the Safety of
Radioactive Waste Disposal
The perspective of a person responsible for
science and technology within an implementing
organisation
Piet Zuidema
Director Science & Technology, Nagra, Switzerland
Session IV: Communication of risk and uncertainties
EURADWASTE '08
20 – 22 October 2008, Luxembourg
Communication: some introductory remarks
� We all agree: communication is a major & challenging issue - not
only in waste disposal (see e.g. stock markets, ….)
� We communicate ‚always & everywhere‘ (intentional, un-intentional)
� Communication involves a broad spectrum of …
- messages & messengers
- communication mechanisms
- types of messages (emotional, factual)
- communication tools & products
- …
� Communication is complex: what, how, to whom, by whom,
when, why, …
2 Zu/201008
CEC_Communication_Zuidema_201008
77

Communication as a scientist …
… to other scientists (within implementing organisation,
with regulator, with experts, with scientific community)
� Tremendous progress in communication techniques (visualisation)
� Take advantage of work done elsewhere (CAD, GIS, tools from
mining & hydrocarbon exploration, ….)
� But: this is the easy part & only useful for ‘specialists’
� … wealth of information & its complexity may distract from the
real issues and confuse the non-specialist
� … thus: tools / messages can often not be used directly for
communication with other target groups (public, politicians,
decision-makers, …)
� p.m.: take advantage of the power of examples of ‘hardware’ (e.g.
URLs) in communicating with the public
3 Zu/201008
CEC_Communication_Zuidema_201008
The different ‘external’ groups and their interaction
Formal communication & interaction and perception by the public
Policy
Politics
4 Zu/201008
Industry
(Implementer)
CEC_Communication_Zuidema_201008
Regulator
(Supervision,
Expertise)
78
Public

Geological repositories …
Differences to other projects from the scientific point of view
� Challenges due to long time scales involved
- ‘Engineering approach’ 1) not possible for critical aspects (e.g. long-term
performance)
- Our job is limited to research, design & implementation (with rather
limited testing possibilities) � demonstration of compliance ‘by paper’
� However: ‘opportunities’ due to inherently favourable phenomena
- Decay of radiotoxicity (but: some very long-lived isotopes)
- Immobility of some of the key isotopes (geochemical behaviour) 2)
- Geological stability (long historical record, natural analogues) 2)
� And: favourable project conditions
- Relatively small volumes of wastes
- Significant financial resources available
- Time available for careful planning & implementation
1) research/design � prototypes & testing � implementation � observation & corrections
2) there is safety & ‘certainty’ for a well designed repository (well chosen site, adequate design)
5 Zu/201008
CEC_Communication_Zuidema_201008
Radioactive decay: powerful, but not complete …
Radiotoxicity of 1 t of spent fuel and 8 t of natural U
6 Zu/201008
CEC_Communication_Zuidema_201008
79
Remaining lifetime
of our sun

Geological long-term stability (plate tectonics: 100 Ma )
7 Zu/201008
CEC_Communication_Zuidema_201008
Geological long-term stability (plate tectonics: 60 Ma )
8 Zu/201008
CEC_Communication_Zuidema_201008
80

Geological long-term stability (plate tectonics: 30 Ma )
9 Zu/201008
CEC_Communication_Zuidema_201008
Geological long-term stability (plate tectonics: 0 Ma )
10 Zu/201008
CEC_Communication_Zuidema_201008
81

Independent evidence: natural analogue
Geochemical immobilisation; example: Cigar Lake (1.4 10 9 a)
11 Zu/201008
CEC_Communication_Zuidema_201008
Limits of predictability of a geological disposal system
12 Zu/201008
CEC_Communication_Zuidema_201008
82

Times & doses in perspective (example: Switzerland)
Some of the porewater
constituents originating
from the time of deposition
are still there today
1) for a well designed repository (well chosen site, adequate design)
13 Zu/201008
14 Zu/201008
CEC_Communication_Zuidema_201008
Geological repositories …
Differences to other projects from the societal point of view
� First of a kind in each country
� Novelty of project (time scales, radioactivity, …)
� Connection to nuclear power (and future of nuclear power)
� Not in my backyard & not in my term of office
� ….
The need for …
� creating trust and confidence (it takes a long time to achieve it, but
can be destroyed within ‘minutes’ – e.g. through poor communication)
CEC_Communication_Zuidema_201008
83

Requirements for trust: overview
� Trust depends upon the specific project
- properties of the system
- the way in which the system is implemented (incl. monitoring,
reversibility, institutional control, …)
- the scientific understanding about the system
- the way in which the system has been assessed
- p.m.: this includes also non-nuclear aspects (EIA, …)
� … and upon decision-making process …
- the existence of rules (and how they were developed)
- the clarity in the sequence of decisions � delineate decisions
- the behaviour of the people / organisations involved
� … and requires the engagement of legitimate stakeholders
- how to identify them (who is legitimated?)
- how to involve them (how are they engaged?)
15 Zu/201008
16 Zu/201008
CEC_Communication_Zuidema_201008
Trust is generated (through the implementer & by science)
… through good projects
� consistent with siting, design & implementation principles
� based on a sound scientific basis
- 'home work' done through adequate RD+D programme & integration of
science (interaction with scientific community)
- achieve acceptance by informed scientific community (reviews, etc.)
� no disagreements due to misunderstandings
- allows proper identification & treatment of uncertainties
- is starting point for all analyses
� assessed with a proper & transparent analysis (incl. documentation)
- working process (incl. QM)
- methods, tools & information / data � abstraction of scientific basis
- discussion of uncertainties & their importance (long time scales, …)
- arguments
- documentation (� 'auditability')
- reviews
CEC_Communication_Zuidema_201008
84

Trust is generated (by the regulator)
… through adequate criteria & guidance and a high quality,
transparent review process
� Understandable overall system requirements
� Transparent siting criteria (safety, interface to other issues (EIA, ..))
� Requirements and guidance for design (more than long-term safety)
- long-term safety issues
- operational aspects
- monitoring/reversibility/retrievability
� Stepwise approach
- development of project
- implementation of repository (role of monitoring, …)
� Review
- scope (& level of detail) in accordance with the stage of the project
- independent, but through interaction with the implementer
- responsive to the needs of society
17 Zu/201008
CEC_Communication_Zuidema_201008
Trust is generated (by policy maker)
… by providing an adequate framework (legislation)
� Reasonable and balanced (and understandable) overall goals
18 Zu/201008
- be clear about the need to solve an issue of national interest with
visible progress within reasonable timescales (stepwise approach)
- consider the needs of society (stepwise implementation, a certain
level of reversibility, consider non-nuclear issues)
- acknowledge the ambitious performance criteria used
� Process
- role of step-wise approach (adaptive staging)
- involvement of all interest groups with clearly defined roles
- maintain momentum & provide stability of process (duration of
project phases longer than duration of ‘term of office’)
CEC_Communication_Zuidema_201008
85

Summary and conclusions
� Communication is essential for progress with repositories
� Communication takes places ‘everywhere & always’ and uses a
broad spectrum of communication channels (incl. ‘behavior’)
� The different groups (implementer, regulator, policy maker) have their
distinct roles & responsibilities – also in their communication
- providing scientifically sound projects
- providing suitable criteria & guidance, performing independent &
balanced reviews
- Ensuring a proper process, interacting with all the interest groups
� International organizations are important through their work &
platforms, e.g.:
- to help harmonize the regulatory framework & corresponding messages
- to ensure proper interaction with scientific community (sound science)
� Trust and confidence will only be achieved [& maintained] if all
parties keep to their roles and take their responsibility (and are
careful in their communication)
19 Zu/201008
CEC_Communication_Zuidema_201008
86
Thank you!

PANEL DISCUSSION
Summary of the Panel Discussion Concluding
Session IV: Communication of risk and uncertainty
Panel members:
Claudio Pescatore (Chair), OECD/NEA
Mario Dionisi (Rapporteur), APAT, Italy
Daniel Galson, Galson Sciences Ltd, United Kingdom
Kaj Nilsson, Oskarshamn Municipalità, Sweden
Klaus-Jürgen Röhlig, Technische Univ. Clausthal, Germany
Nadia Zeleznik, ARAO, Slovenia
This report summarizes the outcome of session IV on communication of risk and uncertainty. The
session included a keynote speech by Dr Piet Zuidema, NAGRA, Switzerland, on "Communicating
safety of radioactive waste disposal – the perspective of a person responsible for science and technology".
This was followed by a panel discussion based on a set of questions.
The questions put forward for discussion were:
� How to communicate risk and uncertainties to a non-technical audience?
� What are the needs and expectations of citizens in this respect?.
� What role can natural and man-made analogues play?
� What level of technical detail is required?
� What is the role of and experiences with Peer Reviews?
� Is harmonisation necessary to reach public acceptance? What has to be harmonised (safety criteria,
dose limits, PA methodologies, timescale for deterministic calculations)?
� What is the key to enable the public to accept residual risks and uncertainties?
� Do we need better communication on geological disposal at the national/EU level?
Future development and implementation of geological repositories is strongly affected by the general
consensus and public acceptance.
The objective of the Session IV was to explore and debate means and approaches to manage the different
aspect of communicating the safety of geological repositories.
After a short address given by the Chairman, the session began with the keynote speech. The presentation
focussed on the different aspects of communication and particularly on the role and responsibilities
of the different actors involved, such as policy makers, regulators, industry and public
in the process of achieving trust and confidence. Besides the challenges that geological disposal
presents, such as for instance the demonstration of safety over the involved long timeframes, there
are also some opportunities in this kind of concept, such as: inherently favourable phenomena (decay,
geochemical behaviour, geological stability, natural analogues ….).
87

One important element in the communication process has been identified in the "generation of
trust". To this aim all should contribute: policy makers through transparency about the need to
solve an issue of national interest, regulators with adequate criteria & guidance and transparent review
process, industry with good projects developed on sound scientific basis.
The chairman suggested the panellists to address the following questions:
Please provide your own definition of what is safety and then address the questions:
o Does the public have a different vision of safety than the technical specialist? Which
do you think it is?
o If the different visions are not well understood and/or if the differences are not
smoothed or eliminated, can risk or safety be effectively communicated?
Do people have equal interest in safety and risk? And in uncertainty and confidence?
Is it only the implementer that has to communicate safety or risk? Is there a role for other
institutional actors?
To what extent credibility of the message is linked to the credibility of the messenger and of
the system?
To what extent are people concerned with the very long-term (thousands of years) versus the
short and medium term (tens to hundreds of years). Are we addressing the two time scales?
How do the above observations on your communication strategy.
The panellist then presented their contribution to the discussion and the debate included not only
issues listed in the questions but ranged more widely.
In the following, the main points made by the panellists and the audience are summarized:
1. The increase of effort on the issue of communication by regulators, implementers and international
organizations has been generally recognized. The increase of tools for communication
(media, web sites…) and consequently increase of information should not distract from
the real issue and confuse the non-specialist. Communication sometimes is not so much a
question of amount of information but rather of quality of information;
2. Communication on such a complex issue like radioactive waste disposal is strongly dependent
on the level of familiarity that the stakeholders have with nuclear. In this respect, an interesting
contribution was given by the representative of a Swedish Municipality (Oskarshamn)
where local population has been familiar with nuclear installations for many years.
Local population living in the area of a possible geological repository is asking information
about the impact of the repository not in a very long-term scale, but rather on their lives and
following generations;
3. Experience from many different projects involving the communication of safety is that terms
like uncertainty and risk assessment are very difficult to understand by the general public:
sometimes it is better to use terms like performance;
Safety functions, such as concentrate and contains, seems to be a good starting point: it is
easier to explain in a few words things like the availability of the waste form, or the impermeability
of a barrier, rather than dose or risk curves. The latter also have the disadvantage
of being related to release rather than containment;
88

4. The regulators could play a very important role in the process of achieving trust on the institutions
by the general public. It was recognized that an early involvement of the regulator in
the projects, already at the site selection and investigation stages, could be a productive approach.
With this regard, not a minor issue is the physical presence of the regulator on the
local site where investigations are carried out: this will help the public to achieve the feeling
that regulator in this process has the role of representative of the public;
5. It is also necessary to recognize that the concepts of risk and safety are perceived by the
general public in a different way than the experts and this is due not only to differences in
educational level but more often to emotional factors. Among the factors that are influencing
the public acceptance, also psychological factors have to be taken into account;
6. Universities and Research Institutes could also play an important role in the communication
process, since they can be seen as "third party", sufficiently “credible” and independent
from policy makers and implementers;
7. It has been generally recognized that one of the most important issue in the development of
a geological repository and consequently in a successful communication, is a clear and
transparent decisional process. Political commitment and national decision about radioactive
waste management is an essential starting point before going to the public;
8. Communication is not a process between one responsible organization and the public. There
are interactions among the different groups: Policy makers, Regulatory Authorities (that
could be more than one), Implementers, and Public. Each one has its own distinct role and
responsibilities in the process of achieving trust and confidence;
9. It was highlighted that, perhaps one of the mistake done in the past, and in some extent is
still present, is to approach the communication issue as "we and them", where "we" is intended
as the experts and "them" is intended as the public. The general approach of communicating
the safety of a radioactive waste disposal should start with creating a feeling that
we, as a nation, have a problem and each one is involved with their respective responsibility
to achieve a solution;
10. Finally, it was noted that a situation where the community has veto rights may favour the
dialogue and the cooperation amongst all involved parties.
89

COMMUNITY RESEARCH in RADIOACTIVE WASTE MANAGEMENT – Partitioning
and transmutation and geological disposal: 6 th Euratom Framework Programme for nuclear
research and training activities (2002-2006)
Introductory keynote:
The Euratom Research and Training Programme in Radioactive Waste Management
Summary
Simon Webster
European Commission, Brussels, Belgium 3
EU support for R&D in nuclear science and technology is channelled principally through
multi-annual Euratom research Framework Programmes (FPs) covering all areas linked to the
peaceful uses of nuclear energy. Radioactive waste management, including both partitioning
& transmutation (P&T) and geological disposal, has been a priority area of research for many
years and remains so in the current programme FP7 (2007-2011). Over the years, FPs have
covered the full range of scientific topics, with a progression from more fundamental research
in the first programmes, to more applied R&D and demonstration in later programmes. In
FP6, €90M were earmarked for this research effort, funding primarily a small number of large
integrating projects grouping all research in key domains in the area of geological disposal
(near-field and engineered barriers, far-field, engineering systems and demonstration, performance
assessment) and P&T. Future support from the Euratom programme will seek to
maximise effectiveness by further enhancing co-ordination with respective efforts in EU
Member States. In the case of P&T this involves closer integration with the research on advanced
nuclear systems and fuel cycles. In the case of geological disposal, though important
research and investigation is still required on a number of key issues, future efforts are increasingly
related to local site conditions and linked to the licensing of a particular repository.
1. Introduction – Euratom Research
The European Commission (EC) is responsible for the planning and implementation of the EU research
programme. For more than three decades, the principle method of providing this support has
been the Framework Programme (FP). This is a shared-cost grant-based programme (projects are
partly funded by the participating organisations), each FP having a duration of at least four years.
These programmes are implemented via calls for proposals published at regular intervals in the
EU’s Official Journal, with the proposals being evaluated by independent experts.
Ever since the start of European integration back in the 1950s there has been a separate Treaty covering
nuclear issues. The Euratom Treaty [1], short for Treaty establishing the European Atomic
Energy Community, was one of the original Treaties of Rome at the inaugural signing in 1957. At
that time, one of the main objectives was to contribute to the formation and development of
3
The views expressed in this paper are those of the author and do not necessarily reflect those of the European Commission
91

Europe's nuclear industries, therefore the Euratom Treaty covers everything that was then considered
important – nuclear safeguards, radiation protection, supply of fissile material, and also research.
The inclusion of research has meant that all EU support for R&D in applied nuclear science
and technology (including radioactive waste, reactor systems – fusion as well as fission, etc.) comes
under a separate legal basis from the rest of EU research, and is covered by its own legally distinct
FP. These Euratom FPs have traditionally been run in parallel with the “non-nuclear” FPs and are
implemented in a similar fashion using the same type of funding instruments, at least as far as research
in nuclear fission and radiation protection is concerned.
Back in the early 80s, during the first FPs, the Euratom Programme accounted for something like
25% of the total EU research spending. Today, this percentage is much less – in FP7, the nuclear
(Euratom) component is some €2.75 billion (over 5 years) compared with some €50.52 billion (over
7 year) for the non-nuclear FP covering the whole of non-Euratom research. The majority of the
Euratom FP7 research budget will be spent on the fusion programme – approx. €2 billion – the rest
is split between the nuclear activities of the EC's own Joint Research Centre (€517 million) and the
programme in fission and radiation protection (€287 million). One of the priority areas of the latter
is, and has been ever since the first FP, radioactive waste management (RWM). Research on geological
disposal of high-level / long-lived waste has always been a part of this effort, though since
FP4 P&T has also been extensively investigated. In the early Euratom programmes, funding was
also available for research on low-level waste management and decommissioning activities, though
these are now considered to have reached a high level of industrial maturity and further such support
at the EU level is no longer necessary.
The research on P&T is examining the potential to reduce the amounts of some of the longest-lived
radionuclides in the most radiotoxic wastes. This involves chemical separation of key radionuclides
followed by nuclear transmutation, either in a sub-critical nuclear reactor coupled to a particle accelerator
or in a critical power reactor. This is a long-term research programme, which is increasingly
being linked with research on advanced reactor systems and associated fuel cycles as part of
the development of the next generation of more sustainable nuclear reactors. Indeed, such research
is fully within the scope of the Sustainable Nuclear Energy Technology Platform (SNE-TP, see Section
2.3) launched in September 2007 and is covered by the strategic research agenda of this platform.
However, it is clear that no matter what the efficiency and effectiveness of these separation
and transmutation processes, there will always be some ultimate waste that must be disposed of by
geological disposal.
2. Supporting Research in RWM
Work in geological disposal initially concentrated on more fundamental aspects of the physical,
chemical and geological processes affecting deep disposal. Projects tended to be smaller and there
was less emphasis on technology and engineering. With the construction of dedicated underground
research laboratories (URLs) in the various national host rock environments, research projects have
become more focussed on the specific conditions prevailing underground, the engineered barriers,
the required engineering systems and associated demonstration experiments and the overall performance
assessment. Since there are only relatively few URLs in Europe, they have naturally become
magnets for all EU research in host rock conditions, which in turn has resulted in enhanced
co-operation between research teams and waste agencies in different EU Member States. The research
often involves large and costly experiments, again encouraging interested research teams to
combine efforts in order to reduce costs. The key role played by URLs means that they are also
important focal points for Euratom FP funding. Until the end of FP5, a total of c. €200M was dedicated
to geological disposal research through successive FPs. In FP6, a further €90M was commit-
92

ted to support for research on RWM in total, of which approximately half went on geological disposal.
In the case of P&T, Euratom support began with the programmes FP4 & 5, which concentrated on
strategy studies and basic processes and feasibility, developing into fully integrated projects in FP6
to push the techniques to a level where demonstration pilot facilities could be envisaged. Up to
€50M in total was spent on this support during FP4 & 5. In FP6, half of the budget allocated to
RWM, i.e. some € 45 M, was devoted to research other than on geological disposal. Though the
majority went on P&T, a significant amount was spent on research on other techniques to reduce
quantities and radiotoxicity of nuclear waste (optimised fuel management in power reactors and
R&D on the fuel cycle in general, including advanced cycles).
2.1 FP4 and FP5
Geological disposal
By the time of FP5 (1998-2002), all key research areas of importance in the safety case of geological
disposal were under investigation in the Euratom programme Many of the projects followed on
from research supported in FP4, with an important development towards large-scale demonstration
projects performed in URLs. Their objectives were to investigate the feasibility of constructing
Engineered Barrier Systems (EBS) and studying in situ the thermo-hydro-mechanical processes in
the EBS and of the actual rock environment. FP5 also saw the first projects studying societal and
public involvement issues associated with waste management, in particular new ways of dealing
with and communicating risk, and local democracy and governance issues associated principally
with site selection. The inclusion of this topic (since retained in FP6 and also in the scope of FP7)
is indicative of a recognised need to deal with waste management issues on a more holistic rather
than purely technical level. The following headings summarise the principal achievements of FP4
& 5 (refer to [2] and [3] for full details).
Development of repository technology: As a result of several large-scale tests (heating, hydration,
etc.) carried out under real conditions in URLs, repository designers and engineers are now increasingly
able to develop operational plans for repositories that can be refined over the years before
construction work begins. Much has also been learned that will be useful when considering how to
monitor real repositories through their operational lives and beyond, which includes both the technical
and societal requirements for such activities – a truly cross-cutting issue where there are diverse
views on what is needed and how and when monitoring data could be used. This is closely
linked to the whole issue of “phasing” repository development so as to establish a broad consensus
of confidence before moving from one step of the disposal process to the next, as well as the issue
of retrievability, which is increasingly being incorporated as a requirement to be studied in national
programmes.
Long-term behaviour of waste forms and containers: Spent fuel and HLW are extremely stable
materials and will remain so in the deep, stable environment provided by geological disposal. As a
result of many years of study, their properties and behaviour are well understood. There are a range
of suitable metals in which to contain these wastes, enabling them to be safely deposited and isolated
inside the buffer and the rock of a deep repository.
Groundwater and radionuclide movement around repositories: Overall understanding of the
most important aspects of radionuclide chemistry in natural waters continues to improve and important
gaps in this knowledge, where there have been unanswered questions for many years, are being
93

filled – one key example being colloid behaviour. In the future, new analytical techniques, such as
for measurement of the Redox conditions in site characterisation, will provide ever-increasingly
detailed and more reliable information, thereby enabling refinement in the models and in the understanding
of processes. However, in the meantime, the safety assessment modellers are generally
satisfied with the adequacy with which they can represent chemical transport aspects.
Safety assessment of geological disposal: Safety assessment is an established, everyday activity.
Today, it takes a practical and conservative approach by overestimating impacts, but future developments
will allow it to become more realistic. In particular, these developments will enable a better
treatment of uncertainties and refinement of the coupling of the various processes involved in
geological disposal of high-level waste.
Public Involvement in repository programmes: Controversial construction projects will always
be affected by the NIMBY (“not in my backyard”) syndrome – none more so than radioactive waste
facilities. Experience has taught the implementing organisations and the proponents in general that
gaining the trust of the local populations is essential, and this can only be achieved via constructive
dialogue, transparency, and involvement and empowerment of local communities in the decisionmaking
process. However, this remains a very complex socio-political issue, which continues to be
linked, especially by opposition groups, to the continued use of nuclear power.
Partitioning & Transmutation
Strategy studies: Analysis of fuel cycle strategies and the balancing of production and consumption
of waste, including the type of research that would be needed to bring the techniques up to a
level of industrial application and the likely timescales involved.
Partitioning: Development of efficient processes for the chemical separation of long-lived radionuclides.
The aim has been to develop detailed flow-sheets for the separation of desired longlived
isotopes using chemical processes. The synthesis of innovative and selective organic extractants
has also been studied with a view to directly extracting minor actinides from high-level radioactive
waste.
Transmutation: Research aimed at gathering all scientific and technical data necessary to carry out
preliminary design studies of an accelerator driven sub-critical demonstrator reactor for transmutation
of the most hazardous radionuclides. In particular, transmutation using a sub-critical accelerator-driven
system (ADS) has been investigated. The practicability of such a transmuter on an industrial
scale requires the operation of an experimental ADS device, and preliminary design studies
of an experimental ADS have addressed critical points of the whole system, such as the proton accelerator,
neutron-producing spallation target unit, cooling systems, reactor housing of the subcritical
core etc. In addition, future R&D needs were identified, safety and licensing issues defined,
costs assessed and a timetable for realisation outlined.
2.2 FP6 – the new funding instruments
In FP6 (2002-2006) the EC introduced new funding instruments to enable a more efficient and effective
structuring of EU research, to reduce fragmentation and to promote European centres of excellence
and mobility of researchers, all in line with the objectives of the European Research Area
(ERA). To attract EU support, research groups were encouraged to join forces in collaborative
partnerships called Networks of Excellence (NoE) and Integrated Projects (IP). A NoE is a means
to promote sustainable integration of key research organisations in a given field. An IP on the other
94

hand focuses more on the product rather than the process, bringing together key research players in
an ambitious project to go beyond the current state of the art in a particular field. Both instruments
stress the need for training of researchers and required specific training programmes to be established
within the projects.
On a purely technical level, the aims of research in RWM, as stated in the Council Decision establishing
FP6 Euratom, were the establishing of a "sound technical basis for demonstrating the safety
of disposing spent fuel and long lived radioactive wastes in geological formations", and "to determine
practical ways of reducing the amount and/or hazard of the waste to be disposed of by partitioning
and transmutation and to explore the potential of concepts for nuclear energy to produce
less waste". However, on a more strategic level, the Euratom programme also actively encourages
more cooperation between research bodies in Europe. This is illustrated in Table 1, which shows
the trends and programme emphasis in recent FPs in the area of geological disposal (similar trends
are evident in P&T).
Table 1. The changing emphasis in geological disposal – comparison of the last four FPs
Framework Programme
Total Euratom
contribution
No. of
projects
95
No. of projects aimed at coordination
and networking
FP4 (1994-1998) €33.5 M 42 2 RS
FP5 (1998-2002) €29 M 43 10 RT, RS
Programme emphasis
1
FP6 (2002-2006) €45 M 17 all major projects I&N, RT, RS
FP7 (2007-2011) - - all major projects I&N, PA, LI
1 RS = repository system behaviour (near-field / far-field basic phenomena)
RT = repository technology / URLs
I&N = integration and networking
PA + LI = performance assessment & licensing issues
The introduction of the new funding instruments in FP6 was an opportunity to improve still further
the degree of collaboration between research players in both geological disposal and P&T. During
FP6, four major IPs (totalling some €27M of EU funding) were launched in the field of geological
disposal, and a further two major IPs (totalling some €29M) were launched in the area of P&T.
These IPs cover all the principal thematic areas listed in 2.1 above: engineering and repository design;
near-field behaviour; far-field studies; performance and safety assessment; partitioning; transmutation.
In addition, a major cross-cutting NoE was funded in the area of actinide science. Project
details and descriptions are provided in Tables 2 & 3.
Though there was initial reticence on the part of the research community to go along this route of
large multi-partner projects, and indeed the added administrative burden has occasionally been considerable,
there is nonetheless consensus regarding the overall benefits, especially from the point of
view of increased networking, integration of practices and results and the development of an harmonised
EU vision on the key issues. These projects not only pushed back the frontiers of knowledge
in Europe, but also greatly enhanced the effectiveness and the efficiency of the overall European
research effort in this field. Both these aspects must be capitalised upon during FP7.

Table 2. FP6 Integrated Projects and Networks of Excellence in RWM
Project title & description 1 Instrument Coordinating
Number of
consortium
organisation
partners 2
NF-PRO – Understanding and
EU contribution
/
total cost
Start date &
duration
physical and numerical modelling
of the key processes in the nearfield
and their coupling for different
host rocks & repository strategies.
www.nf-pro.org
IP
SCK.CEN
(B)
40 (10)
€8M /
€16.8M
1/1/04
4 years
ESDRED – Engineering Studies
and Demonstrations of Repository
Designs. www.esdred.info
IP
ANDRA
(FR)
13 (9)
€7.32M /
€18.1M
1/2/04
5 years
FUNMIG – Fundamental processes
of radionuclide migration.
www.funmig.com
PAMINA – Performance Assess-
IP
FZK-INE
(DE)
51 (15)
€8M /
€15M
1/1/05
4 years
ment Methodologies in Application
to Guide the Development of the
Safety Case. www.ip-pamina.eu
IP GSF (DE) 25 (10)
€4M /
€7.62M
1/10/06
38 months
EUROPART – European research
program for the partitioning of minor
actinides and some long-lived
fission products from high active
wastes issuing the reprocessing of
spent nuclear fuels. www.europart-
project.org/
EUROTRANS – European Research
Programme for the Transmutation
of High-Level Waste in
an Accelerator Driven System.
http://nuklearserver.fzk.de/eurotrans/Start.html
ACTINET-6 – Network for Actinide
Sciences
www.actinet-network.org
IP CEA (FR) 27 (11)
IP
FZK- NUK-
LEAR (DE)
96
32 (15)
NoE CEA (FR) 27 (13)
€6M /
€10.3M
€23M /
€43M
€6.35M /
€10.5M
1/1/04
3-4 years
(work continues
in FP7 project
ACSEPT)
1/4/05
5 years
1/3/04
4-5 years
(work continues
in FP7 project
ACTINET-I3)
1 Refer to http://cordis.europa.eu/fp6-euratom/projects.htm (“management of radioactive waste”)
2 The figures in parentheses indicate number of different European countries represented.
Table 3. Brief descriptions of the major FP6 projects
NF-PRO has been investigating dominant processes and their couplings affecting the isolation of nuclear
waste within the near-field. It has been applying and developing conceptual and mathematical models for
predicting the source-term release of radionuclides from the near-field to the far-field. Results and conclusions
of experimental and modelling work are being integrated in performance assessments. To understand
the performance of the overall near-field system, an adequate insight in both the performance of the individual
near-field sub-systems and their interactions is essential and this constitutes the core of the integration
component of the project. The consortium of 40 partners represents 7 European waste management
agencies, 25 research institutions and 8 Universities.

ESDRED has the overall objective of demonstrating the technical feasibility of deep disposal on an industrial
scale, especially as regards the activities required during construction, operation and closure of a deep
geological repository. The project will also show how these activities comply with requirements regarding
long-term safety, operational safety, safeguards and monitoring, and is a joint research effort by the major
European radioactive waste management agencies (or their subsidiaries).
FUNMIG is a complement to NF-PRO, the main objectives being the fundamental understanding of radionuclide
migration processes in the geosphere, their application to performance assessment and the
communication of the results. An understanding of processes involved in the transport of key radionuclides
and their retardation at the molecular level is fundamental, but this must be scaled up to the dimension
of host rock strata being considered in Europe (clay, granite, salt). The migration processes can then
be studied at scales of interest in performance assessment (PA), and this integration and abstraction to PA
are key issues. The knowledge acquired during the project will be disseminated to the wider scientific
community and other stakeholders by active training and other dedicated knowledge management activities.
A large consortium of research organisations, waste management agencies and universities across
Europe are implementing the project.
PAMINA is the most recent of the large projects and integrates key activities undertaken in previous FPs.
It has the objective of improving and harmonising integrated performance assessment (PA) methodologies
and tools for various disposal concepts of long-lived radioactive waste and spent nuclear fuel in different
deep geological environments. The IP PAMINA aims at providing a sound methodological and scientific
basis for demonstrating the safety of deep geological disposal of such wastes, that will be of value to all
national radioactive waste management programmes, regardless of waste type, repository design, and stage
that has been reached in PA and safety case development. The project is organised into four components
oriented towards research and technological development and one component oriented towards training and
transfer of knowledge.
EUROPART helped to define partitioning methods for the elimination of minor actinides from nuclear
waste flows emanating from the reprocessing of high-burn-up uranium oxide (UOX) or multi-recycled
mixed oxide (MOX) spent fuel. It also defined partitioning methods for the co-recycling of all the actinides
contained in spent nuclear fuels that could arise from future nuclear reactor designs, e.g. hybrid or generation-IV
reactors employing advanced dedicated fuel cycles or ADS (Accelerator Driven System) concepts.
Two principle research techniques were investigated. First, hydrometallurgy that uses the high active
aqueous effluents produced by reprocessing spent fuel using the PUREX process. These liquids contain all
the nuclear wastes, i.e. fission products and minor actinides, and will be processed either using the technique
of solvent extraction or extraction by chromatographic methods. This method will also be considered
for the processing of wastes from future reactor designs. The second technique is pyrometallurgy where
nuclear wastes from the reprocessing of actual or future nuclear spent fuels are dissolved in molten salts
(halides or fluorides) at high temperature (several 100s of degrees centigrade), followed by the separation
of minor actinides using various pyrometallurgical methods (electro-deposition as metals, liquid extraction
using a molten metallic solvent, or selective precipitation as oxides).
97

EUROTRANS focuses on the transmutation (nuclear conversion) of long-lived and/or toxic radionuclides
found in nuclear waste, such as plutonium and the minor actinides, into short-lived or stable elements, and
will initiate work on a European Transmutation Demonstration (ETD) based on the use of accelerator
driven systems. It will carry out a first advanced design of a 50-100 MWth experimental facility to demonstrate
the technical feasibility of transmutation in an Accelerator Driven System (XT-ADS), and produce a
generic conceptual design (several 100 MWth) of a modular European Facility for Industrial Transmutation
(EFIT). Both designs will bear the same fundamental system characteristics in order to allow scalability
between XT-ADS and EFIT. A substantial amount of experimental and theoretical work will be carried
out, including the coupling of an accelerator, a spallation target and a subcritical blanket. Specific work is
devoted to the development of key accelerator components, dedicated fuels, heavy liquid metal (HLM)
technologies and basic nuclear data.
ACTINET-6 is encouraging sustainable integration of European research on the physics and chemistry of
actinides. The goals are, more specifically, to co-ordinate the use of the major actinide research facilities
within the European scientific community, improve human mobility between member institutions (in particular
between academic institutions and national laboratories) and to promote excellence through a process
of selecting R&D projects and support for training activities. ACTINET-6 has a broad participation of
research organisations and academic institutions with expertise in actinides science, as well as effective
links with the user community.
2.3 FP7 and the future
RWM remains a priority research area in the latest Euratom programme – FP7 (2007-2011 [4]).
However, unlike previous Euratom FPs, there are no ring-fenced budgets for the various priority
areas. This is in order to maximise flexibility in programme implementation and ensure that the
limited funds are allocated as effectively as possible. In particular, since some FP6 projects are still
in progress, including in RWM, it was important not to earmark FP7 funding without having a clear
idea of the results from these projects and where there may be a need for continued EU support.
The Euratom FP7 Specific Programme for nuclear research and training activities, approved unanimously
by the EU Member States towards the end of 2006, has this to say on the objective and
scope of the research to be supported in the field of RWM:
“Through implementation-oriented RTD, the activities aim to establish a sound scientific and technical
basis for demonstrating the technologies and safety of disposal of spent fuel and long-lived
radioactive wastes in geological formations, to underpin the development of a common European
view on the main issues related to the management and disposal of waste, and to investigate ways
of reducing the amount and/or hazard of the waste by partitioning and transmutation or other techniques.
Geological disposal: Research and technological development in the field of geological disposal of
high-level and/or long-lived radioactive waste involving engineering studies and demonstration of
repository designs, in-situ characterisation of repository host rocks (in both generic and sitespecific
underground research laboratories), understanding of the repository environment, studies
on relevant processes in the near field (waste form and engineered barriers) and far-field (bedrock
and pathways to the biosphere), development of robust methodologies for performance and safety
assessment and investigation of governance and societal issues related to public acceptance.
Partitioning & Transmutation: RTD in all technical areas of partitioning and transmutation (P&T)
which could be the basis for the development of pilot facilities and demonstration systems for the
98

most advanced partitioning processes and transmutation systems, involving sub-critical and critical
systems, with a view to reducing the volumes and hazard of high-level long-lived radioactive waste
issuing from treatment of spent nuclear fuel. Research will also explore the potential of concepts
that produce less waste in nuclear energy generation, including the more efficient use of fissile material
in existing reactors.”
This, therefore, represents a progression from the objectives of previous programmes. In geological
disposal, the emphasis is now much more on “implementation oriented” research. In the light of the
progress made during FP6, and also in the socio-political arena in a number of countries, these
clearly defined objectives in FP7 represent the next logical phase of support to the development of
geological disposal in Europe. Regarding P&T, research is increasingly integrated into the broad
field of nuclear systems in general, in particular advanced systems such as the generation-IV concepts
and associated fuel cycles. In order to fully achieve the ambitious generation-IV objectives of
increased sustainability through full actinide recycling, many of the techniques investigated in the
area of P&T will need to be fully developed and assimilated with advanced nuclear systems.
Introducing Technology Platforms
A Technology Platform (TP) is a forum bringing together key stakeholders – industry, academia,
research community, national research coordinators, even regulators – in a particular field of research
in order to establish and implement a strategic research agenda (SRA) in this sector. The TP
members decide amongst themselves how best to conduct future research and contribute resources
for the implementation of the SRA. Most importantly, the stakeholders share a common vision regarding
the direction in which the research should go and are willing to collaborate in order to further
the platform’s agenda. A TP belongs to its stakeholders, not to the EC, though the EC can be
instrumental in providing the initial impetus and high-level political support needed at start-up. In
addition, the SRA can be used by the EC to orient the FP calls for proposals in this field, thereby
ensuring the FP remains as effective as possible in this area as well as bringing a significant degree
of support to the platform’s activities.
On 21 September 2007, the first TP in the nuclear field was formally launched in Brussels. The
Sustainable Nuclear Energy TP (SNE-TP [5]) brings together all the principal nuclear research and
industrial stakeholders in Europe in a broad-based TP covering the whole sector of nuclear systems
and safety, including cross-cutting issues such as research infrastructures and education and training.
Industry, including nuclear suppliers, utilities and large users of electricity are all on board, as
are the major research organisations and institutes, academia and the Technical Safety Organisations
(TSOs). The vision document [5], endorsed at a high level by the stakeholders, presents the
prospects for developing nuclear technology from an R&D perspective in the coming years, with a
special emphasis on increased sustainability through the use of fast breeder reactors, cogeneration
of electricity and process heat using very-high temperature reactors, and the continued safe operation
of current light-water reactors. The whole of the fuel cycle (i.e. including recycling and P&T)
is included in the scope of SNE-TP, with the exception of geological disposal. This exclusion is
quite deliberate and reflects the sensitive nature of the disposal issue and the need for the implementing
organisations – the waste management agencies – to keep their distance from any activities
linked to promotion of nuclear technology in order to maximise credibility and trust in the eyes of
the local population at potential host sites. Nonetheless, SNE-TP is ensuring the all-important coordination
of P&T research with that on advanced nuclear systems in general.
A complementary but separate TP is being established in the area of geological disposal. First and
foremost, the main end-users, the waste management agencies, are well defined and there is broad
99

consensus that geological disposal is the only feasible end point for much of today's nuclear waste.
This vision is also shared amongst the other research stakeholders, for example the principal research
institutes, and is reflected in the majority Member States' programmes. There is also a good
degree of co-operation in ongoing projects, and the relatively small number of URLs also promotes
a converging of national research programmes. A TP could also enable an exchange of experiences,
sharing of technology and planning of research tasks of common interest, as well as identifying
issues that are of purely national or bilateral interest.
The important work carried out in the FP5 project NET.EXCEL [6] showed that a greater degree of
integration is possible. A follow-up study – the CARD project [7] – was launched in November
2006 to elaborate further on the possibility of establishing a TP in geological disposal, and reported
back favourably on the idea in March 2008. Work on setting up the platform continues during
2008, piloted by a small executive group led by the Swedish and Finnish radioactive waste management
agencies (SKB and Posiva), and including the German Federal Ministry BMWi and the
French waste agency (ANDRA). Other stakeholders will be involved in the drafting of the allimportant
vision document, which will be opened for public consultation in early 2009 prior to the
official launch of the TP by the end of the year. Any structure to enhance coordination of research
in this field must be capable of handling the different requirements and speeds of the various national
programmes, and platform members should include not only the national waste agencies but
also the major research institutes and the TSOs, who work closely with the regulatory authorities.
3. Conclusions
In 1975, the Euratom programme identified potentially suitable rock formations in Europe, producing
an atlas of hard igneous and metamorphic rocks (granite, gneiss), clay-rich rocks and salt formations
selected on the basis of their stability, low permeability and good containment properties.
Since then, work has been focused on these three geological environments and EU Member States
have progressively developed their own active R&D programmes, supported by Euratom. Basic
R&D, both in the field and laboratory, has been built upon by practical tests and experiments in
specially constructed URLs that have now been operating for more than 20 years. The steps from
concept to implementation will therefore take many decades, and the further operational steps leading
to final closure of these repositories are expected to take at least as long. This slow and cautious
evolution reflects not only the complexity and multidisciplinary nature of the science involved,
but also the need for time-consuming demonstration experiments in host rock environments.
Today, the geological disposal concept is moving towards maturity and implementation.
Therefore, emphasis in FP7 is very much placed on integration of efforts on all remaining key aspects
with an implementation-oriented focus.
Delays in implementing actual repository programmes are now largely a result of the complex
socio-political issues involved and “wait and see” attitudes in some countries. Nowhere are the issues
more crucial than in the actual selecting of repository sites. The first national programmes to
have surmounted this problem are in Finland and Sweden, and both countries should have operating
facilities by 2020. In Finland, mining operations at the actual repository site have been underway
for more than 4 years. Since 2006, a legal framework is in place in France that should allow its national
repository to be in operation by 2025. Other countries have engaged in consultation and review
processes that have led to the selection of phased geological disposal as the reference solution
for the management of their most hazardous radioactive waste.
In the area of P&T, Euratom support has provided a crucial focal point for Member States' national
programmes over the last 10 years, culminating in substantial integrative efforts during FP6. As a
100

esult of the important cross-cutting links with advanced fuel cycles and next generation nuclear
reactors, this research is paving the way for the development of more sustainable nuclear systems.
The Euratom FP6 projects were ambitious undertakings that, through significant investments of
public money, contributed to major technical advances in RWM in general and have led to a restructuring
of the research community in line with the ERA vision. Working together within Euratom
has provided added value by bringing together numerous academic and professional scientists
and a broad range of stakeholders, facilitating networking and the development of a common EU
view. Knowledge management has been a key consideration, and these projects have also contributed
to crucial training needs, effective communication and dissemination of results.
The key to continued success is increased integration of national programmes in all these fields and
enhanced cooperation between stakeholders across the full range of R&D. The TP model is a flexible
mechanism that has already proved its effectiveness in a number of areas, and SNE-TP and the
future TP in geological disposal are crucial initiatives ensuring the coordination of European R&D
across the full scope of RWM R&D activities. They also will enable the EC to target more effectively
the spending of future EU research funds in order to maximise programme added value.
References
[1] The Euratom Treaty: http://eur-lex.europa.eu/en/treaties/index.htm
[2] “Geological Disposal of Radioactive Waste Produced by Nuclear Power… from concept to
implementation”, EUR21224, Office for Official Publications of the European Communities,
2004
(http://ec.europa.eu/research/energy/pdf/waste_disposal_en.pdf )
[3] “EURADWASTE'04 – Radioactive waste management. Community policy and research initiatives”,
Proceedings of the sixth EC Conference, Luxembourg 29 March – 1 April 2004,
EUR21027, Office for Official Publications of the European Communities, 2004
(http://cordis.europa.eu/fp6-euratom/ev_euradwaste04.htm)
[4] Euratom FP7 & SP7 on Nuclear Research (http://cordis.europa.eu/fp7/find-doc_en.html)
[5] Refer to http://www.snetp.eu/
[6] Christer Svemar et al, “NET.EXCEL Thematic Network: Networking for Research on Radioactive
Waste Geological Disposal”, presented at EURADWASTE’04, Session VIII, Luxembourg,
29 Mar – 1 Apr, 2004 http://cordis.europa.eu/fp6-euratom/ev_euradwaste04.htm
[7] Refer to http://cordis.europa.eu/fp6-euratom/projects.htm (click on "management of radioactive
waste"; projects are listed alphabetically)
101

102

SESSION V: Partitioning and transmutation and its impact on geological disposal
Chairman: Dr Ved Bhatnagar, Unit ‘Fission’, European Commission, DG Research
Introduction and objective
Increasingly the buzz of the renaissance of nuclear power is in the air in Europe. This renewed interest
is a manifestation of growing disquiet about carbon pollution, mounting demands for cleaner
air, security, sustainability and diversification of energy supply. However, a lack of more permanent
waste management solutions is one of the most important factors that are hindering the acceptance
of nuclear energy. How does one convince politicians and the public at large that solutions at
hand to dispose of the long-lived waste in underground geological sites are safe? Naturally, the
NIMBY, ‘not-in-my-backyard’, syndrome persists. Perhaps, by partitioning (chemical separation)
and transmutation (radionuclide conversion), it would be possible to extract the energy-bearing
component of waste, reduce its volume (increasing effectively the capacity of repositories) as well
as reduce the long-lived component of waste by transmutation. Such a scheme may contribute to
help alleviate the fear of the public concerning the longevity of the waste to be disposed of. Besides
partitioning processes and transmutation of waste by accelerator-driven systems (ADS), generation
IV safe nuclear reactor concepts that burn waste and produce fuel for further use are poised to contribute
effectively in transmutation activities.
The objective of the session on partitioning and transmutation (P&T) is to instigate a lively debate
about the various options dealing with the waste management for sustainability of nuclear power.
The session will have four invited speakers outlining the P&T activities in the EU together with
scenarios implementing P&T and its impact on geological disposal. The session will be concluded
by a panel discussion entitled Radioactive waste management – burn or bury?
103

104

Summary
An Overview of Partitioning Activities in Europe
J.-P. Glatz, B. Christiansen, R. Malmbeck, E. Mendes,
C. Nourry, D. Serrano-Purroy, P. Soucek
European Commission, JRC, ITU, Karlsruhe, Germany
Sustainable nuclear energy generation includes the requirement to minimize the nuclear waste
produced and thereby the recycling of all actinides. This goal can be achieved with new socalled
grouped separation processes, derived from aqueous techniques or from dry or pyrochemical
partitioning processes. Significant progress was made in recent years for both routes
in the frame of the European research projects PARTNEW, PYROREP and EUROPART, and
since 2008 ACSEPT.
In the frame of the above mentioned European research projects, reprocessing of EBRII type
metallic alloy fuels with 2% of Am and 5% of lanthanides (U60Pu20-
Zr10Am2Nd3.5Y0.5Ce0.5Gd0.5) is being carried out by electrorefining at ITU. An excellent
grouped separation of actinides from lanthanides (An/Ln mass ratio = 2400) had been obtained.
A similar good separation between Am and lanthanides was achieved by CEA in Marcoule using
a liquid fluoride salt – molten Al extraction process.
These results represent the first demonstration of an efficient grouped actinide recovery from
realistic metallic fuels and are therefore an important step in achieving the sustainability goals
of future reactor systems.
1. Introduction
Nuclear energy systems of the future as they were defined by the Generation IV International Forum
(GIF) are supposed to provide a sustainable energy generation for the future; they are supposed:
To meet clean air objectives and promote long-term availability of the systems and effective
fuel utilization
To minimize and manage the nuclear waste produced and notably reduce the long term stewardship
burden in the future, thereby improving protection for the public health and the environment.
To increase the assurance that they are a very unattractive and least desirable route for diversion
or theft of weapons-usable materials.
These requirements induce a changed reprocessing approach compared to the present industrial
process, in the sense that instead of separating and cleaning the fissile materials to a very high degree
and putting everything else to the waste, the long-lived waste constituents should now be recycled.
A “dirty” fuel is accepted including significant changes in the fuel fabrication and handling.
Remote operations will certainly be needed, but at the same time the waste radio toxicity is considerably
reduced.
It is evident that the corresponding fuel cycles will play a central role in trying to achieve these
goals. By creating clean waste streams (containing only the fission products) these technologies can
105

significantly reduce the quantities of long-lived radionuclides consigned to waste and provide an
intrinsic barrier to weapons proliferation.
A new concept based on a grouped separation of actinides is widely discussed in this context.
Achieving this type of separation is of course a real challenge since technologies available today
have been developed to separate actinides from each other.
The new group separation processes can be derived from aqueous or pyrochemical partitioning
processes of minor actinides developed in the frame of partitioning and transmutation (P&T) studies.
Significant progress was made recently for both routes in the frame of the European research
projects PARTNEW, PYROREP, EUROPART [2, 3, 4] and ACSEPT in FP7.
The present paper describes the progress made in Europe in developing the grouped actinide recycling
concept in view of the sustainability goals described above for innovative reactor systems.
In France, the CEA has launched extensive research programs in the ATALANTE facility in Marcoule
to develop the advanced fuel cycles for new generation reactor systems. The concept is to use
all actinides in the energy production process. They are considered as a whole and recycled in a
grouped mode [5]. In this so-called global actinide management (GAM), the actinides are coextracted
in a sequence of chemical reactions (grouped actinide extraction (GANEX)) and immediately
reintroduced in the fuel fabrication process.
The fuels used in the new generation reactors will be significantly different from the commercial
fuels of today. Because of the fuel type and the very high burn-ups reached, pyrometallurgical reprocessing
could be the preferred method. The limited solubility of some of the fuel materials in
acidic aqueous solutions, the possibility to have an integrated irradiation and reprocessing facility
with improved economics and the higher radiation stability of the molten salt media are some of the
arguments in favour of pyro-reprocessing.
3. Results
An efficient and selective recovery of the key elements from the spent nuclear waste is therefore
absolutely essential for a successful sustainable fuel cycle concept. This necessitates that Am and
Cm can be selectively separated from lanthanide fission products, certainly the most difficult and
challenging task in advanced reprocessing of spent nuclear fuel due to the very similar chemical
behaviour of trivalent elements. There are three major reasons to separate actinides from lanthanides:
� neutron poisoning: lanthanides (esp. Sm, Gd, Eu) have very high neutron capture cross sections,
e.g. > 250 000 barn for Gd-157
� material burden: in spent LWR fuels, the lanthanide content is up to 50 times that of Am/Cm
� segregation during fuel fabrication: upon fabrication, lanthanides tend to form separate phases,
which grow under thermal treatment; Am/Cm would also concentrate in these phases
The separation can be derived from aqueous or pyrochemical partitioning processes of MAs. Significant
progress was made recently for both routes in international collaborations in the frame of
the European research projects mentioned above.
3.1 Aqueous Partitioning Schemes
If a so-called double strata concept e.g. proposed by JAERI in the OMEGA (Option Making Extra
Gains from Actinides and Fission Products) project would be adapted, the industrial reprocessing of
commercial LWR fuel with a recycling of U and Pu. In this first stratum based on the well established
PUREX process, an aqueous TBP extraction process could ideally be combined with an ad-
106

vanced aqueous partitioning scheme to make the long-lived radionuclides available for the second
stratum based on fast reactor systems
In aqueous MA partitioning schemes, two main routes are possible, see Fig 1. The optimal strategy
would be of course a process, where MAs are directly extracted from the PUREX raffinate, HLLW.
However at present no extractant, capable of selective and efficient separation of the MAs at high
acidities (>2M HNO3) in a highly radioactive solution containing all FP, among them lanthanide
elements in a mass excess of 20 times compared to MAs could be found. Partitioning of MA involving
co-extraction of Ln and a subsequent separation of the two element groups is therefore the
only viable option at present.
FP
Selective extraction
MA extraction
(org. complexant)
co-extraction of
MA, Ln
MA /Ln
Ln
LWR fuel
Dissolved fuel
PUREX
HLLW
Selective stripping
MA stripping
(aq. complexant)
MA
Am / Cm sep.
Am Cm
Transmutation
U, Pu, (Np)
Ln
107
Selective extraction
high acid
MA extraction
_______ developed
- - - - - future ?
Figure 1: Strategies for the separation of the minor actinides from HLLW
For an efficient recycling scheme, losses of the relevant elements should be as low as possible
(0.2% or less), a compromise between extraction and back extraction has to be made. Therefore the
European research carried out during the last decades has resulted in the development of the combination
of DIAMEX and the SANEX processes [6, 7, 8]. These are based on the co-separation of trivalent
actinides and lanthanides (DIAMEX) by a diamide, followed by the subsequent selective
separation of MAs in the SANEX process.
Among many extractants tested world-wide, the combination of DIAMEX and BTP [9] is shown to
be the best combination for an efficient recovery of MAs from HLLW or transmutation targets.
Diamides do not require feed adjustment, can easily be recycled to the process, do not leave any
residue upon incineration. Concerning the separation of MAs from Ln, BTP has shown to be the
most efficient extractant, giving at the same time the highest separation factor with no feed acidity
adjustment required. Separation factors between MAs and lanthanides up to 80 are reached in a single
stage extraction. These values are considerably improved in a continuous multistage process and
a Am/Cm product containing less than 1% of Ln is obtained. Unfortunately due to a high sensitivity
to hydrolysis and radiolysis, an industrial application of the BTP molecule requires further investigations
to improve these properties.
3.2 Pyro-reprocessing
FP (Ln)

Pyrochemical processes rely on refining techniques in high temperature (500°C-900°C) depending
on the molten salt eutectic used. Typically chloride systems operate at lower temperature compared
to fluoride systems. In nuclear technology, the processes are mainly based on electrorefining or on
extraction from the molten salt phase into liquid metal. The electrometallurgical process was applied
for the first time as an integral part of the system in the Integral Fast Reactor (IFR). Pyrochemical
separation processes for the recovery of uranium and to some extent for plutonium have
been investigated since decades [10] and remains the core process in the present EBR-II Spent Fuel
Treatment Program [11].
The fuels used in the new generation reactors will be significantly different from the commercial
fuels of today. Pyrometallurgical reprocessing could be the preferred method for the advanced oxide
fuels (mixed transuranium, inert matrix or composite), metal fuels or nitride fuels, because of a
limited solubility of some of the fuel materials in acidic aqueous solutions. Other advantages of the
pyro-chemical approach to reprocess advanced fuels, in comparison to hydrochemical techniques,
are a higher compactness of equipment and the possibility to form an integrated system between
irradiation and reprocessing facility, thus reducing considerably transport of nuclear materials. In
addition, the radiation stability of the salt in the pyro-chemical process as compared to the organic
solvent in the hydro-chemical process offers an important advantage when dealing with highly active
spent minor actinide fuel. The fuels will be often irradiated to a very high burn-up and shorter
cooling times would certainly reduce the storage cost.
At the European level in FP5 PYROREP and in FP6 EUROPART projects, research programmes
aimed to increase the level of knowledge in pyrochemistry with respect to process development,
specific waste treatment and confinement. In these projects, the effort was put on basic data acquisition
mainly in molten chloride and on the core process assessment. The focus was mainly on two
promising core processes: electrorefining on solid aluminium in molten chloride and liquid-liquid
reductive extraction in molten fluoride/liquid aluminium. Some original confinement matrices were
also studied for waste conditioning. Finally, integration studies have been initiated in order to assess
and to compare some selected process flowsheets and possibly to redirect R&D programs.
3.2.1 Molten Salt Electrorefining
In support to process development, basic properties of An and some FPs in molten salts (chlorides
and fluorides) and in liquid metal solvents have been extensively studied.
A very important work was done in basic data acquisition in molten chloride media, mainly at ITU
with a comprehensive study of actinides (U, Pu, Np, Am, Cm), lanthanides and some other important
fission products. Thermochemical properties are derived from the electrochemical measurements
and from basic thermodynamic data for instance in the case of Np of NpCl3 and NpCl4 in the
crystal state. It could be demonstrated, that the NpCl3 has a strong non-ideal behaviour in molten
LiCl–KCl eutectic.
In contrast to the IFR concept, where U is deposited on a solid stainless steel cathode and transuranium
actinides on a liquid Cd cathode, the electrorefining processes studied in the frame of the
European projects PYROREP and EUROPART rely on a co-deposition of all actinides on a solid
Al cathode material, because stable actinide alloys are formed; a redissolution of trivalent actinides
can thus be avoided in contrast to e.g. W cathodes. Also the redox potentials on solid cathodes show
a much larger difference in the reduction potential between actinides and lanthanides. Fig. 2 shows
that the reduction potentials for U 3+ , Pu 3+ , Am 3+ , La 3+ and Nd 3+ determined by transient electrochemical
techniques (mainly cyclic voltammetry and chronopotentiometry) on different cathode
materials. On Bi and Cd, the selectivity of the minor actinide recovery seems to be limited due to
the small difference in reduction potentials between actinides and lanthanides,
108

A metallic alloy fuel with the composition U61Pu22Zr10Am2Ln5 (similar to fuels used in the IFR
program) has been reprocessed in a 25-run batch experiment by electrorefining in a LiCl/KCl eutectic
and an excellent grouped separation of actinides from lanthanides (actinide/lanthanide mass ratio
= 2400) was achieved. The results show a high recovery rate of actinides (> 99.9%) and a good
separation from lanthanides (< 1% in the lanthanide product); they represent a first demonstration
of an efficient grouped actinide recovery from realistic metallic fuels.
The above-mentioned processes are all based on metallic fuel materials and require a highly pure
argon atmosphere. Today all commercial reactors are operated with oxide fuels and advanced reactor
systems selected in the GENIV roadmap rely also on oxides as the major fuel option. Pyroreprocessing
as described above, where all actinides are recycled is based on metallic materials,
therefore a head-end reduction step for oxides fuels is needed to convert oxides into metals.
Potential / V vs. Ag/AgCl
-1
-1.2
-1.4
-1.6
-1.8
-2
-2.2
Pu
Am
La
liquid Bi
Pu
Am
La
liquid Cd
Figure 2: Reduction potentials of some actinides and lanthanides on different cathodic materials.
This conversion can be performed chemically, e.g. by reaction with lithium dissolved in LiCl at
650°C. The recovered metal can directly be subjected to electrorefining and the Li2O converted
back to lithium metal by electrowinning. A more elegant method, where the difficult handling of Li
metal and recycling through reconversion from Li2O can be avoided is the so-called direct electroreduction.
The oxide ion produced at the cathode is simultaneously consumed at the anode and
thus the concentration of oxide ion in the bath can be maintained at a low level. A complete reduction
of the actinide elements can be achieved and the subsequent electrorefing to separate actinides
as described in the previous paragraph can be carried out in the same device. At the same time, the
heat generating fission products are removed. The electrochemical reduction process is the more
reliable technique to convert oxides into metal. A process studied by CRIEPI/ITU is schematically
shown in Fig. 3.
109
U
Pu
Am
Nd
La
U
Pu
Am
Nd
La
solid W solid Al

Figure 3: Schematic layout of an electroreduction process developed by CRIEPI/ITU
The fuel is not crushed but loaded as fuel element segments in a cathode basket, e.g. made of Ta.
The anode is made of carbon, the corresponding reactions are:
Cathode: MxOy + 2y e - = xM + yO 2-
Anode: yO 2- - -
+ y/2C = CO2 (g) + 2y e
The molten salt can be either LiCl or CaCl2. In CaCl2 the higher temperature of 1123 K in comparison
to 923 K for LiCl induces a faster diffusion of oxygen ions to the anode. At the same time an
increased initial reaction rate leads to the formation of a thin dense metal layer at the fuel surface
hampering the diffusion of oxygen ions into the salt.
3.2.2 Liquid-liquid reductive extraction in molten fluoride/liquid aluminium
The liquid fluoride salt – liquid metal reductive extraction process is extensively studied by CEA in
the ATALANTE facility in Marcoule [12]. A process has been developed to study the distribution
of actinides and lanthanides in molten fluoride/liquid metal medium. The results obtained with Pu,
Am, Ce and Sm in the (LiF – AlF3)/(Al – Cu) medium revealed excellent capabilities of the system
for separating the actinides from the lanthanides.
110

With a salt composition corresponding to the basic eutectic (LiF – AlF3, 85 – 15 mol %), up to 99
% of Pu and Am could be recovered in a single stage, with separation factors from Ce and Sm exceeding
1000. The distribution coefficients logically decrease as the initial AlF3 concentration increases.
A thermodynamic model has been developed on the basis of the experimental results to
simulate the extraction efficiency versus fluoroacidity. The model clearly reveals the difference in
solvation between divalent and trivalent lanthanides in fluoride media. The experimental results
compared to those obtained without Cu being present in the metallic phase, are summarised in
Table 1.
Table 1 Mass distribution coefficients and separation factors of actinides and lanthanides with and
without Cu in the metallic phase
Al – Cu (78-22 %mole) Al
M DM SAm/M M DM SAm/M
Pu 197 ± 30 0.73 ± 0.21 Pu 273 ± 126 0.78 ± 0.47
Am 144 ± 20 1 Am 213 ± 30 1
Ce 0.142 ± 0.01 1014 ± 213 Cm 185 ± 31 1.15 ± 0.35
Sm 0.062 ± 0.006 2323 ± 488 Ce 0.162 ± 0 .02 1315 ± 289
Eu < 0.013 >11000 Sm 0.044 ± 0.004 4954 ± 1139
La < 0.06 >2400 Eu < 0.03 >7100
La 0.03 7100
The results show that the distribution ratios of Pu and Am have similar high values independent
from the presence of Cu in the metallic phase and in all cases high separation efficiency from lanthanides.
4. Conclusions
Performed in close international collaborations, the research in the European projects PARTNEW,
PYROREP, EUROPART has led to important progress in the field of actinide partitioning:
Considering hydrochemistry technology:
• the processes have reached lab-scale demonstration and are more or less suitable for industrial
implementation
• new large scale facilities for advanced reprocessing demonstrations are needed
• consolidation and optimization are required for some processes
111

Considering pyrochemistry technology:
References
Pyrochemistry is a less mature technology that needs further developments
two reference routes were identified: electrorefining of actinides onto solid aluminium cathode
in molten chloride salts and liquid-liquid reductive extraction in molten fluoride
salts/liquid aluminium
a first demonstration of an efficient grouped actinide recovery for both routes was achieved
Developing supporting analytical techniques in combination with adapted safeguarding
processes is needed
[1] http://www.ne.doe.gov/genIV/neGenIV1.html
[2] H. BOUSSIER et al., Pyrometallurgical Processing Research Programme “PYROREP”. Final
Technical Report. 2003. 143 p. http://www.cordis.lu
[3] C. Madic, M. Lecomte, F. Testard, M.J. Hudson, J.O. Liljenzin, B.Sätmark, M. Ferrano, A.
Facchini, A. Geist, G. Modolo, A.G.Espartero, J. De Mendoza, Proceedings of the International
Conference GLOBAL 2001, September 9-13, 2001, Paris, France.
[4] C. Madic, M.J. Hudson, Proceedings OECD-NEA: 8th. IEM on Actinide and Fission Product
Partitioning and Transmutation, Las Vegas, Nevada, USA 9-11 November 2004
[5] J.-M. Adnet et al., Proceedings of GLOBAL 2005, Tsukuba, Japan, Oct 9-13, 2005, Paper No.
119
[6] C. Madic, M.J. Hudson, J.O. Liljenzin, J.-P.Glatz, R. Nannicini, A. Facchini, Z. Kolarik and
R. Odoj, New partitioning techniques for minor actinides, European report, EUR 19149,
(2000)
[7] C. Madic, F. Testard, M. J. Hudson, J.O.Liljenzin, B. Christiansen, M. Ferrando, A. Facchini,
A. Geist; G. Modolo, G. A.Gonzales-Espartero and J. De Mendoza, "PARTNEW- New Solvent
Extraction Processes for Minor Actinides-Final Report", CEA-report 6066, (2004)
[8] D. Serrano-Purroy, P. Baron, B.Christiansen, R. Malmbeck, C. Sorel and J.-P. Glatz, "Recovery
of minor actinides from HLLW using the DIAMEX process," Radiochimica Acta, 93, 351
(2005)
[9] R. Malmbeck, et al.: Advanced Reprocessing of Irradiated Fuel using the PUREX, DIAMEX
and SANEX –Processes, Proc. Euradwaste'99 - Radioactive Waste Management Strategies
and Issues, Luxembourg, Nov. 15-18 (1999)
[10] J.J. Laidler, et al, Development of pyroreprocessing technology, Prog. Nucl. Energy 31, 1/2,
131-140 (1997)
[11] C. C. McPheeters, R. D. Pierce, and T. P. Mulcahey, Application of the Pyrochemical Process
to Recycle of Actinides from LWR Spent Fuel, Progress in Nuclear Energy, Vol. 31, No. 1/2
(1997) 175-186
[12] J. Lacquement, S. Bourg, H. Boussier, O. Conocar, A. Laplace, M. Lecomte, B. Boullis, J.
Duhamet, A. Grandjean, P. Brossard and D. Warin, Progress of the R&D Program on Pyrochemistry
at CEA Proceedings of GLOBAL 2005 Tsukuba, Japan, Oct 9-13, 2005, Paper No.
153
112

Overview of Activities in Europe Exploring Options for Transmutation
Summary
Dankward Struwe 1 , Joseph Somers 2
1 FZK – PL NUKLEAR, Karlsruhe, Germany
2 JRC - ITU, Karlsruhe, Germany
Different P/T scenarios have been investigated in the Projects RED-IMPACT and PATEROS
and all imply fuel reprocessing and recycling of actinides and possibly fission products in a
fission reactor system. It can be concluded that fast systems are more efficient in destroying
actinides. Therefore, options for transmutation in fast reactors have been investigated in the
EUROTRANS and the ELSY programmes. An objective of analyses of accelerator-driven
sub-critical systems is the development of a plant design named EFIT with Lead cooling using
a fertile free fuel which lead to safety relevant reactivity feedback coefficients being considerably
worse than those of conventional core designs. Therefore, changes in the safety system
architecture of ADS systems have been developed and qualified. Similarly it became necessary
to initiate R&D programmes with regard to the coolants of Lead and Lead/Bismuth
(LBE) within the DEMETRA programme. Additionally a broad development programme has
been set-up to provide a highly reliable accelerator necessary for a successful steady state
power operation of sub-critical systems. For the development of innovative fuels, technologies
and components a test bed providing prototypical conditions becomes necessary. The respective
development work concentrates on the XT-ADS system design. This follow-on design of
the MYRRHA concept is intended to be built in MOL, Belgium as a multi-purpose LBEcooled
irradiation facility with a fast spectrum sub-critical core driven by an accelerator.
1. Introduction
Today, about 2500 tons of spent fuel are produced annually in the EU, containing 25 tons of plutonium,
3.5 tons of the “minor actinides” neptunium, americium, and curium and 3 tons of long-lived
fission products. These radioactive by-products pose a potential hazard which can be assessed by
their radiotoxicity. The radiotoxicity of the fission products dominates during the first 100 years but
decreases to the reference level after about 300 years. The long term radio toxicity, however is
dominated by actinides, mainly plutonium and americium isotopes so that the reference radiotoxicity
level of spent fuel is reached only after more than 100,000 years. Reduction in the radiotoxicity
time span is the basis for the motivation in the quest for options to reduce the volume and the radiotoxicity
of such waste. Various concept design options have been and will be evaluated for transmutation
and/or incineration of the waste.
2. EC co-funded projects for P&T
European society as a whole has expressed a desire for the deployment of innovative processes for
the management of nuclear waste. Therefore, P&T development and the necessary R&D effort were
113

agreed for support through an integrated effort at the European level despite the diversity among the
Member States in terms of the fuel cycle policy. A key criterion for the selection of projects and
activities to be co- funded by the European Union is the “European Added Value”. As shown in
Fig.1 the projects are organised coherently to provide all elements necessary for a better understanding
and control over the various components required for the deployment of P&T at an industrial
scale level in the 2020 to 2040 time scale.
Fig. 1 EC co-funded P&T projects
2.1 The Coordinated Action PATEROS
This Coordinated Action aims at the establishment of a European vision for the deployment of P&T
up to the level of industrial implementation. It should serve as a basis for decisions on advanced
fuel cycles leading to a sustainable nuclear energy and thus is intended to provide input for the
structuring of the corresponding part of the Sustainable Nuclear Energy Technology Platform
(SNE-TP).
Performance characteristics of various options for transmutation systems were evaluated e.g.
Multi recycling in fast reactors (FR) of TRU unloaded from LWRs and, successively, from
FRs.
Reduction of TRU inventory in fuels unloaded from LWRs.
Reduction of MA inventory.
Use of different fuel types and coolants in critical systems as well as the impact of different recycling
(e.g. homogeneous and heterogeneous MA recycling) has been studied [1]. Such strategies
necessitate fuels with variable amounts of MA incorporated in conventional fuels. The results of the
PATEROS programme have been complemented by those obtained in the EISOFAR project for sodium
cooled systems and the ELSY project for lead cooled systems. Though dependent on design
optimisation, the generic transmutation characteristics for different coolants and homogeneous re-
114

cycling fuels do not vary significantly for critical low conversion ratio systems. In general, the introduction
of MA leads to a beneficial effect on the reactivity variation with burn-up but results in
slightly worse reactivity feedback effects concerning the Doppler and the void reactivity. For large
size sodium cooled reactors (SFR) with about 3000 MWt a maximum MA content of 2.5 to 3.5 % is
acceptable from a safety point of view. In the case of a large gas cooled fast reactor (GFR), this upper
limit could become higher i.e. up to about 5 to 7 %. A yet higher MA content (e.g. for U-free
fuels with a MA content > 40 %) can result in a very significant decrease of the delayed neutron
fraction which might result in problems regarding safety. Heterogeneous actinide recycling provides
the possibility to circumvent this disadvantage and - more importantly - to disconnect the minor
actinide cycle from the conventional fuel cycle.
2.2 Fuels for homogeneous recycling
The incorporation of minor actinides into fuels for fast reactors represents an immense challenge.
The reprocessing plants must deliver actinide streams suitable for conversion to solid and further
processing. Traditionally, mixed oxide fast reactor fuels have been prepared from UO2 and PuO2
powders fed from the separated U and Pu streams in the PUREX process enabling flexibility in preparing
fuels with varying Pu enrichment via standard powder metallurgical routes. Introduction of
minor actinides (MA) will immediately require full automation of the entire fabrication process,
including assembly production, transport and storage.
Research on minor actinide bearing fuels is still in its infancy; and major experimental programmes
are needed before such fuels can be licensed for industrial application. The SUPERFACT experiment
[2, 3] represented the first major milestone in this area. A total of eight fuel pins with pair
wise four fuels were manufactured at the JRC-ITU using the sol gel liquid to solid conversion route,
as this method has the distinct advantage that it is nearly dust free, and thereby minimises radiation
dose risk within the facility. The selected compositions represent both homogeneous and heterogeneous
(MA targets) minor actinide recycling in fast reactors (see Table 1). The irradiation was performed
during 360 EFPD (equivalent full power days) in a standard Phenix bundle, within which
the standard MOX fuel operated at linear powers of 430 and 370 at beginning and end of life (BOL
and EOL), respectively.
115

Table 1: Fuels irradiated in the SUPERFACT experiment
Fuel Linear power (BOL) Linear power (EOL) Burn up
(W.cm -1 ) (W.cm -1 ) (at %)
380 325 6,4
(U0,74Pu0,24Am0,02)O1,95
7
(U0,74Pu0,24Np0,02)O1,973 380 325 6,4
(U0,55Np0,45)O1,996 206 283 4,5
(U0,6,Np0,2Am0,2)O1,926 174 273 4,5
The non-destructive examination of the irradiated pins showed a good in-pile behaviour, with much
commonalities between the SUPERFACT and standard MOX pins. The central hole about 1 mm in
diameter was found in the homogeneous pins (see Figure 3), but was not observed for the heterogeneous
pins, where the linear power was lower. The latter high MA content pins exhibited higher
fuel swelling (column length, cladding diameter). Helium release was 4 times higher for the homogeneous
fuels than for standard MOX, and 40 times greater for the pins with high MA content.
Transmutation rates were of the order of 30% in all fuel types, being marginally higher for the homogeneous
modes.
Figure 3: Ceramograph of an irradiated SUPERFACT fuel pellet
2.3 The Integrated Project (IP) EUROTRANS [4]
The strategic objective of the EUROTRANS project is the stepwise approach towards the achievement
of a European Transmutation Demonstration (ETD). A major step in this project is the advanced
design of an approximately 50 to 100 MWth eXperimental facility to demonstrate the technical
feasibility of Transmutation in an Accelerator Driven System (XT-ADS) in a short term, i.e. in
about 10 years. In parallel a generic conceptual plant design with a power of several 100 MWth of a
modular European Facility for Industrial Transmutation (EFIT) will be developed for a potential
realisation in the long-term. The project is structured (see Fig. 4) such that all aspects necessary for
the realisation of the projects objectives are covered by R&D activities.
The major objectives of the six Domains are:
DM0 Management: Management of the IP by the IP Co-ordinator, supported by the Project Office
at FZK
DM1 DESIGN: An advanced design file is being developed for a short-term experimental demonstration
of the technical feasibility of Transmutation in an Accelerator Driven System (XT-ADS).
116

Liquid Pb-Bi (LBE) is used as coolant and for the spallation target. The core is designed with standard
MOX fuel to be loaded with a few minor actinide (MA) fuel assemblies. In parallel, a reference
design for a modular European Facility for Industrial Transmutation (EFIT) is carried out with
a power of up to several 100 MWth, as a basis for a cost estimate and safety studies for an ADSbased
transmutation system. For the EFIT, liquid Pb is used both for the coolant and the spallation
material. A back-up gas cooling option for EFIT will be studied with an inert matrix based MA core
fuel.
DM2 ECATS: With a view to assisting the design of XT-ADS and EFIT, validation experiments on
the coupling of an accelerator, a spallation target and a sub-critical blanket are performed. These
experiments concentrate on the GUINEVERE programme at the SCK/CEN site in Mol and experiments
in the YALINA facility in Byelorussia. Both are essentially fast sub-critical zero power facilities
driven by a continuous wave deuteron beam.
DM3 AFTRA: Design, development and qualification in representative conditions of a U-free fuel
concept for the EFIT is performed while ranking of different fuel concepts such as oxide composites:
(Pu, MA, Zr)O2 or CERCER (Pu, MA)O2+MgO or CERMET (Pu, MA)O2 + Mo according to
their main out-of-pile properties, their in-pile behaviour and their predicted behaviour in normal
and transient operating conditions, and their safety performance in accidental conditions.
DM4 DEMETRA: Heavy Liquid Metal (HLM) technologies and thermal-hydraulics for application
in ADS, and in particular to XT-ADS and EFIT, are improved and assessed. Reference structural
materials are characterised in representative conditions (with and without irradiation environment)
in order to provide the data base needed for design purposes (e.g. fuel cladding, in-vessel components,
primary vessel, instrumentation, target container, beam window).
DM5 NUDATRA: Simulation tools for neutron physics characterisation of ADS transmuter cores
are improved and assessed, including application to shielding designs and the associated fuel cycle.
The activity is essentially focussed on the evaluated nuclear data libraries and reaction models for
materials in transmutation fuels, coolants, spallation targets, internal structures, reactor and accelerator
shielding.
DM0 Management
Project Office
DM3 AFTRA
Fuels
DM1 DESIGN
ETD Design
IP Co-ordinator
DM4 DEMETRA
HLM Technologies
Figure 4: Organisation of the EUROTRANS project in six Domains
The characteristics of both plant designs investigated are listed in Table 2. It can be seen that XT-
ADS represents an advanced design which takes advantage of the significant amount of work already
performed at SCK/CEN Mol, Belgium within the MYRRHA project. In contrast EFIT is to be
seen as a conceptual study on the potential design features of an industrial scale transmuter with
117
DM2 ECATS
Coupling Experiments
EC
DM5 NUDATRA
Nuclear Data

lead cooling (see Fig. 5). Alternatively, gas cooling is investigated as well to enable choices when it
comes to a decision how to proceed in the far term future.
Table 2: Design choices taken for EFIT and XT-ADS
XT-ADS EFIT
Coolant Pb-Bi Pure
Gas)
Lead (backup:
Primary System Integrated Integrated
Power 70 MWth ~ 400 MWth
Core Inlet/Outlet Temp 300°/ 400 o C 400°/ 480 0 C
Target Unit interface Windowless Windowless
Target Unit geometry Off-center Centered
Fuel MOX (except for a few MA (Pu, Am)O2 + MgO (or
Fuel Assemblies)
Mo)
Av. Fuel Power density 700 W/cm³ 150 to 200 W/cm³
Fuel pin spacers Grids Grids
Fuel Assembly type Wrapper Wrapper
FA cross section Hexagonal Hexagonal
Fuel loading Bottom Top
Nominal coolant Forced with mechanical pumps Forced with mechanical
circulation
pumps
Primary coolant circulation
for DHR
Natural circulation Natural circulation
Secondary coolant Low pressure boiling water Superheated water cycle
Accelerator LINAC (power: 2 ~ 5 MW) LINAC (power: ~ 12
MW)
Beam Ingress (1) Top Top
(1) Proton beam specifications of these designs are considered for the so called HPPA (High-Power
Proton Accelerators) with a proton energy ranging from 600 MeV for the XT-ADS with a maximum
bean intensity of up to 4 mA up to the range of 800 to 1000 MeV for the EFIT design with a
maximum beam intensity of up to 20 mA. The recommended proton beam should possess a CWbased
time structure with additional short and well defined (sharp edge) beam interruptions of
about 100 to 200 �s with a repetition frequency in the order of 1 Hz. These beam holes, shutting
down the neutron power source from time to time, should enable continuous and accurate on-line
measurements and monitoring of the reactor sub-criticality. The beam should have a power stability
of 2% and a 10 % beam size stability on the target etc. (see [5]).
The windowless target is the reference solution for the EFIT and the XT-ADS design. In this target
design the proton beam impinges directly on the redirected LBE flow in the lower part of the beam
tube at core centre (spallation zone). Cooling of the target is achieved by forced convection inside
the target unit.
118

Figure 5: Conceptual design of EFIT
The HPPA is presently very actively studied (or even already under construction) for a rather broad
use in fundamental or applied science. Compared to other HPPA applications, many specifications
are similar: 600 MeV final energy with up to 4 mA CW beam for the XT-ADS and 800 to 1000
MeV with up to 20 mA for the EFIT design, 2% beam power stability, 10 % beam size stability on
target, etc. [5]. The reliability requirement, i.e. the number of unwanted “beam-trips”, is rather specific
to ADS and is essentially related to the number of allowable beam trips of duration longer than
1 s, because, if frequently repeated, they can significantly damage the reactor structures, the target
or the fuel, and also decrease the plant availability. Therefore, beam trips in excess of one second
duration should not occur more frequently than five per operation cycle. The chosen strategy to implement
reliability for the accelerator relies on over-design, redundancy and fault-tolerance [6].
This approach requires a highly modular system where the individual components are operated substantially
below their performance limit (“de-rating” principle). In contrast to a cyclotron, a superconducting
linac, with its many repetitive accelerating sections grouped in “cryomodules”, conceptually
supports this reliability strategy.
The proposed reference design for the accelerator, optimized for reliability, is shown in Figure 6
[7]. For the injector, an ECR source with a normal conducting RFQ is used, followed by warm IH-
DTL or/and superconducting CH-DTL structures up to a transition energy still to be optimized
around 20 MeV. Then a fully modular superconducting linac accelerates the beam up to the final
energy.
119

Figure 6: Principle design characteristics of the reference accelerator
2.4 Inert Matrix Fuels (IMF) for transmutation in ADS systems
Fuels developments for the ADS have concentrated on fertile free (inert matrix) fuels for highest
transmutation rates. Some of the tests on these fuels are discussed below.
2.4.1 IMF – MgAl2O4 (EFTTRA T4)
EFFTRA T4 was the first European experiment to investigate the irradiation behaviour of Am in an
IMF fuel [8]. The chosen inert matrix was magnesium aluminate spinel (MgAl2O4) and the Am was
introduced using an infiltration procedure. The resulting fuel consisted of a micro-dispersion of Am
bearing particles, with diameters less than 3 μm. The mass of Am in the pellets was about 10 wt%
which given the density of spinel corresponds to an actinide loading of about 0,2 g.cm -3 The irradiation
was performed in the HFR Petten for 358 EFPD with transmutation and fission of 96 and
28% of the original Am being achieved. Ceramographs of the sample before and after irradiation
(see Fig. 7) indicate a tremendous increase in porosity, with the pores originating at the location of
the Am bearing particles. These bubbles are a result of He build up during irradiation, and to a large
part they contribute to the very large volumetric swelling of the fuel (ca. 20%), which also produced
severe cladding deformation, but without rupture. The main conclusion of the experiment, apart
from eliminating spinel as an IMF, lies in the need to manage effectively He produced in the fuel,
e.g. low density fuels with closed or open porosity, operation at higher temperature to encourage
atomistic He diffusion to the plenum, or ultimately alternative fuel forms (e.g. SPHEREPAC).
2.4.2 IMF - Zirconia
PSI has performed a major investigation on zirconia based fuels. Zirconia has a monoclinic crystal
structure but can be stabilised in the cubic form by the addition of ca. 15 mole% of Y. Furthermore,
all actinides can be incorporated in the cubic structure. Several (Zr,Y,Pu)O2 based fuels, manufactured
by different processes, have been irradiated in the OECD Halden reactor [8], but PIE is not
completed yet. Within the European programme EUROTRANS, two Am based zirconia fuels are
120

eing irradiated in the HELIOS irradiation programme in the HFR Petten (see Table 5). One fuel
contains Pu to increase the temperature. In a parallel experiment (CAMIX), performed by the CEA,
a similar fuel is being irradiated under fast flux conditions in the Phenix reactor [10].
Figure 7: Ceramographs of the EFTTRA T4 pellets before and after irradiation
Table 5: Fuels irradiated in the HELIOS irradiation experiment (begin 2008)
Fuel Composition Am content
(g.cm -3 Pu content
) (g.cm -3 Particle size
) (μm)
1 Am2Zr2O7 + MgO 0,76 - -
2 (Zr,Y,Am)O2 0,76 - -
3 (Zr,Y,Pu,Am)O2 0,76 0,42 -
4 (Zr,Y,Am)O2 + Mo 0,76 - 80-100
5 (Pu,Am)O2 + Mo 0,32 1,28 20-150
2.4.3 IMF Composite Fuels
50 μm
Stabilised zirconia as a matrix has one major disadvantage, namely its thermal conductivity, which
is substantially lower than UO2, and thereby limits the MA loading. Composite fuels, in which actinide
oxide particles are incorporated in a second material of distinctly higher thermal conductivity,
offer a means to overcome this limitation, and also permit other design options, based on the actinide
inclusion size, to improve the fuel performance. The EFFTRA-T4 fuel was the first MA CER-
CER system to be irradiated. Subsequently MgO has also been considered as a host material, especially
as it is readily compatible with the PUREX process for reprocessing. Several experiments are
ongoing in the Phénix reactor. ECRIX used a powder metallurgy to prepare a fuel with a microdispersion
of AmO2-x particles in an MgO matrix. In the CAMIX COCHIX experiment inclusions
of (Zr,Y,Am)O2 are dispersed in an MgO matrix. Here, two different inclusion sizes were selected,
with damage to the matrix being limited for larger particles. Within the FUTURIX FTA irradiation
programme, fuels consisting of (Pu,Am)O2 dispersed in MgO are being irradiated. These irradiation
programmes will be completed shortly. In parallel out of pile investigations, it has been shown that
such MgO based materials exhibit unfavourable evaporation behaviour, which could limit its performance
in transient conditions.
121
50 μm

Table 6: FUTURIX FTA fuels
Fuel name Fuel Composition
Max. linear power T° max. estimated
(W/cm) (°C)
DOE 1 U0,24Pu0,2Am0,03Np0,01Zr0,52 350 940
DOE 2 Pu0,29Am0,07Zr0,64 440 1050
DOE 3 U0,5Pu0,25Am0,15Np0,10N 370 1010
DOE 4 Pu0,23Am0,04Zr0,73N 290 830
ITU 5 Pu0,80Am0,20O2-x + 86 vol%Mo 140 1590
ITU 6 Pu0,23Am0,24Zr0,53O2-x + 60 vol%Mo 130 1510
CEA 7 Pu0,5Am0,5O2-x + 80 vol%MgO 100 1420
CEA 8 Pu0,8Am0,2O2-x +75 vol%MgO 80 1260
A metal matrix (Mo) provides an excellent possibility to obtain composite fuels with highest thermal
conductivity, and ultimately can permit highest actinide loading in the fuel. Mo must be enriched
in 92 Mo to ensure no neutronic penalties, or eventual build up of Tc. Currently, four such fuel
samples are under irradiation in the FUTURIX FTA (Phénix) and HELIOS (HFR Petten) irradiation
programmes.
2.5 Experimental qualification of the on-line monitoring of a sub-critical core configuration
The GUINEVERE project will provide answers to the questions of on-line reactivity monitoring,
sub-criticality determination and establishment of operational procedures for ADS [11]. A continuous
beam is needed for validation of the reactivity monitoring and it is necessary to establish a more
complete data base for a pure lead core. To meet these requirements SCK/CEN modifies the VE-
NUS critical facility located at the Mol site as follows:
Development of a new GENEPI-C accelerator operating in continuous and pulsed mode.
Installation of the accelerator at the VENUS facility and coupling to the new core.
Adaptation of the VENUS facility to host a fast lead core which will be named VENUS-F.
Based on the experience of GENEPI 1-2 development and operation, a new GENEPI-C accelerator
is being developed for continuous and pulsed mode operation. To realise the intended beam trips
the performance of prompt beam interruptions are foreseen with a repetition rate of a fraction of 1
Hz and duration of beam interrupts between a few hundreds �s and a few tens of ms. The necessary
development is carried out at the CNRS laboratories in Grenoble.
A sub-critical core will use two different types of fuel loading in a unique assembly type. It is foreseen
to start the operation of the zero power facility in a critical mode configuration which necessitates
a shut-down system that provides sufficient shut-down reactivity insertion in a time interval
that is sufficiently short to stop the chain reaction before core damage occurs. Moreover the system
has to be intrinsically safe. Therefore, a system has been chosen with the shut-down rods allocated
at the periphery of the fissile fuel assemblies. The rods fall into the core under gravity when receiving
the signal to de-energize electro-magnets which keep the rods out of the core region. Such a
system was already installed at the VENUS facility during the first years of operation.
3. Concluding remarks
122

In the course of the development of new designs for nuclear systems for transmutation it became
obvious that a successful introduction of these systems needs further design optimization. However,
part of the information needed for the design specifications is still uncertain and therefore design
solutions have to consider rather large uncertainties in the design process. For example, if an economically
acceptable solution for an ADS system is envisaged it becomes necessary to launch a
broad R&D program ranging from evaluation of open material property issues to basic features of
the reliable operation of such a complex system. As result of the European research effort around
the EUROTRANS project first answers to the most relevant open questions will be obtained.
The development of MA bearing fuels is still at a very early stage, and will require substantial effort
in the future. A recent road mapping exercise (EISOFAR) has shown quite clearly that a new SFR
deployed in 2020 will necessarily be fuelled with MOX fuel. This reactor together with the XT-
ADS plant can then be used as an instrument to test and qualify fuels bearing MA in homogeneous
or heterogeneous recycling modes. In the meantime the fabrication, critical property determination
and irradiation testing dedicated to specific effects are necessary for the down selection of options.
Furthermore, the development of multidimensional theoretical approaches will also contribute to an
improved understanding of fuel behaviour, and thereby improved means to design essential experiments
along with a reduction in their number and concomitant costs.
References:
[1] C.Fazio, M.Salvatores and W.S.Yang, “Down-selection of partitioning routes and of
transmutation fuels for P/T strategies implementation”, Proc. Int. Conf. GLOBAL 2007,
Boise, Sept. 2007
[2] C. Prunier, F. Boussard, L. Koch, M. Coquerelle, "Some specific aspects of homogeneous
Am and Np based transmutation fuels through the outcomes of the SUPERFACT experiment
in the Phénix reactor", Proc. Global 1993, Seattle, September 1993.
[3] C.T. Walker, G. Nicolaou, "Transmutation of Np and Am in a fast neutron flux: EPMA results
and KORIGEN predictions for the SUPERFACT fuels", J. Nucl. Mater., 218(1995)129
[4] J. Knebel et. Al: “European research programme for the transmutation of high level nuclear
waste in an accelerator driven system – EUROTRANS” Proc. FISA 2006 Luxemburg
(March 2006) EUR 21231
[5] A. C. Mueller: “The PDS-XADS Reference Accelerator” International Workshop on P&T
and ADS Development, October 2003, SCK•CEN Mol, Belgium.
[6] P. Pierini: “ADS Reliability Activities in Europe”, 4th OECD NEA International Workshop
on Utilization and Reliability of HPPA, May 2004, Daejon, Rep. of Korea.
[7] J-L. Biarrotte et al.: “A reference accelerator scheme for ADS applications”, International
Conference on Accelerator Applications, August 2005, Venice, Italy.
[8] R.J.M. Konings, R. Conrad, G. Dasel, B.J. Pijlgroms, J. Somers and E. Toscano, "The EFT-
TRA-T4 experiment on americium transmutation", J. Nucl. Mater., 282(2000)159.
[9] Ch. Hellwig, M. Streit, P. Blair, T. Tverberg, F.C. Klaassen, R.P.C. Schram, F. Vettraino, T.
Yamashita, "Inert matrix fuel behaviour in test irradiations", Journal of Nuclear Materials
352 (2006) 291–299
[10] G.Gaillard-Groléas, F.Sudreau, D.Warin, "PHENIX Irradiation Program on Fuels and Targets
for Transmutation", Proc. Global 2003, New Orleans, December 2003.
[11] P. Baeten et.al.: “The “GUINEVERE“ project at the VENUS facility” Proc. HPPA’05 on
“Utilisation and Reliability of High Power Proton Accelerators” Mol, Belgium (May 2007)
123

124

Impact of partitioning and transmutation on nuclear waste management and the
associated geological repositories
Summary
Enrique M. González-Romero
CIEMAT, Madrid, Spain
Recent Eurobarometers show that radioactive waste management is perceived by the EU27
citizens as the main issue for nuclear energy sustainability. The nuclear research community,
both at EU and worldwide, has performed an intense R&D program during the last 15 years
on advanced fuel cycles, and the associated technologies of partitioning and transmutation
(P&T), to improve this sustainability. In this R&D the optimum utilization of resources and
the reduction of the final wastes to be disposed of have been the main driving motivations.
Reutilization of irradiated fuel in closed fuel cycles is the main option to address simultaneously
both objectives. However there are many ways to combine the available and new technologies
to close the cycle, each of them is described by an implementation scenario. The best
solution might be different for different countries or regions depending on their long term
strategy for nuclear energy deployment and the availability of technologies. Many of these
scenarios had been studied in great detail by international collaborations within the frameworks
of NEA/OCDE, IAEA, the Yucca Mountain and AFCI projects in the USA, of the
OMEGA project in Japan and a large EU program. The EU has financed, within the FP6, two
specific projects to investigate these scenarios, RED-IMPACT and PATEROS, and a number
of other R&D projects to develop the associated technologies.
This paper summarise the rationales behind the possible introduction of these advanced fuel
cycles with P&T, the advantages/difficulties of different types of deployment scenarios, and
the overall potential impact of these technologies on the nuclear waste management and the
associated geological repositories. The technologies of Partitioning and Transmutation and
the detail effects of P&T on the repository performance assessment will be addressed by other
presentations of the same session.
The most recent studies show potential value of P&T as a way to reduce the radiotoxic inventory
at long term and the heat source of the high level waste at short and medium term. This
last potentiality could allow significant enhancement of the final repositories capacity. On the
other hand, the studies show the small effect of these technologies on the dose levels to the
public from the repository, under its normal evolution conditions. Furthermore, the results of
the studies indicate that the new intermediate level waste of the advanced cycles need special
attention, as they could represent and important fraction of the final radiotoxic inventory and
volume to be disposed of in the final repository. Another issue requiring further investigation
is the timeline of new cycle deployment process and the advantages of using regional approaches.
Finally, large efforts are still needed to clarify the social effects and to prepare sufficiently
precise evaluations of the economical effects of the deployment of these advanced
cycles.
125

After a review of the present status, PATEROS has prepared a road-map of the relevant R&D
topics that need to be addressed to clarify the feasibility and optimize the potential advantages
of the different fuel cycles minimizing the costs and new difficulties.
1. Introduction
Nuclear energy generates 30% of the electricity of the EU. Its characteristics, which will be further
enhanced in the future generations of reactors, perfectly match the requirements of reduction of
emissions, security of supply and competitivity, defined by the EC in the SET plan as the priorities
for the future energy supply in the EU. Still, different countries of EU27 have very different attitudes
towards the future use of nuclear energy for electricity generation or other uses.
Independently of the political decision of continuation or phase out nuclear energy, all countries
using nuclear energy to generate electricity are facing the question of the final management of its
spent nuclear fuel and other high level radioactive wastes, HLW. The disposal in stable deep geological
formations has been proven to be a technically viable solution to handle the already available
and future spent fuel or HLW. However the Deep Geological Disposal has not been fully implemented
in any country, although Finland and the USA have approved its construction and identified,
at least, one accepted site for the facility. Other countries like Sweden and France are also
close to identifying an emplacement, but most countries have still not selected a site. Indeed, a special
Eurobarometer of 2008 [1] shows that more than 70% of the EU27 population believes that
“There is no safe way of getting rid of high level radioactive waste" and only 43% out of the 79%
that have an opinion, have understood that “Deep underground disposal represents the most appropriate
solution for long-term management of high level radioactive waste". Finally the study shows
that the main fears are “The possible effects on the environment and health”(51%) and “The risk of
radioactive leaks while the site is in operation”(30%).
In this framework, many countries have defined their present policy for waste management as the
direct disposal of spent fuel, eventually after some interim storage near surface for 40 to 150 years.
However there are two arguments that are driving large interest on searching viable variants and
alternatives to the direct geological disposal: the long term sustainability of nuclear energy and the
minimization of the long term legacy of hazards for future generations.
From the sustainability point of view, the spent fuel of the present Light Water Reactors, LWR,
contains very valuable materials, typically 95% of the Uranium of the fresh fuel and Plutonium with
as much energy potential as 25% of the fissile part ( 235 U) of the fresh fuel. There are several reactor
concepts, particularly Fast Reactors, but also the new generation of thermal LWR, that are able to
use these components of the spent fuel to generate large amounts of electricity. These reactors will
fission this new fuel to generate energy, transmuting it from long-lived actinides to fission fragments
with largely reduced radiotoxicity and, most of then, of much shorter half-life. Some of these
reactors are also able to use higher actinides (Np, Am and Cm) bringing further the utilization of the
energy potential and the minimization of wastes from the spent fuel. A continuous recycling of the
actinides (U, Pu and minor actinides), that can be implemented combining reprocessing technologies
with some advanced reactor concepts, allows to multiply the amount of energy extracted per
ton of mined Uranium by a factor between 30 and 100. It should be noted that, according to the
most recent estimations [2], the Uranium resources at acceptable exploitation cost might be only
enough for 60 to 200 years, with the technologies of the reactors presently in operation. Recycling
will make these resources sufficient for several thousand years, making these components of the
spent fuel high valuable resources.
In order to reduce the amount of long lived radioactive materials sent to the final disposal a process
of separation and recycling, typical of many other industries, has been proposed. This methodology
126

is generically described as Partitioning and Transmutation, P&T. The first step is to separate, or
partition, the spent fuel into different components according to its final use or disposal requirements.
Different options are considered for combinations but the basic components are:
- The irradiated Uranium, with large mass and volume, low specific radioactivity and thermal
load and large energetic potential;
- The transuranium actinides, very small fraction of the total, very radioactive, with very long
half-lives and high thermal load, high energetic potential and proliferation attractive;
- Some selected short lived fission fragments, the Cs and Sr, that include the isotopes producing
most of the thermal load of the spent fuel for the first hundred years ( 137 Cs and 90 Sr); they
are very radioactive during 300 years but of low activity and mass afterwards;
- Some selected long lived fission fragments, the I and Tc, they represent the largest contribution
to the radiotoxicity at very long term after the actinides and are some of the main responsibles
of the dose at long term from the geological disposal; they are a small fraction of the total
and very low specific radioactivity and thermal load;
- The rest of fission fragments, very radioactive and with relevant thermal loads at very short
times, but after less than hundred years they have low specific radioactivity and thermal load;
- The activated structural materials and other intermediate level wastes, of high volume, low
specific thermal load and an specific activity that can be low or medium depending on technological
choices;
After this partitioning some of these groups will be recycled in normal or advanced reactors (including
subcritical ADS). In these reactors the actinides (U, Np, Pu, Am, Cm, …) undergo fission
becoming fission fragments or converted in other actinides, this is described saying that they had
been transmuted. The spent fuel from this transmutation normally still contains significant amounts
of actinides and it is necessary to repeat the Partitioning and Transmutation steps several times. In
addition, the parasitic transmutation of few selected long-lived fission fragments could be done by
neutron capture in these advanced reactors. The present studies indicate that very large reduction
factors (typically 1/100) can be expected for the actinides inventories.
As a consequence of its potential benefits, large R&D initiatives on P&T had been launched worldwide
with large participation of the EU. The Omega project at Japan; the Accelerator Transmutation
of Waste, ATW, Advanced Accelerator Applications, AAA, and Advanced Fuel Cycle Initiatives,
AFCI, at the USA; a large Partitioning and Transmutation program in the EU, and similar initiatives
in Russia and South Korea, efficiently coordinated in several NEA/OCDE working groups
and IAEA projects, had developed these new concepts. In parallel, the Generation IV [3], GNEP
and other initiatives for the preparation of the medium term future of nuclear energy had incorporated
these concepts in their strategies.
2. Background on Partitioning and Transmutation
According to recent global energy scenario surveys [4][5][6][7][9], the increase in energy demand
from emerging countries, the increase on cost of gas and oil and the need of limiting the emission of
greenhouse effect gases will result in a world nuclear installed capacity equal or higher to the present
park.
A standard light water power reactor of 1 GWe discharges about 23 tons of actinides each year;
cumulatively 900 tons in 40 years lifetime of the reactor. One ton of spent fuel of average burnup of
40 GWd/t contains about 10 kg of Pu and 1.5 kg of the other transuranium elements, mainly neptunium,
americium and curium, which are called the minor actinides (M.A.). These transuranium
elements are not only long-lived radiotoxic substances but also major heat sources which affect the
performance of the repository. It is predicted that without partitioning and transmutation of tran-
127

suranium elements, “repository availability may be the major constrain to nuclear energy” [7]. The
amount of spent nuclear fuel accumulated in Europe, is estimated as 37000 tons for year 2000 with
additional 2500 tons of spent fuel being produced every year [8].
A scientifically proven and technologically available solution for this spent fuel is its disposal in
deep stable geological formations. Many countries including many from the EU have selected in the
last decades the direct disposal of spent fuel as their final solution for these wastes, however none
has still built the Geological Disposal and so no spent fuel had so far been directly disposed. The
spent fuel generated so far are stored in the power plants (spent fuel pools or dry storage containers)
or in centralized interim storage plants. Still in the EU several countries like France, Belgium, the
Netherlands, Germany and UK are or have been reprocessing part of their spent fuel.
The amount of spent fuel and HLW already available and to be produced by the future exploitation
of the nuclear power plants, calls for the evaluation of any possible technological solution that
might minimize the burden of their final disposal. In this contest, a large R&D effort has been developed
in the EU, and other countries, to evaluate and develop a complete set of partitioning and
transmutation solutions. The EU effort has been coordinated around the different R&D framework
programs, FP, with special increase in the number of projects and resources in the FP5 and FP6. A
brief summary of these activities in the FP4 and FP5 can be fount in [11]. These projects cover the
technology development, basic data measurement and evaluation for different processes involved in
the P&T technologies like partitioning technologies (aqueous or pyrochemical), transmutation reactors,
subcritical systems and spallation targets, fuels, structural materials and nuclear data for transmutation
systems. Complementary the global evaluation of feasibility, performance, R&D needs
and implications of the different options for the implementation of P&T in advanced fuel cycles had
been performed in the framework of NEA/OCDE expert groups and WPPT (Working Party on Partitioning
and transmutation) and WPFC (Working Party on scientific issues of advanced Fuel Cycles),
and in IAEA projects. A comprehensive summary of most relevant projects, with detailed review
of antecedents and a very complete list of references can be found in [8][12][13][14][17].
The first comprehensive report with a consensus of the possible performance and role of P&T as a
waste management technology was [13]. This study however only evaluated the modification of
inventories and expected consequences. The implications on the actual waste management and the
potential benefits in the final geological disposal had been studied first in [14] and their evaluation
is the objective of FP6 project RED-IMPACT [10], that is including additional realism on the studies
by including considerations on the impact of intermediate level wastes, ILW, and the transition
process from present situation towards the advanced fuel cycle.
3. Advanced fuel cycle scenarios
There are many possible new fuel cycle elements and combinations that had been proposed to implement
the advanced technologies of P&T, each of these configurations is call a fuel cycle scenario.
In first approximation the critical parameter is the mass flows for the different actinides in the
fuels or targets of each reactor included in the fuel cycle. Every single study has defined a new set
of scenarios depending on the particular topics to be highlighted, however many conclusions are
sufficiently generic to be valid for wide families of similar scenarios. A good set of scenarios, from
[13], that has been taken as starting point in many other studies is shown in Figure 1.
Between these scenarios there are, on one hand, those that transmute all the transuranic elements,
TRU, and on the other hand those that only recycle Pu (2), as in the simple closed cycle. Then we
can identify scenarios that are based in a more or less complex single stratum (3a, 3b and 5), where
the transmutation and the generation of electricity is done in the same reactors, and double strata
scenarios (3c and 4) where the electricity generation is performed in reactors with clean fresh fuel
128

(only U and Pu) and there are a small number of transmutation systems dedicated to the minor actinides
plus any remaining Pu. Another important element is the combination of systems with fast
and thermal neutron spectrum and the selection for the TRU or minor actinide transmutation of
critical reactors or subcritical ADS. Finally there is also discussion about the homogeneous transmutation
of minor actinides within the reactors fuels or their “heterogeneous” loading in specific
transmutation targets (H2). Although not shown in this scheme the selection of the fuel nature and
the reactor, recycling, and fuel fabrication technologies have also important influence on the scenario
feasibility and performance.
Figure 1: Advanced fuel cycle scenarios from [13], including Light water reactors, LWR, fast critical
reactors, FR, and accelerator driven subcritical systems, ADS. The figure indicates the flows of
Plutonium, Pu, Uranium, U, transuranic elements (Np, Pu, Am, Cm,…), TRU, minor actinides (Np,
Am, Cm, …), M.A., or all the actinides (TRU+U), An.
4. Partitioning and Transmutation expected performance
The potential benefits of the P&T technologies depends first of all on the capacity of the different
technologies and scenarios to reduce the inventories of critical components of the final high level
wastes as compared with the spent fuel of the open cycle. This performance was studied in detail at
[13] for the eight different fuel cycles described in Figure 1, and with particular attention to the advantages
or difficulties of using different types of fast spectrum systems, fast critical reactors, FR,
and Accelerator Driven Subcritical systems, ADS.
In the corresponding NEA/OCDE expert group was agreed, by consensus, that the principle of sustainable
development requires the fuel cycle of future nuclear energy systems to be closed for plutonium
as well as minor actinides, in order to ensure the production of fission energy with limited
amounts of natural resources (i.e. uranium) and long-lived radioactive waste. It also requires a safe
and cost-effective nuclear energy production. The resource efficiency and waste reduction goals
together can ultimately only be reached by the introduction of advanced reactor systems with a sig-
129

nificant fraction of fast reactors. For economical and deployment reasons, however, a massive substitution
of existing LWR-based by such advanced reactor and fuel cycle technology is not a realistic
near-term scenario.
P&T which could address the high-level radioactive waste issue at mid-term and prepare the ground
for a more resource-efficient nuclear energy system in the future, may become an attractive and appropriate
intermediate strategy on the way to the ultimate goal of the sustainable nuclear energy
system. In this context, the accelerator-driven system (ADS) can play an interesting role as a minor
actinide or transuranics burner.
The principal conclusions from this study that provide arguments to evaluate P&T added value and
which could influence P&T policy development are:
- While P&T will not replace the need for appropriate geological disposal of high-level waste,
the study has confirmed that different transmutation strategies could significantly reduce, i.e. a
hundred-fold, the long-term radiotoxicity, Figure 2, of the waste and thus improve the environmental
friendliness of the nuclear energy option. In that respect, P&T could contribute to a
sustainable nuclear energy system.
- Very effective fuel cycle strategies, including both fast spectrum transmutation systems (FR
and/or ADS) and multiple recycling with very low losses, would be required to achieve this
objective.
- Fully closed fuel cycles may be achieved with a relatively limited increase in electricity cost
of about 10 to 20%, compared with the LWR once-through fuel cycle.
- The deployment of these transmutation schemes need long lead-times for the development of
the necessary technology as well as making these technologies more cost-effective. The full
potential of a transmutation system can be exploited only if the system is utilized for a minimum
time period of about a hundred years.
- Further R&D on fuels, recycle, reactor and accelerator technologies would be needed to deploy
P&T. The incorporation of transmutation systems would probably occur incrementally
and differently according to national situations and policies.
Figure 2: Evolution of the radiotoxicity of the spent fuel in the open fuel cycle and HLW in advanced
scenarios from [8] and gain on radiotoxicity in advanced fuel cycles from [13].
5. Impact on the HLW final disposal
After the evaluation of the reduction of inventories, it was realized that one of the main consequences
will be the effect on the geological disposal of the HLW. A new NEA/OCDE was per-
130

formed where the previous study was complemented by studying the possible conditioning of these
HLW, and evaluating the effect of the P&T technologies in the HLW final disposal. This calculation
included the evaluation of thermal loads, disposal volume of the repository required for the
HLW and performance assessment with estimation of the dose to the representative person from the
most critical group. The study also improved the economical assessment paying particular attention
to the uncertainties and sensibilities on unknown parameters to provide the appropriated understanding
of the potential financial implications of implementing the different advanced fuel cycle
options. Of particular relevance was the study of complementing the actinide transmutation with the
separation of the short lived fission fragments, Cs and Sr, responsible of most of the thermal load at
the reference disposal time.
The main conclusion from this expert group was that advanced fuel cycles offer possibilities for
various strategic choices on uranium resources and on optimization of waste repository sites and
capacities, while keeping almost constant both the radiological impact of the repositories and the
financial impact of the complete fuel cycle. In this sense it should be possible to design, within acceptable
costs, innovative nuclear reactor cycles, which at the same time spare resources and make
the most efficient use of the foreseen geological repository sites.
The main conclusions from this expert group relevant to understand the added values of P&T are:
- Since the actinides are mostly recovered, minor actinide management techniques, in fact, always
reduce the total decay heat of the waste. While the heat reduction at normal disposal
time (50 years) is modest (at most 70%), schemes with minor actinide management show a
significant reduction potential at longer times. A tenfold reduction could be achieved by prolonging
the cooling time from 50 to 200 years.
- Separation and temporarily storage of Cs and Sr can reduce the number of vitrified HLW canisters
25-40%. The interim storage time needed varies from 12 to 32 years depending on the
waste loading. Additional cost of Cs/Sr separation and Cs/Sr interim storage results a 5-10%
increase in total cost.
- Removing and sequestering Cs and Sr in a separate area of the repository or another facility,
would allow a further substantial increase in the drift loading of the repository, up to a factor
of 43 in comparison with the direct disposal case for 99.9% removal of Pu, Am, Cs, and Sr.
131
Decay heat (W/TWh-th)
1.E+04
1.E+03
1.E+02
1.E+01
1.E+00
Fission prod.
(all reactors)
PWR-UOX MOX-UE
ADS-TRU
ADS-MA
FR-metal
FR-carbide
FR-MOX1
MOX-Np
1.E-01
1 10 100 1000 10000
Storage time (a)
Figure 3: Thermal load from different scenarios in [14]. Ratios respect to direct disposal in the
open cycle and the long-term evolution of decay heat from actinides in waste
- Generally, the high level radioactive waste arising from the advanced fuel cycle scenarios
generates less heat than the LWR spent fuel. In case of disposal in hard rock, clay and tuff
formations the maximum allowable disposal density is determined by thermal limitations. Especially
in the case of a fully closed fuel cycle, the considerably smaller thermal output of the
PWR-MOX

high level waste at the considered cooling time of 50 years allows a significant reduction in
the total length of disposal galleries required. Separation of Cs and Sr allows further reduction
in the HLW repository size. For example, in the case of disposal in a clay formation the required
length of the HLW disposal galleries is reduced by a factor of 3.5 when comparing the
waste from the fully closed fuel cycle compared to once-through fuel cycle, and by a factor of
9 when Cs and Sr are separated. Extending the cooling time from 50 to 200 years will result in
a reduction of the thermal output of the high level waste and, consequently, of the required
size of the repository. In the case of advanced fuel cycles this reduction is a factor of about 30.
- In the case of disposal in rock salt the heat generation of the disposed waste contributes to a
fast salt creep and void volume reduction. Therefore, a lowering of the thermal output of the
high level waste forms needs an optimization of the waste packages and of the disposal configuration.
- For all considered repositories the maximum dose resulting from the disposal of the high level
waste from the various considered fuel cycles scenario does not significantly change from one
scenario to another.
- All reactor parks including LWRs produce significant amounts of residual uranium. The use
or long-term storage or disposal of this uranium has to be considered as an integral part of the
waste management.
- Large amounts of the LILW-LL are expected from the reprocessing activities. However, the
uncertainties are very large because no real or experimental data on all the secondary waste
flows exist.
- Total electricity costs are dominated by reactor investment costs and, therefore, do not vary
widely between different advanced fuel cycles schemes. Fuel cycle costs vary between cycles
by factor two but are significantly affected by uncertainties on unit costs for advanced technologies
and processes. At present, the uncertainties on unit price and technologies do not allow
to differentiate the cost of electricity in the different fuel cycles concepts, and the only
statement that can be safely done is that the cost of advanced fuel cycles will not exceed the
present once through cycle by more than ~20%.
- The price of natural uranium has a significant effect in LWR cycles and in particular in oncethrough
cycle, whereas the effect is much slighter in cycles involving advanced reactors due
to the low consumption of natural uranium. The portion of waste management costs including
repository in the total electricity production cost is so low that we could ignore uncertainties
in those waste management steps.
Similar conclusions had been reached in a number of studies of the Separations and Transmutation
criteria to improve utilization of a geologic repository applied to a repository similar to the one proposed
at Yucca Mountain [15]. In these studies, the main conclusion is that the repository capacity,
as defined by the different thermal limitations of the repository after the shutdown of the forced
cooling, can be multiplied by a factor of 5 if Pu and Am are recycled, by a factor 40 if Pu, Am, Sr
and Cs are recycled and by a factor 91 if Pu, Am, Cm, Cs and Sr are recycled.
6. Impact on the Deep Geological Disposal
Both NEA/OCDE studies had been continued within the FP6 RED-IMPACT project. This project
has included the evaluation of the intermediate level wastes, ILW, more complete evaluation of the
waste streams including structural materials and fuel impurities activation as well as the activation
of all the specific components of the ADS (spallation target structural materials and coolant activation,
spallation products activation,…). Also the level of details in the Cs & Sr handling scenarios
and the performance assessment, including in this case the handling of the ILW, has been improved.
132

Finally RED-IMPACT has made a large effort to evaluate the transition process from the present
European situation to hypothetical future scenarios.
The RED-IMPACT [10] analysis allows reaching the following conclusions relevant for understanding
the added value of P&T:
- Thermal power is a critical characteristic of HLW as it strongly affects the required repository
volume (gallery length). In scenarios where Pu or M.A. are largely included in the HLW, after
the 50 years cooling time the contributions from actinides dominate the thermal load of the
HLW and consequently the evolution of HLW thermal power is rather slow and takes about
1000 years to reduce by a factor 10 and about 10000 years for a factor 100. On the other hand,
for scenarios with full Pu and M.A. recycling, fission fragments (mainly 90 Sr and 137 Cs and
their descendants) dominate the thermal power for 300 years in the waste streams and so, the
total thermal power for these scenarios is reduced by a factor 10 in only 100 years and by
more than a factor 100 at year 300 after unloading.
- These reductions show the possibility, for scenarios with full Pu and M.A. recycling, of large
gains in the reduction of the thermal load to the repository and on its associated capacity by
delaying the disposal time 100 to 200 years more. Similar reduction on the HLW thermal
power can be gained at shorter times by separating the Sr and Cs from the HLW. In fact, when
Pu and M.A. are recycled, the heat from the actinides plus 1% of these fission fragments, at
the reference disposal time of 50 years, is only 2% of the total. This effect combined with the
minimization of fissile materials, might help reducing the volume of the repository for granite
and clay formations.
- RED-IMPACT has shown that without specific Cs and Sr management, the HLW disposal
gallery length can be reduced in factors that range from 1.5 to 6 for the granite and clay formations.
If Sr is removed and Cs disposal is delayed by 50 additional years (disposal time was
set as 50 years), then an additional factor 4 can be obtained in some of the geological formations,
bringing the reduction in the disposal gallery length to a factor 13. Higher reduction factors
are not excluded if the Cs disposal would be delayed longer.
- RED-IMPACT results show that there is small or no advantage from P&T on the dose to the
average member of the critical group from the normal evolution scenarios of the geological
repository. This is expected because the main component for the dose in these scenarios is
produced by fission fragments and activation products. On the other hand, P&T have a significant
positive reduction on the dose to the reference group in the low probability human intrusion
scenarios. More details can be found in the presentation [20] of this conference.
- On the other hand, RED-IMPACT has identified that the ILW might seriously compromise
the advantages of P&T for the repository in some geological formations and for large reactor
parks. In these conditions, sending these materials to the geological disposals might require
significant space and might generate additional dose that counterbalance the advantages of the
HLW minimization. The main contributors to the ILW maximum doses are the possible impurities
in the cladding and structural materials, particularly 14 N that could be activated to 14 C. If
this dose could become a significant hazard, the use of low activation steels and a stronger
specification of the impurities content could possible limit the problem. In any case, new repository
concepts and, eventually, new legislation might be needed to properly handle the
ILW, without handicapping the P&T advantages. Concepts of dedicated repositories for ILW
at intermediate depth and with appropriated retention buffers are currently being developed by
JNFL [16].
- The time dependent studies have shown the importance of a correct and sufficiently anticipated
policy of P&T to take full benefit of the advantage of these advanced cycles. In particular,
it is important to start early enough the reprocessing to have accumulated sufficient TRU,
133

Pu or M.A. by the time it is needed in the transmutation plants, without requiring strong peaks
in the utilization of the reprocessing and fabrication plants. However, this will result in significant
stocks of Pu and M.A. separated or in Pu/M.A. rich fuels fabricated in advance. The
studies also show that reaching equilibrium composition is a very slow process and that reducing
significantly the waste inventories in scenarios on reduction of nuclear power will require
very long times, although regional cooperation might reduce this periods of time to the lifetime
of the transmutation plants.
- In addition, the transition scenarios have also shown the relevance of waiting till the appropriate
technology is available. For example, the use of standard reprocessing technologies that
will send all M.A. to the HLW glasses will not be acceptable in scenarios of reduction of the
nuclear energy installed power, where the M.A. inventory reduction is one of the main objectives.
If this was done, the inventory of M.A. immobilized in the glasses will limit the possible
reduction on TRU mass to less than a factor 10 instead of the factor 100 reachable if the all
the M.A. are transmuted before their storage. These considerations depend on the scenario,
and this practice will be much more acceptable in the case of continuous or increased nuclear
installed power.
It must be noted that the benefits provided by P&T will not be obtained by free. Indeed if no optimization,
development and special protections are implemented, the risk of proliferation (reduced
from the final repository) will increase in the different steps of the fuel cycle. Similarly the integral
dose to workers in advanced fuel cycles could be larger. In addition, as already pointed before,
there will be more secondary low and intermediate level wastes from the new steps of the advanced
fuel cycles. Finally, depending on their implementation, advanced fuel cycles might require an increase
of the transports of radioactive material, and the operation of interim storages of different
radioactive streams.
All these aspects must be addressed thoroughly by detailed R&D programs. The PATEROS EU
project has prepared a road-map for this R&D to be developed in the EU, to identify, develop and
demonstrate the options and technologies that could allow obtaining the benefits of P&T without
unacceptable risks or costs in the advanced fuel cycles.
7. Specific Rationales and Added Value of P&T for different scenarios of Nuclear Energy
deployment
7.1 Added value of P&T for a sustainable energy production from Nuclear Fission
The long term sustainability of nuclear energy requires from the technical point of view to maintain
or improve the safety, efficiency and economical competitivity, in a continuous way, and to gain
public acceptance and proliferation resistance from the social point of view. In addition, it has two
technical conditions whose relevance increases with the envisaged duration of the nuclear power
exploitation: to find sufficient fuel (at reasonable prices) and to be able to handle the continuous
generation of radioactive wastes, particularly the HLW. P&T, in its generic sense, is a key element
for these two conditions.
The presently identified Uranium resources with the current technology and at the year 2004 nuclear
electricity generation will be exhausted in 85 years [2]. This time can be reduced by the increase
of Uranium demand foreseen from the increase of its deployment in emerging countries or
can be increased if we take into account all the conventional resources. In total a range of 60 to 270
years is the present projection of U availability, with the second figure implying much higher U
prices. P&T provides a new technology to the nuclear fuel cycle, based on of fast nuclear reactors
and recycling of actinides that will make the presently identified Uranium resources worth 2500
years. In addition, this technology will turn the spent fuel from the present reactors from waste into
valuable fuel, extending even further the possible exploitation period of nuclear fission energy.
134

Figure 4: Spent nuclear fuel (metric tons) as a function of time in the USA, for different prospective
studies and scenarios.
Countries committed to the continuous utilization of nuclear fission and with a large park of nuclear
reactors would face a continuous production of radioactive wastes and a continuous grow in the
number of repositories needed. For example, the amount of spent nuclear fuel accumulated in the
USA as a function of time is displayed in Figure 4 for different evolutionary scenarios considered.
As for Europe, the existing figures for year 2000 [8] indicate an accumulated amount of 37000 tons
with additional 2500 tons of spent fuel being produced every year.
Figure 5: Estimated number of geologic repositories in the USA, for the different scenarios of the
cumulative spent fuel in 2100.
The assessment of the number of geological repositories needed to accommodate the spent nuclear
fuel in geological repositories varies with the different scenarios and solutions adopted for the fuel
cycles and with the implemented radioactive waste and fuel management policies. As an example,
Figure 5 provides an estimation of the number of repositories needed in year 2100 in the U.S.A., for
different scenarios and corresponding amounts of cumulative spent nuclear fuel. The reference indicate
that only one repository is need with P&T, whereas as much as 5-22 could be required with the
direct disposal policy, or otherwise that a repository able to receive the spent fuel of 50 years of
open cycle, could accept the HLW of advanced fuel cycles with P&T, corresponding to 250-1000
years.
The reutilization of energy and the extension of the value of fuel resources from less than 100 years
to several thousands can also be achieved by the simple closed fuel cycle with continuous recycling
135

of U and Pu only. P&T could complement this sustainability axis, reducing the final wastes and the
size or number of the repositories at mid and long term.
7.2 Added value of P&T for a country reducing the nuclear reactors park or phasing out nuclear
A country reducing the nuclear reactors park or phasing out nuclear will have to cope with the legacy
of spent fuels, and the need to define a safe and acceptable strategy to manage them, essentially
based on the implementation of a geological repository. P&T offers the potential to reduce the burden
on a geological repository, in terms of waste mass and volume minimization, and in terms of
significant reduction of the heat load and of the potential source of radiotoxicity.
Indeed for a country phasing out nuclear, P&T will allow to reduce the inventory of long lived radioactive
materials by a factor 100 in activity, heat load, radiotoxicity and transuranium actinide
elements mass, by 200-300 years after the final disposal of the reduced HLW. This will bring proliferation
risks from the repository to the fuel cycle and consume all the energetic resources for
electricity production. In this way, the responsibility on the final use of the energy contained in the
transmuted actinides and the prevention of its possible proliferating application will be assumed by
the present generation, which, by sure, has the appropriate technology and know how. The reduction
of long-lived radioactive isotopes will also strongly reduce the consequences of any remote and
very low probability accident that brings people close to the HLW. Indeed, with P&T the HLW radiotoxicity
inventory will be restored to the level of mined uranium in 300-1000 years instead of the
one million years needed for the spent fuel to reach the same level.
In addition, the P&T will allow reducing the thermal load at disposal time and eventually reducing
the required disposal gallery length and the specifications of the final storage site. The very large
uncertainties on the cost of the new technologies do not allow concluding if the significant cost reduction
on the geological disposal will be compensated by the additional cost of the other elements
of the advanced fuel cycles. A reduction on the dose to the public is not to be expected, but it is also
not needed as any, properly designed, Deep Geological Disposal site will bring the dose well within
the allowed limits for general public and, most often, largely under the natural background.
The fastest way to reduce the actinides inventories in the fuel cycle, for a fixed installed power, is to
use inert matrix (or low fertile content) transmutation fuel loaded with all the transuranium elements.
Fast ADS systems allow using such extreme fuels. However if a country decides to phaseout
independently it will find that using P&T by approximately as much time as the nuclear energy
was deployed, the reduction on the transuranium actinides and radiotoxicity inventory will be limited
to something like a factor 4. Otherwise, if the inventories reduction factors have to reach 20-50
the required P&T time might become 2-3 times longer than the operation of present nuclear power
plants [18]. This trade-off can be avoided using a regional approach to implement P&T for a country
reducing the nuclear park.
Several studies within the WPFC/NEA/OCDE have shown that a “regional approach” can be beneficial
to support the deployment of P&T strategies aiming at waste minimization. In fact, to benefit
from the recognized potential of P&T strategies, it is necessary to develop sophisticated technologies
for the fuel cycle and to develop new facilities for fuel reprocessing and fabrication and innovative
reactor systems. It does not seem probable for countries which have a stagnant or are phasing
out nuclear energy, to cope with this major endeavour in isolation. On the other hand, [19] and
similar studies show that multinational or regional solutions involving, at least, one country with
sustained utilization of nuclear energy and closed fuel cycle, allow to reduce the spent fuel and
HLW inventories of countries shutting down or reducing the park, within the present century with
benefits and without jeopardizing the cycle at global scale.
136

7.3 Added value of P&T for a country without installed nuclear energy
High level wastes and their final disposal in a long term repository as well as proliferation risks are
always quoted as critical issues which strongly dominate public opinion. As such, countries without
installed nuclear capacity but in the process to reconsidering their energy policy and domestic energy
mix due to severe concerns on environmental protection and climate changes on one hand, and
security of supply and high energy costs on the other hand are often recalcitrant to put forward and
support nuclear energy as a viable option vis-à-vis public acceptability.
As already stated P&T can provide large reduction of the TRU mass inventory and therefore P&T
development up to full demonstration of feasibility and effectiveness can provide elements to policy
makers in order to successfully propose nuclear energy in the country.
Furthermore, P&T is a challenging R&D field with significant synergies and common areas of development
with other advanced beneficial technologies such as nuclear fusion, radioisotopes production,
nuclear medicine, etc. and leads to spin-offs in the fields of accelerators, spallation sources,
liquid metal technology, etc.. This is also consistent with the European Research Area policy for
synergism among research programs and activities in the EU.
In addition P&T is an exciting area of investigation which can draw new generations and maintain
and develop competence in the currently stagnating field of nuclear energy research; it may be also
seen as a training ground for young researchers. Finally, countries without installed nuclear energy
may have nuclear industries interested in developing new nuclear systems – both critical and subcritical
- which exhibit high performances as far as waste minimization through fuel recycling
and/or TRU transmutation. P&T technologies also prepare industry to contribute in new reactor
development and help to retain and distribute nuclear experience and know-how.
8. Conclusions
Partitioning and Transmutation applies to the nuclear fuel cycle the general principle of most sustainable
industries of classification (partitioning) and recycling (transmutation) of the components
that are useful or dangerous for the population or the environment. With reasonable extrapolation of
the performances of present technologies, P&T will allow largely reducing the long term burden of
the spent fuel and high level waste, and in this way it can contribute to significantly improve its
management.
The main objectives of P&T are the transuranium actinides, Neptunium, Plutonium, Americium and
Curium. P&T can provide a reduction larger than 100 on the mass of these transuranium elements
sent to the final disposal, reaching the same reduction factor on the radiotoxicity for the total of all
the high level wastes after few hundred years. This large reduction on the inventories provides significant
reduction of the consequences of low probability accidents, like human intrusion, and drastically
reduces the potential proliferation interest of the repository. The inventory reduction also
implies that the radiotoxicity reaches the level corresponding to the Uranium mined for the fabrication
of the fuel within 1000 years, whereas the spent fuel in the open cycle will take several times
100000 years to reach the same level.
In order to reach this reduction, very high separation efficiency (99.9% for U and Pu and 99% for
M.A.) is required in the reprocessing technologies. The feasibility of this reduction also requires the
use of fast neutron system for all or, at least, a part of the transmutation systems. Fast critical reactors
or ADS could both be used in this sense. In addition, it is also necessary to find a final use or
final disposal for unused Uranium.
This reduction of the transuranium actinides results, in addition, in a large reduction of the thermal
load of the wastes send to the geological disposal. Indeed, if the recycling is applied to all the tran-
137

suranium elements, the thermal load is largely dominated by the Cs and Sr fission fragments. Their
short half live, 30 years, provides a fast decay of the thermal load send to the repository. If all the
high level wastes are sent to the repository after 50 years cooling time, the gain on thermal loads,
although depends on the details of the fuel cycle, will be modest. However delaying the disposal
time for 100 or 200 years provides, in the case of fully closed cycles with P&T, large reductions
(above a factor 10) on the heat load to the repository. Alternatively, the separation of these two elements
from the high level waste produces immediately the same thermal load reduction factors. In
this second case, a specific interim storage is required for the Cs and Sr.
The thermal load is the critical parameter determining the capacity of granite or clay deep geological
repositories. The reduction of thermal loads from P&T can allow to increase the repository capacity
or to reduce the disposal gallery length needed. If the HLW disposal is delayed by 100 years
or Cs and Sr are separated by at least 100 years and P&T had been applied to all transuranium elements,
the disposal capacity can increase by more than a factor 10, and in optimal conditions it is
possible to reach a factor 50. Even if there is not such optimization, the capacity can increase by
factors from 2 to 6. These gallery length reduction factors can be used to reduce the cost of the repositories
or to reduce the number of sites.
Special attention will be required to the selection of low activation structural materials and very low
level impurities content, and to the optimization of the intermediate level waste management, as
otherwise their increase in the advanced fuel cycles could seriously compromise the advantages of
P&T for the repository in some geological formations and for large reactor parks.
The P&T technologies will not significantly alter the dose to the average public person from the
normal critical groups. Without P&T this parameter is designed to be more than one order of magnitude
below the regulatory limits and the natural radiation background on the surface. This dose is
mainly expected from the fission fragments and activation materials and so P&T will not provide
significant reduction, except if Iodine is transmuted. On the other hand, attention should be put to
avoid that the new ILW do not significantly increase this dose, particularly in the case of large reactor
parks.
Different options are possible to implement P&T integral Fast Reactors, Double Strata with ADS,
phase-out with ADS and others. The best solution for a country or region depend on its plans for the
future use of nuclear energy and the present and future fuel cycle available technologies.
For countries planning for a sustainable use of the nuclear energy from fission, P&T is a key component.
It is needed to extend the potential use of nuclear fuel from less than 200 years to several
thousands and simultaneously reduce the amount of high level radioactive waste per unit of energy
generated, avoiding the need of large number of repositories along the time of exploitation.
P&T is also useful in case of reduction or shut down of nuclear energy, as it allows reducing the
burden for final storage, both from the size and the duration points of view. The collaboration
within countries in a region will reduce the R&D efforts required to implement this technology for
countries reducing nuclear energy exploitation, and could make more affordable the P&T option for
countries reducing the use of nuclear energy.
P&T technologies are attractive also from the point of view of countries considering launching a
new program of nuclear energy. It provides the framework for minimizing the difficulties of the
back end of the fuel cycle and allows the progressive development of nuclear technologies. Indeed,
P&T share several technologies with proposed future generation reactors (spectrum, coolant, fuels,
materials…).
P&T provide interesting potential benefits for different EU countries from the point of view of sustainable
development of nuclear energy and waste minimization, reduction of TRU inventory as
unloaded from LWRs and the reduction of M.A. inventory.
138

P&T require the development of new technologies for many components of the nuclear fuel cycle:
reactor technologies, accelerator technologies for ADS, fuel fabrication, advanced reprocessing
technologies, coolant and material compatibilities. No real show stopper has been found on any of
these fields, but the R&D effort needed is large and a clear planning and intensive efforts are required
to make these technologies industrially deployable by the 2040-2050, when present estimations
indicates they will be required. The next critical step of this R&D is the design and construction
of demonstration plants for an advanced fast system optimized for actinide transmutation, possibly
a subcritical system, an advanced reprocessing facility and an advanced fuel fabrication facility.
References
[1] Attitudes towards radioactive waste (Fieldwork February–March 2008/Publication June 2008).
Special Eurobarometer 297 / Wave 69.1 – TNS Opinion & Social. European Commission.
[2] Uranium 2005. Resources, Production and Demand Nuclear Energy Agency, OECD (2006).
[3] A Technology Roadmap for Generation IV Nuclear Energy Systems. GIF-002-00. U.S. DOE
Nuclear Energy Research Advisory Committee and the Generation IV International Forum
(2002)
[4] World Energy Outlook 2006. International Energy Agency (2006)
[5] World Energy Council (2000)
[6] The Future of Nuclear Power – An Interdisciplinary MIT Study, MIT Report (2003)
[7] Climate Change 2007: Mitigation of Climate Change. Working Group III contribution to the
Intergovernmental Panel on Climate Change Fourth Assessment Report. Summary for Policymakers.
Formally approved at the 9th Session of Working Group III of the IPCC, Bangkok,
Thailand, 30 April – 4 May 2007. IPCC (2007)
[8] European Technical Working Group on ADS, Report (2001)
[9] The Role of Nuclear Power in Europe, World Energy Council (2007)
[10] Red-Impact: Impact of Partitioning, Transmutation and Waste Reduction Technologies on the
Final Nuclear Waste Disposal. W. von Lenza, R. Nabbi and M. Rossbach (Eds.) (2008). FZ
Jülich, Report Vol. 15.
[11] Partitioning and Transmutation: Towards an easing of the nuclear waste management problem.
EUR 19785. EURATOM (2001), and Overview of the EU research projects on partitioning
and transmutation of long-lived radionuclides. EUR 19614. EURATOM (2000)
[12] Overview on the existing studies on P&T/C and related technologies. W. Gudowski et al. Deliverable
1.1 of RED-IMPACT. FP6 contract FI6W-CT-2004-002408.
[13] Accelerator Driven Systems (ADS) and Fast Reactors (FR) in Advanced Nuclear Fuel Cycles
– A Comparative Study, OECD/NEA Report (2002)
[14] Advanced Nuclear Fuel Cycles and Radioactive Waste Management. NEA 5990, Nuclear Energy
Agency, OECD (2006)
[15] Separations and transmutation criteria to improve utilization of a geologic repository. R.A.
Wigeland, et al. NUCLEAR TECHNOLOGY Vol. 154, p 95 (2006)
[16] Development of safety assessment for radioactive waste disposal. T. Shimizu et al. GLOBAL
2007, Boise (USA) September 2007
[17] Physics and Safety of Transmutation Systems, OECD/NEA Report (2006)
[18] TRU transmutation studies for phase-out scenarios based on fast neutron ADS systems E.
Gonzalez et al., AccAPP-ADTTA'01, Reno, Nevada, USA, 11-15 November (2001). Detailed
139

Phase-Out TRU Transmutation Scenarios studies based on Fast Neutron ADS Systems, E.
Gonzalez and M. Embid-Segura. 7th NEA Information Exch. Meeting on Actinide and Fission
Product P&T, Jeju (Republic of Korea), 14-16 October 2002.
[19] Partitioning and Transmutation Potential for Waste Minimization in a Regional Context, Salvatores
M. et al., 8th NEA Information Exch. Meeting on Actinide and Fission Product P&T,
University of Nevada, Las Vegas 9-11 November, 2004.
[20] Impact of Advanced Fuel Cycle Scenarios on Geological Disposal. J. Marivoet et al.. Proc.
Euradwaste '08 Conference (2008).
140

Impact of Advanced Fuel Cycle Scenarios on Geological Disposal
Summary
Jan Marivoet 1 , Miguel Cuñado 2 , Simon Norris 3 , Eef Weetjens 1
1 SCK·CEN, Mol, Belgium
2 Enresa, Madrid, Spain
3 NDA RWMD, Harwell, United Kingdom
The impact of advanced nuclear fuel cycles on radioactive waste management and geological
disposal has been evaluated within the Red-Impact project. Five representative fuel cycles,
which were considered in equilibrium, were identified and the resulting waste volumes and
compositions were estimated. Repository designs for disposal in hard rock and clay formations,
developed by national radioactive waste management agencies for today's waste types,
were used as reference concepts. After a 50 years cooling time, the heat generated in the highlevel
radioactive waste arising from advanced fuel cycles is significantly lower than that in
spent fuel from the present "once through" fuel cycle. This would allow the dimensions of a
geological repository to be comparatively reduced. The impact of advanced fuel cycles on the
radiological consequences in the case of the expected evolution or reference scenario is rather
limited. This is because the maximum dose in this scenario, which is calculated to occur a few
tens of thousands of years after the disposal of the waste and is associated with radionuclide
transport in groundwater, is essentially due to mobile fission and activation products; as geological
disposal systems are very effective at retarding the migration of actinides, the contribution
of the actinides to the effective dose is limited. The associated intermediate-level waste
contains considerable amounts of mobile activation products; these species persist in giving
relatively high post-closure doses. On the other hand, for the variant human intrusion scenario,
calculated doses to a geotechnical worker resulting from inadvertent intrusion into a
high-level waste repository are significantly reduced in the case of advanced fuel cycles, because
of the much lower actinide content of the waste. However, it should be noted that the
realism of human intrusion scenarios, and the weight that could be placed on the associated
outcome, is strongly debatable.
1. Introduction
During the last 15 years, various studies investigating the role and feasibility of partitioning and
transmutation (P&T) within nuclear fuel cycles have been carried out. The European Commission
(EC) has initiated several research projects within its various framework programmes on the possibilities
of applying P&T techniques to reduce the inventories of long-lived isotopes in radioactive
waste [1,2]. The final objective of these EC-funded research projects is to lay the groundwork for
future sustainable nuclear fuel cycle strategies involving transmutation in a dedicated waste-burning
accelerator driven system (ADS) or in future Generation IV fast neutron reactors (FR). The Nuclear
Energy Agency (NEA) has conducted a series of studies on P&T systems [3-5], which focused on a
review of the progress in the separation of long-lived actinides and fission products, the options for
their transmutation, and the possible benefit for the management of the radioactive waste. The International
Atomic Energy Agency (IAEA) has published reports on the implications of P&T on
nuclear fuel cycles and waste management [6,7]. Within the Generation IV International Forum [8],
141

10 nuclear nations (Argentina, Brazil, Canada, France, Japan, Korea, South Africa, Switzerland,
UK and US) and Euratom (European Union) are working together to develop advanced reactor
technologies for commercial deployment in the 2015 to 2030 timeframe.
The main objective of the Red-Impact project was to evaluate the impact of the introduction of advanced
fuel cycles for electricity generation on radioactive waste management, and, more specifically,
on geological disposal. The project started in March 2004 and ended in September 2007. The
project was divided into 6 work packages: (1) waste management and transmutation strategies, (2)
industrial deployment scenarios, (3) assessment of waste streams, (4) waste management and disposal,
(5) economic, environmental and societal assessment, and (6) synthesis and dissemination of
results. The results obtained in the Red-Impact project are summarised in a synthesis report [9]. The
outcomes of work packages 1 to 3 are presented at this conference by E. González [10]. The current
paper gives an overview of the main results obtained within work package 4, which considered
waste disposal in two hard rock formations, two clay formations and one salt formation. However,
complete impact analyses were only done for one repository in granite (hard rock) and for one repository
in clay. Therefore, this paper focuses on results obtained for those two repositories.
1.1 Considered fuel cycles
During the first months of the Red-Impact project, work package 1 identified five representative
fuel cycles, which were considered in equilibrium. These fuel cycles are:
fuel cycle A1: the reference fuel cycle, which is the "once through" cycle based on pressurised
water reactor (PWR) plants with uranium oxide fuel;
fuel cycle A2: fuel cycle based on PWR plants with uranium oxide fuel, the generated Pu is recycled
once as mixed oxide (MOX) fuel;
fuel cycle A3: fuel cycle based on a sodium cooled fast neutron reactor with MOX fuels, in
which Pu is multi-recycled;
fuel cycle B1: fuel cycle based on a sodium cooled fast neutron reactor with MOX fuels, in
which all the actinides are recycled;
fuel cycle B2: fuel cycle based on PWR plants with uranium oxide fuel, the generated Pu is recycled
once as MOX fuel, the minor actinides and the Pu in the spent MOX fuel are recycled in
a fast neutron ADS.
1.2 Considered repository systems
Repository designs for disposal in granite and clay formations developed by Spanish (Enresa) and
Belgian (ONDRAF/NIRAS) national radioactive waste management agencies for today's waste
types were used as reference concepts.
The repository design considered for disposal in granite is a horizontal in-gallery disposal concept,
which is shown in Fig. 1. The repository is assumed to be excavated in a generic granitic formation
somewhere in Spain.
The repository design considered for disposal in clay is described in detail in the SAFIR 2 report
[11]. This reference repository is assumed to be excavated in the Boom Clay formation at the Mol
site. At that site the host formation is about 100 m thick, of which an 80 m thick central zone is very
homogeneous clay. The repository will have a central access facility consisting of at least two vertical
shafts and two transport galleries. The disposal galleries will be excavated perpendicular upon
the transport galleries. The high-level waste (HLW) and spent fuel (SF) canisters will be placed one
after the other or with some spacing between two canisters (to respect temperature limitations) in
142

the centre of the galleries. A bentonite backfill will be placed between canisters and the gallery
walls. Because the Boom Clay is a plastic clay, a concrete lining is required to avoid convergence
of the gallery walls during the operational phase of the repository. As an example, Fig. 2 shows a
scheme of a disposal gallery configuration for uranium oxide SF.
DISPOSAL CONCEPT
• Deep disposal
• Crystalline rock
• Carbon steel waste package
• Horizontal emplacement
• Bentonite buffer
143
Steel liner
Bentonite
blocks
Waste package
Figure 1: Longitudinal section of a disposal gallery (Spanish concept for disposal in granite).
Figure 2: Gallery configuration for uranium oxide spent fuel disposal (SAFIR 2 concept [11] for
disposal in clay).
The functioning of geological disposal systems in granite or clay can be explained in relatively simple
terms with the help of safety functions [11]:
physical containment: a watertight container, also called overpack, is isolating the radioactive
waste from groundwater during the strongly transient initial phase (re-saturation processes, heat

elease, strong radiation, pressure rebuilding, etc.) after repository closure; as long as this safety
function is effective, no release of radionuclides from the waste form can occur;
slow release: after container failure, groundwater comes in contact with the conditioned waste
and degradation of the waste matrix and leaching of radionuclides will start; various physicochemical
processes, such as corrosion resistance of the waste matrix, precipitation, sorption or
co-precipitation will strongly limit radionuclide releases into the buffer;
retardation: radionuclides dissolved in the groundwater in contact with the waste will start to
migrate through the buffer and the host formation; because of the very low hydraulic conductivity
of bentonite and clay of the host formation (in the case of disposal in clay), groundwater
flow in the repository's barriers is about negligible and radionuclide transport will be mainly
diffusive; furthermore, many radionuclides will be sorbed onto minerals of the buffer and the
host formation; retardation delays the releases and drastically limits the amounts of radionuclides
that are released into the environment per unit of time; many radionuclides will have decayed
before reaching an aquifer, from where they can reach the human environment.
2. Methodology
The methodology adopted for analysing the impact of advanced fuel cycle scenarios on geological
disposal consisted of two steps: (1) an analysis of the adaptations needed to accommodate the new
waste streams in the disposal concept, and (2) an assessment of the radiological consequences of the
radioactive waste repository.
It was first verified whether the available repository concepts are applicable for the disposal of the
main high-level and long-lived waste types arising from the advanced fuel cycles. Several waste
characteristics were taken into consideration such as volume, composition, thermal output, gamma
and neutron emission, retrievability aspects, and criticality; it appeared that the MOX SF arising
from fuel cycle A2 is the most difficult-to-handle waste type in the Red-Impact waste inventory.
Variants of the disposal concepts allowing the accommodation of MOX SF have already been developed
by the waste agencies. Consequently, the HLW types arising from the advanced fuel cycles
do not require considerable adaptations to the available repository concepts, but the dimensions of
the disposal galleries can be adjusted to the thermal output of the new heat-generating waste types.
The analyses reported in Section 3.1 will thus focus on the influence of the thermal output of the
HLW and SF on the dimensions of the geological repository.
The second step of the analyses consists in an assessment of the radiological impact of the geological
disposal of SF, HLW, and intermediate-level waste (ILW) arising from the considered fuel cycles.
In a safety assessment of a geological repository a representative set of possible evolution scenarios
is considered. The assessments made in the framework of the Red-Impact project focused on
the radiological consequences in the case of the expected evolution scenario (this scenario is also
called normal evolution or reference scenario, and is associated with radionuclide transport in
groundwater); this scenario assumes that the engineered and natural barriers of the repository system
will function as expected. Additionally, in order to illustrate the impact of the reduction of the
actinide inventory of the waste, we also considered a variant human intrusion scenario - this assumes
that a geotechnical worker handles a core taken from a borehole drilled through the repository
and that the core contains fragments of the disposed waste.
144

3. Results
3.1 Impact of the thermal output of the waste on the repository's dimensions
The minimum lengths of the galleries for disposal of SF and HLW are derived from heat dissipation
calculations by ensuring that the temperature limitations are respected; in the disposal concepts considered
in this study these are that the temperature has to remain below 100 °C in the bentonite
buffer for granite and at the interface between the gallery liner and the host formation for clay. As
the thermal output of the ILW canisters is negligible, the length of the ILW galleries is determined
by the number and size of the disposal containers. An overview of the estimated lengths of the SF
and HLW disposal galleries is given in Table 1.
Table 1: Estimated lengths of the SF and HLW disposal galleries
Fuel cycle scenario
Granite
A1 A2 A3 B1 B2
SF + HLW gallery length (m/TWhe) 8.89 11.12 5.52 3.63 4.49
relative SF + HLW gallery
length
Clay
(-) 1.00 1.25 0.62 0.41 0.51
allowable thermal output (50 a) (W/m) 353 332-376 365 379 379
SF + HLW gallery length (m/TWhe) 5.92 5.74 3.48 1.88 2.89
relative SF + HLW gallery
length
(-) 1.00 0.97 0.59 0.32 0.49
3.2 Evaluation of the radiological impact in the case of the reference scenario
The dose to a member of the reference group in the case of the reference scenario, which is associated
with radionuclide transport in groundwater, was calculated by making a number of simplifying
assumptions. For granite it was assumed that the canisters will fail between 1300 and 10 000 years,
and that waste matrix lifetimes are 72 000 years for HLW, 10 million years for uranium oxide SF
and 1 million years for MOX SF. For clay the canister lifetime was assumed to be 2000 years, and
the waste matrix lifetimes 100 000 years for HLW and 200 000 years for SF. For granite sorption
on the buffer and on minerals of the host formation was taken into account, whereas for clay sorption
on the buffer was neglected. The calculated doses, normalised per produced electricity, are
shown in Figs. 3 and 4 for a repository in granite and clay respectively. The radiotoxicity released
from the repository into the environment was also calculated. Table 2 compares the radiotoxicity
released over a 10 million years period with the initial radiotoxicity in the disposed waste. In most
recent safety cases, a time cut-off of 1 million years is used; however for the purpose of Red-Impact
it was considered more appropriate to consider a longer time scale of 10 million years to illustrate
that also at the very long-term no considerable change in the radiological impact has to be expected
from P&T.
Table 2: Comparison between initial radiotoxicity in the disposed SF and HLW (50 years cooling
prior to disposal) and the radiotoxicity released from the geological repository into the environment
over a 10 million years period.
Fuel cycle
Initial radiotoxicity (50 Released radiotoxicity (10
a)
Ma) Containment factor
(Sv/TWhe) (Sv/TWhe) (-)
145

A1 3.79E+08 270 7.12E-07
A2 3.51E+08 54.4 1.55E-07
A3 1.93E+08 19.9 1.03E-07
B1 8.74E+07 14.8 1.69E-07
B2 1.43E+08 15.1 1.06E-07
Dose (Sv/yr-TWh(e))
1.E-08
1.E-09
1.E-10
1.E-11
1.E-12
1.E-13
Scenario A1
Scenario B2
Scenario A2
1.E-14
1.E+03 1.E+04 1.E+05 1.E+06 1.E+07
Time (years)
146
Scenarios A3 & B1
Scenario A3
Scenarios A3 & B1
Scenario B2
Figure 3: Total doses due to disposal of SF and HLW for the 5 considered fuel cycle scenarios
(disposal in granite).
Dose via river pathway (Sv/year/TWh(e))
10 -10
10 -11
10 -12
10 -13
10 -14
10 -15
10 -16
10 -17
10 2
A1 SF
A2 HLW and MOX SF
A2 ILW
A3 HLW
B1 HLW
A3/B1 ILW
B2 HLW
B2 ILW
10 3
10 4
129 I management 126 Sn peak actinides
10 5
Time after canister failure (years)
Figure 4: Total doses due to disposal of SF, HLW and ILW for the 5 considered fuel cycle scenarios
(disposal in clay).
10 6
10 7

3.3 Evaluation of the radiological impact in the case of the variant human intrusion scenario
In the considered human intrusion scenario it is assumed that a core from exploratory drilling is
subjected to laboratory analysis by a geotechnical worker (although some difficulty might be expected
in successfully coring the waste because of the presence of a thick metallic container). A
number of activities during laboratory analysis of core material can give rise to exposure via inhalation,
ingestion and external irradiation. The core inspection scenario defined by Kelly and Jackson
[12] is used. The calculated doses to a geotechnical worker are shown in Fig. 5. Also shown in this
figure is the dose to a geotechnical worker examining natural uranium ore; the Cigar Lake uranium
ore is used here as a reference, which has an average uranium grade of 8%.
Dose (Sv)
1x10 2
1x10 1
1x10 0
1x10 -1
1x10 -2
1x10 -3
1x10 -4
1x10 -5
10 2
A1 SF
A2 HLW
A2 MOX SF
A3 HLW
B1 HLW
B2 HLW-ADS
B2 HLW-MOX
B2 HLW-UOX
Upper ICRP IL
Lower ICRP IL
Cigar Lake NA
10 3
147
10 4
Time (year)
Figure 5: Doses to a geotechnical worker for the 5 considered fuel cycle scenarios
(IL: intervention level; NA: natural analogue).
4. Discussion
High-level radioactive waste arising from advanced fuel cycles generates significantly less heat
than does the equivalent amount of spent fuel arising from the "once through" fuel cycle. The
smaller thermal output of the waste allows a reduction in the size of the SF and HLW repository
needed per unit produced electricity. For instance, the greatest reduction in needed length of disposal
galleries is observed for fuel cycle B1 (fast reactor with multi-recycling of the actinides), a
reduction with a factor 2.5 (granite) to 3.2 (clay) in comparison with the reference fuel cycle A1.
The reduction factor of the needed surface area of the repository can even be higher than that for the
gallery length, when the distance between two disposal galleries is also optimised, respecting the
minimum distance between two galleries imposed by geomechanical limitations.
The calculated doses in the reference scenario are, for all considered fuel cycles, far below the dose
constraint of 0.3 mSv/a proposed by ICRP [13]. The impact of the application of P&T in a fuel cycle
on the maximum dose resulting from the disposal of the generated spent fuel or HLW is rather
limited, because the maximum dose is essentially due to mobile long-lived fission and activation
products. The amount of generated fission products is about proportional to the heat produced by
10 5
10 6

nuclear fission in the reactors. The different neutron spectra in fast reactors or ADSs can have some
influence on the generated amount of some fission products; this explains the somewhat higher dose
due to 135 Cs at 3 million years for fuel cycles A3 and B1 in the case of disposal in granite. The calculations
indicate that the amount of iodine that is going into the repository, which depends on the
fraction of spent fuel that is reprocessed, strongly influences the maximum dose. The higher disposal
density, which can be considered in the case of disposal of HLW from advanced fuel cycles,
can result in a small decrease in the calculated dose from radionuclides (e.g. Se, Zr, Tc, Pd, Sn) for
which the release is controlled by a solubility limit. Very long-term doses, i.e. after a few million
years for disposal in clay, are calculated to be lower in the case of advanced fuel cycles, because
smaller amounts of actinides are present in the HLW arising from these fuel cycles.
Table 2 shows that the fraction of the disposed radiotoxicity that is released into the environment
over a 10 million years period ranges between 10 -6 and 10 -7 . These values confirm the excellent performance
of a geological disposal system.
On the basis of the analysis reported above, maximum doses associated with future high-level waste
repositories could strongly depend on the applied iodine management. If iodine is captured at future
reprocessing plants and immobilized in matrices for which the stability is comparable to today's
borosilicate glasses, doses due to these iodine-wastes would still dominate total doses resulting
from a repository in which the main long-lived waste types, i.e. SF, HLW, ILW and iodine-waste,
arising from a fuel cycle would be disposed of; the resulting peak doses for fuel cycles with reprocessing
of the spent fuels would even be higher than for the reference fuel cycle A1. As a consequence,
if the iodine were to be captured at the reprocessing plants, it would be necessary to develop
extremely stable waste matrices, possibly in combination with long-lived containers, for conditioning
and packaging the iodine waste.
Although the ILW contains relatively high amounts of mobile fission and activation products such
as 14 C, 36 Cl and 129 I, the maximum ILW dose is, in the case of disposal in clay (see Fig. 4), calculated
to be about one order of magnitude lower than the maximum HLW dose; in the disposal concept
considered, the diffusive transport of radionuclides through the 40-m-thick natural clay barrier
ensures that the releases of mobile radionuclides from the host clay formation into the surrounding
aquifers are spread over several tens of thousands of years. An important radionuclide occurring in
ILW is 14 C, of which considerable amounts were calculated to be generated in a fast reactor (due to
activation of impurities in structural metals) or an ADS (because of nitride fuel). The 14 C peak is
clearly visible on the ILW dose curves for disposal in clay at about 20 000 years. The development
of low activation materials might reduce the generated amount of 14 C. Considering sorption of C on
the cementitious materials used in the near field of the repository or on minerals of the clay host
formation would also reduce the 14 C doses calculated in the evaluations.
Regarding calculated doses in the variant human intrusion scenario, Figure 5 shows that only for 3
HLW types, arising from the advanced fuel cycles B1 and B2, does the dose in the geotechnical
worker scenario reduce to less than the dose associated with an intrusion into a Cigar Lake uranium
ore body within a 1 million years time scale. For the other HLW and SF types the calculated doses
to a geotechnical worker remain (much) higher than the dose associated with an intrusion into the
Cigar Lake ore body. Consideration of the intervention levels of 10 and 100 mSv, which were proposed
by ICRP [13] when human intrusion could lead to doses to those living around the repository
site, can be used to contextualise the dose values calculated for the various SF and HLW types. Applying
a simple criterion of ranking when the calculated dose for each SF and HLW type is bounded
by the ICRP intervention levels allows a ‘ranking’ of the high-level waste and spent fuel types. It
should be noted that the applicability of intervention levels to human intrusion doses is questionable
148

and any related conclusions drawn from these intervention levels should therefore be treated cautiously.
Table 3 gives the estimated 'required isolation times' for the main SF or HLW types arising
from each of the 5 base fuel cycle scenarios. In order to verify whether the radiotoxicity present in
the disposed waste can be used as an alternative indicator, Table 3 also gives the time after which
the radiotoxicity in the disposed waste drops under the radiotoxicity in fresh uranium required for
the fuel production in fuel cycle A1. Comparison of the times obtained in Table 3 from the 100 mSv
intervention level with the times derived from the radiotoxicity shows that these times are often of
the same order of magnitude.
Table 3: Ranking of high-level waste and spent fuel types, on basis of calculated dose from geotechnical
worker scenario in comparison with ICRP intervention levels; for comparison the
time after which the radiotoxicity in the disposed waste drops under the radiotoxicity present
in the fresh uranium required for the fuel production of fuel cycle A1 is also given
Comparator
ICRP 10 mSv ICRP 100 mSv
Fuel cycle
intervention intervention Radiotoxicity
Waste type
level
level
Fuel cycle B1: HLW ~ 40 000 a ~ 1000 a ~ 300 a
Fuel cycle B2:
HLW-UOX
HLW-ADS
~ 1000 a
~ 70 000 a
~ 250 a
~ 13 000 a
~ 300 a
Fuel cycle A3: HLW > 1 Ma ~ 70 000 a ~ 24 000 a
Fuel cycle A1: UOX SF > 1 Ma ~ 100 000 a ~ 200 000 a
Fuel cycle A2:
HLW
MOX SF
~ 100 000 a
> 1 Ma
~ 20 000 a
~ 200 000 a
~ 90 000 a
In case of fuel cycles generating different HLW and SF types, waste types can occur in which a lot
of radioactivity is concentrated, but of which only a very small number of waste packages are generated;
this is the case for the MOX spent fuel in fuel cycle A2 and the vitrified HLW arising from
the pyro-reprocessing of ADS spent fuel in fuel cycle B2. The results given in Table 3 show that the
times obtained for the 100 mSv intervention level for the waste types of which the largest quantities
are generated correspond much better with the times estimated on the basis of the radiotoxicity generated
in the considered fuel cycle. It should be noted that the use of a variant scenario such as inadvertent
human intrusion scenarios in any calculations to assess the post-closure consequences of a
geological repository is contentious. For example, the considered geotechnical worker scenario assumes
that, in spite of the fact that the drill-bit has to perforate a thick-wall metallic container, the
geotechnical workers will not become aware of the presence of very exotic materials in the underground
and will not take precautions when examining the extracted cores. It is important to strike a
balance between quantitative calculations and qualitative arguments when deciding the priority that
should be given to the assessment of the human intrusion pathway in any forward decision-making
process concerning long-term radioactive waste management.
149

5. Conclusions
The HLW arising from advanced fuel cycles generates less heat than the spent fuel arising from the
reference once-through fuel cycle. This would allow the dimensions of the geological repository to
be reduced.
The total doses due to geological disposal of spent fuel, HLW and ILW are dominated by contributions
from mobile fission and activation products (actinides being immobile due to very high sorption
and solubility limitation). As a consequence the transmutation of actinides in fast reactors or
ADS would have little impact on the resulting doses in the case of the considered reference scenario.
On the other hand, the removal or capture of 129 I upon reprocessing could have a much
stronger impact on the maximum dose. Considerable amounts of mobile activation products, such
as 14 C, are calculated to be generated in advanced fuel cycles and, consequently, to give a significant
contribution to the total dose; this observation confirms the importance of developing low activation
fuel matrices and construction materials for future nuclear reactors.
The radiotoxicity in the HLW from advanced fuel cycles, after a few centuries cooling time, is drastically
reduced by the transmutation of the actinides. Although this would lead to a considerable
reduction of the potential hazard in the case of inadvertent human intrusion, the use of this variant
scenario as a decision making tool for long-term radioactive waste management is highly debatable.
6. Acknowledgements
This project has been co-funded by the European Commission and performed as part of the 6 th
Euratom Framework Programme for nuclear research and training activities (2002-2006) under contract
FI6W-CT-2004-002408.
References
[1] V. Bhatnagar, S. Casalta and M. Hugon (2005) Partitioning and Transmutation Research in the
Euratom 5 th and 6 th Framework Programmes. Proc. 8 th Information Exchange Meeting on
P&T, Las Vegas, 9-11 November 2004. Nuclear Energy Agency, Paris.
[2] V. Bhatnagar and G. Van Goethem (2007) Overview of the EU Activities on Partitioning and
Transmutation Research in the Euratom 6 th and 7 th Framework Programmes. Proc. 9 th Information
Exchange Meeting on P&T, Nîmes, 25-29 September 2006. Nuclear Energy Agency,
Paris.
[3] Actinide and Fission Product Partitioning and Transmutation; Status and Assessment Report
(1999) Nuclear Energy Agency, Paris.
[4] Accelerator-driven Systems (ADS) and Fast Reactors (FR) in Advanced Nuclear Fuel Cycles:
A Comparative Study (2002) Nuclear Energy Agency, Paris.
[5] Impact of Advanced Nuclear Fuel Cycle Options on Waste Management Policies (2006) Nuclear
Energy Agency, Paris.
[6] Accelerator Driven Systems: Energy Generation and Transmutation of Nuclear Waste: Status
Report (1998) International Atomic Energy Agency, Vienna, Report TECDOC 985.
[7] Implications of Partitioning and Transmutation in Radioactive Waste Management (2004) International
Atomic Energy Agency, Vienna, Technical Reports Series No. 435.
[8] A Technical Roadmap for Generation IV Nuclear Energy Systems (2003) US Department of
Energy, Nuclear Energy Research Advisory Committee, Washington, Report GIF-002-00.
150

[9] W. von Lenza, R. Nabbi and M. Rossbach (Eds.) (2008) Red-Impact: Impact of Partitioning,
Transmutation and Waste Reduction Technologies on the Final Nuclear Waste Disposal. FZ
Jülich, Report Vol. 15.
[10] E. González (2008) Impact of partitioning and transmutation on nuclear waste management
and the associated geological repositories. Proc. Euradwaste '08 Conference (in preparation).
[11] SAFIR 2: Safety Assessment and Feasibility Interim Report (2001) ONDRAF/NIRAS, Brussels,
Belgium, Report NIROND 2001-06 E.
[12] M.Kelly and C.P. Jackson (2007) User Guide for SHIM Version 5.0. Serco Assurance Report
to UK Nirex, Report SA/ENV-0557.
[13] Radiation Protection Recommendations as Applied to the Disposal of Long-lived Solid Radioactive
Waste (2000) ICRP Publication 81, Annals of the ICRP, 28, 4.
151

152

PANEL DISCUSSION
Summary of the Panel Discussion Concluding
Session V: Partitioning and Transmutation and its impact on geological disposal
Panel members:
Ved Bhatnagar (Chair), European Commission, Brussels
1 Gianluca Benamati, Member of Italian Parliament
2 Bernard Boullis, CEA, France
3 Enrique Gonzalez, CIEMAT, Spain
4 Philippe Lalieux, ONDRAF/NIRAS, Belgium
The objective of the panel discussion 'Radioactive Waste Management – Burn or Bury' was to
instigate a lively debate about the various options dealing with the waste management for sustainability
of nuclear power and to seek synergy between geological disposal and partitioning and
transmutation technologies. The debate became interesting and lively by drawing parallels from the
back-end of the human life-cycle in which the custom of 'burning' and 'burying' of human-waste in
different communities was put forward as a point of animated reference.
After a brief introduction by the chairman the panellists were invited to make a '2-slide introductory
presentation' outlining briefly their views on this topic. Subsequently, the floor was open for discussion
with questions from the audience and answers from the panel members.
Following is a brief summary of the introductory remarks by the panellists:
Mr Benamati, a member of the Italian parliament, pointed out that the European public opinion is
divided 50-50, for and against nuclear energy. Most importantly, 39 % of the sampled people that
are against nuclear energy could change their opinion if a permanent and safe solution for nuclear
waste could be found. He pointed out that the European public, in general, was not well informed
of different possibilities of waste management. He said that the public opinion could look quite differently
at the problem if the nuclear waste is transmuted (burned) to decrease its half-life before
disposal. He believes that politicians should be better informed about the merits of nuclear energy
and composite waste management solutions. Personally, he believes that a strong synergy between
the “burn and bury” lobby is the best way forward. He said that he is committed to promote the research
in the field of partitioning and transmutation at national level supporting the efforts of the
EU, in order to positively complement the R&D activity on geological disposal. The combined results
of those two technologies could be very fruitful for the pacific use of nuclear energy.
Mr Boullis first outlined the principles of 'The 2006 French Act' concerning the nuclear waste management,
namely (i) continue recycling to decrease the amount of nuclear waste, radio-toxic inventory
etc., (ii) establish a retrievable geological repository as the reference option for the final waste
management and (iii) the year 2012 as a time point for reassessment of industrial potential of diverse
options of P&T leading to a prototype of the selected process by 2020 and (iv) definition of a
repository by 2015 with a view to its operation by 2025. He also pointed out the options that can be
considered for adoption in future: (i) U and Pu recycled with fission products (FP) where minor ac-
153

tinides (MA) are destined for geological disposal (GD), (ii) Heterogeneous recycling of MA and
U+Pu where FP are destined for GD, (iii) Homogenous recycling of MA and U+Pu where FP are
destined for GD, (iv) Co-extraction and recycling of U and Pu where FP and MA are destined for
GD and (v) the double strata approach where U and Pu are recycled in conventional reactors and
MA are transmuted in a dedicated accelerator driven system (ADS) where FP and residual MA are
destined for GD. Selection of an option is scheduled in 2012. Mr Boulis presented a scenario of
French nuclear power plants (NPP) up to 2060 involving life extension techniques of present-day
Gen II plants, operation of Gen III NPPs from 2012, operation of Gen III+ NPPs from 2020 and
Gen IV fast NPPs from 2040. He warned that the solution of the back-end of the nuclear fuel-cycle
must be considered with urgency otherwise one risks accumulating huge stockpiles of the spent
fuel. In this regard he alluded to the establishment of new pilot facilities at La Hague (FR) site for
quantities of fast reactor MOX fuel fabrication (tons) and MA pins (kgs).
Mr González emphasised the need to consult the public when developing waste-management policies
that incorporate P&T. This requires developing new technological options involving significant
financial and manpower resources. However, each country will have to consider its own constraints
for deployment of these technologies if it's overall nuclear policy so permits. He highlighted
the importance of detailed planning in erecting and utilisation of advanced facilities so that
best results are obtained. He also reiterated that public acceptance of nuclear requires the participation
of all stakeholders in decision-making.
Mr Lalieux shed some light on the key elements of the spent fuel that drive the geological disposal
community. These are (i) decay heat (fission products: ~ 150 y, actinides: 10000 y) and (iii) radio-toxic inventory
of the nuclear material in the waste (actinides: for more than 10000 y). He believes that the
contribution of P&T in reducing the long-term burden of repositories is limited if it is restricted to
actinide management. Benefits of separation of heat bearing elements to aid in increasing the repository
capacity should be matched with the risk of the generation of secondary waste and occupational
doses to technicians. He recognised the role of P&T in sustainability of nuclear energy but
this in itself should not impact on delaying the geological disposal related decisions by public authorities.
There were many questions that were posed by the audience for panellists to discuss. Mr Miguel
Cuñado of ENRESA, Spain was critical of P&T and warned of possible activation products which
may be more harmful if appropriate measures are not taken in selecting appropriate materials. He
was also unhappy that reprocessing plants are releasing iodine into the sea though it is done under
the purview of regulatory authorities in a controlled manner and within the limits set for such releases.
Mr Jan Marivoet of SCK-CEN (BE) said that P&T offers a number of very interesting perspectives
for waste management of future fuel cycles. He also indicated that the interaction between
P&T community and waste management organisations is highly desirable. Mr Jordi Bruno
of Spain said that we are not two scientific communities that are in confrontation (although we are
chasing the same funding) but politicians are using P&T to drag their feet, and they appear to take a
lifetime to make up their minds in approving the geological repositories. The panellist Mr Lalieux
added that it is important to balance the role of P&T and to avoid delaying disposal-related decisions.
In the chairman's view, the progress and prospects, challenges and recommendations for P&T discussed
during the P&T session can be summarized as follows:
154

Progress and Prospects:
• Separation of main heavy metals (U, Pu) and heat bearing components (e.g. Cs, Sr, Am) before
disposal increases the repository capacity (3-100 times) in certain geological media.
• Storage of Cs and Sr for 100-300 years in specialized (calcinated) waste forms is recommended.
Due to the long-lived Cs-135 isotope, after storage, disposal of this waste form
would also be required.
• Transmutation/burning of separated MA (in ADS or FR) reduces the ‘long-term burden’ on
repositories. This may aid the GD community in securing a ‘broadly agreed political consensus’
of waste disposal in geological repositories.
• Transmutation of MA has also a favourable impact in the unlikely occurrence of ‘human intrusion
scenarios’.
• The maximum eventual dose to human beings from a geological repository in ‘normal scenarios’
is likely to be due to fission products though evidence is appearing that MA also can
be mobile under certain conditions.
Challenges:
• Advanced partitioning processes are to be reinforced towards pilot and test facilities for optimized
separation processes leading ultimately to industrial facilities requiring strong political
support and huge financial and human resources.
• Dedicated efforts have to be made in developing and selecting low activation materials in
reducing the intermediate level waste and reducing secondary waste streams in the processes
of P&T so that it does not put undue burden on the safe disposal of additional secondary
waste produced.
• A close cooperation between the two (P&T and GD) communities in defining unified and
coherent systems for effective waste management is highly desirable.
Recommendations:
• The GD community should take into account the requirements and accommodate the waste
streams emanating from the advanced (MA) reprocessing systems and support development
of appropriate waste forms for geological disposal.
• Keeping in mind the natural decay of nuclear waste, a careful roadmap and planning (taking
account of the time needed e.g. for regulatory authority approvals) should be made so that
there is no mismatch between the schedules of partitioning, transmutation and disposal technologies.
In conclusion, P&T is essential for the sustainability of nuclear energy. GD is indispensable for radioactive
waste management. Both communities should work together for the future of nuclear energy.
References
[1] Gianluca Benamati: Member of Italian Parliament, Heavy-liquid-metal technologists turned
politician since 2007, a former director of ENEA, Brasimone Nuclear Centre, Italy.
[2] Bernard Boullis: CEA Director, Advanced Fuel Cycle and Waste Management Research Programs
relating to future options for the back end of the fuel cycle. Formerly, member of the
La Hague-plant process-design team, in-charge of the CEA Radio-Chemistry department and
of the ATLANTE research facility.
155

[3] Enrique González: Head of Nuclear Division, CIEMAT, Spanish Energy, Environmental and
Technological Research Centre. Experimentalist in Nuclear data and neutronics of Fast systems,
Leader of some of the Nuclear data experiments at CERN and of the Nuclear data Domain
in FP6-Eurotrans project.
[4] Philippe Lalieux: Director, Long-Term Nuclear Waste Management, ONDRAS / NIRAS Belgian
Agency for Radioactive Waste Managem
156

Sessions VI to VIII: Geological disposal
Introduction and objectives
Euratom support to research in geological disposal – a new dawn?
At the end of the Sixth Framework Programme (FP6), Euratom programmes had been supporting
research in the field of geological disposal of radioactive waste for more than three decades. In recent
years, the scientific community has gradually come to the conclusion that generic research and
the knowledge in most areas of the safety case have reached sufficient maturity to enable step-wise
implementation of geological disposal, even though actual progress has been slow in most Member
States. During FP6, the EC launched a number of IPs and an NoE in key areas of repository systems
and processes common to multiple host rock types. The objective of this new approach to research
has been, on the strategic level, to further increase co-operation, integration, and co-ordination of
efforts at European level while, on the technical level, building on past research in order to reduce
uncertainties and thereby increase safety margins and confidence in the soundness of geological
disposal in general.
The purpose of the sessions on geological disposal will be to take stock of the progress made by
Euratom FP6 research projects in their respective scientific fields and in integrating research at
European level. The objective of Euratom FP7 (2007-2011) in the area of geological disposal is to
support implementation-oriented R&D on all remaining key aspects, and therefore the panel discussions
organised in the areas of near- and far-field processes will provide a key opportunity to
review whether and which research is still needed. Equally, the presentation and panel discussions
on the strategic co-ordination of R&D for waste disposal in the EU should establish the state of play
on the relevance and readiness of the key R&D stakeholders to set up a technology platform for the
development and implementation of a strategic research agenda in this field. The results of this exercise
will no doubt have an important impact on the content and implementation of Euratom FP7
work programmes and future Euratom framework programmes, possibly heralding a new dawn for
Community support for R&D in the area of geological disposal.
157

158

SESSION VI: Near-field processes
Chairman: Dr Simon Löw, ETH Zürich, Switzerland
159

160

Advances in Integrating European Research on the Near-Field System
Alain Sneyers 1 , Bernd Grambow 2 , Pedro Hernan 3 , Hans-Joachim Alheid 4 , Jean-François Aranyossy
5 , Lawrence Johnson 6
Summary
1 SCK•CEN, Mol, Belgium; 2 SUBATECH, Nantes, France
3 ENRESA, Madrid, Spain; 4 BGR, Hannover, Germany
5 ANDRA, Châtenay-Malabry, France; 6 NAGRA, Wettingen, Switzerland
In the 1970's, national and international research programmes were established with the purpose
of investigating various options for the long-term management of radioactive waste. Today,
broad consensus exists in the scientific community on geological disposal as the ultimate
solution for the safe management of the radioactive wastes. Through deep disposal, radioactive
waste is isolated from the biosphere by multiple engineered and natural barriers. The Integrated
Project NF-PRO (Sixth Framework EURATOM Programme, European Commission)
has investigated key-processes in the near-field of a geological repository for the disposal of
high-level vitrified waste and spent fuel. In previous EC Framework Programmes, the scope
of individual projects was restricted to specific near-field components and processes. NF-
PRO represents a major step forward since it is the first EU project bringing together all areas
of expertise required for evaluating the performance and the evolution of the near-field system.
Under NF-PRO, interactions between different experts have increased and major advances
have been made with respect to the phenomenological understanding of processes influencing
the performance of the near-field system. NF-PRO's work programme concentrated
on outstanding key-issues and consisted of a wide range of research topics and methodological
approaches including laboratory and in situ experiments, detailed process modelling as
well as broader assessments of the performance of the near-field system. As part of NF-PRO,
models and data on specific and coupled near-field processes were translated to concise data,
parameters and models as input to broader assessments of the performance of the near-field.
The level of integration achieved by NF-PRO, in particular the pooling of all fields of expertise
required for development of a comprehensive and phenomenological understanding of the
behaviour of the near-field system and its evolution in time and space is unique and has not
been achieved in preceding Framework Programmes. The scientific output by NF-PRO is particularly
relevant to repository development programmes in EU Member States where information
derived from the project can be applied in the safety case. This paper gives a summary
overview of key scientific advances made by NF-PRO.
1. Introduction
Since 1984, the European Commission has supported international research on the management of
radioactive waste through the multi-annual EURATOM Framework Programme. In 2002, the Sixth
Framework Programme (FP6) was launched. As part of FP6, new instruments for co-operation, notably
the Integrated Projects and the Networks of Excellence, were established. These instruments
allowed to broaden the scope of research as well as to enlarge the research consortia. Also, the av-
161

erage level of funding of research projects increased substantially. As a result, co-operations between
research groups that had worked independently in previous EU-supported Programmes were
strengthened. Accordingly, the Sixth Framework Programme represented a major step forward in
the structuring of European RD&D Programmes.
This paper discusses progress made in research on the management of radioactive waste by the FP6
project NF-PRO. NF-PRO is a multidisciplinary FP6 Integrated Project investigating the barrier
performance of the near-field of geological repositories for high-level waste disposal. The NF-PRO
consortium consists of 40 leading nuclear research organisations, radioactive waste management
agencies/implementing organisations, universities and consulting companies. For a full and in-depth
account of work performed, the reader is referred to the NF-PRO Synthesis Reports and corresponding
references to the detailed NF-PRO technical reports herein [1-5].
2. The near-field system
Facilities for the geological disposal of radioactive waste are designed to provide safety by minimising
the release of radionuclides from the repository system over extended periods of time. Multiple
engineered barriers enclose the disposed waste packages. These barriers constitute the nearfield
system and play an essential role in securing the overall safety of geological disposal. The
near-field is a complex environment consisting of several components including the waste form, the
waste canisters, backfills, seals, plugs and the region of the host rock immediately surrounding the
repository system. Repository construction and operation as well as waste emplacement will affect
chemical and physical conditions and therefore the confinement properties of the disposal system.
After repository closure, the near-field environment will evolve as a result of chemical interactions
between different repository components, heat generation and radiation. Safety evaluations have to
take into account the combined effects of processes occurring in the near-field in order to assess
their impact on radionuclide mobility and thus the effectiveness of the safety functions of isolation
and retardation.
3. Scope and objective of research addressed by NF-PRO
The Integrated Project NF-PRO addresses the near-field of a repository for the geological disposal
of vitrified high-level radioactive waste and spent fuel. NF-PRO’s project scope takes into account
various repository concepts/designs that are currently under investigation in EU Member States.
The host rocks addressed by NF-PRO are salt, granite and clay.
The principal objective of NF-PRO is to improve the understanding of key processes in the nearfield.
NF-PRO is the first European research project bringing together all fields of expertise related
to the near-field system. Accordingly, NF-PRO represents a major step forward in the developing of
a scientific and technical basis for radioactive waste management.
4. Summary of advances made by NF-PRO
4.1 Advances in research related to key processes affecting the waste form
The waste form constitutes the first engineered barrier in a geological repository. Upon contact with
groundwater, the waste matrix will slowly dissolve leading to the release of radionuclides. The
mechanisms controlling the alteration of the waste matrix are complex and depend, amongst others,
on the geochemical conditions prevailing in the near-field. NF-PRO has investigated processes affecting
radionuclide release from the waste form under near-field conditions.
162

Advances in research on vitrified high-level waste
The dissolution of vitrified waste is generally described as a two-stage process involving a high initial
dissolution rate, during which near-field materials become increasingly saturated with silica released
from the waste glass, followed by a low residual dissolution rate. Prior to NF-PRO, glass dissolution
models and parameters were mostly based on experimental observations obtained from
relatively simple subsystems, for example the subsystem glass-claywater. Under NF-PRO, integrated
experiments were conducted to investigate glass dissolution processes in more realistic nearfield
conditions. Different vitrified waste glasses (SON 68 and blended Magnox glass) were
brought in contact with various compacted near-field materials including the bentonite buffer (Volclay
KWK) and corrosion products (magnetite) released from metallic repository components. The
experimental matrix allowed for simulating different stages in the evolution of the near-field system.
Conditions representative for the long-term evolution were reproduced by adding amorphous
silica to saturate silica sorption sites on the corrosion products and the clay buffer. Experimental
results obtained by NF-PRO confirm that, after the saturation of the aqueous phase adjacent to the
glass with silica, the glass dissolution rate decreases by several orders of magnitude. It was found
that corrosion products from metal-based near-field components may retard the decrease in glass
dissolution rate as they provide additional adsorption sites for dissolved silica. Experimental results
were used to calibrate and to improve glass corrosion models in the presence of corrosion products
and clay: a satisfactory agreement between the modelling and experimental results was obtained.
In addition to research on the key mechanisms affecting glass dissolution under integrated nearfield
conditions, NF-PRO investigated various factors potentially influencing the mobility of radionuclides
released from vitrified waste. In particular, the effect of dissolved carbonate on the release
of rare earth elements and uranium was investigated and new data were obtained on the uptake
of Eu in newly formed clay minerals secondary phases such as powellite and calcite.
Advances in research on spent nuclear fuel
An important area of research dealt with by NF-PRO concerns the investigation of key processes
affecting the release of radionuclides from spent nuclear fuel under geological disposal conditions.
In spent fuel, radionuclides are inhomogeneously distributed. Consequently, the release of radionuclides
from spent fuel is typically described by two terms:
The Instant Release Fraction (IRF), which corresponds to the fraction of the radionuclide inventory
that segregated partially from the UO2 or MOX matrix during irradiation in the reactor.
This fraction is located in parts of the fuel with low confinement properties, in particular
at the grain boundaries and in the fuel microstructures. The IRF is anticipated to be immediately
released in repository conditions upon contact with groundwater following breaching of
the spent fuel canister/cladding.
The fraction of the radionuclides that are embedded in the spent fuel matrix and which will
dissolve slowly in contact with groundwater.
In previous Community–supported Programmes, a substantial amount of work has been performed
to increase knowledge on the Instant Release Fraction. As part of the project "Spent fuel Stability
under Repository Conditions" (SFS) (Fifth Framework Programme), a model was developed to describe
and to quantify the IRF fraction [6]. Work by NF-PRO has built on the outcome of SFS and
has addressed a number of remaining uncertainties and open questions. The following key results
were obtained by NF-PRO regarding the Instant Release Fraction:
163

Prior to NF-PRO, the question whether the process of solid state � radiation-enhanced diffusion
to the release of fission products from spent fuel was unresolved. A synthesis of modelling
and experimental data as part of work by NF-PRO has shown that the diffusion process is
very slow and therefore will not significantly contribute to the IRF.
Under NF-PRO, a new micromechanical model was developed to re-assess the impact of the
ingrowth of He on the spent fuel microstructure and radionuclide diffusion processes. Conclusions
from this work have shown that He accumulation may result in the formation of bubbles
in the fuel matrix. For UOX fuel however, the bubble pressure will remain under the threshold
value to crack the fuel. Conclusions from this work have allowed diminishing the conservatism
of former IRF values of key safety relevant radionuclides for PWR UOx fuels.
Until recently, few data were available on the IRF inventory for high burn up UOx fuel and
MOX fuel. NF-PRO has delivered new data on the IRF of spent UOx fuel as a function of the
burn-up. These data have reduced conservatism of former IRF values of key safety relevant
radionuclides for PWR UOx fuels. Also, estimates of the grain boundary inventory for spent
MOX fuel were made by performing leaching experiments on fragmented and powder samples
of SBR MOX.
NF-PRO has investigated various processes affecting the dissolution of the disposed spent fuel in
the presence of near-field materials. The following results are reported:
Experimental work has been performed to investigate the dissolution of �-doped UO2 under
reducing conditions. Radiolysis effects were shown to be less important for fuel dissolution
than previously thought. Indeed, the radiation field of 3000-10000 y old fuel has no effect on
the oxidative dissolution of fuel. Isotope dilution tests have indicated that spent fuel corrosion
in absence of radiolysis effects is still rate and not solubility controlled.
Experiments to investigate the dissolution of UO2 in the presence hydrogen gas confirmed
that hydrogen gas is an active reductant that lowers the dissolution of uranium in the presence
of Volclay. This is an important result since the inhibiting effect of H2 on the dissolution of
the spent fuel matrix confirms that, in the normal evolution scenario, low matrix dissolution
rates are applicable as long as H2 is present.
Investigation of the effect of trace constituents in groundwater on gamma radiation-induced
corrosion of UO2 indicates that low concentrations of bromide in repository groundwater may
offset to some degree the protective effect by hydrogen with respect to the release of certain
radionuclides.
Integration of the work on the Engineered Barrier System (next paragraph) has led to a better understanding
at process level of the evolution of glass and spent fuel under near-field repository conditions.
More specifically, due to the slow dissolution rate of the overpack, the waste may be protected
much longer than previously assumed. This constitutes a non-negligible safety reserve, making
the reference source term models more conservative than recognized before the start of
NF-PRO. The proposed reference source term models were not fundamentally changed in the project,
but the underlying processes are now better understood.
164

4.2 Advances in research related to the chemical evolution of the engineered barrier system
The Engineered Barrier System (EBS) isolates the disposed radioactive waste from the geosphere.
Diverse processes including the uptake of groundwater, chemical interactions between different
near-field components and the formation of secondary phases and corrosion products, influence the
confinement properties of the EBS and control radionuclide transport in the near-field. NF-PRO has
investigated a range of processes affecting the chemical evolution of the EBS, in particular the evolution
of the composition of porewater in the bentonite barrier, concrete-bentonite interactions, the
corrosion of metal-based repository components including the interaction of corrosion products
with bentonite, and radionuclide mobility in the bentonite barrier.
Progress in research related to processes affecting the bentonite porewater composition
Clay-based (bentonite) buffer materials are widely applied in repository concepts in granite and
clay host rocks. The bentonite buffer will progressively hydrate and act as a low permeability barrier
limiting and controlling radionuclide transport. An important part of the work by NF-PRO has
been dedicated to the investigation of the porewater chemistry and processes and mechanisms having
an impact on the evolution of the bentonite porewater composition. Changes in porewater composition
are particularly relevant to evaluate - amongst others - the dissolution of the waste matrix,
radionuclide mobility and canister corrosion rates. These compositional changes can be initiated by
internal factors, for example by a thermal gradient or by chemical interactions between different
repository components, or by external factors, for example changes in the composition of groundwater
entering the near-field system from the host rock following alteration of the host rock. NF-
PRO has investigated the mechanisms affecting porewater chemistry and radionuclide mobility to
evaluate the time-dependent evolution and the overall performance of the near-field system.
Eh and pH are key parameters with respect to chemical conditions in the near-field: Eh and pH have
major effects on concentrations of aqueous species (for example HS - , HCO3 - and Cl - ), which may
affect corrosion of metal-based engineered barriers and/or enhance radionuclide solubility. Until
recently, obtaining reliable data on redox conditions in compacted bentonite was very difficult since
different laboratory techniques tended to disturb the system and introduce sampling artefacts into
the measured data. Major advances have been made by NF-PRO regarding the direct measurement
of Eh and pH in compacted bentonite. Electrodes capable of producing accurate measurements of
Eh and pH in compacted bentonite were developed and used to assess the response of the bentonite
buffer to changes in pH (pH 11.7) and redox potential (aerobic) of contacting external groundwater.
Using cells with a total length of 20 mm, oxidising and alkaline perturbations were measured only
in the first 5 mm of the cells. pH-values re-established to values around 9 (the initial pH value was
equal to ~8.5) in the first cm of the bentonite and the extent of redox pertubations is limited. These
experimental observations indicate that redox and pH conditions are buffered very effectively in
compacted bentonite.
NF-PRO has investigated the interaction between saline porewater and the bentonite buffer. In particular,
the amount of water absorbed in different types of bentonite (MX-80 and FEBEX bentonite)
was measured as a function of porewater salinity. It was confirmed that the distribution of the water
among external and internal (interlamellar) sites depends on the pore water salinity. Results from
experimental work indicate that the amount of internal water decreases with increasing salinity.
Additional experiments have led to the conclusion that the amount of water absorbed in bentonite
depends on the dominant cation in the smectite exchanger. Experimental results with homoionised
FEBEX bentonite show that maximum water absorption occurs with magnesium or calcium as the
dominant exchangeable cations. Minimum water absorption was observed with potassium as domi-
165

nant exchangeable cation. This change in the interlayer charge, together with the ionic strength of
the pore water, has an impact on the swelling capacity of the bentonite.
In the framework of NF-PRO, laboratory experiments were performed to investigate the effect of a
thermal gradient on the bentonite pore fluid composition and the hydration of unsaturated bentonite.
A compacted bentonite column was heated to 100°C at the bottom end while diluted granite water
was injected at the top surface, which was kept at ambient temperature. After 7.6 years, the column
was dismantled and analysed. Experimental results indicate that carbonate behaviour is consistent
with the dissolution of calcite leading to oversaturation of the porewater with respect to gypsum,
which precipitates in the upper part of the column, the latter resulting in a decrease in sulphate. Observations
from these experiments confirm that pH buffering can be explained by dissolution/precipitation
reactions.
Advances in studies related to corrosion
In a geological repository, the container isolates the disposed waste from groundwater for a designated
period of time. Corrosion will affect the containment properties of the container. A variety of
corrosion experiments have been conducted under NF-PRO in which key issues were addressed related
to iron and steel corrosion rates and interactions between corrosion products and the bentonite
buffer bentonite properties. Work by NF-PRO involves investigations on interactions between bentonite
and corrosion products from carbon steel canisters, or in the case of the KBS-3 concept, copper
canisters with cast iron inserts, and the effects of these interactions on the properties and potential
effect on the performance of the engineered barrier system.
Various experiments were carried out to identify the nature of corrosion products that are formed at
the container-bentonite interface. These experiments have shown that magnetite, which is formed
through an intermediate Fe(OH)2 phase, is a prevailing corrosion product.
A range of iron corrosion experiments have been conducted by NF-PRO to obtain long-term corrosion
rate data. Experiments performed in bentonite water or slurry at different temperatures have
shown that, after an initial stage of enhanced corrosion, the corrosion rates decreases to values below
5 �m/year. This indicates that that the corrosion rate is controlled by the formation of a corrosion
product layer. In the long term, it seems that the corrosion rate is mainly controlled by the
properties of the corrosion products. The initial corrosion rate increases with higher temperatures
but slows down after steady-state has been established.
Advances in studies related to canister-bentonite interactions
An important part of work by NF-PRO involved questions related to the effects of copper and iron
corrosion products on the bentonite properties. Laboratory tests under different experimental conditions
indicate typical diffusion depths in compacted bentonite in the order of less than 1 mm for
copper canisters to 5 to 6 mm for steel canisters.
Corrosion experiments performed under NF-PRO have shown that the physical properties of the
bentonite are modified by increased iron concentrations in the bentonite. In particular, this may increase
hydraulic conductivity, possibly decrease swelling pressure and decreased cation exchange
capacity. In the long-term, the physical properties of the bentonite may be affected by the conversion
of montmorrilonite to non-swelling Fe-silicates (smectite). Results from corrosion experiments
were modelled with the reactive-transport code PHREEQC.
166

Several experiments were performed to obtain information on the effect of canister corrosion products
on radionuclide sorption. Experiments were conducted to study sorption of Cs onto magnetite,
assuming that this is the main corrosion product in anaerobic conditions. The results showed that
sorption of Cs was negligible up to pH 8.5. Maximum sorption values were reached around a pH of
12. A significant increase in Cs sorption was observed with decreasing the ionic strength. Linear
sorption isotherms were observed for a Cs concentration within the range used (i.e. up to 10 -6 M).
As expected, Cs showed very minor sorption on the oxide. All experimental results could be satisfactorily
described with a simple model. The model developed was also capable of reproducing
sorption of Cs on magnetite in more complex solutions (bentonite and cement synthetic pore water).
Advances in studies related to concrete-bentonite interactions
Experiments investigating the evolution of pore water composition due to the interaction with concrete
have provided further evidence for the buffer capacity of bentonite. Experimental results have
shown that using an external solution with pH 13.5, the bentonite pore water reaches a maximum
pH of 9 after experimental durations of 596 days, indicating that pH is buffered by geochemical
processes in the bentonite. Further experiments were performed to obtain information on mineralogical
changes due to interactions between bentonite and high pH groundwater. At high pH´s (pH
� 13.5) minor mineralogical changes were found to occur in the bentonite. These changes are limited
to an alteration zone of a few millimeters from the interface after two years of experiments. The
most important changes are related to partial dissolution of montmorillonite (at expected rates of 10 -
12 to 10 -13 mol m -2 s -1 ), formation of saponite-type clay and the precipitation of zeolites (in some of
the experiments) and brucite. The most relevant change in all these experiments conducted under
repository conditions (diffusion experiments) consisted in a reduction of the CEC from 25 to 50%
of the initial value close to the interface with the concrete. Modelling of the interaction between
bentonite and concrete has been successful and was capable of reproducing the main processes,
such as partial pore blocking, brucite precipitation, minor montmorillonite dissolution, and the replacement
of Mg- by K-montmorillonite.
Advances in research concerning radionuclide transport in bentonite
NF-PRO's programme of work included a range of experiments investigating radionuclide transport
in bentonite materials. These experiments were performed to develop a mechanistic understanding
of the sorption and the diffusion of anions and cations in compacted bentonite. The experimental
programme allowed assessing the effect of the degree of compaction and porewater chemistry,
combined with the microstructure of bentonite.
Experiments were performed to determine the sorption of Ni(II), Eu(III) and U(VI) onto Namontmorillonite
as a function of carbonate concentration and pH. Data obtained from these experiments
indicate that Ni(II) sorption onto montmorillonite is rather insensitive to the presence of inorganic
carbon whereas the presence of inorganic carbon decreases sorption of Eu(III) and U(VI)
on montmorillonite. Experiments using Volclay bentonite at pH 8 show that Th sorption on bentonite
decreases in the presence of organic matter.
Radionuclide diffusion processes in bentonite were investigated on the basis of a series of throughdiffusion
and out-diffusion studies with 36 Cl - , 22 Na + and 85 Sr 2+ and 134 Cs + using Volclay KWK bentonite.
Experimental data indicate that values of effective diffusion coefficients for both cations and
anions depend on ionic strength of water in pore spaces. In the case of anions, an increased ionic
strength leads to an increased transport rate whereas in the case of cations, a decrease of transport
rate was observed. This is explained by difference between concentration gradient in reservoirs and
167

eal concentration gradient in samples. This observation is particularly important from a performance
assessment perspective since it suggests that the values of effective diffusion coefficients can
be site specific, i.e., different for groundwater with low and high ionic strength.
Under NF-PRO, various experiments were performed to assess the influence of iron corrosion products
on radionuclide mobility in the near-field. In particular, the question whether corrosion products
can reduce and immobilize elements such as Tc(VII), Se(VI), Se(IV), Np(V) was addressed.
For selenium, experimental data show that under anoxic conditions, Se(IV) and Se(VI) in groundwater
can be reduced by iron and/or iron corrosion products (FeCO3, Fe3O4 and green rust) to insoluble
Se(0) and Se(-I) (as FeSe2).
4.3 Advances in research related to Thermo-Hydro-Mechanical and coupled processes affecting
the near-field system
During the operational phase of a repository and for some time after closure, the near-field is subject
to strong thermal, hydraulic and mechanical gradients. These gradients will evolve towards
equilibrium. Processes controlling this evolution are strongly coupled and depend to some degree
on local chemical conditions.
In most repository concepts considered by EU Member States, buffer or backfill materials are emplaced
around the disposed waste canisters. In granite and clay host rocks, bentonite is applied as
buffer material while in rock salt, crushed salt is used. After emplacement of the bentonite buffer,
hydraulic gradients lead relatively rapidly to the hydration of the bentonite. Buffer and backfill materials
are associated with important safety functions since they limit transport of radionuclides that
may be released from the waste matrix and provide a favourable chemical and mechanical environment
with respect to other barriers (e.g. the waste form, the canister/overpack and the host rock).
An important part of the work performed by NF-PRO involved the investigation of the combined
effects of thermal, geomechanical and hydrological processes on the behaviour of bentonite buffer
and crushed salt backfill. A comprehensive analysis of the THM (C) processes in the EBS has been
carried out as under NF-PRO, taking into account various types of buffer material (bentonite, salt),
different scales (from laboratory to real scale), and several degrees of saturation and thermal states.
Under NF-PRO, a large number of experiments were performed to evaluate parameters and processes
that influence the hydration of bentonite buffer.
Progress in research related to the impact of THM processes on the bentonite buffer
In a first series of experiments, the evolution of the hydration of the bentonite buffer and the THM
evolution of the near-field during the transient period was investigated. Experimental work as part
of NF-PRO has provided a large body of qualitative and quantitative information on individual
processes relevant for the THM behaviour of the bentonite during the saturation-thermal phase.
This information includes the potential threshold gradient for water flow in bentonite, dependence
of permeability on temperature, water vapour movement. Data and information at parameter/process
level has been included in THM models in order to improve the predictive capabilities
for the transient THM phase. In a second series of experiments, tests were conducted in order to
assess the impact of the transient THM period on the long-term properties of the bentonite buffer.
These experiments have confirmed that the bentonite swelling capability remains basically unchanged
after several years of hydration with thermal gradients. Moreover, no mineralogical
changes were observed.
168

Results obtained by NF-PRO have contributed to increase the experimental knowledge base on T-
H-M processes in the near-field and have provided additional support to the basic assumption that
the mechanical and chemical properties of the bentonite are such that the required safety functions
are fulfilled. Nevertheless, a number of open questions were identified. For example, it has not yet
been fully clarified whether the thermal gradient will hinder or significantly delay full saturation of
the bentonite buffer. Another unresolved issue concerns the homogenisation of density gradients
that are generated in the long term during saturation. These uncertainties are to be addressed by future
research.
Advances in research related to crushed salt backfill
An important part of work within NF-PRO focuses on the investigation of crushed salt backfill. A
limiting factor regarding the use of salt-based granulates as backfill is associated with their high
initial porosity and permeability, resulting in a low sealing capacity and mechanical integrity immediately
after emplacement. Consequently, crushed salt backfill will only attain its low permeability
through either compaction by drift convergence or by pre-compaction before emplacement.
Work by NF-PRO includes investigations to support long-term predictions of compacted salt behaviour,
the optimization of the compaction process and the study of the long term compaction development
at low differential stresses. The use of bentonite as additive to crushed salt to ease the
compaction and reduce permeability has been tested. Experimental work has demonstrated that adding
about 15 to 20 percent of bentonite reduces the permeability of crushed salt by four orders of
magnitude and eases compaction significantly. The application of this type of backfill is restricted
to low temperature (< 100°C) areas. Alternatively pre-compacted salt bricks can be used as backfill.
The performance of the salt bricks mainly depends on the characteristics of the interfaces between
the bricks and between the bricks and the surrounding rock salt. These characteristics have
been determined by shear tests and modelling. Moisture seems to accelerate the healing of interfaces
with time. Microphysical models have been developed to explain the compaction and permeability
behaviour.
Studies on gas migration in a saturated buffer
The migration of gas through the bentonite buffer is a key topic with respect to the long-term safety
of disposal. In a geological repository, gases are generated by radioactive decay and the corrosion
of steel. As part of NF-PRO, a full-scale in situ experiment (LASGIT) has been developed to evaluate
the consequences of the release of gas from a canister in a KBS-3 type repository. The LASGIT
experiment (Äspö Hard Rock Laboratory, Sweden) consists of a full-scale simulation of a canister
surrounded by bentonite buffer disposed vertically in granite rock at a depth of 420 meters below
the surface. Under NF-PRO, a preliminary set of hydraulic and large-scale gas injection tests were
performed in the hydrated buffer in view of investigating the response of the bentonite buffer. Highquality
test data derived from LASGIT were applied to test and to validate modelling approaches.
Results obtained so far are in line with assumptions made in safety assessment on the capability of
the bentonite buffer to reseal after gas breakthrough.
4.4 Advances in research related to the initiation and the development of the excavation damaged
zone
The host rock is a natural barrier isolating the geological repository from the biosphere. Repository
construction will induce changes in the host rock in the immediate vicinity of the excavated zone
leading to the development of an excavation damaged zone (EDZ). As a result, confinement properties
of the host rock may be locally altered, especially in the area close to the disposal galleries and
169

access shafts, providing higher permeability for water and gases. The EDZ could act as a preferential
pathway for radionuclide transport and may play an important role in the overall performance of
a repository, particularly in scenarios with early canister and seal failure. The extent and the characteristics
of the affected zone may vary significantly.
Prior to NF-PRO, a large number of studies were published regarding EDZ in terms of phenomenology
in different host rocks. Nevertheless, a need was identified for a deeper knowledge of EDZrelated
phenomena in order to establish a better basis for abstracting a PA model of transport in the
EDZ from a detailed scientific description of the processes.
Characterisation of the EDZ
As part of NF-PRO, conventional methods and techniques were applied to study the initiation of the
EDZ and to measure key characteristics and associated parameter values of the EDZ. In addition to
in situ measurements, laboratory tests on rock samples were performed. Experimental data were
compared with results from numerical modelling studies. New techniques for in situ EDZ characterisation
were developed and improved: these include seismic-acoustic monitoring, ultra-sonic logging,
seismic and geoelectric tomography and very high resolution ultrasonic logging.
Short-term evolution of the EDZ
Important progress has been made by NF-PRO with respect to investigations of the phase of repository
operation (short-term evolution of the EDZ). During this phase, ventilation of drifts may cause
desaturation and induce thermo-hydro-mechanical effects and/or chemical perturbations in the host
rock. In general, results from experimental work by NF-PRO on indurated and soft clay rocks indicate
that the extent of the affected zone in the host is very limited. The size of the altered host rock
in the experimental gallery at -445m in the Callovo-Oxfordian formation (Meuse/Haute-Marne
URL, Bure, France) and in the ventilation II experiment in the Opalinus Clay (Mont Terri URF)
typically is in the range of 20 to 60 centimetres. In situ experiments to study the zone affected by
oxidation around the Test Draft in the HADES URF (Mol, Belgium) evidence that the oxidised
zone extents less than one metre in the Boom Clay. The affected zone is associated with open fractures
that form during excavation. Diffusion of oxygen in the Boom Clay is a very slow process,
which is negligible relative to instantaneous oxidation along fractures.
Long-term evolution of the EDZ
Investigations under NF-PRO on the long-term evolution of the EDZ focussed on self-sealing and
gas transport processes in clays and salt host rock.
Laboratory and in situ experiments performed under NF-PRO on plastic and indurated clays have
provided new quantitative data on self-sealing of the EDZ (i.e. the decrease of permeability with
time) and have expanded knowledge on different factors (thermal, mechanical, chemical…) involved.
Short-term microscale processes contributing to self-sealing include swelling of smectite
minerals, mechanical closing due to the plasticity of clay minerals and creep. In the long term, precipitation
of minerals such as carbonates plays a role in self-sealing processes. Other factors such as
resaturation and thermal impact will lead to a quasi-complete self sealing of the EDZ. Observations
made on plastic clays (e.g. Boom Clay) indicate that self-sealing may result in a progressive restoring
of permeability to values comparable to those of undisturbed rock, although residual microscale
evidence of features initially formed in the EDZ network may remain. In indurated clays like the
Callovo-Oxfordian clay, open fractures of the EDZ progressively close and very low permeability
170

close to values of the undisturbed host rock may be attained although, locally, the porosity may remain
greater than in the undisturbed rock. In the very long-term, a reconsolidation may occur along
fissures and/or micro-fissured zones.
Investigations concerning gas transfer along the EDZ
Work in the framework of NF-PRO on gas transfer along the EDZ focussed on the characterisation
of gas transfer properties, in particular gas permeability, gas threshold pressure, capillary curves of
the EDZ before, during and after hydro-mechanical self-sealing. Also, studies were performed to
investigate the consequences of impairments due to an excessive gas pressure. These investigations
focussed on gas transfer processes in plastic and indurated clay and rock salt. Results obtained have
demonstrated that fast self-sealing of the EDZ occurs in plastic clays. At the microstructural level,
preferential pathways for gas migration may exist. However, effective hydro-mechanical selfsealing
is expected to occur relatively rapidly during resaturation, in particular before significant
gas pressure build-up occurs.
4.5 Advances in research related to process couplings and integration in Performance Assessment
NF-PRO has carried out integrated analyses of the near-field evolution. These analyses have been
performed for different reference disposal concepts and host rocks and have confirmed that the
near-field displays a high degree of robustness and redundancy. Results from these evaluations are
summarised in [5] and have allowed identifying remaining key uncertainties and priority areas in
future research for these systems.
5. Conclusion
Experimental studies and modelling work by NF-PRO have provided new insights in and information
on key process affecting the overall performance of the near-field system. NF-PRO has a major
strategic impact on European disposal-related research since the Integrated Project has effectively
contributed to strengthening the scientific-technical basis for geological disposal.
References
[1] Grambow et al. 2008: RTDC-1 Final Synthesis Report, Dissolution and release from the
waste matrix. EU NF-PRO Project FI6W-CT-2003-02389.
[2] Arcos, D., Hernán, P., De La Cruz, B., Herbert, H.-J., Savage, D., Smart, N.R., Villar, M.V.,
Van Loon, L.R. 2008: NF-PRO RTDC-2 Synthesis Report, Deliverable (D-N°:D2.6.4) EU
NF-PRO Project FI6W-CT-2003-02389.
[3] Villar et al. 2008: RTDC-3 Synthesis Report. EU NF-PRO Project FI6W-CT-2003-02389.
[4] Aranyossy J-F., Mayor, J.C., Marschall, & Plas, F. 2008: EDZ development and evolution.
RTDC-4 Synthesis Report. EU NF-PRO Project FI6W-CT-2003-02389.
[5] Johnson, L. et al. 2008: RTDC-5 Synthesis Report, Deliverable (D-N°:5.2.3), EU NF-PRO
Project FI6W-CT-2003-02389.
[6] Poinssot, C. et al. Final report of the European Project Spent Fuel Stability under Repository
Conditions. CEA-R-6093. ISSN 0429-3460, 2005, pp. 103.
171

172

Challenges of assessing long-term performance of nuclear waste matrices in
repository near-field environments – insights from the NF-PRO and MICADO
projects
Summary
Karel Lemmens 1 , Bernd Grambow 2 , Kastriot Spahiu 3 , Yves Minet 4 ,
and Christophe Poinssot 4
1 SCK·CEN, Belgium
2 SUBATECH, France
3 SKB, Sweden
4 CEA-Marcoule, Nuclear Energy Division, France
For the first time, waste form behaviour (borosilicate glass and spent fuel) has been
studied under near-field conditions in a fully integrated fashion. The results obtained on
glass in the NF-PRO project confirm the fundamental understanding of the dissolution
mechanism also for very compacted near-field conditions in presence of material
boundaries such as glass/iron corrosion product and glass/clay. NF-PRO has also reduced
important uncertainties in the long-term performance of spent fuel under near-
field conditions. A quantification of the remaining uncertainties in parameters and models
is under way in the MICADO project. It has been shown that the instant release fractions
(IRF) will not increase with time. Radiolysis effects were shown to be less important
for fuel dissolution than previously thought. Although NF-PRO has improved our
phenomenological understanding of the glass/SF dissolution processes, it is still difficult
to reliably predict waste form performance under near-field constraints.
1. Introduction
In a geological repository for high-level vitrified waste and spent fuel, the waste matrix represents
the first engineered barrier isolating the disposed waste from the biosphere. The mechanisms of dissolution
of waste forms in groundwater are very complex. The analysis of experimental data is even
more complex since the rate controlling process changes with time and geochemical boundary conditions.
Dissolved silica plays a key role in the glass dissolution process. The presence of near-field
materials will influence dissolved silica concentrations and the timing of changes in rate limiting
steps. In the case of spent fuel, the waste matrix has a complex structure. Both slow release from the
matrix and fast release from the gap and grain-boundaries must be considered. Redox conditions
and radiation are of critical concern for spent fuel stability.
In the Fifth Framework Programme (FP5), the main emphasis has been on the integration of data
from experimental studies into modelling. Waste form related work performed in NF-PRO (FP6)
focussed on remaining open questions, in particular on those issues that entail significant uncertainties
with respect to the source term in an integrated assessment, considering the thermal, chemical,
hydrological and mechanical evolution of the near-field. The quality of key experimental data and
the different approaches and underlying hypotheses for spent fuel performance are being evaluated
173

in the European coordinated action project MICADO. This project assesses uncertainties governed
by the divergence between the various models and the experimental databases, and uncertainties in
predictions that arise by comparing the outcomes of the various models.
The waste form is of course only one of the barriers within a disposal concept characterised by the
superposition of multiple partly interdependent barriers. Barrier functions concern the establishment
of favourable geochemical conditions, limiting water access to the waste, limiting the transfer of
contaminants to the geosphere etc. The relative role of the waste form within these barrier systems
has been illustrated in the integration research component of NF-PRO [1]. The calculations show
that, compared to the absence of any barrier, the presence of a suitable waste form in its container
and the low solubility of key radionuclides already reduce hypothetical dose contributions at the
near-field/geosphere interface by up to five orders of magnitude. The importance of the waste form
depends on the stability of the waste form and on the disposal concept. In a disposal concept in
which the transfer of contaminated groundwater to the biosphere takes longer than the time of degradation
of the waste form, the waste form stability will have less significance than when nearfield/biosphere
transfer rates are fast. On the other hand, if, as it appears today, a waste form is stable
for hundreds of thousands or even for more than a million years, then it will provide important
safety margins and it sustains the safety case in any disposal concept.
2. Methodology
The methodology of the waste form oriented research within NF-PRO consisted in combining bibliographic
surveys with modelling and experiments, to obtain missing data in integral near-field environments
for various host rock formations, and the evolution with time. Integration with the other
research components (processes in the engineered barrier and in the engineering disturbed/damaged
zone) was attempted by a performance assessment oriented approach. The programme started from
a review of previous projects on glass and spent fuel and considering the material choices, boundary
conditions and scenarios fixed by the other research components of NF-PRO. Work on glass was
organised in a work package oriented on experiments, and another one on the geochemical modelling
of the observed interactions. Work on spent fuel was organised in a work package dealing with
the evolution of spent fuel under normal and early failure scenarios, focussing mainly on the instant
radionuclide release fraction (IRF), and a work package concerning the behaviour of the fuel matrix
under near-field conditions. The methodology of the still ongoing MICADO project consists in
bringing together model developers and potential users, experimenters and people with overall system
understanding to (1) select models (2) to select and review a common spent fuel/UO2/MOX
experimental database, (3) to apply the models with or without re-parameterisation to the repository
relevant part of the database, and assess deviations and interdependence of the model parameters,
(4) to evaluate all relevant information and approaches for spent fuel performance assessment, including
uncertainty propagation to the overall safety analyses, and assess simplification strategies to
translate detailed mechanistic models into overall system codes, and (5) to provide an independent
regulator point of view on the appreciation of the effects of documented model uncertainties on predictive
uncertainties for the repository safety.
3. Results
We provide only a summary of the main results. For more details, we refer to reference [2].
174

3.1 Vitrified waste
The simplified reference source term for vitrified waste at the start of NF-PRO was the r0-rr model.
In this model, glass dissolution is described to occur in two stages: a first stage of high dissolution
rate (r0), which lasts up to the saturation of the metallic overpack products with silica, and a second
stage of low residual dissolution rate (rr or rres), which lasts until the glass is completely dissolved.
The processes at the basis of this reference source term model were studied in a programme with
three components
(1) Glass-water interaction, covering the determination of some basic parameters of glass dissolution
and the effect of dissolved carbonate on the release of rare earth elements and U.
(2) Radionuclide immobilisation in secondary phases.
(3) Validation of key mechanisms of glass dissolution in integrated near-field conditions.
We present the main results for each of these components.
Determination of basic parameters of glass dissolution
The dissolution rates r0 and rres are well known for the R7T7 glass (corresponding to the SON68
standard reference glass), but this was not the case for the UK “blended Magnox-UO2” glass, which
contains more magnesium and aluminium than the R7T7 glass. These parameters were determined
within NF-PRO. The blended Magnox-UO2 glass behaves less favourable than the R7T7 glass:
both the forward rate and the residual rate are higher for the blended Magnox-UO2 glass. The reason
for the higher residual rate is probably the formation of secondary magnesium phases, triggering
the glass dissolution. The lower stability of the blended Magnox-UO2 glass was confirmed by
the integrated tests, which are discussed further. The reference source term model can now be applied
for the blended magnox-UO2 glass, provided sufficient data are available concerning another
important parameter, i.e. the exposed surface area of the glass.
Effect of dissolved carbonate on the release of rare earth elements and uranium
Long-term dissolution tests with glass GP WAK1 in synthetic Opalinus and Konrad clay pore water
solutions showed no clear effect of carbonates on the release of rare earth elements and uranium
from the glass in the carbonate concentration range of 73-98 mg/l, but this conclusion cannot be extrapolated
as such to higher carbonate concentrations. Crystalline secondary phases were observed,
such as powellite, barite, calcite, CaSO4 and clay-like Mg(Ca,Fe) silicates, but there were no distinct
uranium phases in the gel. This means that carbonate concentrations do not have much impact
on the radionuclide retention in the gel. This should be taken into account if radionuclide retention
is considered explicitly in the source term model. In the reference source term model used for NF-
PRO, radionuclide retention is not considered explicitly. In this case, retention in the gel contributes
to the safety margin.
Role of radionuclide immobilisation in secondary phases
Under conditions typical for a deep geological nuclear waste repository, secondary alteration phases
are formed during dissolution of the waste glass, once their solubility limit has been reached. Radionuclides,
which have been released from the waste matrix may co-precipitate with these secondary
phases and form thermodynamically stable solid solutions, such as amorphous gel layers and
various crystalline phases (see previous paragraph). It was found that trivalent actinides (Am, Pu,
Cm) can be structurally incorporated into the host minerals powellite, calcite and clay minerals,
forming solid solutions. The quantitative understanding of this solid solution formation has im-
175

proved, with amongst others the determination of activity coefficients for Eu in powellite. The fact
that Am, Pu, Cm can be structurally incorporated into the host minerals is a universal observation.
If one succeeds in developing further thermodynamic data for the aqueous – solid solution equilibria,
these data will be relevant to every system.
Validation of key mechanisms of glass dissolution in integrated near-field conditions
The dissolution of glass in near-field conditions is believed to be driven by a set of key mechanisms,
i.e. the transport of water into the glass, followed by ion exchange, and transport of silica out
of the glass. The transport of silica out of the glass is influenced much by the near-field, by means
of diffusion into and sorption on overpack corrosion products and backfill clay, and by advection or
precipitation in the near-field. Prior to NF-PRO, this model, and the parameter values involved,
have been developed based on tests of the various simplified subsystems (e.g. glass/clay water, clay
water/magnetite etc.). The objective of the integrated glass dissolution experiments in NF-PRO was
to validate this glass model and the underlying key mechanisms in realistic near-field conditions,
where all processes are coupled in a realistic set-up. The results were used to support geochemical
models, which were then applied for long term predictions. Tests were performed with glasses
SON68 and the blended Magnox-UO2 glass. The dissolution of these glasses was followed in reactors
where the glass was in contact with (1) a layer of compacted Volclay KWK, or (2) a layer of
magnetite powder, followed by a layer of Volclay. These tests allowed seeing the impact of direct
contact with the clay, and the impact of the presence of a layer of magnetite between the glass and
the clay. Parallel tests were performed with addition of amorphous silica to the Volclay or magnetite.
These tests allowed assessing the effect of saturation of the Si sorption sites, which simulates
the long term. Although the reference long term in situ temperature is 50°C or less, the tests were
performed at 90°C, to accelerate the processes.
The fundamental understanding of the glass dissolution in aqueous solutions was confirmed:
(1) Initial fast dissolution of the glass matrix
(2) accumulation of dissolved glass constituents in solution
(3) accumulation of less soluble glass constituents on the glass surface by forming a surface
layer of solid reaction products (including formation of a so called “gel layer”)
(4) slow-down of reaction rates due to the accumulation of dissolved silica in the aqueous
phase adjacent to the dissolving glass or within the pore water of the gel
(5) approach of a residual reaction rate which can be up to 10000 times lower than the initial
rate
(6) retardation of process (5) by adsorption of dissolved silica on iron corrosion products .
The quantitative reproduction of the data with the geochemical codes was nevertheless not always
satisfying. The rate increasing effect of magnetite can be simulated, but a better description of the
combination of geometric, hydrodynamic, thermodynamic and kinetic constraints is necessary.
The silica/clay system must be better understood (kinetics, solubility, sorption, temperature
effect).
The hypothesis of instant sorption equilibrium on the magnetite must be revised.
In addition to silica sorption, silica precipitation on the magnetite may take place.
The pH evolution at the glass interface must be better understood.
176

3.2 Spent fuel
The release of radionuclides from spent fuel is classically described by two terms:
The Instant Release Fraction (IRF) is assumed to be immediately released under repository
conditions upon contact with groundwater. This concerns mainly the fraction of the radionuclide
inventory located in the gap between fuel and cladding, and in the grain boundaries.
The radionuclides that are embedded in the spent fuel matrix will dissolve slowly as a result
of the matrix alteration.
The main results related to these two terms are as follows:
Instant release fraction and long term fuel evolution
The programme focussed on obtaining new experimental IRF data and on updating the IRF model
developed in the SFS project (FP5, [3]). This model considered both the initial IRF after irradiation
and its potential increase with time. Significant results have been obtained on both aspects.
First, regarding the characterisation of the IRF of “young” irradiated fuels, new IRF values have
been obtained both for high burn-up UOX and for MOX fuel, even though the measured values may
be overestimated due to inclusion of fractions from oxidized fuel surfaces. In particular, these results
seem to evidence that in case of water contact to the centre of the fractured fuel, the rim zone
of high burnup UOX fuel contributes less to the IRF, compared to the fuel in the centre of the fuel
rod, although the opposite was expected. Release of 36 Cl form UOX spent fuel may be much faster
than previously anticipated.
Second, one of the critical questions was whether this IRF will remain constant or grow with time
due to decay enhanced diffusion and fracturing by helium ingrowth. It was demonstrated that this
diffusion process is so slow that it will not significantly increase the IRF. Furthermore, a new micromechanical
model was developed to assess the impact of He accumulation due to decay. Results
show that He accumulation will lead to bubble formation, but for UOX fuel, the critical bubble
pressure will most probably not be sufficiently high to fracture the fuel. Hence the present day surface
area and IRF values of UOX fuel are expected to remain roughly constant with disposal time,
at least in absence of water. The situation for MOX fuels is more complex and still needs to be assessed.
So, NF-PRO has significantly reduced the uncertainties regarding the IRF quantification and a new
pessimistic and realistic set of instant release values has been provided for a number of potentially
mobile radionuclides.
Spent fuel matrix dissolution in the presence of a corroding container
Both studies of the dissolution of UO2 doped with emitters to simulate the radiation field of spent
fuel of a few thousand years, and of real spent fuel were carried out under different conditions
which simulate European deep repository conditions. The following results were obtained:
Tests with doped material have shown very low dissolution rates, calculated by the isotope
dilution method, while the measured 238 U concentrations are very low. For performance
assessment this may imply that under reducing conditions, solubility controlled dissolution
models cannot be used to describe long-term corrosion, and the radiation field of 3000-
10000 y old fuel does not promote the oxidative dissolution of fuel.
Tests with doped UO2 in the presence of bentonite confirm the counteracting effect of dissolved
hydrogen on the UO2 dissolution, but also suggest important sorption of U on the
clay. Hydrogen is assumed to slow down U(VI) dissolution, but not U(IV) dissolution. The
latter is assumed to continue until saturation of the sorption sites of the bentonite.
177

4. Discussion
Tests with dissolved 238 Pu in H2 containing solutions without a UO2 surface (homogeneous
�-radiolysis) confirm literature results at much higher H2 concentrations, i.e. that no consumption
of oxidants occurs in the bulk solution in absence of a catalysing surface.
Tests under -radiation with spent fuel and UO2(s) in H2 saturated low ionic strength NaCl
solutions with 10 -4 or 10 -3 M Br - have contributed to the understanding of the importance of
fuel surface processes in these conditions. H2 considerably lowers the amount of oxidants
produced by �-�-radiolysis by reacting with the oxidising OH-radical to produce water and a
reducing H-radical. Br - ions block the beneficial effect of H2 in the bulk solution.
Presence of Br - ions can result in a decrease, or increase, or have no influence on the production
of H2O2 (usually an oxidant) by �-radiolysis in dilute groundwaters. This should be
studied further in the future, since �-radiolysis is expected to dominate at the time of fuel
contact with groundwater in the normal evolution scenario.
Important progress was achieved on the modelling of the radiolysis effects at the fuel/nearfield
interface, by integrating a modified radiolytic model into a transport code and accounting
for the presence of H2 via the corrosion potential of the UO2 matrix. An important result
is that in H2 saturated media, redox fronts at the fuel surface break down, because the rate of
consumption of H2 by radiolytic or catalytic reactions is much lower than the rate of its
transport to the surface.
The study of the oxidative dissolution rates of UO2 in phosphate containing solutions shows
that the enhancing effect of phosphate as compared to carbonate may be larger in solutions
containing both ligands, due to the absence of secondary phases.
The study of the waste form dissolution processes and the exchange of information with the other
research components of NF-PRO have allowed a better description of the expected evolution scenarios
and the associated safety margins. The findings provided no compelling arguments to fundamentally
change the reference source term models. The underlying mechanisms are better understood
now, though, and missing elements have been better identified.
The results show that the glass reference source term model and the underlying expected evolution
scenario is conservative for several reasons that were less evident at the start of NF-PRO:
1. The reference source term models implicitly assume very short overpack life times, necessary
to overcome the thermal phase. In reality, the life time of the overpack will be much
longer, as suggested in [4]. Depending on the disposal design, the expected time for complete
overpack corrosion can be some tens of thousands up to 250 000 years or more. Even
if the overpack is locally perforated, intruding water will be trapped in a confined volume.
This means that the release of radionuclides by the waste form will start much later or be
slower than assumed in the reference evolution scenario and source term models.
2. The glass source term model assumes that the glass dissolves at the forward rate as long as
the magnetite is not completely saturated and does not take into account the saturation of the
magnetite by Si from the clay.
3. The glass source term model does not take into account the possibility that the dissolved
iron may diffuse at least partially into the clay to be sorbed or integrated into secondary
phases, rather then form a compact magnetite layer at the interface with the glass [5]
The life time calculations for vitrified waste have shown that the silica sorption on a magnetite layer
can shorten the life time of the glass at most by a few centuries. When we compare this short effect
with the long expected overpack life time, it is likely that the overall effect of the overpack will be
178

favourable. Using more conservative sorption parameters for the corrosion product layer would not
much change this conclusion.
The model for spent fuel is split in two independent terms: the IRF which is environmentindependent
and only related to the fuel properties, and the matrix contribution which depends on
the geochemical conditions. Significant improvement has been made on both terms within NF-PRO
even though the matrix model does not take into account any effects of (uranium) sorption on the
fuel matrix dissolution rate. Sorption of uranium on the near-field materials is indeed expected to
have a short term effect only. For spent fuel, the updated expected evolution scenario looks as follows:
Prior to the failure of the container in some ten thousands of years or more, the temperature begins
to drop to ambient values and a strongly reducing geochemical near-field environment is
established, governed by hydrogen formation and a redox potential close to the stability field of
water. The fuel inside the canister will evolve slowly due to radioactive decay: helium bubble
formation will occur at lattice defects, void spaces like grain boundaries or alpha damaged sites.
This will, however, have no significant effect on the spent fuel structure and surface area for
UOX fuel. For MOX fuels, this still needs to be assessed.
After container failure, water will slowly enter the void volume inside and will come in contact
with the fuel rod. It may take a long time until the container becomes filled with water since
physically the container will still be there, even if some defects have occurred. The entering water
will react to a much larger extent with the inner container surface (corrosion rate in the range
of μm/yr) than with the spent fuel matrix (corrosion rate in the range of 0.1 nm/yr). This leads
to H2 generation and pressure buildup limiting further water inflow. The details of this process
cannot be quantified today, so the conservative assumption is made that immediately after container
failure the spent fuel will come in contact with water.
The stability of the fuel cladding is not taken into account either. In reality, very few claddings
will have failed initially. Most claddings will remain tight for thousands of years, providing absolute
confinement of the fuel for this period. But again, this process cannot yet be quantified,
so its barrier function is ignored. Release of activation products from the cladding has also to be
considered, but this was outside the scope of the NF-PRO project.
There may, however, be a time where the cladding has failed before the container has become
filled entirely with water. In this case, the spent fuel may come in contact with hot water vapour.
The results of NF-PRO suggest that mobile fission products from the IRF will become
mobilized, but essentially very little or no attack of the fuel matrix will occur. These results are
probably influenced by reflux conditions in the reactor. It is not clear yet whether the data can
be applied directly to under-saturated conditions in a repository.
Sooner or later, the container will be filled entirely with hydrogen saturated and Fe 2+ rich
groundwater. Then, the IRF, containing a few percent of the mobile radionuclides like 129 I, 36 Cl
or 135 Cs, is immediately released. For these fractions, and with the conservative hypothesis of
instantaneous cladding failure, the fuel can not be considered as a confining waste form. There
is high confidence that for UOX fuel the IRF at that moment will still be the same as today.
Nevertheless, the largest fractions of the mobile nuclides like 129 I are bound to the fuel matrix.
They will become released only after the aqueous dissolution of the fuel matrix. Considering
the long containment times before corrosion leads to container failure, matrix dissolution will
only occur when the dose rate of and radiation will have dropped to insignificant levels, and
when the specific activity of the fuel will have dropped well below the threshold of 30
MBq/g, below which no accelerating effect of radiolysis on spent fuel corrosion is expected [3].
Under these strongly reducing conditions chemical non-oxidative dissolution of the fuel will
occur, but even under U(IV) saturated conditions, fuel corrosion may continue to occur, al-
179

though with a slow rate. The data indicate that only some hundred μm of the fuel will corrode in
1 million years.
Whether or not radiolytic effects are detrimental to spent fuel stability depends to a large extent
on the time of container failure. In the accidental scenario of container failure within the first
centuries after disposal, radiolysis may prevail. In this case, bromide concentrations in
groundwater might make the stabilizing hydrogen effect insignificant.
It was long thought that radiolysis will provide oxidizing conditions at the fuel surface, even in
overall reducing environments. This has motivated various studies of secondary phase formation
of U(VI) solid phases like phosphates or studtite in NF-PRO. It has now been shown that in
hydrogen saturated environments reducing conditions will remain dominant directly at the fuel
surface. Therefore, only U(IV) containing alteration products are expected to form.
5. Remaining uncertainties and recommendations for the future
5.1 Vitrified waste
The integrated tests and modelling have shown that the key mechanisms allowing a qualitative and
semi-quantitative description of the glass dissolution processes are known. The application to specific
disposal sites will have to take into account the specific characteristics of the disposal design
and host rock characteristics. Glass dissolution in cement based near-field conditions has not been
studied in NF-PRO. The related phenomenological description and process understanding still
needs more work than for the other systems at neutral pH. In all reference concepts, the overpack is
expected to last much longer than just the thermal phase, and thus will provide a non-negligible
safety reserve. The formation of metallic corrosion products does not fundamentally change the
glass dissolution, because sorptions on metallic corrosion products and on bentonite are similar.
The current models predict faster glass dissolution as long as the magnetite layer is not saturated
with silica, but the total sorption capacity of the magnetite is too low to have an important effect on
the total glass life time. Detailed modelling of the interactions between glass and corrosion products
is not yet possible, but a detailed description would most probably not lead to significantly shorter
life time predictions. Moreover its application in the safety assessment may be not straightforward
because the formation of a thick corrosion product layer is unlikely. Instead, iron rich clay minerals
are expected to be formed in the absence of glass [5]. The long term effect of metallic corrosion
products is probably negligible, because of saturation of the sorption sites by silica. Furthermore,
site densities of magnetite used in the experimental studies are expected to be much higher than in
reality, since dense layers of very little porosity will be formed. In the reference source term model,
the maximum dissolution rate is applied as long as the magnetite layer is not saturated. Omitting
this magnetite layer (i.e. replacing it by a bentonite layer) would imply a longer glass life time.
Hence the reference source term model is conservative in this respect. One should take this into account
when setting future research priorities. Nevertheless, the programme has suggested two potential
long term effects of metal corrosion that may be worthwhile further investigation: (1) the
precipitation of iron rich silica phases partially replacing the magnetite layer (this may include also
coprecipitation phenomena), and (2) a local pH increase due to actively corroding iron.
Apart from this, a better characterization of the long term dissolution processes and measurements
in conditions more representative for specific disposal designs are still necessary.
The programme has given further evidence that retention of specific radionuclides in secondary
phases is likely to contribute to the overall safety, but this process is not included in the reference
source term model. Including it would require much more research in this area, taking into account
the different conditions depending on the disposal site and design.
180

The exposed glass surface area (cracking factor) is a very important parameter to which little attention
has been given in the past European programmes. Because this parameter is relatively independent
from the disposal site and design, there is a common interest for further study on the European
level.
An important remaining methodological uncertainty is related to the risk of the use of extrapolated
dissolution rates in performance assessment. Residual rates of the order of magnitude 10 -4 g/m²d or
lower have been measured both in simple test conditions and in more realistic conditions (with presaturated
media). Measurements under real long-term conditions are by definition impossible. The
use of the low residual rates in performance assessment therefore relies on the assumed system understanding.
A compromise is necessary between (over)conservative estimations based on direct
laboratory measurements, and possibly too optimistic estimations based on extrapolations or hypotheses.
The lower the reference residual rate, the more solid the scientific basis must be. With the
current knowledge, residual rates about 1000 times smaller than the (well known) maximum rates
are defensible. The selection of reference dissolution rates for spent fuel faces similar problems.
5.2 Spent fuel
The work in NF-PRO has reduced the uncertainty for the time evolution of the instant release term.
Important remaining uncertainties concern the release of 129 I, 14 C and 36 Cl from high-burnup UOX
fuel, mass transfer processes in the grain boundaries for all types of fuel, the micromechanical impact
of bubble accumulation for higher He content (MOX fuel), and specific release from Pu clusters
in MOX fuel.
Although NF-PRO has confirmed that the matrix dissolution rate will be very low, a good understanding
of the underlying processes is still lacking. The proportionality of the matrix dissolution
rate with activity has been invalidated for reducing conditions in the normal evolution scenario.
The fuel matrix dissolution under reducing conditions is considered being controlled either by solubility
constraints or by slow kinetic rate laws. NF-PRO data seems to indicate kinetic control but
the mechanisms are not known. Consequences for performance assessment are quite different for
the two cases. Future research should find out which is the best description. If kinetics is dominating,
it should be better known, and the reaction products should be identified. To explain the low
rates under reducing conditions, the mechanistic understanding of the competing effect of radiolytic
H2O2 and H2 and other reducing species and surfaces must be improved using doped materials
that simulate aged spent fuel. H2 will be present in the normal evolution scenario at least as long as
the overpack is not entirely corroded, i.e. possibly a few hundreds of thousands of years, depending
on the disposal design. On the longer term, the reducing conditions and the corresponding slow matrix
dissolution should be based on other reducing components. Particular attention should go to the
matrix dissolution mechanisms for MOX fuel, for which data are still limited and allow no realistic
prediction of fuel performance in a repository under H2 saturated, Fe 2+ rich conditions.
Surface area is a key factor in any matrix dissolution concept, if the dissolution is driven by kinetics.
Differences in the specific surface area of UO2 and spent fuel need to be better understood to
reduce uncertainties in drawing conclusions from UO2 behaviour on spent fuel performance.
Uncertainties in spent fuel models are still high. Concerning the IRF, although a more deterministic
model is available since the SFS project [3], uncertainties are still high for high-burnup UOX and
MOX fuels and the micromechanical evolution of fuel pellet. The reference model describing matrix
alteration at the start of NF-PRO was the Matrix Alteration Model (MAM), developed within
SFS. The radiolytic models for UO2/water interface reactions include numerous reactions of radiolytic
species and associated rate constants both for pure water and for the interaction with the fuel
181

surface. The coupling between water radiolysis and surface interaction is strongly model dependent.
The transition from radiolytic controlled to chemically controlled dissolution is only represented in
few models and the effect of hydrogen is represented only with many ad-hoc assumptions. The effect
of Fe 2+ and H2, and certain parameters like bromide concentrations on fuel dissolution is experimentally
still rather poorly documented, and models describing this effect are not verified experimentally.
Surface area and porosity evolution is not well considered either in the models. No
mechanistic model for MOX fuel dissolution exists. The work of MICADO is concentrated on UO2
fuel.
The current source term models do not yet include the effect of spent fuel/material interfaces (spent
fuel with bentonite, clay, iron or cement). The experiments of NF-PRO suggest that these near-field
materials have at least a short term effect on the matrix dissolution rate. Compared with the large
database on simple spent fuel/UO2/water systems the obtained data are still scarce and less conclusive.
Future research should study this more in detail, and include coupling effects for realistic flow
conditions.
In the normal evolution scenario of long-term hydrogen generation by container corrosion, unsaturated
conditions may prevail in the vicinity and inside of the disposal container of spent fuel for
tens of thousands of years. It is known that radiolysis of thin water films sorbed on solids under under-saturated
conditions may involve quite different radiolysis behaviour than bulk water irradiation.
NF-PRO has provided new data on the spent fuel behaviour under unsaturated conditions, but
the results are not conclusive and new experiments are necessary to enhance the confidence in the
system understanding for all stages in the evolution scenario.
6. General conclusions
In general, the knowledge of glass and spent fuel dissolution is sufficient to propose source term
parameter values (in general dissolution rates) to performance assessment with a specified level of
conservativeness. Further work should be balanced against the necessary degree of conservativeness.
Hence, the selection of future experimental work should be based on its potential impact on
the overall performance of the system. It is important that close interaction with the other system
components and the integration by performance assessment is continued. This evaluation should be
part of the selection process of future research programmes.
References
[1] Alonso, J. et al., Report Deliverable D-N° 5.1.7, EU NF-PRO Project FI6W-CT-2003-02389
(2006).
[2] Grambow, B. et al., RTDC-1 Synthesis Report, Deliverable D-N° 1.6.3, EU NF-PRO Project
FI6W-CT-2003-02389 (2008).
[3] Poinssot, C. et al. Final report of the European Project Spent Fuel Stability under Repository
Conditions. CEA-R-6093. ISSN 0429-3460 (2005)
[4] Johnson, L. et al., RTDC-5 Synthesis Report, Deliverable D-N° 5.2.3, EU NF-PRO Project
FI6W-CT-2003-02389 (2008).
[5] Arcos, D. et al., RTDC-2 Synthesis Report, Deliverable D-N° 2.6.4, EU NF-PRO Project
FI6W-CT-2003-02389 (2008).
182

Key Processes affecting the Chemical Evolution of the Engineered Barrier System
David Arcos 1 , Jaime Cuevas 2 , Ana Maria Fernández 3 , Horst-Jürgen Herbert 4 , Pedro Hernan 5 , Luc
Van Loon 6 , Pedro Luis Martin 3 , David Savage 7 , Nick Smart 8 and Maria Victoria Villar 3
Summary
1 Amphos XXI, Spain
2 Universidad Autonoma de Madrid, Spain
3 CIEMAT, Spain
4 GRS, Germany
5 ENRESA, Spain
6 Paul Scherrer Institut, Switzerland
7 Quintessa Limited, UK
8 Serco, UK
The NF-PRO project, completed at the end of 2007, has tackled key issues in the chemical
evolution of the EBS, through a four-year programme of laboratory experiments and computer
modelling, involving some 18 participants from 8 countries in the European Union and
beyond. Laboratory experiments ranged in physical scale from the ‘micro’ (millimetre scale)
to ‘macro’ (metre scale ‘mock-up’ experiments), whilst computer simulations aided interpretations
of laboratory experiments in ‘real’ time, and enabled the extrapolation of experimental
results to the timescales of relevance to waste isolation (up to a million years). Although not
strictly a component of NF-PRO, natural systems evidence has been used to place experimental
results into the relevant time context.
1. Introduction
The engineered barrier system (EBS) plays a key role in the long-term isolation of high-level radioactive
wastes (HLW) in deep geological repositories. Consequently, it is important to have a good
understanding of its chemical evolution with time. After repository closure, compacted clays acting
as ‘buffers’ around waste packages will resaturate and swell, forming a diffusive transport barrier
for potential canister corrodants, and in the long-term, for the migration of waste constituents.
Upon contact with saturating groundwater, canister metals will slowly corrode, initially under aerobic
conditions, and ultimately, anaerobically. In the long-term, the stability of swelling clays in the
EBS may be challenged by chemical interactions with other barrier materials, such as cement and
concrete, in situ groundwaters, and canister metals. Ultimately, the perforation of canisters by corrosion
will lead to radionuclide migration through the clay buffer via diffusion, and retardation by
sorption within the clay.
The principal objective of NF-PRO for the near-field evolution component was to establish the scientific
and technical basis for evaluating the safety function ‘containment and minimisation of release’
of the near-field of a geological repository for high-level radioactive waste and spent fuel. To
this end, NF-PRO investigated realistic processes and process couplings affecting the isolation of
183

nuclear waste within the near-field through experimental studies and both interpretative and predictive
modelling.
pH
14
13
12
11
10
9
8
7
High-pH external
solution added
6
100 200 300 400 500 600 700
Elapsed time (days)
184
External solution
5 mm
10 mm
20 mm
DIF-6
Figure 1: pH evolution of a diffusion cell experiment where pH of the external solution was shifted
to more alkaline conditions (from 8.3 to 11.7). Experimental data from [1].
2. Clay hydration, swelling and pore fluid evolution
Pore water characterisation is extremely difficult in samples of highly compacted bentonite, so that
typically pore water compositions are obtained by means of geochemical modelling. However,
there are potentially large uncertainties associated with some parameters, especially pH and redox
potential. In NF-PRO, a series of experiments was designed to enable measurements of these two
parameters in addition to full major element analysis, in highly compacted bentonite. These experiments
were aimed at the measurement of Eh and pH in bentonite pore water and their response
to changes in the composition of contacting external water [1]. The results indicate that shifting
redox conditions of the contacting external water to more oxidising conditions were only reflected
in the first 5 mm. When the conditions of this experiment were shifted to anaerobic conditions, the
Eh in the first 5 mm of the bentonite decreased to values equal to those of the rest of the bentonite.
These experimental results can be explained by reversible geochemical processes occurring in the
bentonite that partially buffer the redox of the system, likely enhanced by the high bentonite/pore
water ratio, impeding the penetration of the oxidising front further inside the bentonite. Similar behaviour
was identified when shifting the pH of external water to more alkaline conditions (pH =
11.7), where an increase in pH was only recorded in the first 5 mm of compacted bentonite. However,
in this case, after a sharp increase (Fig. 1), the pH in the first 5 mm of bentonite decreased
gradually until values between 9.5 and 10 were reached. This evolution can be explained by the
buffering effect induced by interaction with the bentonite minerals.
The effects of a thermal gradient on the hydration of unsaturated bentonite was examined in an experiment
consisting of a 60 x 7 cm column of unsaturated FEBEX bentonite at an initial dry density
of 1.64 g/cm 3 [2]. The bottom end of the column was heated to 100 ºC, while the top surface was in
contact with dilute, granitic-type water (I = 0.005 M) under an injection pressure of 1.2 MPa, to en-

sure hydration of the bentonite. After 7.6 years, the experiment was dismantled and several parameters
analysed and compared with similar experiments running for 0.5, 1 and 2 years.
Chloride concentration (mmol/100g)
7.0
6.0
5.0
4.0
3.0
2.0
1.0
0.5 years
1 year
2 year
7.6 years
0.0
0 10 20 30 40 50 60
Distance from the heater (cm)
Exchangeable cation (meq/100g)
185
50
Na Mg
5.0
Ca K
4.8
45
4.6
40
4.4
4.2
35
4.0
3.8
30
3.6
25
Na 3.4
3.2
20
3.0
0 10 20 30 40 50 60
Distance from the heater (cm)
+
Mg 2+
Ca 2+
K +
Figure 2: a) Chloride concentrations in the bentonite along the experimental columns for different
duration tests obtained by aqueous leaching at 1:4 solid to liquid ratio [2]. The thick black line is
the initial chloride concentration in the pore water; b) Cation exchange population along the experimental
column for the 7.6-years test [2]. Thick straight lines are the reference values of the
cation exchange population in the FEBEX bentonite.
At the end of the experiment, it was discovered that the bentonite was not fully saturated. Full saturation
occurred only in the first 10 cm of the column near the hydration surface; whereas the bottom
end close to the heater was found to be less saturated than at the beginning of the experiment. The
hydration of the upper part of the column caused an increase in bentonite swelling, leading to an
increase of the porosity (interlamellar porosity) associated with the expansion of the interlamellar
space. This process caused a decrease of the bentonite dry density in the upper part of the column
and the opposite effect in the bottom of the column, where a decrease of water content was recorded.
Although no mineralogical alteration was observed after the experiment, different bentonite-water
interaction processes controlling the chemistry of the system were detected. Hydration
caused dilution and advective transport of chlorides (Fig. 2a) and sodium, transport of sulphates
controlled by gypsum solubility and dissolution/precipitation of carbonates. As a consequence, a
modification of the average cation exchange population in smectites was observed, increasing the
MgX2 and KX contents in the bottom of the column (Fig. 2b) at the expense of sodium and calcium.
The migration of chloride through the column can be understood by comparing the results of the
experiments of different duration (Fig. 2a). From this, it can be seen that as hydration proceeds,
fronts of chloride developed (the hydration water is less saline than the initial pore water), [2].
However, an additional effect can be seen for the 7.6 years experiment, where the chloride concentration
of the front increases significantly when it reaches the bentonite near the heater. The reason
for such behaviour is related to: i) there is no possibility for further migration of chloride, so it accumulates
at the bottom of the column due to water movement; and ii) water evaporates due to the
higher temperature in this part of the column and water vapour migrates upwards, thus increasing
the chloride concentration of the remaining pore water.
Assuming that chloride behaves as a conservative element in the system, the geochemical processes
associated with the heating/hydration experiment (sinks and sources for other elements) can be
evaluated by normalising all other chemical components with respect to chloride. The results of
this comparison show that in the upper part of the column carbonate increases, due to the dissolution
of carbonate minerals (calcite). This is consistent with the increase of pH recorded in this sec-
Exchangeable K (meq/100g)

tion of the column [2]. Moreover, the increase in calcium associated with the dissolution of calcite
leads to the oversaturation of pore water with respect to gypsum, which precipitates in the upper
part of the column, leading to a decrease in sulphate. At the bottom of the column, water evaporation
induces the precipitation of anhydrite and calcite, leading to the replacement of Ca by Mg in
the exchange position of the montmorillonite. Throughout the column, there is a clear relationship
between sodium and sulphate, which is more pronounced in the warmer part of the system. This
can be interpreted by precipitation of sodium sulphate minerals (e.g. mirabilite, Na2(SO4)·10H2O),
although precipitation of sodium sulphate-carbonate (e.g. burkeite, Na6CO3(SO4)2) cannot be disregarded,
and would account for the decrease in carbonate in this part of the system. The precipitation
of these phases would lead to a decrease in the sodium occupancy in the exchange sites of the
montmorillonite, being replaced by Mg.
3. Canister corrosion
A variety of iron corrosion experiments have been conducted under NF-PRO to extend and increase
confidence in long-term corrosion rate data. Experiments have been conducted with steel in the
form of coupons, wire, and powder, in pure and synthetic bentonite pore waters, and also in the
presence of either compacted or slurried bentonite. Temperatures were in the range 30 to 100 °C
with experiment durations over two years. Some experiments were performed in a temperature gradient.
Corrosion rate data have also been obtained for steel in compacted bentonite as a function of
external chloride concentration, temperature, and pH. Typical results show that after an initial stage
of enhanced corrosion, the rate decreases to values of ~1 �m/yr (Fig. 3). These data were obtained
using steel wire samples that had been initially pickled in acid to remove residual surface oxide.
The results are interpreted as anodic control of the corrosion rate exerted by the formation of a corrosion
product film on the surface of the iron. The results obtained suggest that during anaerobic
corrosion of carbon steel, two types of corrosion films are formed; one tightly adhering to the metal
surface, and one formed by precipitation of dissolved iron on the surface of the metal. It seems that
the growth of the first layer begins when oxygen combines with metal to form an initial thin oxide
layer. Such an oxide film will grow into the metal by spontaneous formation of oxide matrix sites
on the metal side of the metal/film interface. This oxide film will thicken until an equilibrium is
established after which it will continue to grow slowly into the metal at a constant rate. After the
formation of this layer, the corrosion rate is determined by its dissolution rate. This was also confirmed
by findings that higher temperatures increase the initial corrosion rate but not after steadystate
has been established (Fig. 3). The corrosion rate is generally slightly lower in experiments
without compacted bentonite, than in comparable tests with solid bentonite present. This suggests
that the presence of the clay influences the corrosion reactions (such as the dissolution rate of the
corrosion product film) occurring on the surface of the steel. All the data (including those obtained
outside NF-PRO) clearly suggest that the anaerobic corrosion rates of iron in clay environment will
decrease to values below 1 �m/year after formation of a thin corrosion layer on the surface of metal.
4. Interaction of clay with other barrier components
4.1 Steel
Prior to NF-PRO it was thought that an iron/steel canister would corrode to produce large volumes
of iron corrosion products in situ. Stresses in excess of the sum of the lithostatic and hydrostatic
load could arise in the near-field as a result of the volume of these canister corrosion products, because
canister corrosion products have a lower density than steel. In NF-PRO, several tests have
been performed to obtain information about the corrosion products at the bentonite/steel interface,
both with and without a temperature gradient.
186

Figure 3: Example of results from anaerobic corrosion experiments on steel in bentonite slurry and
compacted bentonite at 30 ºC and 50 ºC [3].
The results of temperature-gradient experiments show that lepidocrocite (�-FeOOH) and goethite
(�-FeOOH) formed in all cases coating the bentonite, whereas close to the carbon steel magnetite
was present, except in the experiment under oxic conditions. The results suggest that the amount of
oxygen trapped in bentonite was sufficient for the initial formation of Fe(III)-oxyhydroxides which
were then transformed to magnetite. In other experiments without a temperature gradient and with
corroding iron in compacted bentonite, the oxide minerals magnetite, hematite and goethite were
identified on the surface of the corrosion coupons using laser Raman spectroscopy. The higher oxidation
state minerals may have been due to experimental artefacts (e.g. residual oxygen at the start
of the experiments). Only a thin layer of magnetite was observed on the surface of the corroded
wires in compacted bentonite, where the total surface area of the steel was greater (100 times that of
the experiments with coupons) and so the consumption of any residual oxygen would have been
faster. There was no evidence for the presence of any iron oxide or oxyhydroxide phases in the
bentonite matrix, despite the fact that a local concentration of iron in some parts of bentonite increased
up to 20 %. No discrete iron-rich clay phases were observed. This can be explained by a
very high tendency of iron (II) ions to sorb on clay edges or by the formation of meta-stable amorphous
‘gel’ precursors, which are not easy to detect by conventional analyses.
The results of corrosion experiments at 100 °C have been modelled ([4]) using the reactiontransport
code, PHREEQC, indicating that during the one-year simulation time frame, steel corrosion
leads to an increase in the iron concentration in pore water and this iron diffuses into the compacted
bentonite from the steel-bentonite interface. Magnetite is predicted to precipitate as the only
corrosion product at the steel surface. The most relevant iron retention process in bentonite is sorption
on montmorillonite edge sites, which limits the extent of iron penetration into the bentonite to
less than 5 mm from the steel-bentonite interface. However, the amount of iron sorbed on the
montmorillonite surface is very low (less than 0.0002 % of the surface available sites, which is
equivalent to 5.5×10 -4 mg Fe/100 g of bentonite). Cation exchange is also an important retention
mechanism; although retained iron by this process is half that retained by surface sorption. In this
187

case, the iron diffusion front in the bentonite can reach a distance of 10 mm from the steel-bentonite
interface, but cronstedtite is limited to precipitate close to this interface and the maximum amount
of cronstedtite predicted in this location is 0.0006 wt %.
Other modelling studies carried out under NF-PRO have emphasised the need to address reaction
kinetics to estimate the long-term degree of bentonite alteration due to interaction with iron [5]. For
example, it is likely that the sequence of alteration of bentonite by Fe-rich fluids will proceed via an
Ostwald step sequence. Although natural systems evidence is not completely analogous to waste
package corrosion scenarios, the low-temperature diagenesis of iron-rich sedimentary rocks shows
that chlorite is the common Fe-silicate in ancient sandstones, but does not occur in recent sediments
(< 1 m.a.), whereas the mixed ferrous-ferric silicates, odinite and cronstedtite occur in recent sediments,
but do not occur in ancient sediments. It may be concluded that although chlorite is the most
likely stable Fe-silicate phase in the iron-bentonite system, its formation is kinetically inhibited, and
occurs through an Ostwald step process via odinite, cronstedtite, and/or berthierine precursors.
These processes of nucleation, growth, precursor cannibalisation, and Ostwald ripening to address
the issues of the slow growth of bentonite alteration products have been incorporated into models of
bentonite alteration under NF-PRO. This, together with incorporation of processes of iron corrosion,
diffusion and sorption of Fe 2+ ions in the iron-bentonite system in the model has enabled the
extrapolation of the results of short-term corrosion experiments to the long-term [5].
Results obtained during NF-PRO have thus challenged the previous wisdom that iron corrosion
would lead to the development of thick corrosion product layers and accumulation of iron in situ. It
is clear from experiments with compacted bentonite conducted under NF-PRO that this is not the
case over experimental timescales, in which only thin corrosion product layers develop, with iron
diffusing and sorbing readily through the bentonite, driven by the concentration gradient between
the internal and external boundary of the bentonite.
4.2 Cement and Concrete
18-month duration batch experiments at 30 and 60 °C showed that the main process observed during
contact of FEBEX bentonite with pH 13.2-13.5 pore fluids was the dissolution of montmorillonite
[6]. Calculated dissolution rates are similar to those measured in other work (10 -12 to 10 -13 mol
m -2 s -1 ). The precipitation of secondary minerals, such as zeolites (K-chabazite and K-phillipsite),
and a mixture of Mg-rich mica (celadonite) and other smectite-type clays was also observed. It has
to be stressed that the main reaction is the formation of K-zeolite and not the conversion of smectite
to mica (illitisation). Batch tests of bentonite interacting with a pH 11 solution showed negligible
reactivity. Calcium silicate hydrate minerals (CSH) were not detected, but equilibrium calculations
based on the speciation of the experimental pore fluids were consistent with the achievement of an
equilibrium state between montmorillonite and a CSH-gel.
Diffusion experiments with cement pore fluids produced a mineralogical alteration of approximately
2.5 mm in the bentonite, but only in those experiments using young cement-derived water
(pH=13.5). This alteration zone was a mixture of poorly ordered Mg-rich minerals (brucite, hydrotalcite
and tri-octahedral Mg-smectite) (Fig. 4). The alteration zone observed in the diffusion experiment
did not evolve with time, mainly due to the reduction of porosity that led to a sharp decrease
in diffusivity. The results of these experiments were successfully modelled with the reaction
transport code, Raiden-3 [7], exhibiting all the main characteristics of the experiment, including
pore blocking, brucite precipitation, minor montmorillonite dissolution, and ion exchange of Mg-
by K-montmorillonite throughout the length of the bentonite.
188

Figure 4: SEM image of the alteration zone from a cement-bentonite diffusion experiment showing
the presence of brucite and tri-octahedral Mg-smectite [6].
4.3 Saline groundwaters
The interaction of waters of different salinities with MX-80 bentonite has minor effects on the mineralogical
composition of bentonite [8], with only minor amounts of montmorillonite being dissolved
leading to the precipitation of pyrophyllite. However, the most relevant changes occur in the
octahedral layer of smectite, where Mg is replaced by Al. This replacement has two effects: 1) a
decrease of the interlayer charge; and 2) an increase in Mg in solution leads to the replacement of
Na by Mg in the interlayer space. The swelling pressure of bentonite is expected to depend on the
ionic strength of pore water, as already identified by the results of other experiments prior to NF-
PRO. This was also recognised in NF-PRO, where experiments conducted with MX-80 bentonite
showed that swelling pressure decreases as ionic strength increases [8]. Moreover, a clear correlation
exists between interlayer charge and swelling pressure. In pure water, swelling pressure increases
with increasing interlayer charge, whereas in low salinity solutions, swelling pressure decreases
with interlayer charge (water uptake capacity is lower for multivalent cations than for Na).
Finally, in high salinity solutions, swelling pressure is subject to an initial increase, and then a decrease.
This different behaviour depends upon specific mineralogical transformations.
5. Radionuclide Transport
An experimental programme was designed in order to contribute to an improved understanding of
the basic processes underlying diffusion/retention of charged and uncharged radionuclides in compacted
bentonite. This programme included studies on pure clay minerals (montmorillonite, illite,
kaolinite) and bentonite in both dispersed and compacted form. From the behaviour of the radionuclides
in the pure clay systems, the behaviour in the complex systems was evaluated (‘bottom-up
approach’).
189

5.1 Sorption
As an example of the type of work carried out, one component of the sorption studies focused on
the influence of carbonate on the sorption of radionuclides. Carbonate is the main inorganic ligand
present in bentonite pore water and is known to form soluble complexes. Experiments were performed
to determine the sorption of Ni(II), Eu(III), and U(VI) in montmorillonite (SWy-1) and bentonite
(MX-80) as a function of carbonate concentration and pH ([9]; [10]) in a pH range from 7 to
9.5. The experiments were performed by contacting a NaHCO3 solution with a suspension of purified
Na-montmorillonite, the main mineral present in bentonite. The experimental data indicate that
Ni(II) sorption onto montmorillonite is rather insensitive to the presence of inorganic carbon at levels
up to 20 mM and pH values below 9, within the experimental uncertainties associated with the
measurements. Only at very high inorganic carbon concentrations (0.1 M) was a more pronounced
effect on sorption observable. On the other hand, a clear effect of inorganic carbon on Eu(III) and
an even more pronounced effect on U(VI) sorption on montmorillonite was obtained. Model predictions
were carried out using the 2SPNE SC/CE model [11] with the assumption that metal carbonate
complexes do not sorb. In the case of Ni(II), the data and the model predictions agree within
the uncertainty of the data. For Eu(III) and U(VI), the model prediction underestimate the measured
data.
The sorption of cations on dispersed and compacted bentonite in synthetic pore water was studied.
The sorption of 22 Na + and 85 Sr 2+ on dispersed and compacted material was found to be very similar,
indicating that compacting the material does not result in a decreased accessibility of the ion exchange
sorption sites.
5.2 Diffusion
A set of through-diffusion and out-diffusion studies with 36 Cl - [12]; 22 Na + and 85 Sr 2+ [13] and 134 Cs +
have been performed with Volclay KWK bentonite. In the case of 36 Cl - both the dry density (�) of
the bentonite (� = 1300–1900 kg m -3 ) and the composition of the background electrolyte (I = 0.01-
1M NaCl) were varied. In addition, mass balance calculations for 36 Cl - and stable Cl - were made.
The results show that anions are excluded from the interlamellar space of the montmorillonite, so
that diffusion takes place in the interparticle pore space of the bentonite. Due to external negative
charges, anions are also partially excluded from this interparticle pore space. The extent of exclusion
depends on the ionic strength of the pore water. Furthermore, the interparticle pore space also
depends on the bulk density of the bentonite. There is a relationship between the effective diffusion
coefficient of anions, De A [m 2 s -1 ] and the diffusion accessible porosity, � [-], given by:
D A A m
e Dw
where Dw A is the diffusion coefficient of anion A in bulk water and m is a constant, depending on
the porous medium (Fig. 5, right). This relationship covers results obtained by work performed in
NF-PRO [14], earlier work on MX-80 and FEBEX (Ca) bentonite. Because the effective porosity
depends on the ionic strength of the pore water and the bulk dry density of the bentonite, also the
effective diffusion coefficient depends on the pore water composition and bulk dry density. At high
ionic strength, diffusion becomes faster. Dilution of the pore water results in a decrease of diffusion
of anions in the bentonite. At high bulk dry density, diffusion is slow whereas at low bulk dry
density, diffusion is faster. In the case of the cations, only the bulk dry density was varied. The
solution used was a synthetic pore water [12]. In the case of cations, the situation is different.
Studies on the diffusion of cations that adsorb via an ion exchange process on montmorillonite
190

( 22 Na + , 85 Sr 2+ , 134 Cs + ) showed that these cations could migrate both in the interlamellar and the interparticle
pore space.
D e [m 2 s -1 ]
10 -9
10 -10
Na-montmorillonite
Na-bentonite
10-11 0.01 0.1 1
I [M]
22 Na +
D e [m 2 s -1 ]
191
10 -10
10 -11
Na-montmorillonite
Na-bentonite
0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Figure 5: Dependence of the effective diffusion coefficient of 22 Na + and 85 Sr 2+ for Na-bentonite and
Na-montmorillonite (� = 1,900 kg m -3 ) on the ionic strength of the contacting solutions. Data for
Na-montmorillonite taken from [15].
The distribution of the cations between these two pore spaces depends on the chemical composition
of the pore water. Studies on pure Na-montmorillonite [15] showed that at high ionic strength, diffusion
occurred more via the interparticle pore space. At low ionic strength, diffusion takes mainly
place in the interlamellar space. Because the concentration gradient in the interlamellar pore space
is much higher as compared to the gradient in the interparticle pore space, diffusion in the interlamellar
pore space results in much higher diffusive fluxes. Dilution of the pore water thus leads to
an increase of the diffusive transport of cations in bentonite, whereas concentrating the pore water
slows down diffusion. This is the opposite effect to that observed for anions.
The diffusion results obtained for 22 Na + and 85 Sr 2+ on bentonite for a given composition of the pore
water are in good agreement with the data obtained for pure montmorillonite (Figure 5). This supports
the view that montmorillonite determines the transport of these cations in bentonite. Both
22 Na + and 85 Sr 2+ diffuse mainly through the interlamellar space of the montmorillonite.
6. Conclusions
The NF-PRO Project has made significant improvements to knowledge in many aspects of relevance
to the geochemical evolution of the EBS, most notably with regard to: direct measurement of
key geochemical parameters in compacted bentonite; understanding the redistribution of chemical
components during re-saturation of compacted bentonite under heating; the long-term corrosion
rates of steel under anoxic conditions; the alteration of bentonite due to interaction with other barrier
components; and radionuclide sorption and diffusion in compacted bentonite. Moreover, the
knowledge gained in some of the processes addressed through different experiments, resulted in the
identification of new issues that could be relevant for the long-term evolution of the near field, such
as: clarification of the different processes involved in the retention of iron in bentonite during steel
I [M]
85 Sr 2+

corrosion; up-scaling of clay alteration processes; and an extension of advances in knowledge of
sorption and diffusion to other radionuclides of interest to performance assessment.
7. Acknowledgements
Quintessa are grateful for supporting funding from the UK Nuclear Decommissioning Authority
(Radioactive Waste Management Directorate) to help prepare this paper.
References
[1] Eh and pH in compacted MX-80 bentonite. Muurinen, A. and Carlsson, T., NF-PRO RTD2
Deliverable 2.2.14 European Commission, Brussels, Belgium, 2007.
[2] Villar, M.V., Fernández, A.M. and Gómez-Espina, R., NF-PRO RTD2 Deliverable 2.2.7
European Commission, Brussels, Belgium, 2006.
[3] Experimental studies of the interactions between anaerobically corroding iron and bentonite.
Carlson, L., Karnland, O., Oversby, V.M., Rance, A., Smart, N., Snellman, M., Vähänen, M.
and Werme, L.O., Physics and Chemistry of the Earth, 32, 334-345, 2007.
[4] Geochemical evolution of near field. Process modelling. Arcos, D., Grandia, F. and Tremosa,
J., NF-PRO RTD2 Report D2.6.6, European Commission, 2007.
[5] Modelling iron-bentonite interactions. Savage, D., Watson, C., Benbow, S. and Wilson, J.,
Applied Clay Science, in press.
[6] Reactivity/transport investigations on the detailed understanding of the chemical reactions at
the bentonite-concrete interface. Cuevas, J., Fernández, R., Vigil, R., Rodríguez, M., Leguey,
S. and Cuñado, M.A., EC NF-PRO deliverable D2.4.5, European Commission, 2007.
[7] The Diffusion of Cementitious Water in Bentonite: A Raiden 3 Simulation. NF-PRO RTD
Component 2, WP 2.6. Watson, C.E., Hane, K., Savage, D. and Benbow, S., Quintessa Report
QRS-1137AB-1, Quintessa Limited, Henley-on-Thames, UK, 2005.
[8] Summary of GRS results in NF-PRO. Herbert, H.-J., Kasbohm, J. and Sprenger, H., NF-PRO
RTD2 Deliverable D2.2.4 and D2.4.1 European Commission, Brussels, Belgium, 2007.
[9] Effect of carbonate on the sorption of Ni(II), U(VI) and Eu(III) on montmorillonite: Methodology
and experimental results for Ni(II). Bradbury, M.H. and Baeyens, B., NF-PRO RTD2
Deliverable D2.5.1, European Commission, Brussels, Belgium, 2005.
[10] Report on sorption measurements. Bradbury, M.H. and Baeyens, B., NF-PRO RTD2 Deliverable
D2.5.14, European Commission, Brussels, Belgium, 2006.
[11] A mechanistic description of Ni and Zn sorption on Na-montmorillonite. Part II: Modelling.
Bradbury, M.H. and Baeyens, B., Journal of Contaminant Hydrology, 27, 223-248, 1997.
[12] Diffusion and retention of radionuclides in compacted bentonite. Part I: Determination of
pore water composition of compacted bentonite for different bulk dry densities. Van Loon,
L.R., Müller, W., Glaus, M.A., Baeyens, B. and Bradbury, M.H., NF-PRO RTD2 Deliverable
2.5.4 European Commission, Brussels, Belgium, 2005.
[13] Diffusion and retention of radionuclides in compacted bentonite: Part II: Diffusion and retention
of 22 Na + and 85 Sr 2+ in compacted bentonite at different bulk dry densities. Van Loon, L.R.,
NF-PRO RTD2 Deliverable 2.5.13 European Commission, Brussels, Belgium, 2006.
[14] Anion exclusion effects in compacted bentonites: towards a better understanding of anion diffusion.
Van Loon, L.R., Glaus, M.A. and Müller, W., Applied Geochemistry, 22, 2536-2552,
2007.
[15] Diffusion of 22 Na and 85 Sr in Montmorillonite: Evidence of interlayer diffusion being the
dominant pathway at high compaction. Glaus, M.A., Baeyens, B., Bradbury, M.H., Jakob, A.,
Van Loon, L.R. and Yaroshchuck, A., Environmental Science and Technology, 41, 478-485,
2007.
192

Impact of Thermo-Hydro-Mechanical Processes on Repository Performance
Summary
Patrik Sellin 1 , Hans-Joachim Alheid 2
1 SKB, Sweden
2 BGR, Germany
The purpose of the RTD C3 in the NF-Pro project was to improve the degree of understanding
of the thermo-hydro-mechanical and geochemical (THM(C)) processes in the near-field in an
integrated way to enhance the predictive capability of the existing models, especially in relation
to their link with safety functions and the assessment of long-term safety. The focus was
put on a few selected THM(C) processes:
The saturation of a bentonite buffer
Gas migration in/through a bentonite buffer
Compaction properties of a crushed salt backfill
The main focus of the work was to improve the handling these processes in long term performance
assessment, but issues on EBS manufacturing has also been addressed.
1. Introduction
The primary aim of the engineered barriers is to isolate the radioactive waste from the rock and particularly
from the groundwater. Buffers and backfills (of some disposal concepts) will be exposed to
high temperature, significant temperature gradients, and pressure exerted by the canisters and by the
convergence of the emplacement drift. These factors combine to create conditions that determine
the short- and long-term performance and chemical stability and form the basis of the design of the
buffer/backfill with respect to composition, density and dimensions.
A number of projects and large scale experiments carried out prior to NF-Pro (for example FEBEX,
PROTOTYPE, OPHELIE, BAMBUS) have provided very valuable information and a significant
amount of data that is being used for the validation of THM numerical models. However the duration
of these tests has been yet relatively short compared with the slow evolution of the processes
involved. Also a number of questions remained in relation with some aspects such as the long-term
evolution of the bentonite properties and the effects of gas generation in a clay buffer. In the case of
buffers made with compacted salt grit, there are still open questions concerning the compaction
process at low porosities and development of permeability in the buffer. Clay-salt mixtures may
help to guarantee an instant sealing of the buffer even at times when compaction is not yet completed
but the THMC behaviour of such mixtures were not well analysed. Joints between blocks
and with the rock are possible pathways that also may have to be considered.
For these reasons, a comprehensive analysis of the THM (C) processes in the EBS have been carried
out in RTD Component 3 in NF-Pro, considering different types of buffer material (bentonite,
salt), different scales (from laboratory to repository scale), and different degrees of saturation and
thermal states. This paper is a brief summary of some of the activities and conclusions from the
work within the RTDC [1].
2. Objectives
193

Bentonite barriers are very important for disposal concepts, especially in granite, and its long term
evolution and final state need to be predicted in order to give credit to them as radionuclide transport
barriers. Water uptake leads to the swelling of the bentonite, sealing the gaps between the bentonite
blocks, and its extrusion into the small fractures intersected by the disposal drift. After the
initial THM transient the bentonite becomes a relatively homogeneous fully saturated barrier with
very low permeability that ensures that radionuclide transport through it is controlled by diffusion.
In addition the swelling pressure of the saturated bentonite ensures that the canister remains in position
without sinking. The swelling pressure also ensures that the buffer has a self-sealing capability.
For a safety case the transient THM phase is of interest in order to obtain good estimates of the
thermal evolution of the near and far field and the hydration of the bentonite barrier. It also defines
the conditions in the EBS for the long-term evolution when canister may fail and radionuclide
transport may occur.
A key purpose of the buffer is to serve as a diffusive barrier between the canister and the groundwater
in the rock. An important performance requirement on the buffer material is to not cause any
harm to the other barriers. Gas build-up from corrosion of canister iron could potentially affect the
buffer performance in four ways:
1. Permanent pathways in the buffer could form at gas break-through. This could potentially lead
to a loss of the diffusive barrier.
2. If the buffer does not let the gas through, the pressure could lead to mechanical damage of the
other barriers. The main concern is damages to the near field rock and the buffer itself.
3. The gas could dehydrate the buffer.
4. A gas phase could push water with radionuclides through the buffer along gas-generated pathways.
The expected evolution of a HLW/ SF repository in salt is the achievement of a tight inclusion of all
wastes within a time period of about a century. Therefore, the normal evolution scenario will not
lead to any release at all. Nevertheless, it cannot be totally excluded that some brine penetrates to
the wastes in an early phase, either through an improper seal or coming from an undetected brine
inclusion in the salt rock. Such disturbed evolution scenarios, unlikely though they may be, are the
subject of most PA calculations for repositories in rock salt.
The purpose of the RTD C3 in the NF-Pro project was to improve the degree of understanding of
the thermo-hydro-mechanical (THM(C)) processes described above in an integrated way to enhance
the predictive capability of the existing models, especially in relation to their link with safety functions
and the assessment of long-term safety.
The main focus of the work was to improve the handling these processes in long term performance
assessment, but issues related to the manufacturing of the EBS has also been addressed .
3. Treatment in performance assessment
The safety assessment of a repository system relies on the safety functions performed by the barriers
of the system, both in near and in the far field. As there is no practical way to control the evolution
of the repository system after closure, it is necessary to construct and to operate the repository
system in a way that ensure that the evolution of the system does not impair the performance of the
required safety functions. This means that it is necessary to have a sufficient understanding of the
processes which control the evolution of the system, to be able to anticipate that the variables of
each barrier are and remain in the ranges for which they can perform their assigned safety functions,
for the relevant time frames [2].
194

In the safety assessment the THM(C) evolution is usually not considered explicitly in the quantification
of radionuclide transport. During the THM transient the canisters provide complete containment,
so that radionuclide releases could only start after near field conditions approach a steady
state and the other near field barriers are fully available to perform their assigned safety functions.
The former fundamental assumptions require that the safety assessment be adequately supported by
a THM(C) analysis.
For repositories in salt rock the near field is expected to remain unsaturated in all the time frames of
the safety assessments. Nevertheless, in some altered scenarios, groundwater could ingress into the
near field and later be expelled by the convergence of the salt rock, providing a mechanism for the
transport of the contaminants. For these repository concepts, the THM evolution of the repository
system has to allow the compaction of the backfill and the near field salt rock so that the permeability
of these barriers decrease in due time to the required low values.
In all repository concepts considered in NF-PRO, the generation of gases in the near field is expected.
The most important volumes of gases would originate in the reaction between metals present
in the near field (i.e. canister parts) and water in the reducing conditions prevailing in the near
field after closure. In the safety assessment it is necessary to verify that gas pressures do not reach
values which may put at risk the integrity of the repository barriers; furthermore, the overpressure
due to gases should not increase the transport of contaminants, either dissolved in groundwater or in
a gas phase, above acceptable levels. Both gas generation and gas transport on the one hand, and the
behaviour of the repository barriers against the gases, are dependent on the THM(C) conditions
In summary, the main objectives for THM(C) analyses in integrated safety assessment are:
- To provide confidence that THM(C) processes do not raise undue uncertainties on the performance
of the safety functions in the appropriate time frames.
- To define the near field THM(C) environmental conditions for other analyses needed to support
the safety assessment (i.e. canister lifetime).
At a general level, THM(C) analysis is important in particular to support the following basic assumptions
done in PA:
- The buffer and/or the backfill act as a barrier for radionuclide transport when required
- The properties of the repository barriers are not impaired during the transient thermal phase
(i.e. the eventual alterations are not critical regarding the barrier functions).
- The interactions between the different engineered barriers and between the EBS and the host
rock are not detrimental to each other.
4. Activities
To study the saturation process in a bentonite buffer, laboratory tests have been carried out with the
following specific objectives:
To evaluate the parameters and processes influencing the hydration kinetics in the clay barrier
and to understand it, especially with respect to the effect of thermal gradient and its long-term
evolution.
To improve the understanding of water flow under low hydraulic gradients similar to those
expected towards the end of the saturation process, determining the existence of threshold or
critical gradients for different bentonite dry densities.
To investigate the movement of moisture –liquid and vapour fluxes– within compacted bentonite
under both thermal and hydraulic gradients. This is of particular importance in assessment
of the advective movement of chemical species and the potential build-up of concentrations
in the region close to the canister during the heating phase.
195

To determine the impact of temperature on the hydro-mechanical properties of the bentonite
and the reversibility of the modifications observed.
Laboratory work has also been done to optimize both the design of the single hole granite probes
and the interpretation of the TDR measurements made in the FEBEX in-situ tests.
Tests were done on small cylindrical specimens to investigate the effect of the hydraulic gradient on
the permeability of bentonite. The results from the test with low gradients could indicate if there
exists a critical gradient for the bentonite. The critical gradient is the hydraulic gradient below
which flow occurs but it is not Darcian. The possible threshold hydraulic gradient depends on the
dry density and the injection pressures applied.
FEBEX bentonite compacted to dry density 1.65 g/cm 3 with hygroscopic water content was hydrated
with granitic water in 40-cm long cylindrical cells under isothermal conditions and under
thermal gradient for five years to investigate the effect of the thermal gradient on hydration. The
initial saturation of compacted bentonite takes place quicker under thermal gradient than at laboratory
temperature. Afterwards, the water intake is higher for the sample tested under room temperature.
In both tests, the 10 cm of bentonite closest to the hydration surface seem to have reached
steady-state conditions with respect to relative humidity (Figure 1).
Relative humidity (%)
100
90
80
70
60
50
40
30
0 10000 20000 30000 40000 50000
Time (hours)
RH1
RH2
RH3
Relative humidity (%)
196
100
90
80
70
60
50
40
RH1
RH2
RH3
30
0 10000 20000 30000 40000 50000
Time (hours)
Figure 1: Evolution of relative humidity in the isothermal test (left) and the test performed under
thermal gradient (right) during infiltration (sensor 1 placed 30 cm from the bottom, sensor 2 at 20
cm and sensor 3 at 10 cm). The thick vertical lines indicate periods of failure
A heating/hydration test in a 60-cm long bentonite column was dismantled after 7.6 years operation.
The final average degree of water saturation was 92 percent, and an important gradient of water
content and dry density was originated along the column, what indicates that the state analysed is
still a transient one. Those gradients condition the hydro-mechanical properties of the bentonite,
leading to an inhomogeneous distribution of swelling capacity which nevertheless, after the long
treatment, still remains in values close to those expected for the untreated bentonite.
Systematic small scale tests to study the movement of moisture have been performed on MX-80
bentonite. In these tests a number of different thermal and thermal-hydraulic gradients have been
applied across the samples and transient measurements of moisture content, temperature, dry density
and ion concentrations have been made. Figure 2 illustrates chemical concentration variations
for a thermal gradient test. The movement of moisture due to drying processes and chemical ions
due to the so called heat pipe process can be clearly observed.

Distance from heater surface (mm)
100
90
80
70
60
50
40
30
20
10
0
0.00E+00 1.00E-03 2.00E-03 3.00E-03 4.00E-03 5.00E-03 6.00E-03 7.00E-03 8.00E-03
Concentration (mol/kg)
197
Initial
1 Day
3 Day
7 Day
15 day
30 Day
Figure 2: Chloride distribution of T-test (initial degree of saturation 65 %)
Tests were also done to investigate the effect of temperature on HM-properties. In one set of tests
FEBEX bentonite was compacted at dry densities 1.50 and 1.60 g/cm3 and saturated with deionised
water under vertical stresses from 0.1 to 3.0 MPa at temperatures from 30 to 80°C while the vertical
strains were measured. In another FEBEX bentonite compacted at dry density 1.50, 1.60 and 1.70
g/cm3 was saturated with deionised water under constant volume conditions and temperatures from
20 to 80°C while the swelling stresses developed were measured. Afterwards permeability was determined
in the same specimens by imposing hydraulic gradients to the saturated samples. The
swelling under load tests have shown that the effect of temperature on the swelling capacity is
smaller than the effect of the vertical load applied during hydration or the effect of initial dry density.
The Febex Mock-up is an experiment at almost full scale and under controlled boundary conditions.
The aim is to know and understand the long-term behaviour of a clay barrier submitted to thermal
and hydraulic gradients, and to validate and verify the near field THM models under controlled
boundary conditions. The mock-up test surpasses the space-scale limitation of the laboratory tests,
by adoption of the actual dimensions of the repository, but it does not prevent the time-scale limitation.
The duration of the test, related to the operative life of the repository, makes it possible to extrapolate
the future behaviour of the clay barrier from the experimental transient state [3] (ENRESA
1997). The final evaluation will be made after the dismantling. However, the test is stilla a valuable
instrument to understand and evaluate the behaviour in the near field, and the generated data-base
has value in itself. In particular, it has verified most of the hypothesis on the THM processes in the
transient phase of the clay barrier, especially in the presence of water vapour.
The Febex “in situ” test consisted of a full-scale simulation of a HLW disposal facility, in accordance
with the ENRESA AGP Granito (Deep Geological Disposal, Granite) reference concept. The
test comprises two electrical heaters, of dimensions and weight equivalent to those of the canisters
in the concept, in a 2.28 m diameter drift specifically selected and excavated in granite. The entire
space surrounding the heaters is filled with blocks of compacted bentonite to complete the 17.4 m
of barrier for the test section. This test zone was closed with a concrete plug. The test was installed

in the underground laboratory in Grimsel (Switzerland). The hydration of the buffer has progressed
more or less as expected although the rate of hydration became slower than initially predicted by
THM models after approximately 3 years. NF-PRO has extended the in-situ test data base for three
years more, which helped to assert the previous observations as well as the comparison with the
computational model for a longer period of time. However the information provided is still limited
for the judgement the long-term performance of the clay buffer due to its slow evolution.
The modelling within the component has been carried out of several groups with different objectives
using different modelling tools. The focus has been on:
Analysis of the FEBEX in-situ test
Analysis of the FEBEX mock-up
Addition of new processes into existing models
Analysis of moisture movement in TH-tests
THM-analysis of a deposition hole for spent fuel
The experimental work on gas migration in bentonite has been focussed around the Lasgit experiment
in the Äspö Hard Rock Laboratory. Lasgit uses a full scale KBS-3 deposition hole A full-scale
canister has been modified for the Lasgit experiment with twelve circular filters of varying dimensions
located on it’s surface to provide point sources for gas injection, mimicking potential canister
defects. These filters can also be used for hydraulic testing and to inject water during the hydration
phase. A large scale bentonite experiment means that a substantial amount a time needs to be devoted
to the saturation. Despite this, during NF-Pro a preliminary hydraulic test and a preliminary
gas injection test have been performed in the Lasgit. A preliminary model has also been developed
to understand how the fluid flow is coupled to the deformations that may take place in the Lasgit
experiment.
The compaction and permeability behaviour of crushed salt/bentonite mixtures have been studied
experimentally both at room and elevated temperatures through:
Principal compaction behaviour with strain controlled oedometer tests (85/15 and 90/10 mixtures
at T = 70° and 100 °C).
Advanced knowledges of the spatial state of stress with triaxial compaction tests (85/15 and
90/10 mixtures at T = 50° and 100 °C, 85/15 mixture at T = 70°C).
Porosity/permeability behaviour with combined compaction/permeability tests at room temperature
(80/20 mixtures with a maximum grain size diameter of 8 and 16 mm, respectively, at
T = ~32 °C).
Analysis with respect to earlier results.
The permeability of 80/20 or 85/15 mixtures of crushed salt and bentonite is reduced at room temperature
in comparison with pure crushed salt up to four orders of magnitude for equal void ratios.
At 90/10 mixtures, the reduction is only two orders of magnitude.
The laboratory investigations on the behaviour of pre-compacted salt bricks were focused on the
following issues:
Triaxial compression and permeability tests at different confining pressures on highly precompacted
crushed salt samples.
Gas injection tests on fluid saturated salt bricks with respect to 2-phase properties (e.g. permeability
and threshold pressure).
Long term hydrostatic compaction tests with and without added brine, respectively.
198

Shear-tests at increasing normal pressure between highly pre-compacted crushed salt blocks
and host rock (rock salt) including wetting with brine.
The load bearing capacity and dilatancy behaviour of the saltbricks (‘dry’ state) differ significantly
from the behaviour of the compact natural rock salt due to reachable larger deformations and relatively
high load bearing capacities whereby local dilatancy is overlapped by the decrease of porosity
during loading. It is important to note that the strength of the salt bricks is drastically reduced
when moisture is present. Hydrostatic compaction experiments clearly demonstrate that besides the
loading conditions the water content is the key factor for the compaction processes.
Shear tests were performed to investigate contact properties between salt brick surfaces and the
rock salt. Whereas at dry conditions only some friction occurs, significant strengthening is observed
when moisture is present because of activation of cohesion.
The development of a microphysical basis for the constitutive models of the compaction behaviour
and the evolution of transport properties of crushed salt is considered crucial for extrapolating results
from laboratory to in-situ conditions. In NF-Pro special attention was paid to the (controversial)
role of water [4]. An oedometer compaction cell with about 20 mm diameter was used for testing
sieved granular salt samples using the stress relaxation method. This allowed compaction creep
rate vs. axial stress data to be acquired for an individual sample at a range of near-constant porosities.
Stress relaxation data obtained for dried salt samples show that the compaction creep rate
measured at a given porosity is highly
18
sensitive to applied stress ( ε ∝ σ
Figure 3. Compaction creep data for granular salt tested in
presence of saturated NaCl solution as pore fluid.
199
� ) but
insensitive to grain size. If exposed to
the lab air, from which 25-50 ppm of
water is typically adsorbed by dry
granular salt, the samples become
considerably weaker, i.e. faster creep
rates are observed at a given stress and
porosity. When tested in the presence of
saturated NaCl solution as pore fluid,
the samples show much higher
compaction creep strain rates than airdry
samples (Figure 3). The creep rate
also becomes strongly sensitive to grain
size across the whole region of stress
investigated. It is evident from the
experiments, that dislocation
glide/creep and pressure solution both
operate in the samples, with the former being most important in truly dry samples and the latter becoming
important in brine-bearing samples at low stresses and fine grain size. However, an additional
water-sensitive process also seems to operate in lab-dry and wet samples.
5. Conclusions
The studies of the saturation of a bentonite buffer point to the existence of a threshold gradient for
water flow in bentonite. When the effect of a threshold gradient is included in the modelling of the
barrier behaviour, the trend with a very low hydration in hydration tests under thermal gradient can
be well reproduced, but the inclusion of this process makes the model underestimate the hydration

under isothermal conditions. If the existence of a threshold hydraulic gradient will hinder the full
saturation of the barrier is still open question.
The laboratory, the mock-up and the in situ tests results have proven that the thermal gradient delays
saturation and that it has a major effect on water distribution inside the bentonite. To explain
this observation, several physical processes have been tested, among which the thermo-osmosis effects.
The inclusion of this process in the model improves the predictions of saturation in the hotter
areas but not near the hydration front.
The mock-up and the in situ tests have provided long-term (up to 10 years) data bases on the evolution
of the thermal, hydraulic and mechanical parameters of the clay barrier, information that is still
very limited for judging the long-term performance of the clay barrier due to its slow evolution.
Also, a data base on water intake and distribution inside the clay for different periods of time, as
well as on porosity and geochemical changes has been obtained for several laboratory tests. The
state of the bentonite after 7.6 years of being submitted to barrier conditions is known: no mineralogical
modifications have been observed and the swelling capacity has not been altered.
The effect of temperature on HM properties is better known and has been quantified for several
boundary conditions. Trends of behaviour and factors affecting have been revealed: the swelling
capacity decreases with temperature only when the bentonite is compacted at high densities and the
overload is high; the permeability increases with temperature approximately in the proportion expected
by water viscosity changes.
New developments have been made that have significantly improved the capabilities of the THM
formulation used in the analyses. The new developments have been proposed based on strong
physical basis. Therefore, the better description of the clay behaviour obtained using the new
mathematical formulation is not due to the simple fitting of experimental data, but it is the results of
the development of a more complete and reliable formulation.
One successful gas migration test has been performed in the Lasgit setup. The buffer in the Lasgit
experiment is very close to full saturation, but still quite far from stress equilibrium. Therefore, all
results so far should be taken with some caution. However, the work has demonstrated that it is possible
to construct and operate a full scale gas migration experiment. The obtained results also indicate
that the previous understanding of the gas migration process remains valid when the assumptions
are tested in a relevant scale. Observations from Lasgit include:
The hydraulic testing of the buffer material pre-gas injection indicates a hydraulic conductivity
of ~7·10 �14 m/s, which is comparable to values obtained from laboratory scale experiments.
The same hydraulic behaviour is observed post-gas injection, which indicates that no permanent
conductive channels are formed in the bentonite during the gas injection.
In the gas injection test, gas entry pressure was ~800 kPa, which is much lower than what is
observed in laboratory scale experiments (the most likely reason for this is that the clay is not
in stress equilibrium).
The peak gas pressure in the gas injection test was 5.7 MPa, which is slightly above the total
stress in the Lasgit experiment. This could indicate that the gas pressures in the repository
would remain at reasonable levels. However, no firm conclusions should be drawn at this
stage.
Lasgit is planned to operate for many years in the future and more gas injection tests will hopefully
give a better understanding of the gas migration process in bentonite.
200

The investigations performed in NF-PRO on salt backfills have lead to a better understanding of the
mechanical and hydraulic behaviour of the materials especially in the range of low permeability.
Different constitutive laws have been tested and it was found that the Zhang model is appropriate
for describing salt bricks. Use of the Spiers model requires some more experimental work.
It was found that highly compacted crushed salt differs in its hydraulic properties from naturally
grown rock salt, even at comparable porosities. While rock salt is practically tight, the compacted
samples always showed a measurable permeability. It seems to be the case that a network of connected
pores remains present even in highly compacted salt grit, while it does not exist in natural
rock salt.
The compaction behaviour highly depends on the moisture content of the backfill material. This is a
well-known fact, but the investigations have contributed to enhancing the understanding of this
phenomenon and quantifying it, especially at low porosities between 1% and 10%.
Bentonite as an additive to crushed salt backfill is a possibility to enhance the compactibility since
it acts as some kind of lubricant between the salt grains. Additionally, the permeability of the material
mixture is reduced in comparison with pure crushed salt at the same state of porosity.
6. Acknowledgements
This project has been co-funded by the European Commission and performed as part of the sixth
Euratom Framework Programme for nuclear research and training activities (2002-2006) under contract
FI6W-CT-2003-02389.
References
[1] Huertas F., Villar M. V., Garcia-Siñeriz J. L., Sellin P., Stührenberg D., Alheid H-J., (2007)
Thermo-hydro-mechanical (geochemical) processes in the engineered barrier system European
Commission EUR ##### EN.
[2] Johnson L., Alonso J., Plas F., Pellegrini D., Bildstein O., Van Geet M., Becker D., Sellin P.,
Cormenzana J.L., Nordman , H., Lehikoinen J., Sillen X., Weetjens E., Schnier H., Vokal A.,
Hodgkinson, D., Serres C., Norris S, Amme M., BauerA C., Mathieu G., Hautojärvi A (2008)
Understanding and Physical and Numerical Modelling of the Key Processes in the Near-Field
and their Coupling for Different Host Rocks and Repository Strategies European Commission
EUR ##### EN.
[3] ENRESA (1997). Evaluación del comportamiento y de la seguridad de un almacenamiento
geológico profundo en granito. Publicación técnica 6/97. Madrid. 179 pp.
[4] Zhang, X, Grupa, J, 2006:‘NF-Pro deliverable 3.5.7 Compaction behaviour and permeability
of low porosity compacted salt grit (dry and wet)’ NRG 21146/06.77412, 15.12.2000.
201

202

Disturbed and damaged zones around underground openings - effects induced
by construction and thermal loading
Peter Blümling 1 , Jean-François Aranyossy 2 , Lanru Jing 3 , Xiang Ling Li 4 , Paul Marschall 1 ,
Tilmann Rothfuchs 5 and Tim Vietor 1
Summary
1 Nagra, Switzerland
2 Andra, France
3 KTH, Sweden
4 EURIDICE, Belgium
5 GRS, Germany
The excavation of underground openings will lead to stress redistributions around the cavities
and may induce rock failure in their vicinity. The size of such an excavation disturbed or
damaged zone (EdZ / EDZ) and the impact on hydraulic properties is controlled by the local
(secondary) stress state, the pore pressure development and the physical properties of the host
rock. Besides the geometry of the opening itself and the excavation/support techniques used, a
significant impact on the geometry and characteristics of the EDZ is caused by the heterogeneity
of the host rock, the presence and frequency of any natural discontinuities and the potential
anisotropy of the rock mass.
Laboratory testing and large-scale in-situ mine-by experiments provide an understanding of
the time-dependent development and potential self-sealing processes of the EDZ for different
host rocks. Numerical blind predictions and back-calculations (e.g. within the EC projects
Modex-Rep, CLIPEX and NF-PRO) have increased the confidence in the detailed understanding
of the underlying processes.
After emplacement of waste in a repository and backfilling of the tunnels, the rock experiences
thermal loading. Stress and pore pressure will change and may alter the EDZ. New EC
projects (TIMODAZ, THERESA) have been initiated to extend the geoscientific data bases
for an in-depth understanding of THM coupled processes and to provide advanced modelling
capabilities for assessing the evolution of the EDZ before and after closure of the repository
structures.
1. Introduction
The designing of repositories for radioactive waste depends on host rock properties, state parameters
(e.g. stress, pore water pressure, water saturation) and the waste inventory for disposal. The repository
design has to be able to provide passive safety, which means that, even for very long timescales,
radionuclides have to be isolated or their transport retarded in such a way that guidelines and
regulations are met (e.g. dose to man). The key mechanisms of radionuclide transport and the potential
transport paths from the disposal cells into the biosphere have to be assessed in detail. The
potential transport paths comprise the porespace of the intact rock, transmissive natural fractures,
the backfilled underground structures and the immediate area around those backfilled openings.
203

Zones around such openings which exhibit altered properties are called the excavation damaged
zone (EDZ) or excavation disturbed zone (EdZ). The definition of these technical terms was provided
during the Cluster Conference “Impact of the excavation disturbed or damaged zone (EDZ)
on the performance of radioactive waste geological repositories” in Luxembourg in November 2003
by Bernier et al. (2005) [1], stating that: (1) the excavation disturbed zone (EdZ) is the zone where
only reversible mechanical and hydrological processes occur without resulting in major changes in
flow and transport properties, while (2) the excavation damaged zone (EDZ) is a zone with significant
irreversible processes and significant changes in flow and transport properties. These changes
may result in a several orders of magnitude increase in flow permeability and enhanced flow along
preferential flow paths, thus creating a significant impact on the performance of a repository. Thus,
the EDZ may be relevant with respect to the long-term safety of a repository, but not the EdZ.
The potential influence of the EDZ on safety was the reason for a number of very comprehensive
experiments in underground laboratories in different host rocks such as crystalline rocks, clays,
claystones, argillite and salt. It became clear that the creation of the EDZ during the construction
process is important, but that changes during ventilation, heat generation by the waste and the resaturation
process are also crucial for the evaluation of its impact on repository performance.
Recently, the EC supported integrated project on near-field processes (NF-PRO) was completed
allowing important conclusions [2] to be drawn regarding the creation and development of such
zones. Active EC projects within the 6 th Framework Programme – TIMODAZ (Thermal Impact on
the Damaged Zone around a Radioactive Waste Disposal in Clay Host Rocks) and THERESA
(Thermal-Hydrological-Mechanical-Chemical Processes for Application in Repository Safety) – are
underway, focusing on thermally induced changes around repository structures.
2. Methodology
The evaluation of the influence of the EDZ on repository safety requires investigation of different
stages in the lifetime of a repository:
Construction phase:
Stress redistribution and interaction with natural inhomogeneities lead to the creation of the EDZ.
Rock strength and material anisotropy control the size and seriousness of the damaged zone. Further
important factors are related to the construction process, including the applied excavation
method, the speed of excavation and the rock support measures.
Transition phase:
The ventilation of the tunnels and the associated changes in relative humidity may alter the rock.
After waste emplacement, the resaturation of the near-field (host rock and buffer) will take place; at
the same time, heat-generating waste will lead to thermal loading of the near- and far-field. In addition,
corrosion of the canisters will cause gas generation.
Long-term phase:
Human-induced disturbances of stress, pore pressure and temperature will disappear. On the other
hand, the gas overpressures due to corrosion of the canisters and degradation of organic matter
could enhance the expulsion of contaminated porewater along the backfilled tunnels and the EDZ.
Furthermore, hydro-chemical couplings in the tunnel near-field, such as redox phenomena and cementwater/rock
interactions could change the flow and transport properties of the EDZ.
204

The evaluation of EDZ behaviour in repositories is based on:
Laboratory investigations
In-situ tests in underground rock laboratories
Material constitutive laws
Conceptual modelling
Numerical modelling
2.1 Laboratory investigations
Laboratory investigations are carried out on rock samples either from boreholes mainly drilled using
the double or triple core barrel technique or from blocks collected during tunnel excavation and
conditioned with special care. Samples from clay formations need special treatment to avoid alteration
of the rock due to unloading, drying or swelling and different techniques and equipment have
been developed for this purpose.
a) b)
c)
Deviatoric
XRCT Consolidation
T° loading
XRCT
loading
1 2 4
n Permeability measurement
Figure 2.1: a and b) Examples of new triaxial hollow cylinder test equipment from ENPC (a) and
GRS (b) which allow simulation tests to be carried out with mechanical and thermal loading similar
to the evolution that will be encountered around disposal galleries for heat-emitting radioactive
waste. c) During the tests, different techniques are used to study the fracturing and the sealing/healing
processes under different THM conditions (μCT, XRCT, water/gas permeability measurements,
etc.).
After sample preparation, standard tests (e.g. uniaxial or triaxial compression tests, oedometer tests,
creep tests) are carried out, but more sophisticated tests are required for adequately simulating the
stress paths which are representative for EDZ development in a repository. Figure 2.1 shows the
205
hollow
sample
inner
packer
central
heater
outer
jacket
inlet
(gas / water)
3
axial load
axial load
hollow cylideric sample
outlet
(gas / water)
filter
extensometer
temperature
sensor
outer
heater
pressure /
volume
oil
pump

layout of cells developed by ENPC within the framework of the TIMODAZ project and by GRS
within the German programme (part of the THERESA project) which allow investigation of different
processes that are important for setting up and parameterisation of the necessary constitutive
laws.
The envisaged large-scale laboratory test of the THERESA project will be carried out in GRS' large
triaxial MTS test rig on a salt cylinder with a central borehole representing a disposal cell in a repository.
The test rig comprises a triaxial cell designed for a maximum cell pressure of 50 MPa and a maximum
axial load of 4600 kN (~70 MPa), a heating system for generating temperatures up to 150°C, a hydraulic
system for fluid supply up to 15 MPa, a data acquisition system and various instruments for
measurement and control of test parameters, such as axial/radial stress, axial/radial strain, gas/water injection
pressure, inflow and outflow and temperature. The maximum sample size is 280 mm diameter
and 700 mm length. Figure 2.1b shows a schematic representation of the test assembly for the envisaged
laboratory experiment. The relevant processes to be expected in a salt repository around a
HLW disposal cell in terms of EDZ evolution, such as disposal cell excavation, backfilling, heating
and cooling, will be simulated in this experiment. Particular attention will be given to the initiation,
development and self-sealing/healing of the EDZ in the salt sample.
2.2 In-situ tests in rock laboratories
Small-scale laboratory experiments are very helpful in providing an insight into the constitutive
laws for rocks and their parameterisation but fail to characterise rock mass behaviour which may be
controlled by natural discontinuities. Therefore, comprehensive large-scale in-situ tests are necessary
to evaluate the behaviour of the EDZ on a 1:1 scale (Fig.2.2).
A number of mine-by tests [2, 3, 4] have been carried out where investigation areas have been
highly instrumented before the tunnelling or shaft sinking process was started, allowing the detection
of changes in the near- and far-field of the construction site. The placing of the equipment was
guided by predictive model calculations which helped to identify the crucial locations for measurements.
The layout of the planned Praclay heater experiment at Mol, Belgium, is shown in Fig. 2.2 b and c.
Besides the detection of construction-induced damage, the focus of this experiment is on the timedependent
evolution of the site during heat generation in the transient phase. Temperature, pore
pressure, stress changes and displacement will help in understanding the situation when heatgenerating
waste is disposed of in the host rock [14].
3. Results
The creation of the EDZ in different host rocks is controlled by the rock properties (e.g. uniaxial
compressive strength), the stress field and the pore pressure [2, 5]. It is important to note that the
complete stress path during the construction of tunnels needs to be evaluated [5, 6] as fracture generation
and hence the EDZ may strongly depend on this. Failure around the opening may occur under
extension (spalling) due to unloading or under uniaxial / triaxial loading conditions, as shear
failure or as a combination of both. In general, the failure mode is independent of rock type but is
controlled by the relationship of the in-situ stress state and the failure envelops. Spalling is the main
failure process of crystalline rocks due to their high shear strength, while soft clay tends to show
shear failure [5]. Indurated argillaceous materials such as claystone show both failure types.
206

a)
b) c)
Figure 2.2: a) Large-scale in-situ heater test (Praclay heater experiment) to be performed in HA-
DES, Mol; b) the heater test is instrumented to monitor the THMC evolution and c) accompanied by
long-term seismic measurements to study the EDZ evolution.
3.1 EDZ in clay rock
In the national programmes in Belgium, France and Switzerland, different clay host rocks have
been evaluated, ranging from more plastic clays such as the Boom Clay to overconsolidated claystone
such as the Opalinus Clay. The latter is strongly influenced by material anisotropy, which has
a significant influence on the failure mechanisms around underground openings. Texture analysis of
the Jurassic Opalinus Clay shows that clay minerals are well aligned with the bedding planes [7].
The same applies for large carbonatic shell fragments which can be shown to be the origin of microcracks
that lead to macroscopic failure in triaxial experiments [8].
In the Mont Terrri Rock Laboratory, bedding plane slip dominates the failure mechanisms from
borehole scale to gallery scale. Figure 3.1 shows the fracture pattern around a recently overcored
101 mm borehole. After removal of the previously installed packer system, the lower part of the
borehole had been injected with dyed resin. An analysis of the collapse structures shows a distinct
sequence of fracture evolution in the borehole EDZ. Bedding plane failure is followed by shear
cracks, limiting the failure to the sides of the borehole. The latter links up to form a continuous zone
of cracks that limits the highly deformed region. The stack of plates is then pushed into the opening
and progressive bending leads to the development of cracks along their centres.
207

)
a)
e)
d)
f)
d)
e)
f)
c)
Figure 3.1: Overcoring of the strike-parallel SELFRAC [9] borehole at Mont Terri; a) conceptual
model; b) modelling result for volumetric strain from Nagra (2002)[3]; c) 300 mm diameter slice
of the overcore. The old borehole position is visible in the centre (the circular element is a section
of the resin injection pipe); d) regularly spaced fractures in the collapsed region that follow the
bedding planes; e) non-continuous cracks that limit collapsed regions cross-cut the bedding
planes; f) cracks along the central region of the collapsed zone also cross-cutting the bedding.
Sigma 1 orientation is sub-vertical.
The self-sealing behaviour which is obvious for plastic clays was also investigated for indurated
clays in laboratory tests. Here, this process is dominated by the combined impact of recompaction
and clay swelling. A self-sealing test on a large fractured Opalinus Clay sample (L/D= 600 mm /
260 mm) from the Mont Terri Rock Laboratory was performed by GRS within the framework of
NF-PRO. As a result of coring and a long storage time without sufficient confinement, the sample
was already highly damaged and desaturated before testing started (Fig. 3.2). Along the bedding
planes, multiple macro-fractures developed through the whole sample more or less parallel to its
axis. This damaged state may represent the situation in-situ near the drift wall. The large-scale sealing
test focused on examining the effect of normal stress and water resaturation on gas permeability
along fractures.
The fractured sample had a high initial permeability of 5·10 -14 m 2 , which was measured with a radial
stress of 3 MPa and an axial stress of 19 MPa. Keeping the axial stress constant, the radial stress
normal to the fractures was increased stepwise to 18 MPa, resulting in a drastic decrease in permeability
down to 10 -19 m 2 , which is around five orders of magnitude lower than the initial value.
When the radial stress was reduced again to its initial level, the measured permeability of 10 -16 m 2
was still two orders of magnitude lower than the initial value.
208

Another key factor controlling the sealing behaviour of a damaged clay rock is its swelling potential.
Clay minerals in the rock matrix will expand into interstices when wetted with water. After a
water injection phase of about two months, the gas permeability was reduced by two orders of magnitude
from 1·10 -16 m 2 to 1·10 -18 m 2 . Because of the large sample size and the relatively short water
injection time, the fractures in the sample might be only partly resaturated and resealed by swelling
clay and water. Therefore, the fractures might be easily reopened by relatively low injection pressures.
An additional stress increase from 3 to 6 MPa reduced the permeability significantly below
10 -20 m 2 .
Ka
Figure 3.2: Sealing of fractures in claystone (Opalinus Clay) by recompaction (core size: diameter
260 mm, length 600 mm).
3.2 EDZ in rock salt
Permeability K (m 2 )
1E-13
1E-14
1E-15
1E-16
1E-17
1E-18
1E-19
1E-20
0 2 4 6 8 10 12 14 16 18 20
Radial stress σ 2 = σ 3 (MPa)
During recent years, R&D in rock salt has concentrated on in-situ investigation of the excavation
damaged zone (EDZ). Three related projects were performed in the Asse salt mine. The first was
the ALOHA project in which the extent and hydraulic properties of an EDZ under the floor of a
drift were investigated [10]. The second was the EC-funded BAMBUS II project [11] which concentrated
on EDZ anisotropy and self-sealing; the third project ADDIGAS [12] within the ECfunded
NF-PRO project was completed in 2007 and investigated the effectiveness of EDZ removal
and the subsequent evolution of a new EDZ.
Depending on stress conditions and history, the EDZ can extend one to two metres into the rock salt.
Permeability can increase by several orders of magnitude up to the range of 10 -14 to 10 -13 m 2 , usually being
below 10 -20 m 2 . Figure 3.3 shows a compilation of permeability measurement results from BAM-
BUS II and ADDIGAS.
209
axial stress = 19 MPa
loading
unloading
Opalinus Clay at Mont Terri
Large sample: D=260mm, L=600mm
permeability at p=0.7MPa
permeability at p=0.5MPa
permeability at p=0.3MPa

EDZ removal by additional excavation is effective in the sense that it takes years for a new EDZ to
evolve. This is also shown in Figure 3.3, where the measurement results for the ADDIGAS site A
which were obtained 2 months and 16 months after removal of the uppermost metre of salt below
the floor agree well with the respective data obtained at depths between l and 2 m without EDZ cutoff.
The long-term evolution of the EDZ, particularly its healing or sealing, can only be assessed by
numerical simulations on the basis of models adequately validated with experimental data.
Within the THERESA project, which aims to evaluate and improve numerical modelling capabilities
for assessing the long-term evolution of the EDZ in a salt repository by adequate consideration of
the relevant thermal-hydraulic-mechanical (THM) processes, large-scale laboratory tests are envisaged
to provide specific data on EDZ healing/sealing. This will allow verification of the existing models and
codes on the basis of these data, rather than on the basis of the limited and incomplete data that currently
are available from earlier in-situ measurements.
Figure 3.3: Permeability as a function of depth below the floor in the AHE drift (BAMBUS II) and
in ADDIGAS.
The existing constitutive models for the thermal-hydraulic-mechanical (THM) behaviour of the
EDZ/EdZ in rock salt and data available from earlier laboratory testing and other in-situ measurement
results obtained within the framework of NF-PRO have been reviewed, compiled and discussed with
regard to usability for calibration of the different models by the THERESA project partners BGR,
CIMNE, DBE-TEC, FZK, GRS, IfG, NRG and TUC [13].
4. Conclusions
�
Permeability / m2
1.E-12
1.E-13
1.E-14
1.E-15
1.E-16
1.E-17
1.E-18
1.E-19
1.E-20
1.E-21
New Floor ADDIGAS A/C
1.E-22
0. 0 0.5 1.0 1.5 2.0 2.5
Depth Below Floor / m
Host rocks for radioactive waste repositories are generally selected for their ability to minimise the
transport of radionuclides. Together with specially designed engineered barriers, they provide an
effective and redundant multi-barrier system. In many concepts for the geological disposal system,
the host formation is considered as the main barrier providing the basis for the key safety functions.
Hence, preventing unnecessary damage to the host formation is one of the objectives of repository
design. A proper evaluation of the excavation damaged zone (EDZ) in the host formation is thus an
important item for the long-term safety of underground disposal.
210
AHE drift 2001
Site A 2004 - 2 months after EDZ removal
Site A 2006 - 16 months after EDZ removal
Site B 2006 - original 20 year-old floor

In any case, the excavation of the cavities of a geological repository (disposal drifts, transport galleries
and access shafts) and its later operation inevitably lead to the creation of a damaged zone
(DZ) around the engineered part of the disposal system in clay formations [1], while such zones
may not exist in hard rock under preferential stress conditions.
Numerous laboratory and in-situ tests in different host rocks and underground laboratories (e.g. in
the URLs HADES in Belgium, Meuse/Haute Marne in France, Mont Terri in Switzerland, Tournemire
in France, ÄSPÖ in Sweden and the Asse mine in Germany) resulted in adequate constitutive
laws for describing the formation of the EDZ during construction. However, EDZ development is a
dynamic problem that depends on changing conditions, varying from construction and the opendrift
period to the initial closure period and the entire heating-cooling cycle of the decaying waste
until the long-term period. It was shown that the time-dependent changes during the desaturation /
resaturation process and the effect of self-sealing in clay formations (EC FP5 project SELFRAC
[9]) are important aspects which are understood in principle.
The TIMODAZ and THERESA projects are now focusing on studying the combined effect of the
EDZ and the thermal output from the waste on the repository host rock. The influence of the temperature
increase on the evolution of the EDZ as well as the possible additional damage created by
thermal loading will be studied.
6. Acknowledgements
The work presented was carried out within several research projects (e.g. SELFRAC, NF-PRO,
TIMODAZ, THERESA) which were or are co-funded by the European Commission as part of the
5 th and 6 th EURATOM Framework Programmes.
References
[1] Bernier F., Tsang C., Davies, C. (2005): Geohydromechanical Processes in the Excavation
Damaged Zone in Crystalline Rock, Rock Salt, and Indurated and Plastic Clays. In: International
Journal of Rock Mechanics and Mining Science - Elsevier Science B.V Amsterdam,
42:01(2005), p. 109-125.- ISSN 1365-1609.
[2] Aranyossy, J.F., Mayor, J.C., Marschall, P., Plas, F., Blümling, P., Van Geet, M., Armand, G.,
Techer, I., Alheid, H.J., Rejeb, A., Pinettes, P., Balland, C., Popp. T., Rothfuchs, T., Matray,
J. M., De Craen, M., Wieczorek, K., Pudewills, A., Czaikowski, O., Hou, Z. & Fröhlich, H.
(2008): EDZ development and evolution (RTDC 4) - Final Synthesis Report to EC (RTDC 4
NF-PRO).
[3] Nagra (2002): Projekt Opalinuston – Synthese der geowissenschaftlichen Untersuchungsergebnisse,
Nagra Technical Report, NTB 02-03, Wettingen, Switzerland.
[4] Andra (2005). Dossier Argile – Référentiel du site de Meuse/Haute-Marne – Tome 2: Caractérisation
comportementale du milieu géologique sous perturbation. Rapports Andra, Châtenay-Malabry,
France.
[5] Blümling, P., Bernier, F., Lebon, P. & Martin, C.D. (2007): The Excavation-Damaged Zone
in Clay Formations – Time-dependent Behaviour and Influence on Performance Assessment,
Physics and Chemistry of the Earth 32 (2007), 588–599.
[6] Konietzky, H., Blümling, P. & teKamp, L. (2003): Opalinuston - Felsmechanische Untersuchungen,
Nagra Internal Report, Nagra, Wettingen, Switzerland.
211

[7] Wenk H.-R., Voltolini, M., Mazurek, M., Van Loon, L. R. & Vinsot, A. (2008): Preferred orientations
and anisotropy in shale: Callovo-Oxfordian shale (France) and Opalinus Clay (Switzerland),
Clays and Clay Minerals; June 2008; v. 56; no. 3; p. 285-306.
[8] Klinkenberg, M., Kaufhold, S., Dohrmann, R. & Siegesmund, S. (2007): Microstructural investigation
of Opalinus Clay – Proposal of a carbonate distribution model. – Abstract, 3rd International
Meeting: Clays in Natural & Engineered Barriers for Radioactive Waste Confinement,
ANDRA, Lille, France, p. 345.
[9] Bernier, F., Li, X.L., Bastiaens, W., Ortiz, L., Van Geet, M., Wouters, L., Frieg, B., Blümling,
P., Desrues, J., Viaggiani, G., Coll, C., Chanchole, S., De Greef, V., Hamza, R., Malinsky, L.,
Vervoort, A., Vanbrabant, Y., Debecker, B., Verstraelen, J., Govaerts, A., Wevers, M.,
Labiouse, V., Escoffier, S., Mathier, J.F., Gastaldo, L., Bühler, Ch. (2007). Fractures and selfhealing
within the excavation disturbed zone in clays (SELFRAC). Final report to EC
(contract N°: FIKW-CT2001-00182), EUR 22585.
[10] Wieczorek, K., and Zimmer, U. (1998): Untersuchungen zur Auflockerungszone um Hohlräume im
Steinsalzgebirge, Final Report, GRS-A-2651, Braunschweig, Gesellschaft für Anlagen- und Reaktorsicherheit
(GRS) mbH.
[11] Bechthold, W., Smailos, E., Heusermann, S., Bollingerfehr, W., Bazargan Sabet, B., Rothfuchs,
T., Kamlot, P., Grupa, L, Olivella, S., Hansen, F.D. (2004): Backfilling and Sealing of Underground
Repositories for Radioactive Waste in Salt (BAMBUS-II Project), EUR 20621, Commission
of the European Communities.
[12] Jockwer, N. and Wieczorek, K. (2007): Excavation Damaged Zones in Rock Salt Formations, 6 th
Conference on the Mechanical Behaviour of Salt (SaltMech6) -Understanding of THMC Processes
in Salt Rocks, Hannover, May 22 - 25, 2007.
[13] Compilation of existing constitutive models and experimental field or laboratory data for the
thermal-hydraulic-mechanical (THM) modelling of the excavation disturbed zone (EDZ) in rock
salt, THERESA project Deliverable D5, 28 August 2007.
[14] Praclay heater experiments, operational document, EURIDICE, 2005.
212

Summary
Near-Field Processes - The Challenge of Integration into
Performance Assessment
Lawrence Johnson 1 , Delphine Pellegrini 2 , Jesús Alonso 3 ,
Frédéric Plas 4 and Maarten Van Geet 5
Nagra 1 , Switzerland; IRSN 2 , France; Enresa 3 , Spain;
Andra 4 , France; SCK.CEN 5 , Belgium
Over the past decade, the EU has strongly promoted coordination of R&D in relation to nuclear
waste disposal. In the area of engineered barrier (EB) processes there have been numerous
focused projects dealing with waste form processes (SF and HLW), canister corrosion,
and bentonite behaviour, as well as system-wide process-related projects such as gas production
and transport and coupled heat and moisture transport in the EBS. The NF-PRO project
dealt with all of these aspects together in a single large project for EB systems in crystalline,
clay and salt host rock repositories. Furthermore, one component of NF-PRO focused on performance
assessment (PA), which was a daunting challenge given the diversity of systems and
EBS design concepts. The PA-related work was focused on providing context for the experimental
and modelling studies in the sense of defining expected in situ conditions and boundary
conditions, as well as giving a frame of reference for the contributions of the different
parts of the system to overall repository performance. In the end, the PA specialists were able
to bring back to their own national programmes useful data and process level models for integration
into their future integrated PA studies. In addition, the incorporation of a PA group
within the project helped to build some bridges between specialist groups. Some lessons from
the project work are outlined from the perspective of the PA Component of NF-PRO.
1. Introduction
A fundamental objective of all disposal systems for nuclear waste is to provide safety by minimizing
the release of radionuclides from the disposal system. This is done by designing a disposal system
in which the waste is completely contained for some suitable period and, following breaching
of containment, is largely retained within the disposal system as a result of slow radionuclide transport
and strong chemical retention processes. Many processes can influence, either positively or
negatively, the effectiveness of these safety functions of isolation and retardation. The objective
of the EU NF-PRO project is to improve the understanding of and the technical basis for the contribution
to these safety functions provided by the near-field system. The near field is considered to
comprise the engineered barrier system (EBS) and the region of rock immediately surrounding the
repository excavations (i.e. mainly the excavation-disturbed zone). NF-PRO builds on a strong
foundation of previous related EU projects (note the following is a partial list) in the areas of waste
form processes [1, 2], canister corrosion [3], bentonite behaviour [4, 5], and system-wide processrelated
projects such as gas production and transport [6] and safety assessment [7], in addition to
213

that provided by various national safety cases. Research as part of NF-PRO addresses the near-field
processes for repository concepts in clay, granite and salt host rocks that are currently under investigation
within the European Union and is focused on assessment of many of the most significant
safety-related issues. The NF-PRO consortium consists of 40 leading European research organizations,
waste management agencies/implementing organizations and technical safety organizations.
The structure and areas of concern are shown in Figure 1.
Fig. 1 NF-PRO’s main project components and structure
2. Complementarity – national safety cases vs. EU projects
In recent years, substantial progress has been made in the scientific understanding of key processes
affecting the overall performance of the near-field system. This has been achieved both within various
EU projects that have focused on particular components or processes associated with disposal
systems, and within national programmes where all aspects must be considered in a safety case. National
projects typically work on a single system – a specific host rock (and often a specific site),
with a particular design for the excavation and layout and for the EBS that are tailored to the rock
engineering and geochemical requirements and to the relative contribution to isolation and retention
required for the system components in question. The safety case integrates all this information and
strives for completeness, full assessment of uncertainties and identification of areas in which more
focused studies are needed for the next stage of repository licensing.
One benefit of EU projects is that they demand, in contrast, a certain level of horizontal integration
through cooperation of scientific groups across many national programs. Although there are clearly
great differences in the EB materials, the repository design and the host rock in the various programs,
there are also sufficient similarities in a number of areas that the cross-pollination is valuable.
The exposure to studies done by other groups helps to counter the inward-looking tendencies
of national safety cases. In addition, one may confront challenging new results that may bring into
214

question certain assumptions, as well as providing, eventually, a larger set of studies that aid in
reaching scientific consensus on various issues. Unlike the national safety cases, EU Projects such
as NF-PRO do not attempt to achieve completeness in the sense of covering all relevant areas at an
appropriate level of depth. Instead, the project scope is a natural outcome of the respective interests
of the organizations involved and the project must seek to develop appropriate shared goals and
also accept the inability to tackle the completeness question, which is best left to the safety cases of
the national programs that may put the NF-PRO results to further use.
One of the distinguishing features of NF-PRO also lies in the communication that has been established
in the project between groups doing detailed process investigations (the four hexagons on the
sides in Figure 1) and those doing broader assessments of disposal systems (i.e. Performance Assessment)).
Here again one sees the different roles played by those who are discipline-focused and
seek ever greater detailed understanding, which complements the perspective of those who try to
provide a balanced synthesis of information that clarifies how all the repository components and
associated processes interact.
3. NF-PRO findings from a PA perspective
The main findings of NF-PRO seen from the perspective of PA are summarised here. Specialists in
the various scientific disciplines may have different views from PA specialists regarding the significance
of various project achievements, but from the PA perspective, it is considered that both
the quality of the scientific contributions and the impact on the system performance are relevant and
these determine the emphasis presented here. It is emphasized that the summary of findings and recommendations
for future work must be considered as guidelines that need to be evaluated in the
context of the particular disposal system in question and that the absence of a comprehensive total
system PA analysis in the project (due to the NF focus), requires that other considerations may also
influence the focus and even the necessity for R&D in a given area. The details of the work are discussed
in Grambow et al. [8], Arcos et al. [9], Villar et al. [10], Aranyossy et al. [11], and Johnson
et al. [12].
4. Release of radionuclides from the waste matrix
HLW
Studies performed within NF-PRO have improved understanding of the impacts of corrosion products
and clay on the long-term dissolution rate of HLW. These studies indicate significant transient
impacts on the rate (the rate may be initially similar to the high initial rate in distilled water), but a
number of results suggest that the long-term rate will nonetheless be very low. During this transient
only a relatively small fraction of the glass would dissolve before the residual rate is attained. The
details of the transient are of interest to the extent that understanding something about the reactions
involved may help to explain better the processes that control the long-term rate.
The initial or transient glass dissolution rate (about 100 to 1000 times higher than the long-term or
residual rate) is often used in assessment calculations, where it is applied to the entire duration of
the dissolution process and is referred to as a “pessimistic” value or is used in a sensitivity analysis
context. In the context of communicating the results of safety assessments, it appears that it would
be consistent with the findings to use two dissolution phases accounting for the initial rate and then
the residual rate. Use of the initial rate for the entire duration can be considered solely as an illustration
of how the system behaves if the waste form dissolves very rapidly, i.e. as an illustration of
215

system robustness, rather than as some sort of alternative behaviour of the system that has a reasonable
expectation of occurring.
For a near field in which cementitious materials are used in significant quantities, the dissolution
rate of HLW glass is poorly known and requires further study in order to assess the radionuclide
retention capacity of HLW glass at high pH.
Regarding the relative significance of the dissolution rate of HLW glass, it has been shown in safety
assessment calculations that for clay host rock disposal systems, in the present state of knowledge,
there would be almost no difference in dose between the cases in which the initial or the residual
rates are used. This is the result of the strong retention of the radionuclides in the host rock, resulting
in substantial decay during transport. Nonetheless, depending on the degree to which an essentially
independent barrier function is considered necessary, there is an argument to be made for
strengthening the scientific basis for and reducing the uncertainty in the low long-term dissolution
rate.
SF
Over the past few years, significant evidence has accumulated that has shown that the potential oxidative
effects of alpha radiolysis may be significantly mitigated by the presence of dissolved hydrogen
and Fe(II). Although a qualitative explanation for these observations can be envisaged, quantification
of the details in terms of the rate constants is clearly difficult. In addition the minimum concentrations
of H2 or Fe(II) needed to mitigate the effects of �-radiolysis are not certain and it is possible
that other dissolved species may also play a role. It is clear that a better understanding of the
radiolytic processes at the SF interface with porewater is needed.
NF-PRO studies have reduced the uncertainties regarding the time evolution of the instantaneous
released fraction (IRF) of the major safety-relevant radionuclides for moderate burnup fuel. There
is now confidence that the different mechanisms previously identified as potentially leading to an
increase with time of the IRF (� self-irradiation enhanced diffusion and helium accumulation) are of
little relevance. The impact of helium accumulation for higher burnup UO2 and MOX fuel may still
warrant further study. Regarding measurements of the IRF, data are still lacking in particular for
high burnup and MOX fuel. More studies should be done, as present models appear to involve
somewhat pessimistic assumptions.
The relative importance of the IRF vs. the matrix dissolution rate has been explored in studies
within and outside of NF-PRO. For IRF values in the range of 5 to 10% (typical values in some assessment
studies), the IRF dominates the dose, with I-129 being by far the most important radionuclide
from the dose perspective. Nonetheless, this is based on matrix dissolution models based on
uranium solubility that estimate that the fuel dissolution process takes millions of years or more.
The assumption of a dissolution control by alpha-radiolysis and thus a dissolution time of ~100,000
years would result in a decrease in the relative contribution of the IRF but also in an increase in
dose. There is thus an important incentive to more clearly understanding the factors that can counter
alpha-radiolytic oxidative dissolution of spent fuel.
The rim region of the fuel pellets contains higher fission products concentrations (due to higher
burnup) and a different microstructure (smaller grains and large pores) compared with the rest of
the pellet. These differences raise doubts regarding the applicability of the UO2 dissolution data to
the rim region. It is necessary to clarify whether rim leaching behaviour is very different from the
216

est of the pellet and whether it can be treated in a dissolution model in the same manner as the rest
of the pellet.
5. Chemical processes in the EBS
Studies of corrosion of steel in bentonite, including in dense bentonite, show that long-term anaerobic
corrosion rates are less than 5 �m/a, consistent with other studies over the past ten years. This
long-term anaerobic corrosion rate can thus be considered well established.
Iron corrosion products may influence the properties of bentonite by forming new phases (e.g. magnetite
and Fe-silicates), and by influencing mechanical and retardation properties of bentonite.
Some studies of corrosion of steel in bentonite indicate that only a very thin film of magnetite forms
and that considerable Fe(II) is transported into the clay, possibly altering its hydraulic and swelling
properties. This represents a different picture from that envisaged based on steel corrosion studies
in solution, which suggested progressive buildup of a thick layer of magnetite. Although transport
of Fe(II) was not studied in NF-PRO, coupled reactive transport modelling of Fe-bentonite interactions
was performed and this raised a number of questions. Clearly the impacts of sorbed Fe(II) on
subsequent sorption of radionuclides should be further studied, as well as the impacts of neomineral
formation in bentonite (in particular if montmorillonite is slowly replaced by non-swelling
clay) and the rate of transport of Fe(II) through bentonite, to determine the extent of the affected
domain. There is presently substantial work going on in this area that should help to further address
these issues in the context of the safety functions of the buffer.
Electrode measurements during iron corrosion experiments carried out under NF-PRO demonstrate
that corrosion potential values are in accord with the evolution of hydrogen. Nevertheless, the
chemical effect of hydrogen gas on near-field conditions and thus on corrosion remains an uncertainty.
Future experimental work should focus on the acquisition of pore water composition data
during iron corrosion processes to help define and increase confidence in estimates of near-field
redox and pore fluid chemistry.
In addition, redox reactions on iron or steel may play an important role in retention of some redoxsensitive
radionuclides. This has been previously shown with reduction of Tc(VII) and has now
been demonstrated in NF-PRO with the observation of reduction of Se(VI) and Se(IV) on iron and
iron corrosion products (siderite and magnetite) to insoluble Se(0) and Se(-I) (as FeSe2). The solubility
of Se as a result of these reactions should be evaluated for possible incorporation into databases
for radionuclide transport.
In the sorption area, the question of the relationship between values obtained in dispersed vs. compacted
systems was answered only partly. It is still unclear why distribution coefficients (Kd) for
cesium on compacted bentonite of higher density are higher from the evaluation of diffusion curves
than from batch sorption experiments. Further research is needed to explain this discrepancy. In the
context of sorption values for radionuclides used in PA studies, there are significant differences between
the values used for the same nuclides in different PA studies for nominally very similar conditions.
Continuing effort should be put into clarifying the degree to which these arise from true scientific
uncertainties as opposed to different levels of ‘conservatism’ used in the selection of assessment
data. Considering the fundamental importance of chemical retention phenomena in all safety
cases, there remains an ongoing need for basic studies that will link spectroscopic evidence for
sorption, detailed process models (e.g. surface complexation) and empirical sorption studies (Kds)
to continue to build a strong scientific foundation.
217

Regarding diffusion studies performed in NF-PRO, the most important question with direct implications
for PA is the apparent difference in effective diffusion coefficients (De) for neutral HTO and
positively charged species, suggesting that positively charged species can move in compacted bentonite
faster than neutral HTO. This effect was identified a long time ago, and has been ascribed by
some to a transport mechanism for cations in clay materials called “surface diffusion”. In similar
experiments, some groups do measure De values for cations much higher than for HTO while others
do not. Although the differences are not large in the context of the uncertainties that they introduce
into transport and dose calculations, they indicate a lack of understanding that warrants some
further study.
Modelling of the interactions of concrete with bentonite has been further studied within NF-PRO
using pure kinetic approaches and equilibrium-based approaches. Depending on the approach, the
mineral conversion is either slight and may extend over a few metres distance or substantial and be
limited to a decimetre range. The calculated effects and extent of the alkaline perturbation within
bentonite are highly sensitive to the feedback of chemistry on transport (i.e. clogging), to the model
for mineral buffering (e.g. montmorillonite dissolution), to the diffusion rate, as well as to the position
and dimension of the alkaline reservoir (renewal of aggressive solutes by diffusion). Further
studies need to be performed to reduce the uncertainties, notably regarding kinetic and equilibrium
thermodynamic data. Overall, it would be worth considering more realistic assessment of the effect
of chemical interactions between near-field components in systems with heterogeneities, e.g. in
terms of component interfaces, including impacts of the thermal transient.
Some studies regarding the effect of an alkaline plume on the safety-relevant properties (sorption,
porosity and diffusion coefficients) of the bentonite or clay rock would be useful to put into perspective
the relevance of the alkaline plume for repository safety. If these effects are found to be
negligible (similar Kd values and fluxes) or even beneficial (clogging of porosity that may reduce
mass transport) the efforts to be spent studying the alkaline plume should be smaller than if there
appear to be strong detrimental effects on safety-relevant properties bentonite and host rock.
6. THM processes in the EBS
A number of laboratory and URL experiments at various scales involving hydration and heating of
bentonite have been performed over a period of many years and these were modelled within NF-
PRO.
The “standard THM model” was able to make good predictions for THM behaviour of bentonite in
FEBEX “mock-up” and “in-situ” experiments. Trends are well predicted but the calculated longterm
hydration rates are greater than the measured values, the main disagreement being that after 3
years, hydration of the hotter areas of the bentonite proceeds more slowly than predicted by the
model. Further model testing should be performed, as the ability of the “standard THM model” to
explain the long-term experiments results seem not to have been fully exploited. The effect of parameter
uncertainty on predictions should be thoroughly analysed, in order to explore the full capabilities
of the “standard THM model” to predict results. Furthermore, when developing the model
and adding a new process or using a different set of parameter values, the whole experimental database
should be considered, rather than a single experiment. This would allow drawing general conclusions
regarding the occurrence of any new postulated processes.
Another uncertainty relates to the degree of heterogeneity in density that may remain in bentonite in
the long term. This uncertainty regarding the final state of the barrier was already identified at the
start of NF-PRO, and NF-PRO results confirm this uncertainty, but it must be taken into account
218

that in the experiments the bentonite columns were far from full saturation. Experimental data
available on this topic should be identified and new experiments proposed to clarify the expected
range of density variations in the bentonite buffer after saturation.
It is noted that a number of THM studies involving hydration and heating (e.g. FEBEX) were in
progress long before the NF-PRO project began. The comparison of models with experimental results
from these studies has proven valuable. In some cases, reasonable agreement has been obtained,
but in other cases there appear to be significant difficulties in obtaining agreement between
experiments and models. These differences should be explored further. Nonetheless, from the assessment
perspective it is not clear how significant these problems are when one considers the
somewhat tenuous relationship between the key safety functions of buffer and the processes being
modelled in these laboratory and field studies. It is worth reinforcing that safety functions of a bentonite
buffer are related principally to permeability, swelling capacity, and sorption, which are
properties that can be quantified. If these are within a suitable range, then the safety function can be
achieved. The THM models in general do not adequately characterize these parameters – post-test
analysis is clearly also needed. Thus one must judge adequacy of barrier performance using a number
of approaches and the predictive capability of THM models is only one aspect to be considered.
For salt repositories, porosity evolution around canisters and in seals has been an important issue in
PA. Results in NF-PRO show that at relatively low pressure (10 - 20 MPa), the water content in
granular salt is high enough to allow effective compaction to occur. Experiments indicate that pressure
solution may be the dominant deformation mechanism, though the mechanistic basis and role
of recrystallization are not yet fully understood.
Investigations performed in NF-PRO have led to a better understanding of the mechanical and hydraulic
behaviour of salt backfill materials especially in the range of low permeability. Different
constitutive laws have been tested and it was found that the Zhang model is appropriate for describing
salt bricks. Use of the Spiers model requires some more experimental work.
It was found that highly compacted crushed salt differs in its hydraulic properties from naturally
grown rock salt, even at comparable porosities. While rock salt is practically impermeable, the
compacted samples always show a measurable permeability. It seems to be the case that a network
of connected pores remains present even in highly compacted salt grit, while it does not exist in
natural rock salt.
The compaction behaviour of salt highly depends on the moisture content of the backfill material.
This is a well-known fact, but the investigations have contributed to enhancing the understanding of
this phenomenon and quantifying it, especially at low porosities between 1% and 10%.
Bentonite as an additive to crushed salt backfill is a means to enhance the compactibility since it
acts as some kind of lubricant between the salt grains. Additionally, the permeability of the material
mixture is reduced in comparison with pure crushed salt at the same state of porosity. Such a mixture
can be used in low-temperature regions of the repository.
219

7. EDZ characterization and evolution in clay rock and salt
Studies within NF-PRO provide further important evidence of the occurrence of relatively rapid
self-sealing of the EDZ in clay host rocks. The hydro-mechanical self-sealing occurs irrespective of
the extent of the initial EDZ development. The kinetics depend, however, on the type and intrinsic
characteristics of the rock. For plastic clays, high water content, swelling capacity and plasticity of
clay minerals favour a quick and efficient self sealing, more rapid than in indurated clays.
Observations made on the Boom Clay seem to demonstrate that the EDZ self-sealing is accompanied
by a decrease of the overall porosity, a reduction of the connectivity of the micro-fracture network,
and consequently, a re-consolidation of the rock. Thus essentially complete self-sealing may
occur, with a total disappearance of the initial EDZ microfracture network.
For indurated clay rocks such as Opalinus and Callovo-Oxfordian Clays, the open fractures of the
EDZ are progressively closed, mainly as a result of hydro-mechanical effects, but they remain detectable.
This closing process has been shown to be sufficient in situ in Opalinus Clay and in Callovo-Oxfordian
core samples to restore a very low hydraulic permeability (close to that of the undisturbed
rock), but locally, the porosity may remain greater than that of the undisturbed rock.
Placed in the general context of the reference disposal systems, results confirm that the self-sealing
processes will start rapidly after the closure of the disposal cells, especially during the re-saturation
period. In addition to the hydro-mechanical loading, the self-sealing process will be helped by the
swelling of the clay-based (bentonite) buffer material (if included in the disposal concept), expected
before a significant chemical degradation of the metallic and /or cementitious material.
The impact of the long-term chemical interactions on the hydraulic properties of the rock and on the
solute and gas transfer through a possible remaining EDZ have not yet been fully assessed and integrated
in terms of disposal system performance. The alkaline perturbation from cementitious materials
may be significant and may extend several decimetres. It will likely contribute to a decrease in
the porosity (and permeability) of the “long term” EDZ, after the hydro-mechanical self-sealing
process (whilst possibly increasing the porosity of the concrete components). The specific characteristics
of this “new EDZ" are not precisely determined, thus the impacts on mass transport should
be further evaluated, in particular for the transport of gas from the repository system.
As noted in previous national PA studies, despite rather extreme representations of the EDZ in radionuclide
transport calculations, a continuous EDZ with a high permeability makes a negligible
radiological impact relative to the already low impact arising from transport directly through the
clay host rock. These findings have not eliminated the need for a sound scientific understanding of
EDZ-related phenomena. This is largely because PA models of the EDZ are simplified representations
of the relevant phenomena and there is a desire for a more convincing representation of these
processes, i.e. a more transparent basis for abstracting a PA model of transport through the EDZ
from a detailed scientific description of the processes. Clearly, due to the conservative assumptions
made in present PA studies, there is an expectation that more realistic representations of the EDZ in
radionuclide transport calculations will not significantly change the conclusions regarding the negligible
radiological consequences of an EDZ.
220

8. Final remarks
In relation to the safety functions (i.e. isolation, retardation) and the critical processes that contribute
to them and thus provide the foundation for safe repositories, it is clearly important to develop a
good understanding of the relevant phenomena at a fundamental level. For example, it is clear that
chemical retention by sorption on barrier materials must be well enough understood that safety assessment
models do not rely solely on empirical Kd values. Such simplified models of retention
should be supported by fundamental studies of sorption (e.g. spectroscopic evidence and detailed
chemical modelling) which can provide evidence of a sound understanding as well as a basis for
quantifying the uncertainties inherent in moving from a detailed process model to an abstracted Kd
model. This does not mean that detailed mechanistic understanding is required for every radionuclide
for all conditions, nor does it mean that detailed mechanistic models are required for all phenomena.
The extent to which the fundamentals in a given area should be better understood depends
on many factors, including the significance of the phenomenon in relation to safety functions, the
time frame of occurrence of the phenomenon in the context of repository evolution, the information
available from other lines of evidence (e.g. natural analogues), the safety margins and degree of robustness
in the disposal system and the specific scenarios that need to be considered in safety analysis.
There is clearly no recipe available for selecting the areas of work that require the most intense
focus in terms of improving fundamental understanding; however, the factors noted above all need
to be considered and the appropriate balance in the R&D must be found through collaboration between
those who can develop and evaluate models for the complete disposal system and those who
understand processes at a sufficiently detailed level that they can characterize the detailed model
and data uncertainties.
A trend that is clear over the past decade of performance assessment models is the increased emphasis
on realistic models that make both the quality of the system and of the understanding of the
system clearer. Many years ago over-conservative models were the norm, based on the idea that it is
necessary only to demonstrate that the system can give results below a dose limit, thus if this can be
done with conservative models, this should be sufficient to argue safety. However, reviewers increasingly
demand a well-grounded understanding of all the key phenomena so as to have a rigorous
basis for assessing uncertainties. Pursuing this is clearly beneficial in that it helps to clarify
what is known in relation to all processes that may influence release and transport of radionuclides
(and also illustrating which processes do not have an important influence). The demands on national
projects and on their partner research institutions accordingly increase. It also leads to a continuing
(and necessary) debate regarding how much detail needs to be known in various areas. No
simple formula can be found to answer this question. Nonetheless, the recent acceptance of several
national safety cases after rigorous review by independent scientific groups and regulators provides
evidence that the approach is effective and that the scientific basis for geological disposal is sound.
With continued effort, the residual uncertainties can be further reduced and we can move forward to
repository implementation.
References
[1] Poinssot et al. 2005: Final report of the European Project Spent Fuel Stability Under Repository
Conditions, FIKW-CT-2001-00192 SFS, CEA Report CEA-R-6093.
[2] Van Iseghem, P. 2008: A Critical Evaluation of the Dissolution Mechanisms of High-level
Waste Glasses in Conditions of Relevance for Geological Disposal (GLAMOR), FIKW-CT-
2001-00140, European Commission Report EUR 23097.
[3] Smailos, E. et al. 2004: Final Report, Long-Term Performance of Candidate Materials for
HLW/Spent Fuel Disposal Containers, FIKW-CT-2000-00004, European Commission.
221

[4] Alonso, J. et al. 2004: Bentonite barriers in Integrated Performance Asseessment (BENIPA),
FIKW-CT-2000-00015, EUR 21023.
[5] Huertas, F. et al. 2005: Final Report, FEBEX-II, Full-Scale Engineered Barriers Experiment
for a Deep Geological Repository for High-Level Waste in Crystalline Host Rock – Phase
II, EUR 21922.
[6] Rodwell, W. et al. 2003: A Thematic Network on Gas Issues of Deep Repositories for Radioactive
Waste (GASNET), FIKW-CT-2001-20165, EUR 20620.
[7] Becker, D. et al.2002: Testing of Safety and Performance Indicators (SPIN), FIKW-
CT2000-00081, EUR 19965.
[8] Grambow et al. 2008: RTDC-1 Final Synthesis Report, Dissolution and release from the
waste matrix. EU NF-PRO Project FI6W-CT-2003-02389.
[9] Arcos, D., Hernán, P., De La Cruz, B., Herbert, H.-J., Savage, D., Smart, N.R., Villar, M.V.,
Van Loon, L.R. 2008: NF-PRO RTDC-2 Synthesis Report, Deliverable (D-N°:D2.6.4) EU
NF-PRO Project FI6W-CT-2003-02389.
[10] Villar et al. 2008: RTDC-3 Synthesis Report. EU NF-PRO Project FI6W-CT-2003-02389.
[11] Aranyossy J-F., Mayor, J.C., Marschall, & Plas, F. 2008: EDZ development and evolution.
RTDC-4 Synthesis Report. EU NF-PRO Project FI6W-CT-2003-02389.
[12] Johnson, L. et al. 2008: RTDC-5 Synthesis Report, Deliverable (D-N°:5.2.3), EU NF-PRO
Project FI6W-CT-2003-02389.
222

PANEL DISCUSSSION
Summary of the Panel Discussion Concluding
Session VI: Near-field processes
Panel members and authors:
Simon Loew (Chair), ETH Zürich, Switzerland
Peter Blümling, NAGRA, Switzerland
Lawrence Johnson, NAGRA, Switzerland
Karl Lemmens, SCK-CEN, Belgium
David Savage, Quintessa Ltd, United Kingdom
Pierre Van Iseghem, SCK-CEN, Belgium
This report summarizes the views of an expert panel on key remaining research issues in near-field
processes which are critical for the long term safety of the geological disposal of high level radioactive
wastes. The panel first met on the occasion of the Euradwaste '08 Conference and later continued
the discussion in order to create this summary report.
The questions put forward for discussion were:
� Is there a common understanding of the most important safety functions of the Engineered Barrier
System (EBS) in a HLW repository in salt, clay and granite? How would co-location of different
waste types (e.g. HLW and L/ILW) influence these safety functions?
� Which time scales should be considered in the detailed study of these safety functions? What are
the contributions and limitations of “long-term” laboratory and in-situ experiments?
� Which (coupled) processes, parameters and boundary conditions critical for these safety functions
need further research?
� What is the impact of physical and chemical heterogeneity of key properties of the EBS on the
performance of the individual barriers at site scale?
� How can we demonstrate that THMC coupled processes reach stable steady state conditions
during the first 10’000 years and that we can describe them reliably?
� Which implementation oriented studies need to be addressed in future research projects? Which
key R&D stakeholders should have which function in the development of such projects?
Safety Functions of the Engineered Barrier System
The initial idea of the panel was to relate open research questions to safety functions of the engineered
barrier system (EBS), which is composed of waste form, canister, overpack/buffer and seals,
surrounded by the host rock in the immediate neighbourhood (i.e. the near-field). However, this
idea had to be abandoned, because, due to differences in host rock properties and repository concepts,
there are different requirements on the performance of the EBS, and therefore the safety functions
of the EBS are differently defined in the national programmes. This even holds for safety
functions attributed to the EBS in similar host rocks.
223

There has been a significant change in performance assessment (PA) over the past years, in that today
these safety functions are broken down to a lower level for some of the barriers to more clearly
show the safety functions are fulfilled for possible evolutions of the system.
The safety function of a barrier depends also on the considered radionuclide. For example, retention
by the waste form is most important for the species that are not retarded by the other barriers.
For waste glass, the current life time estimation for neutral pH conditions is several hundreds of
thousands of years. For spent fuel, the life time may be well over a million years. Retention by the
waste form can be particularly important for a species like iodine-129, which is found partly in the
instant release fraction of the spent fuel. A larger or smaller amount of instantly released iodine can
make a significant difference for the dose peak maximum.
Even though the safety functions vary between host rocks, waste forms, radionuclides and national
repository concepts, the future research investigations should be iteratively linked with the progress
made in performance assessments.
Consideration of Time Scales in Future Research
Based on the current status, safety assessments should cover periods up to 10 6 years. The objective
of future research investigations is to support the PA investigators in setting reliable bounds to the
time dependent development of the EBS and the disposal system as whole.
In future research investigations various time scales need to be considered and the evidence brought
together, as no process can be studied for this full time frame of interest. Data from lab and in situ
experiments always need to be placed in context with those from natural systems, i.e. they should
be compared with natural analogues within the same project (i.e., it is unrewarding to have separate
programs for experimental and analogue studies).
For example, diffusion of radionuclides in clay host rock, studied in laboratory and field investigations
can provide data over about ten years for non-sorbing radionuclides (cm to dm transport distance).
Isotope profiles of similar species can give related information over hundreds of thousands
of years over tens of meter transport distance. In this example, the requirement for comparison is to
ensure that repository-induced perturbations (changes in hydraulic properties caused by temperature
increases or gas transport) are small enough that the results of the short-term and long-term data can
be used together to support the transport model.
In this context the panel also gives high priority to further investigations of processes that affect the
EBS during the transient period with high gradients (e.g. hydraulic, thermal or chemical gradients)
that could alter the properties of the EBS and thus influence their long-term development and performance.
Future lab and field-scale experiments and model exercises related to these transients
should be done under more realistic (not oversimplified) conditions.
Extrapolation of lab test data, even when the tests are carried out over tens of years, to long timescales
is not straightforward. Natural systems studies and the NF-PRO investigations on steel corrosion
in compacted bentonite show, that growth of the thermodynamically most stable solids is often
kinetically inhibited. In the NF-PRO project, a computer model using the Ostwald step approach
was developed to upscale the experimental results and address long-term alteration processes.
This approach is relevant to long-term alteration processes affecting not only clay, but also
metals and glass. Further research investigations on temporal upscaling methods and limitations are
required.
224

Future Research on Coupled Processes
Most transformations and transport mechanisms in the EBS are driven by processes that show various
degrees of physical (thermo-hydro-mechanical) and chemical coupling. In the past, such transformation
and transport mechanisms in the near-field have been experimentally investigated under
strongly simplified conditions. In future research investigations both “simple” and coupled processes
need further investigations.
Future investigations on the study of uncoupled or weakly coupled processes shall focus on fundamental
mechanisms and the lifetime of individual components of the engineered barriers, for example
the long term corrosion and dissolution processes and rates of spent fuel and glass, or the reaction
of silicon with the near-field materials. For these long term processes methodologies need to
be developed on how to evaluate the long term waste form behaviour for different reference disposal
concepts and scenarios.
Processes with significant THMC coupling, that need further research investigations, are related to
bentonite swelling; bentonite interactions with steel and cement; gas generation and gas transport in
buffer, host rock, shafts and seals; self-sealing of bentonite and clay host rocks; reactive mass transfer
under two phase conditions and formation of reactive barriers; and processes related to drying
and re-saturation of the near-field. Again, experimental and modelling investigations focusing on
more realistic conditions at the detailed and system levels should have high priority, and can be
supported by recent advances in computing. Post-test analysis of large scale validation experiments
is as important as model validation, and complex interacting boundary conditions in the near-field
impacting the waste form behaviour and nuclide transport need to be studied further.
The description of such processes with numerical models is very difficult because of the complexity
of the models and the difficulty to specify the appropriate parameters and boundary conditions. It is
therefore essential to also work towards simplifying these coupled process models in an adequate
way to allow for appropriate sensitivity studies by mapping the possible parameter ranges for the
most important parameters.
Future Research on Heterogeneous Systems
The most important heterogeneities in the EBS, that deserve further investigations, are expected to
occur along the interfaces between the different components, e.g. the canister/buffer interface, and
the buffer/rock interface. Heterogeneities within individual engineered barriers are less important,
mainly because the homogeneity of these barriers can be directly influenced by the fabrication process.
The heterogeneities observed in the bentonite buffer during “short term” experiments relate to
resaturation transients with differential wetting, and are expected not to have any impact on the diffusive
radionuclide transport occurring at later times. The impact of heterogeneities along interfaces
between EBS components on the performance of the whole EBS system could be evaluated
with large scale experiments and numerical models.
Heterogeneities in the near-field host rock can be natural and repository induced. Excavation induced
heterogeneities (EdZ and EDZ) are known to exist for decades and have been studied in the
past intensively in various host rocks, but there are still some open questions on how to address the
uncertainty in large scale behaviour in a reliable manner. Much less is known on the properties of
geological heterogeneities in clay rocks, because these rocks appear as homogeneous at first glance.
However, more recent studies in clay rocks (both indurated and soft) show that they contain many
geological discontinuities which impact the mechanical rock mass behaviour and potentially some
important THM coupled processes in the near-field. Therefore more experimental data is needed to
understand the impact of clay stone discontinuities on the performance of the near-field.
225

226

SESSION VII: Repository technologies, actinides and far-field migration processes
Chairman: Dr Jörg Hadermann, PSI, Switzerland
227

228

Summary
ESDRED – An Integrated European Project
Focused on Technology Development
Wolf K Seidler, Jean-Michel Bosgiraud
Andra, France
ESDRED (Engineering Studies and Demonstrations of Repository Designs) is an Integrated
Project focused on technology development. Over the course of the last 56 months the participants
have designed, fabricated, tested and shown a number of static and dynamic demonstrators
related to the emplacement of HLLL radioactive waste all of which met or exceeded
the design specifications. Other work related to the formulation of low pH cement/concrete/shotcrete
mixes and subsequent application/demonstration of these for repository
construction and for rock support. Still other work related to the formulation of various
material mixes which were then demonstrated as prefabricated or cast in place engineered
barriers (or backfill).
The results achieved within ESDRED are now on display, or ongoing, at various underground
locations including Äspö in Sweden as well as Mont Terri and Grimsel in Switzerland. Andra’s
demonstrators have recently been moved to the Saudron Technological Centre near the
Bure URL in France while the Ondraf / Euridice work can be seen at the MOL facility in Belgium.
DBE TEC’s vertical emplacement demonstrator is set up in the turbine hall of a former
thermal power station located near Landesbergen, Germany. All these demonstrators contribute
to public confidence building, regarding the societal issues related to underground disposal
of radioactive waste.
1. Introduction
The Integrated Project known as ESDRED has been a joint research and development effort by major
national radioactive waste management agencies (or subsidiaries of those agencies) and by research
organisations. ESDRED was co-ordinated by the French National Radioactive Waste Management
Agency (Andra). The original five year budget of EURO 18.4 million, of which 7.3 million
was provided by the European Commission (EC), was overspent because several of the participants
elected to do either more work or more elaborate work than originally envisaged.
The 13 participants (Contractors) in this project, from 9 European countries, are:
Radioactive Waste Management Agencies: Technological R&D Organisations:
ANDRA, France AITEMIN, Spain
ENRESA, Spain CSIC, Spain
NAGRA, Switzerland DBE TECHNOLOGY, Germany
NDA (Originally NIREX), United Kingdom ESV EURIDICE EIG, Belgium
ONDRAF/NIRAS, Belgium GRS, Germany
POSIVA, Finland NRG, the Netherlands
SKB, Sweden
229

2. Methodology
ESDRED has been focused on technology and has had three main objectives.
The FIRST OBJECTIVE was to demonstrate, at an industrial scale, the technical feasibility of
some very specific activities related to the construction, operation and closure of a deep geological
repository for high level radioactive waste. This part of the work was organised inside four (4)
Technical Modules (and numerous work packages) and essentially involved the conception, design,
fabrication and demonstration of equipment or products for which relevant proven industrial counterparts
(mainly in the nuclear and mining industry) do not exist today. Execution of the work was
often by third party sub-contractors (especially the detailed design, fabrication and testing of new
equipment) although, depending on the participant, more or less of the work was done in-house.
Each of the four technical Modules involved from 3 to 7 participants thus ensuring that the knowhow
and experience from several different national disposal concepts could be integrated into the
work. The results of the work within each of these Modules are provided in the next Section.
A SECOND and equally important ESDRED OBJECTIVE was to promote a shared European vision
in the field of radioactive waste disposal technology. This was accomplished through the IN-
TEGRATION process, which is one of the key objectives that identify EURATOM’s 6 th Framework
Programme. Among other things integration resulted from working together, sharing information,
comparing input data and functional requirements, learning from one another’s difficulties,
developing common or similar tender documents and bidder lists, jointly developing courses and
workshops and coordinating demonstration activities whenever possible.
Generally at least 2 Integration meetings were convened annually so that all ESDRED participants
were updated on the progress of the work in all the Modules. Whenever practical these meetings
were combined with the demonstration of a particular piece of new equipment, process or construction.
The THIRD OBJECTIVE focused on training and communication. Over the life of the project the
participants wrote many articles, presented numerous technical papers at international conferences,
held 2 workshops, developed and presented 17 university lectures, and finished up by organising an
international conference on the operational aspects of deep geological disposal. Also a web site
(www.esdred.info) was created and maintained over the life of the project. This site will be kept on
line until sometime in 2010.
3. Results
3.1 Overview
Details of the results, organised by Module are presented in the next 5 Sections. As stated earlier
the work within ESDRED was primarily focused on Technology Development. In many instances
the participants were able to develop and demonstrate equipment, processes or systems for which
there are no industrial equivalents anywhere. Among others these include:
A simplified and partially scaled down version of an air cushion transporter,
A 1:1 scale air cushion emplacement system for 43t spent fuel canisters,
A 1:1 scale water cushion emplacement system for 45t spent fuel Super Containers,
230

A 1:1 scale pushing robot for horizontal emplacement of 2t vitrified waste canisters,
A 1:1 scale transport shuttle, docking table, transport shielding cask and second generation
pushing robot for horizontal emplacement of 2t vitrified waste canisters (to be completed by
year end),
A 1:1 scale system for emplacement of vitrified waste and spent fuel canisters in long vertical
boreholes; including the rail mounted transport and the borehole locking equipment (to
be completed by year end),
14 four ton rings/discs of highly compacted bentonite/sand material for use as prefabricated
buffer material (used in the construction of horizontal disposal cells); including the mould
utilised to produce them and the custom built equipment needed to manipulate them,
A system for backfilling the annular gap between a canister and a circular tunnel with
granular bentonite while achieving important in situ design characteristics of the backfill,
The formulation of various backfill materials which were subsequently successfully tested
on surface in a short mock-up of a disposal tunnel, to backfill the annular gap between a circular
tunnel and a Super Container. Subsequently a specially formulated grout was tested for
the same purpose in a 30m long mock-up,
Formulation of low pH cement mixes for incorporation in shotcrete and then used to construct
a 1m and also a 4m long closure plug. The first was loaded to destruction; the second
loading test is ongoing. Low pH shotcrete was also tested as rock support material.
Field testing of wireless monitoring techniques using seismic tomography is still ongoing.
3.2 Module 1 - Buffer Construction Technologies for Horizontal Disposal Concepts
Within Module 1 Andra was able to successfully design the mould (Fig. 1) and the necessary formulation
(70/30 of MX80 bentonite/sand) and thereafter produce 4 ton bentonite rings (Fig. 2) to be
used as an engineered barrier. Ondraf/Niras (Fig. 3) demonstrated backfilling of the annular gap
between a waste canister and the disposal drift wall using a variety of wet and dry products. Nagra
(Fig 4) developed the product and the technique for backfilling disposal drifts with bentonite pellets.
The evolution over time and the performance of bentonite based seals, particularly in relation
to gas permeability, was assessed by GRS and is in fact on-going beyond ESDRED. Finally nonintrusive
monitoring techniques based on seismic tomography were also developed and demonstrated
by the NDA.
Figure 1: 100 Ton Mould for Pressing
Sand/bentonite Rings (Andra)
231
Figure 2: 4 Ton, 2.25m diameter bentonite/sand
ring after 45 000 tons of pressure & stripping
from mould (Andra)

Figure 3: Reduced scale mock-up after backfill
testing (Ondraf / Niras & Euridice)
232
Figure 4: Double Auger (green) Placement of
Bentonite Pellets Around a Canister (Nagra)
Rock support tests, gas seal tests and non-intrusive monitoring tests were all conducted underground.
On the other hand the EBS rings and the backfilling tests were conducted on surface at reduced
and at full scale. The only underground in situ EBS test is the hydraulic plug still to be completed
in the Praclay Gallery at MOL, Belgium. This work will be conducted by Ondraf/Niras and
Euridice.
3.3 Module 2 – Waste Canister Transfer and Emplacement Technology
In Module 2 two participants (Andra and DBE Technology) were able to design, fabricate and
demonstrate the equipment needed for the Transfer and Emplacement of Waste Canisters
weighing between 2 and 5 tons. The respective equipment is designed for emplacement in either
horizontal or vertical disposal boreholes with very small annular clearances between the canister
and the wall of the disposal boreholes. A desk study related to retrievability was produced by the
third participant, NRG, based on the 2 emplacement concepts.
Figure 5: First prototype Pushing Robot & 2t canister
(Andra)
Figure 6: Second prototype Pushing Robot
(Andra)
The horizontal emplacement equipment (Fig. 5) which was produced can be seen by the public at
Andra’s Prototype Exhibition Hall in Saudron near the Bure URL in France. The current second

generation pushing robot is shown in (Fig. 6). The vertical emplacement equipment (Fig. 7) can
currently be seen at DBE TEC’s temporary test facility at Landesbergen, near Hannover Germany.
Most of the documents related to the design, fabrication and testing are public.
Before being put on display, either at Saudron in France or at a facility to be determined in Germany,
both emplacement systems will have been functionally demonstrated in surface facilities using
inert waste canisters that were otherwise accurate geometrically and with regard to mass.
Figure 7: Tilting frame of vertical emplacement system including transport cart and locomotive
(DBE TEC)
3.4 Module 3 - Heavy Load Emplacement Technology
Heavy Load Emplacement Technology for horizontal disposal concepts was the only focus of
Module 3. The participants active in this work (SKB/Posiva & Andra) each successfully produced
a machine for emplacing 43/45 ton waste canisters in horizontal bored disposal tunnels while maintaining
only a very small annular gap between the canister and the walls of the tunnel. One machine
was based on air cushion technology while the other used water cushions. The latter machine was
subsequently adapted to demonstrate the emplacement of packages of 4 pre-assembled bentonite/sand
rings produced in Module 1, weighing 17 tons per package.
233

Figure 8: Demonstration of emplacement
of 43 ton Spent Fuel canister using Air
Cushion Emplacement Equipment (Andra)
Figure 9: Water Cushion Emplacement Machine with
45 Ton Super Container in Background (SKB /
Posiva)
The air cushion machine (Fig. 8) is on public display at Andra’s Prototype Exhibition Hall in Saudron
near the Bure URL in France. At time of writing the water cushion machine (Fig. 9) is set up
underground on the -220m level at the Äspö URL in Sweden. Design details, test results as well as
recommendations for future enhancements are available in the various project reports which have
been designated for public access.
3.5 Module 4 - Temporary Sealing (using low pH cement) Technology
The work in Module 4 consisted first of designing low pH cement formulations and then of preparing
several concrete designs suitable for the construction of sealing plugs and for rock support. In
both cases shotcreting was used as the construction method. A short plug and a long plug were subsequently
constructed in 2 different URL’s. The short plug (Fig. 10), constructed at Äspö in Sweden,
was very quickly loaded to failure (i.e. until it started to slip) by pumping pressurised water
behind the plug. It was monitored during the entire process. The longer plug (Fig. 11), constructed
at Grimsel in Switzerland, is currently being loaded using the pressures generated by the swelling of
bentonite blocks during rehydration.
Figure 10: Short Plug Construction at
Äspö (Enresa / Aitemin / CSIC)
Figure 11: Long Plug in Background with Monitoring
Equipment at Grimsel (Enresa / Aitemin /
CSIC)
234

3.6 Module 5 – Training and Communication
This is an area to which the participants committed many man-days and significant costs. Often the
participants were called upon to work very closely together in order to get the job done. Among
other things:
An ESDRED Web Site was set up early in the Programme and updated on a regular basis.
Among many other things this site provided an annual summary of activities completed,
made important documents (including Proceedings for example) available to the public, advertised
important events, etc. This site will be maintained until sometime in 2010.
A Masters level course (17 lectures) was developed by the ESDRED partners and presented
to the students of the University Politehnica of Bucharest, Romania, in November 2006.
This involved 8 of the 13 ESDRED participants.
Two workshops focusing on R&D related to low pH cements were organised by the concerned
ESDRED partners and attracted an international audience and authors.
ESDRED representatives also participated and contributed world wide to workshops organised
by others.
Media events were organised often around the various demonstrators.
Joint papers were written by representatives from several different national agencies.
Technical articles appeared in Professional Journals, in in-house magazines, in subcontractor
bulletins and journals; sometimes written by ESDRED personnel and sometimes
by others.
Broad dissemination of ESDRED results undoubtedly helped to build confidence in the disposal
concepts being considered in the European nuclear area as well as to bring the representatives of the
national agencies closer together. As a minimum a better understanding of the issues was shared
and a broader knowledge of the available solutions was imparted. Informal networks of engineers,
contractors, suppliers and experts were established on a European Scale.
Finally the crowning highlight of the Project came in June of 2008 when an “International Technical
Conference on Practical Aspects of Deep Radioactive Waste Disposal” was organised by ES-
DRED partners Andra and GRS in conjunction with the Czech Technical University of Prague
(CTU) and RAWRA the Czech national waste management agency. This very successful event,
which also included a special Student Session, was held in the facilities of the CTU, Faculty of
Civil Engineering, Centre of Experimental Geotechnics, June 16-18, 2008. Nineteen of the 38 papers
and posters were related directly to ESDRED, to the national agencies represented in ESDRED
or to the sub-contractors that had been engaged by ESDRED. There were papers and posters from
13 countries with registrants from 19 different countries. Total registration (over 130 attendees) exceeded
the objective fixed 2 years earlier.
4. Discussion
The main work accomplished in Module 1 proved that it was possible to prefabricate large (2.3m
diameter) bentonite/sand rings and discs with the desired characteristics and which could be safely
manipulated and placed in a disposal cell. It also showed that it was possible to cast in place various
engineered barriers (and obtain the desired characteristics) consisting of bentonite, of bentonite/sand
mixtures, of bentonite pellets, of cement grout and other materials to be used for filling
small annular voids or for backfilling disposal drifts and voids around canisters.
235

The various reports produced by the partners in Module 1, most of which are available to the public,
could enable interested parties:
To design a sand/bentonite mixture which, when compressed into a ring or produced as pellets,
is suitable to be used as an engineered barrier around waste canisters with appropriate
physio-chemical characteristics,
To design and fabricate a mould for producing large EBS rings as well as all the related
stripping and handling equipment,
To formulate various wet and dry materials for use as a backfill and to evaluate different related
placement options,
To design a borehole seal for which the relative permeability to gas and water is optimised
and to have an understanding of the performance of such a seal over time,
To evaluate whether non-intrusive monitoring based on seismic tomography is suitable to
their particular application.
Similarly the results of the work completed in Module 2 would enable interested parties to adapt, to
their own repository conditions, the ESDRED equipment and system designs for the disposal of
canisters weighing less than 10 tons. Designs are available for both vertical and horizontal disposal.
By integrating the Andra and the DBE TEC designs, either track or trackless transport could be possible.
The various reports produced by the partners in Module 2, most of which are available to the public,
could among other things assist interested parties:
To design a pushing robot for emplacing canisters in horizontal boreholes with less than 25
mm of clearance between the outside of the canister and the inside of the disposal drift,
To consider a specific design of ceramic pads or sliding runners as part of their canisters
with a view to decreasing friction and creating a non-corrosive interface between the canister
and any metal lining installed in the disposal cell,
To design a transport vehicle complete with transport shielding cask, docking table, and interlocking
gamma gates,
To design a borehole locking device for vertical emplacement,
To design a transport cask and associated emplacement device which incorporates a unique
tilting device to move a horizontal canister into a vertical hole.
The Module 2 final report will also include recommendations for future modifications and/or additional
test work all aimed at further improving the original ESDRED designs.
The work completed in Module 3 clearly showed that either air or water cushion technology can be
used to transport heavy canisters up to 45 tons (heavier loads would simply require more cushions)
in situations where minimising the annular clearance between the canister and the inside face of the
disposal tunnel is important for technical and/or financial reasons. The choice of air or water depends
on the specific conditions. For example water cushions cannot be used in a repository in clay
for obvious reasons. The various public reports produced by the SKB/POSIVA and by Andra, the 3
partners in Module 3, could help interested parties to adapt the ESDRED designs to their own
needs for the emplacement of heavy loads. The final reports also provide recommendations for future
modifications and/or additional test work all aimed at further improving the original ESDRED
designs.
A variety of Module 4 project reports that are available to the public, describe in detail the process
used to develop the various low pH cements which would meet the project needs. The use of one or
236

more of these special cements to produce shotcrete for the construction of retaining plugs and for
rock support is also described. Other reports describe the test plans and the execution related to the
construction of the 2 plugs noted in Section 3.5. The short plug constructed at Äspö was monitored
during loading and after failure it was demolished and tested. The related results have been evaluated
and lessons learned. However the long plug at Grimsel is still intact.
Most of the materials produced within Module 5 including the Proceedings of the 2 Workshops and
the International Conference in Prague, as well as the course provided at the University Politehnica
in Bucharest Romania, can be downloaded from the ESDRED web site.
5. Conclusions
Clearly objectives like designing a common European Repository or even designing Repositories
for various Europe countries that are all similar did not fall within the ESDRED mandate. In the end
the legal framework, the national waste management programme, the existing and the perceived
constraints, the stakeholder expectations as well as the physical setting are different in each country.
On the other hand by working together the participants were able to observe first hand that they
were all facing the same basic challenges and that they all shared a concern for the same fundamental
issues that are key to the design of a safe geological HLLLW repository. In other words it
quickly became clear that there already existed a common European view regarding deep geological
disposal of high level radioactive waste, which was simply reinforced by the ESDRED experience.
The participants in ESDRED believe that they have made a valuable contribution to the body of
technical knowledge related to deep geological disposal of HLLL radioactive waste. They have
done this by fabricating and demonstrating various pieces of equipment which did not previously
exist and which are now available to agencies globally as reference solutions that can be further improved
and/or adapted to very specific applications. By communicating widely and by maintaining
certain tangible results of the ESDRED work for ongoing public viewing, the partners have already
increased the level of the public confidence in the industry’s ability to safely dispose of HLLL radioactive
waste, and they will continue to do so in the future.
At the end of the ESDRED Project (January 31, 2009) the Final Reports produced by each of the 4
Technical Modules and the Final Project Report will be placed on the ESDRED web site
(www.esdred.info/) where they can be downloaded. The European Commission on its site will also
display a number of ESDRED reports (http://cordis.europa.eu/fp6-euratom/lib_projects.htm)
6. Acknowledgements
This project has been co-funded by the European Commission and performed as part of the sixth
Euratom Framework Programme for nuclear research and training activities (2002-2006) under contract
FI6W-CT-2004-508851.
237

238

Achievements of the ESDRED project in Buffer Construction Technology
Chris De Bock 1 , Jean-Michel Bosgiraud 2 , Hanspeter Weber 3 , Tilmann Rothfuchs 4 , Jan Verstricht 5 ,
Brendan Breen 6 , Mark Johnson 6
Summary
1 ONDRAF-NIRAS, Belgium
2 ANDRA, France
3 NAGRA, Switzerland
4 GRS, Germany
5 EURIDICE, Belgium
6 NDA, United Kingdom
ESDRED is a technological integrated European project within the context of the 6th Framework
Program of EURATOM. It is the aim of this 5-year project (2004 to 2008) to demonstrate
the technical feasibility, at an industrial scale, of a number of specific technologies related
to the construction, operation and closure of a deep geological repository for storage of
spent fuel and long-lived radioactive waste. The project is divided into a number of modules,
of which Module 1 is principally dedicated to the construction and/or emplacement, in a horizontal
configuration, of the buffer around a disposed high-level waste package. In addition,
Module 1 aims to test the performance of seals (saturation rate and gas permeability) and certain
seal installation aspects, and to advance the state-of-the-art of the application of nonintrusive
monitoring techniques in geological repositories. The partner organisations working
within Module 1 are: ONDRAF/NIRAS (Belgium), ANDRA (France), NAGRA (Switzerland),
GRS (Germany), EURIDICE (Belgium) and the NDA (United Kingdom).
1. Introduction
R&D projects before ESDRED Module 1 had already studied several bentonite materials and their
conditioning to investigate their application as key materials in buffers, backfill or seals in geological
repositories (e.g. RESEAL in Mol, or BOS in Grimsel). Most of these projects had been performed
on a small or intermediate scale, and only a few addressed the emplacement techniques related
to these components. The demonstration testing at full scale that had been performed was basically
limited to the vertical disposal configuration (e.g. work by SKB related to the KBS-3V design).
From these projects, it was becoming clear that the horizontal configuration would pose specific
challenges that had never been sufficiently addressed, whereas the horizontal configuration
was getting increased consideration in the disposal concepts (reference or alternative) of a number
of EU countries. Module 1 was therefore in the first place set up to advance the state-of-the-art of
the technology related to the construction of the buffer or the backfill around a horizontally disposed
high-level waste package.
239

2. Methodology
In general, the work within ESDRED Module 1 passed through the following sequence:
(1) definition of requirements and gathering of input data, (2) computer modelling and/or laboratory
testing, (3) mock-up or in-situ testing.
For monitoring, it was first assessed that the work should be devoted to non-intrusive monitoring,
instead of wireless monitoring, mainly because of the more general interest to all partners. Then, it
was assessed that micro-seismic technology offered the best opportunity to advance the state-of-theart
level of technology related to the application in geological repositories. Moreover, it was possible
to identify a URL (MontTerri) in which testing could begin without delay. After that, the work
on monitoring followed the sequence described in the preceding paragraph.
3. Results
3.1 Buffer with prefabricated rings (ANDRA)
Disposal concept / design
In 2003, ANDRA developed a concept for horizontal storage of Spent Fuel (SF) Canisters & Vitrified
Waste Canisters, in which an engineered barrier is placed between the canister and the host
rock, acting as a buffer. This concept is illustrated below (see Figure 3.1.1) and described in detail
in ANDRA’s Dossier 2005 [1]. However, in the latter document the prefabricated engineered barrier
is no longer a part of the concept for the disposal of Vitrified Waste and has been maintained
only for the disposal of Spent fuel Canisters.
Vitrified waste canisters and SF canisters are emplaced in horizontal drifts (also called “Disposal
Cells”), approximately 40 m long (see Figure 3.1.1). The same type of horizontal cells is used for
both types of canisters. Disposal cell diameters vary with canister diameters, but remain 2.5 - 3 m.
Figure 3.1.1 Concept of SF disposal cell with Engineered Barrier
Each horizontal disposal cell (for SF or Vitrified Waste canisters) is similarly composed of (from
the exterior to the interior – see Figure 3.1.2):
a steel liner (supporting the wall of the cell excavated in clay), made of carbon steel, approximately
30 mm thick, perforated with holes in order to allow bentonite resaturation
with water progressively seeping from the host rock,
an annular layer of buffer material, 800 mm thick (in radius),
a steel sleeve, made of carbon steel, 25 to 40 mm thick, which holds inside 3 to 22 disposal
packages (depending on length and thermal activity of canister).
240

Figure 3.1.2 Cross section of a disposal cell for Vitrified Waste canister
Role & requirements related to the buffer material
The main function of the engineered buffer is to act as a confining barrier insuring diffusive mode
for water transport; since the canister over pack loses with time its structural integrity and the primary
package glass is then fully exposed. Resaturation allows the engineered buffer to fill any
voids that may be created in order to give the system a hydraulic conductivity of less than 10 -11 m/s.
The mechanical role of the engineered buffer comes from the swelling pressure that results from its
resaturation. This swelling pressure must eventually reach a state of equilibrium with the in situ
ground pressures but it must not exceed the tensile strength of the argillite. The engineered buffer
also plays a small and ever decreasing thermal role.
The buffer material used is a bentonite (70%) – sand mixture (30%). All bentonite materials do not
behave in the same way. MX-80 was chosen since it is well characterized. Furthermore, it offers
high plasticity and is forecast to be available on the market over the long term.
The buffer mixture was subsequently characterized (bench scale testing of different samples with
variable water content and different compaction pressure values) in order to optimize the fabrication
process and to check the compatibility of the cold-compacted material with the required swelling
pressure and conductivity.
Design & fabrication of the mould
It was decided to design and fabricate 10 rings and 2 discs with an outer diameter of 2.25 / 2.30 m,
an inner void (rings only) of 80 cm, a thickness of 50cm and weighing approximately 4 tons each.
Those rings or discs were later assembled, packed and transported in sets of 4.
The compacting pressure deemed optimum (80 MPa) and the size of the rings led to the selection a
press capable of exerting a compacting force of at least 45 000 tons. The only press available in
Europe capable of providing such a compacting force turned out to be the Aubert & Duval press at
Issoire (south of Clermond-Ferrand, France). The mould (see Figure 3.1.3) was subsequently designed
to cope i) with the required pressure and ii) with the geometrical dimensions / functional
241

limitations of the selected press. The fabrication of the various pieces of the mould involved some
very demanding casting and machining.
Figure 3.1.3 the mould during factory commissioning operations at Creusot-Loire facilities
Fabrication of the rings and discs
The fabrication took place in 3 steps (3 pressing campaigns) between June and December 2006.
Stripping of the rings / discs turned out to be a serious technical challenge, but in the end the 10
discs and 2 rings were produced during the last pressing campaign, in less than 16 hours (including
mobilization and demobilization of personnel and equipment). The Figure 3.1.4 below shows the
mould under the press arcade in the Aubert & Duval factory in Issoire, just before the start-up of the
compaction process.
Figure 3.1.4 the 60 000 t press in Issoire
Figure 3.1.5 the 4 ton / 2.25 m cold compacted
buffer ring
The final product turned out to be of very good quality (samples were taken from one of several
extra rings fabricated during the 2 first campaigns in order to check the homogeneity of the
compacted material) with a smooth surface finish. Figure 3.1.5 shows one ring just after stripping
and before conditioning.
242

Handling process
Lifting of the ring or disc from a horizontal position to a vertical one (for assembly in sets of 4) is
carried with a suction cup device (Figure 3.1.6) in conjunction with a tilting frame, while the final
lifting and transportation (inside a special container) is assured by a special yoke (see Figure 3.1.7).
Figure 3.1.6 lifting of disc by a suction cup handling
device
243
Figure 3.1.7 Deposition of the 4
pre-assembled rings on their
transport container lower part
3.2 Backfilling of a horizontal annular gap (O/N and EURIDICE)
�
Disposal concept / design
In 2003, resulting from the redesign process after the review of the SAFIR 2 report [2], O/N
adopted the horizontal Supercontainer as its reference concept for the disposal of HLW. In the Supercontainer
concept, the waste canisters are placed within a carbon steel overpack, which is surrounded
by a high pH concrete buffer (see Figure 3.2.1). A high pH material was chosen for the
buffer with the aim to create a corrosion-protective environment for the overpack.

Figure 3.2.1 Schematic representation of Supercontainer concept (radial and axial cross-sections)
Requirements related to the backfill around the Supercontainer
In the O/N disposal concept for high level waste, the backfill component fulfills two functions:
1. The primary function of the backfill is to prevent a cave-in of the disposal drift, which might
damage the Supercontainer or distort the host rock surrounding the drift.
2. A secondary function of the backfill, applicable only to the disposal of spent fuel, is that the
backfill minimizes the existence of potential escape paths out of the Supercontainer for the
filler material present around the fuel assemblies.
Next to these functions, there are also constraints on the backfill component. Together, the functions
and the constraints determine the requirements for the backfill component. Two types of constraints
can be discerned:
1. constraints related to (long-term) safety:
• not disturb the corrosion-protective characteristics of the buffer (in Supercontainer);
• not act as a thermal isolator;
• not introduce organic materials that can give rise to migration-enhancing complexes;
• no excessive expansion or shrinkage or chemically attack the disposal drift wall;
2. constraints related to feasibility:
• exhibit the physical qualities that will allow it to be pumped or projected into the gap;
• achieve the needed industrial performance of the process;
• dust generation and water run-back should remain very limited;
• limit backfill strength, to keep the option of retrievability open as much as possible.
Backfill material
Because of its straightforward chemical compatibility with the disposal concept and the perceived
better opportunities for achieving the industrial performance, the grout injection technique is considered
by O/N to be the reference solution for backfilling disposal galleries. The development of a
grout fulfilling the requirements was a key process that was entrusted to BASF Construction
Chemical Belgium (formerly DEGUSSA). It resulted in a backfill grout, composed of:
244

1. Bonding material: cement (CEM I 52.5N HSR LA), and powder CaCO3 type Carmeuse
2. Calibrated river sand with grain size between 0 and 4 mm
3. Superplasticizer Glenium®, a polycarboxylate ether-based material
Alternative backfill solutions are still being considered by O/N. A materials survey, accompanied
by pre-testing using the dry-gun projection technique (August 2005), rendered the following list of
granular material for possible use as backfill: pure sand (SiO2), pure bentonite (MX-80), sandbentonite
mixture (25/75), sand-cement mixture (90/10), and bentonite-cement mixture (85/15).
Reduced-scale mock-up testing (grout)
On June 30 th 2006, the grout injection backfill technique was tested on a 5 m long, reduced-scale
(2/3 rd scale) mock-up of a disposal cell. The centre of the mock-up was equipped with a heater,
representing the waste, locked in a steel tube filled with sand to simulate the Supercontainer and its
thermal inertia. The initial average temperature of the tube surface was a stable 40°C. It took
about 100 minutes, at a target rate of 5 m 3 /h, to fill up the annular void. After the setting of the
grout, the result was investigated by means of taking borehole samples and cutting a slice of the
mock-up (see Figure 3.2.2). It was noticed that a 100% void filling with a homogeneous backfill
had been achieved. Based on the above, it was concluded that the test had been a success.
The following characteristics of the backfill were measured on the borehole samples:
• density: 2190 kg/m 3
• thermal conductivity: 3 W/°C-m (if 9% water content), 1.6 W/°C-m (if dried)
• compressive strength: 12 MPa
Figure 3.2.2: Reduced-scale mock-up for grout backfill testing, after slice-cut (Nov. 2006)
Reduced-scale mock-up testing (dry granular materials)
The projection of dry granular materials with a dry-gun was tested on a similar 5 m long, reducedscale
mock-up, but without a central heater. The projection nozzle was mounted on a vehicle, running
on rails, and specially designed for mechanical robustness (see Figure 3.2.3). All selected
backfill materials were tested between June and October 2006. Borehole samples were taken from
the resulting backfill. It was observed that the machine operated without failure under mechanically
very harsh conditions, and achieved a linear pace of about 2 to 3 m/h. No problems with dust
generation or water runback were encountered. The annular void was 100% filled with a backfill
245

which was relatively homogeneous in terms of density and water content and which exhibited a
relatively steep and regular slope, although generally the result was more positive for the bentonitebased
materials than for the sand-based materials. Based on the above, it was concluded that the
test had been a success. Measured results are summarized in Table 3.2-1. The thermal conductivity
is relatively low for some cases and probably on the edge of what would be acceptable. The tests
did not address the issue of chemical compatibility with the Supercontainer concept.
Figure 3.2.3 Granular materials backfill test configuration and nozzle in
operation (June-Oct. 2006)
Table 3.2-1. Summary of measured operational and materials data in the dry-gun backfill tests
Tested material Density
[kg/cm3]
Pure sand
(SiO2)
Sand/cement
(90 / 10)
not measured
(but typically 2.1
saturated and 1.6 dry)
comparable to pure
sand
Pure bentonite 1.391
(MX-80) 1.043 (dry)
Bentonite/sand 1.442
(75 / 25) 1.092 (dry)
Bentonite/cement
(85 / 15)
Full scale mock-up testing
1.528
1.131
(dry)
Water content
[%]
246
Thermal conductivity
[W/m-°C]
not measured not measured
(but typically 2.7
saturated and 0.35
dry)
Compression
strength
[MPa]
not applicable
7.0 2.944 ± 0.133 not measured
(but typically
2 to 5 MPa)
33.4 0.619 ± 0.011 0.200
32.0 0.842 ± 0.017 0.110
35.1 0.653 ± 0.016 0.670

On April 8 th 2008, the grout injection backfill technique was tested on a 30 m long, full-scale mockup
of a disposal cell. This time, the heater inside the sand-filled tube was set to obtain an initial average
tube surface temperature of 60°C. It took about 6 hours, at an average rate of 15 m 3 /h, to fill
up the annular void. It is currently foreseen to investigate the resulting backfill through borehole
sampling and cutting a slice of the mock-up (see Figure 3.2-2). It will depend on this investigation
whether the test can be called a success. An important lesson already from this test is that the logistical
needs behind the backfill operation increase considerably if longer sections of disposal drift
are taken. The possibility to satisfy those needs in underground conditions may turn out to be the
determining factor in the limitation of the section length.
3.3 Buffer of granular material in horizontal configuration of waste container resting on prefabricated
buffer blocks (NAGRA)
Disposal concept / design [3] [5]
The engineered barrier system foresees a massive steel canister and a bentonite backfill. The bentonite
consists of a hybrid system: the canisters are emplaced on a pre-fabricated pedestal of bentonite
blocks and the remainder of the emplacement tunnel is backfilled with a bentonite granulate.
Underground transport of the waste packages is in the reference case by rail systems using transport
casks (shielding) for the disposal canisters. The emplacement is done by specially designed equipment
that allows remote handling. Backfilling of the emplacement tunnels is also performed by remote
handling. All of the emplacement equipment is on rail tracks and is powered by electric drive
and winches since the emplacement tunnels are inclined between 4 and 6% according to the subhorizontal
host rock layer.
The following figures illustrate the emplacement sequence:
• Transfer from surface to the repository level is carried out by common rack locomotives. In
the central area at repository level the wagon carrying the transport cask (shielding) is
shunted to a tunnel locomotive (Figure 3.3.1, left);
• A pedestal of bentonite blocks is positioned on the emplacement trolley at the enlarged
branch tunnel of each emplacement tunnel (Figure 3.3.1, right). The branch tunnel is
equipped with double track and a lock;
• The transport cask with a waste canister is positioned beside the emplacement trolley. After
all preparations are completed operators leave the lock. All the subsequent activities will be
carried out using remote operations. The canister is now pushed off the transport cask by the
hydraulic device 1 (hydraulic wagon) and moved to the emplacement trolley by the transload
equipment (Figure 3.3.1, right)
• The emplacement trolley is driven by gravity and controlled by a winch locomotive within
the lock up to the emplacement position (Figure 3.3.2, left). At emplacement position pedestal
and canister are lowered subsequently and the emplacement trolley is pulled back to the
lock
• After a waste canister has been emplaced, the remaining tunnel is backfilled with bentonite
granulate using twin augers and a wagon which is pulled back continuously by winches
while backfilling (Figure 3.3.2, right).
247

Figure 3.3.1 Emplacement of SF-Canister - Start niche and lock to the emplacement tunnel
(schematic)
Bentonite granulate
Emplacement trolley Bentonite granulate
SF Canister and pedestal of bentonite blocks
248
Backfilling device with twin auger
Figure 3.3.2: Emplacement trolley with spent fuel canister at emplacement position (left); Wagon
emplacing bentonite granulate in disposal tunnel after a waste canister has been emplaced (right)
Results within ESDRED
The objectives of Nagra’s bentonite emplacement testing within the EC-supported project ESDRED
were as follows:
• Testing and demonstrating of suitable granular buffer installation techniques on a full scale
in surface facilities;
• Verification if the requirements can be fulfilled.
The general objectives and the experiences from previous experiments lead to an ambitious project
specific target value for the emplacement dry density of about 1500 kg/m 3 . In previous experiments
Nagra executed various tests with different bentonite types for buffer material. Because of its favorable
properties with respect to technical emplacement, bentonite granulate is used in Nagra’s reference
concept. Although small-scale laboratory experiments were performed several years ago, a
large scale test was felt to be advantageous to improve confidence that the required dry density of
emplaced bentonite granulate can actually be reached.
For this project Wyoming bentonite from Amcol Speciality Minerals was used. The sodiumbentonite
MX-80 was delivered in a conditioned, slightly granulated state to improve the pourability
and the granulation. The production of the granulated material was done in the Rettenmaier facilities
in Holzmühle, Germany. During the granulation of the bentonite, an increase of the bulk
grain dry density from 1.17 g/cm 3 to 2.10 g/cm 3 with simultaneous halving of porosity was
achieved.
The built twin auger system (Figure 3.3.3) to emplace the granulated buffer material has a total
length of about 9 m and a weight of 1350 kg. The length of the two auger casings is 7.0 m, the diameter
of the tubes are 0.2 m. The feed rate can be controlled by the auger turning speed. The rotat-

ing screwing motion of the auger moves the materials to the end of the outer casing tube where the
material either falls off the end of the auger freely or can push the material out into the existing bentonite
mass. The maximum feed rate is actually 7 m 3 of granular bentonite material per hour. The
maximum filling volume of the steel cylinder was about 7 m 3 , resp. about 10 tons of granular bentonite
material.
As part of the testing, a comprehensive laboratory program was executed to investigate the performance
of the overall emplacement system. Bentonite samples were taken from each “big bag”
before filling into the auger for grain size distribution, water content and bulk density. After every
filling operation of the steel cylinder, the following parameters were determined:
• global bulk wet density
• particle size distribution measurements of the granular bentonite material before emplacement
out of the “big bags” and after emplacement sampled at selected points at the outer
surface of the steel cylinder
• water content measurements of the granular bentonite material before and after emplacement
In addition, other properties of the granular bentonite as mineralogy, swelling pressure, thermal
conductivity, etc. were determined in the laboratory of ETH Zurich (Switzerland) and Clay Technology,
Lund (Sweden).
Figure 3.3.3 Experimental setup (left) and emplacement of bentonite granulate with twin auger
system (right)
After each emplacement test, the bulk wet density of the whole steel cylinder was measured. The
bulk density is the net weight of emplaced buffer material over the total volume of the emplaced
bentonite. The bulk densities of the granular bentonite material show only small changes for different
admixtures of fine granular bentonite and coarse granular bentonite material. The water content
increased only slightly during the test runs from 5.0 % to 5.8 %. The results are very promising as
the required densities can be reached reliably (Nagra 2007 [4]). Figure 3.3.4 provides a summary of
the results.
249

1.650
1.600
1.550
1.500
1.450
1.400
1.350
1.300
1.250
bulk wet density bulk dry density
A B C Cw D Dw E Ew
A 100 % coarse rounded granular material, embedded in two layers
B 92 % coarse, 8 % fine, two layers
C 85 % coarse, 15 % fine, two layers
Cw 85 % coarse, 15 % fine, two layers
D 70 % coarse, 30 % fine, two layers
Dw 70 % coarse, 30 % fine, repeat run, two layers
E 64 % coarse, 28 % fine, 8 % briquettes, two layers
Ew 64 % coarse, 28 % fine, 8 % briquettes, repeat run, only one layer
Figure 3.3.4: Results of the emplacement tests, reached total bulk (wet) density and dry density with
different granulate fractions
3.4 Seal saturation rate and gas permeability testing (GRS)
Seal concept
Currently, highly compacted bentonite buffers are studied in the frame of several concepts for the
final disposal of high-level waste (HLW). In 2000, GRS started investigations on the suitability of
moderately compacted clay/sand mixtures as a sealing material in clay repositories [6]. Such mixtures
may represent a reasonable alternative to highly compacted bentonite buffers, especially for
the safe closing of repository rooms containing gas generating waste, since they will act as a gas
vent thereby avoiding the development of undesired high gas pressures in the disposal cells. In a
clay repository, this granular sealing material may be used as buffer and/or as sealing backfill in
disposal boreholes or disposal drifts containing either Nuclear Spent Fuel (NSF) or vitrified highlevel
waste (HLW). The buffer/backfill material will be installed in drifts or boreholes as a slightly
compacted embankment.
Functional characteristics of the seal material and test programme
In contrast to highly compacted buffers, moderately compacted clay/sand mixtures exhibit: (1) a
high permeability to gas in the unsaturated state, and (2) an adequate low permeability to water
or self-sealing potential against water, respectively due to swelling of the clay minerals after water
uptake from the rock. Because of their low cohesion they also show a low gas entry/break-
250

through pressure in the saturated state enabling gases to migrate out of the disposal cell at reasonably
low pressure without any damage of the host rock.
The objective of the test programme is to test and demonstrate that the sealing properties of
clay/sand mixtures determined preliminarily in the laboratory can be technically realized and maintained
under repository relevant in-situ conditions. The test programme consists of three steps: (1)
laboratory investigations/mock-up tests for selection of suitable seal material mixtures and for testing
material installation techniques, (2) scoping calculations for in-situ test preparation with an assessment
of seal saturation under in-situ conditions, and (3) in-situ tests at the Mont Terri Rock
Laboratory (MTRL).
Laboratory Investigations for the Selection of Suitable Material Mixtures
In the GRS-laboratory in Braunschweig, different clay/sand mixtures with mixing ratios between
35clay/65sand and 70clay/30sand have been investigated with regard to their sealing performance
[7]. These investigations have shown that the functional requirements given above are best met by
35clay/65sand and/or a 50clay / 50sand mixtures (see Table 3.4-1) which therefore have been selected
for further testing under in-situ conditions at the MTRL.
Large-Scale Mock-up at GRS’s Laboratory in Braunschweig/Germany
Before going in situ, both the installation techniques and the required saturation time for the material
mixtures being considered were to be investigated and optimized in large-scale mock-up tests in
GRS’s laboratory in Braunschweig.
Table 3.4-1. Measured material parameters at installation conditions (averages in parentheses)
Sample
Gas permeability
under dry
conditions
[m 2 ]
35/65 1.2E-13
50/50 7.5E-14
Initial water permeability
at full saturation
[m 2 ]
3.3E-17 - 9E-18
(5.2E-18)
1.1E-18 - 4.3E-18
(2.2E-18)
Gas
breakthrough
pressure
[MPa]
0.4 - 1.1
(0.75)
0.4 - 2.8
(1.83)
251
Gas permeability
after gas
break-through
[m 2 ]
1.1E-17 - 1.6E-17
(1.4E-17)
5.5E-18 - 6.2E-18
(5,9E-18)
Swelling
pressure
[MPa]
0.2 - 0.4
(0.28)
0.3 - 0.5
(0.35)
The mock-up tests (Figure 3.4.1) are performed in vertically arranged steel tubes and they are designed
as a full-scale replica of the envisaged in the in-situ experiments (Figure 3.4.2). The lower
fluid injection volume is filled with a porous material (stone chips or sand). At top of the porous
medium a filter frit is placed for ensuring a homogeneous distribution of the injected water over the
entire borehole cross section. Above the filter frit, the clay/sand-seal is installed in several layers to
a height of 1 m. The predetermined installation density of about 1.9 g/cm 3 for the 35/65 clay/sand
mixture has been realized by using an electric vibrator for material compaction. A further filter frit
is installed above the seal for water and gas collection. The whole test set-up is sealed against the
ambient atmosphere by a gastight packer on top of the upper filter frit. In contrast to the in-situ experiments
where instruments are not installed in the seal itself to avoid any negative impact on the
sealing properties, the mock-up is equipped with sensors in the test tube wall to monitor the pore
and total pressure evolution at three levels along the seal. Additionally, two total pressure sensors
are installed at the bottom of the upper frit. Furthermore, the mock-up is equipped at the inlet and

the outlet with some pressure and flow sensors to enable determination of the material permeability
to gas and water as well as the gas break-through pressure and the remaining permeability to gas
after gas break-through.
Figure 3.4.1: Mock-up test in
GRS’s laboratory
Data Acquisition
Rack and Valve
Panel
252
Concrete
Plug
Packer
Filter Frit
SB Seal
Filter Frit
Fluid Injection
Volume
Fluid Injection Borehole
Figure 3.4.2: Design of the SB experiment in a
test niche at the MTRL
Scoping calculations
Scoping calculations were performed to assess the time needed to reach full saturation in the mockup
and in-situ experiments [7]. The calculations were performed with the CODE_BRIGHT on basis
of the material data determined in the laboratory investigations and data taken from the literature.
According to the results of the scoping calculations the saturation time determined for the 35/65
clay/sand mixture amounts to about 170 days for the mock-up tests and 300 days for the in-situ experiment
and for a 50/50 clay/sand-mixture 570 days for the mock-up and 1050 days for the in-situ
experiment for a seal length of 1 m and a water injection pressure of 1 MPa.
Test results
As outlined above the total pressure in the mock-up experiment is measured at top of the seal or
the upper filter frit bottom, respectively. As can be seen in Figure 3.4.3 the maximum pressure and
thus full seal saturation seems to be reached after 38 months of testing. The first water breakthrough,
indicating a situation close to full seal saturation, was observed in September 2007, after
about 29 months of testing. A first assessment of the seal permeability to water yielded a value of
about 1E-18 m 2 which is in very good agreement with the data determined at the small samples
used in the laboratory (see Table 3.4-1). This result confirms the functional requirements excellently.
The long time until water break-through, however, exceeds the predicted saturation period of
about 170 days significantly, but this fact is of lower significance for the seal behaviour in a real
repository since much longer saturation times are to be expected here. Nevertheless, in order to improve
the models used in the scoping calculations and to enhance the process understanding, the
discrepancies between modelling and measurement results are to be clarified.

Pressure [bar, abs.]
3,4
2,9
2,4
1,9
1,4
0,9
sensor 1
sensor 2
03.04.05
12.10.05
22.04.06
31.10.06
11.05.07
19.11.07
29.05.08
Figure 3.4.3: Evolution of total pressure at
top of the mock-up seal
Pressure [bar]
253
2,1
1,9
1,7
1,5
1,3
1,1
0,9
0,7
27.10.05
26.02.06
28.06.06
28.10.06
27.02.07
30.06.07
30.10.07
Sensor 1
Sensor 2
29.02.08
30.06.08
Figure 3.4.2: Evolution of total pressure at top
of the in-situ seal
Figure 3.4.2 shows the evolution of the total pressure in one of the respective in-situ experiment
with the same material mixture for comparison. After about 24 moths of testing, the total pressure
stabilizes at values similar to those determined on the laboratory test samples (compare the value of
0.2 – 0.4 MPa Table 3.4-1) and that observed in the mock-up. Thus, similar sealing properties as
observed in these preceding investigations can be expected in the in-situ experiments. Although the
pressure seems to have reached its final value in this in-situ experiment the respective water breakthrough
did not take place during the hitherto testing period. Testing will thus be continued in the
forthcoming months including the determination of the gas entry/break-through pressure of the
saturated seal and the remaining gas permeability after gas break-through.
3.5 Non-intrusive monitoring development and testing (NDA)
Context
Under ESDRED Module 1, a programme was implemented to non-intrusively monitor an existing
Nagra experimental demonstration, the HG-A experiment, at Mont Terri. Following consideration
of repository concepts, monitoring objectives and potential non-intrusive techniques, ESDRED
Module 1 partners concluded that cross-hole seismic tomography was the most promising technique
to investigate changes in the backfill and saturation conditions of a micro-tunnel.
The HG-A experiment is one of a number that have been undertaken in the Opalinus Clay rocks of
the Mont Terri underground rock laboratory (URL) in Switzerland. The HG-A experiment was designed
to mimic the evolution of a sealed disposal tunnel, replicating the phases of buffer saturation
and gas generation (following corrosion processes of the disposal canister). The aims of the HG-A
experiment are to identify gas migration and to measure gas migration through the host rock (the
Opalinus Clay geology) and along the engineered seals of a filled tunnel.
Investigations in the HG-A experiment focus on a 1m diameter micro-tunnel, which was back-filled
with a coarse grained sand mixture ( 2 - 6 mm grain size) and closed with a hydraulic megapacker.
There was nothing in the tunnel to represent the wasteform or a metallic waste package. The
study covered several phases, representing backfill emplacement, saturation of the micro-tunnel and
gas generation. For the non-intrusive studies, the degree of saturation, gas storage and pressure
build-up, and its effects on the geophysical data were monitored over the various phases of the HG-

A experiment, using cross-hole seismic tomography. The aim of the experiments was to investigate
differences in the seismic data due to variations within the micro-tunnel (both its content and physical
condition). Signal generation and recording take place in two boreholes drilled perpendicular to
the micro-tunnel.
Experimental set-up for non-intrusive monitoring
The 1 m diameter micro-tunnel, which constitutes the target to be monitored, is situated between
two water-filled boreholes. One borehole is inclined upwards and the other downwards. The microtunnel
is oriented perpendicular to the plane of the boreholes. Figure 3.5.1 provides an illustration
of the schematic layout. A high frequency P-wave sparker source fired every 0.25 m in the downward-directed
borehole was used to generate seismic waves recorded on a 24-channel hydrophone
array located in the upward-directed borehole. The hydrophones were spaced at 1 m intervals. By
shifting the array three times at intervals of 0.25 m and repeating each shot, a 96-channel hydrophone
array with 0.25 m element spacing was synthesized. The energy from each source shot was
also recorded on eight vertical-component geophones distributed around the micro-tunnel wall
within the plane of the boreholes; Figure 3.5.2 shows these installed within the HG-A micro-tunnel.
So far, seismic tomography measurements have been performed to study six sets of conditions
within the micro-tunnel:
1. air-filled,
2. dry sand-filled,
3. 50 % water-saturated sand-filled,
4. nearly fully water-saturated sand-filled,
5. fully water-saturated sand-filled (several months later),
6. fully water-saturated sand-filled and pressurized to 6 bars.
Figure 3.5.1: Schematic layout of the non-intrusive
seismic tomography experiment at Mont Terri
254
Figure 3.5.2: Geophones installed on wall of
empty HG-A micro-tunnel

Results of non-intrusive monitoring experiments to date at Mont Terri
The results of seismic investigations on the HG-A experiment at Mont Terri suggest that cross-hole
travel-times do not enable information to be gathered about the state of a 1 m diameter micro-tunnel
that lies midway between source and receiver boreholes separated by distances of 5-30 m. In contrast,
data recorded on geophones mounted within the micro-tunnel provide diagnostic information
about the micro-tunnel fill and the state of the micro-tunnel EDZ. Changes in the micro-tunnel fill
and EDZ result in marked variations in seismic wave arrival times, polarities and waveforms. A full
waveform inversion of the combined geophone and hydrophone data will be undertaken as part of
this study to fully assess the imaging capabilities of seismic tomography for this type of application.
3.6 Hydraulic seal material selection and installation test (EURIDICE)
At the Mol Underground Research Facility (named “HADES”), EURIDICE is currently installing
the different components for the PRACLAY Heater Test, in which the thermal behavior of a disposal
gallery will be simulated at near real scale. A gallery ("PRACLAY Gallery") has been constructed
to serve as a dummy disposal gallery. To obtain the undrained initial and boundary conditions
– to which the real disposal galleries will also be subjected – an annular seal composed of
compacted bentonite will be installed between the heated zone and the access gallery.
Figure 3.6.1 provides an illustration of this.
Figure 3.6.1 The seal will be installed between the heated and saturated part of the dummy disposal
gallery ("PRACLAY Gallery"), and the accessible part (linked to the Connecting Gallery)
The objective to obtain the undrained experimental condition requires the seal to be as impermeable
as possible at its location at the interface Boom Clay/seal. This annular seal also provides an interesting
opportunity to study the feasibility of a hydraulic cut-off of any preferential pathway to the
main gallery through the EDZ and the liner with a seal in a horizontal drift (horizontal seal). This
implies minimizing the longitudinal hydraulic gradient along the repository gallery.
A swelling bentonite will be used for the seal material, which should provide:
• a permeability one or two orders of magnitude lower than that of the Boom Clay itself, i.e.
not larger than 10 -14 m/s (~ 10 -21 m²) at its saturated state;
• a suitable swelling pressure, but not larger than the in-situ lithostatic pressure at long term;
the target value for the PRACLAY seal is 4.5 – 5 MPa.
255

This seal will be implemented as ring, assembled with pre-compacted bentonite blocks, taking the
place of the concrete gallery lining over a length of 1 m. A stainless steel confining structure encases
the bentonite ring. The central section will be closed with a steel plate and is equipped with
several flanged holes to allow the feed-through of the instrumentation and heater wiring and tubes.
Figure 3.6.2 provides an illustration of this.
Although there are actually no strict criteria for the choice of bentonite, we selected the MX80 for
several reasons: its favorable physico-chemical characteristics (including swelling pressure and hydraulic
conductivity) and the ample availability of experimental data and modeling parameters. In
addition, the experience from other URLs will allow a knowledge exchange on MX80. The MX80
will be compacted to a desired initial dry density on the basis of numerical scoping calculations taking
into account the interaction with Boom Clay as well as the technological voids resulting from
installation.
Due to the complex working conditions, the original Seal design required a major overhaul, so that
the final contractor only as been appointed in September 2008. The actual construction is planned
for end 2008 / beginning 2009.
Figure 3.6.2 (left) The bentonite ring is encased in the stainless steel confining structure,
consisting of ring flanges installed at the upstream (side of the heated and saturated dummy
disposal gallery) and downstream side of the seal and a cylindrical structure with openings
and stiffeners. The steel rings are already installed and are necessary to give support to the
clay sidewalls. (right) View from the downstream side of the seal. The four openings allow
the feed-through of the instrumentation and heater cables
256

4. Conclusions
Half-way through the final project year, Module 1 has achieved most of its objectives:
• ANDRA has succeeded in the cold compaction of a MX-80 bentonite / quartz sand mixture
prepared as a powder, to obtain the prefabricated buffer rings described in their Dossier
2005 report to the French Government. ANDRA has also successfully tested the handling of
these buffer rings and their rigidity.
• ONDRAF/NIRAS has demonstrated the feasibility of backfilling the annular void around a
horizontally disposed high level waste package, using two different emplacement techniques:
(1) projection of a dry granular material, for which sand, cement, bentonite and mixtures
thereof were used, (2) injection of a custom-made high pH grout designed to have the
required thermal, chemical and physical characteristics. Both emplacement techniques have
been tested on reduced-scale mock-ups. The grout injection technique was also tested on a
30-m long full-scale mock-up. The result still needs to be evaluated.
• NAGRA, using auger technology and a reduced-scale steel model of a horizontal drift with a
waste container disposed on a bed of prefabricated bentonite blocks, has tested the emplacement
of a range of bimodal mixtures of granular bentonite. NAGRA succeeded in
achieving the desired dry density of the emplaced buffer material.
• GRS is satisfactorily running performance tests on four seals of different bentonite-sand
composition in boreholes at the Mont Terri URL. These in-situ tests, and also the preceding
laboratory mock-up testing, indicate that the envisaged sealing function will be confirmed
by the actual tests, but at much lower seal saturation rates than predicted by computer models.
The foreseen nitrogen gas injection tests, after full seal saturation, will therefore most
probably be done after the project end date.
• NDA is executing a test program to advance non-intrusive monitoring based on cross-hole
micro-seismic tomography. In cooperation with the Swiss Federal Institute of Technology
(ETH Zurich), a full wave inversion code and an anisotropic model of the clay test environment
have been developed to interpret the seismic echoes.
The installation of a seal in the Mol URL is the only demonstration test that remains to be done.
5. Acknowledgements
This project has been co-funded by the European Commission and performed as part of the sixth
Euratom Framework Programme for nuclear research and training activities (2002-2006) under contract
FI6W-CT-2004-508851.
References
[1] ANDRA Dossier 2005 “Architecture and Management of a Geological Disposal System”,
[2] “SAFIR 2: Belgian R&D Programme on the Deep Disposal of High-Level and Long-Lived
Radioactive Waste – An International Peer Review”, NEA-OECD report, issued 2003;
[3] Nagra, 2002. Projekt Opalinuston – Konzept für die Anlage und den Betrieb eines geologischen
Tiefenlagers. Nagra Technical Report NTB 02-02. Wettingen, Switzerland.
[4] Nagra, 2007. Plötze, M., Weber, H.P., 2007. ESDRED: Emplacement tests with granular
Bentonit MX-80. Nagra Working Report NAB 07-24. Wettingen, Switzerland.
[5] Fries, T., Claudel, A., Weber, H., Johnson, L., Leupin, O. 2008. The Swiss concept for the disposal of
spent fuel and vitrified HLW. Proceedings of the ESDRED Prague Conference, June 2008.
257

[6] Miehe, R., Kröhn, P., Moog, H., (2003): Hydraulische Kennwerte tonhaltiger Mineralgemische
zum Verschluss von Untertagedeponien (KENTON). Gesellschaft für Anlagen-
und Reaktorsicherheit (GRS) mbH, Köln, GRS-193.
[7] Rothfuchs, t., Jockwer, N., Miehe, R., Zhang, C.-l. (2005): Self-sealing Barriers of
Clay/Mineral Mixtures in a Clay Repository SB Experiment in the Mont Terri Rock Laboratory
Final Report of the Pre-Project, Gesellschaft für Anlagen- und Reaktorsicherheit (GRS)
mbH, GRS-212
258

New Transport and Emplacement Technologies
for Vitrified Waste and Spent Fuel Canisters
Wilhelm Bollingerfehr 1 , Wolfgang Filbert 1 , Jobst Wehrmann 1 , Jean-Michel Bosgiraud 2
Summary
1 DBE TECHNOLOGY GmbH, Germany
2 ANDRA, France
In the context of the Integrated Project “ESDRED” - Engineering Studies and Demonstration
of Repository Designs [1] - funded by the European Commission, transport and emplacement
technologies for radioactive waste packages were subject of industrial demonstration tests.
Two national programmes - the German reference concept, which comprises the emplacement
of spent fuel elements in self shielding casks in horizontal drifts and the emplacement of vitrified
waste in deep vertical boreholes in a repository in salt rock, and the French concept,
which considers the emplacement of vitrified waste into horizontal cells in a repository in
clay (Argillites), were considered. In order to align the emplacement technologies for all categories
of heat-generating waste, the emplacement of spent fuel elements in deep vertical
boreholes was investigated as an alternative to the reference concept. For this alternative concept
and for the French concept, suitable transport and handling equipment was developed,
manufactured and will be tested under repository-relevant conditions on an industrial scale.
This paper briefly describes both disposal concepts, presents the transport and emplacement
systems developed for the waste packages of both emplacement concepts, and illustrates the
set-up of the demonstration test facilities. First results of the demonstration programme are
presented as well.
1. German reference concept for the disposal of heat-generating waste
In Germany, rock salt was selected in the early 1960s as the preferred host rock for a repository for
heat-generating waste because of its unique geohydrologic, thermal, and geomechanical properties
as self-healing impermeable rock. A large number of salt domes with huge dimensions, many of
them principally suitable to host a repository, exist in Northern Germany. In 1977, at the end of a
time-consuming selection process, the salt dome in Gorleben was selected for further exploration
regarding its suitability to host a repository for HLW. The corresponding reference concept for the
disposal of heat-generating radioactive waste (Fig. 1) anticipates the emplacement of canisters containing
vitrified waste in deep vertical boreholes, whereas spent fuel is to be disposed of in selfshielding
POLLUX ® casks in horizontal drifts inside a salt mine [2]. The POLLUX ® casks, carbon
steel casks weighing 65 tons each, are laid down on the floor of a horizontal drift at a depth of
840 m. The spaces between the casks and the drift walls are back-filled with crushed salt. In the
vertical disposal concept, which allows a temperature of max. 200 °C at the contact surfaces between
waste canisters and host rock, unshielded canisters with vitrified high-level radioactive waste
(HLW) are emplaced in boreholes with a depth of up to 300 and a diameter of 60 cm. In order to
facilitate the fast encapsulation of the waste by the host rock (rock salt), the boreholes are not lined.
Obtaining a license to construct a repository in Germany requires previous demonstration to the
competent authority that the level of protection (dose or risk) can be met with a high level of confi-
259

dence. For waste canister transport and handling systems, the proof of compliance with the regulatory
requirements can be provided by means of full-scale demonstration and reliability tests. The
transport, handling, and emplacement techniques of the POLLUX ® cask were subjected to successful
demonstration and in-situ tests performed in the 1990s. As a result, the atomic energy act was
amended in 1994. For waste canisters with high-level reprocessing waste the proof is still pending.
Figure 1: German reference concept for the disposal of heat-generating waste
2. French concept for the disposal of vitrified waste
In France, clay (Argillite) was selected as the host rock suitable for a geological repository for heatgenerating
waste (long-lived high activity), in the Bure area (Meuse Department, east of Parisian
Basin), following a long and difficult site selection process which lasted from 1993 to 1998. The
local construction of a URL (main facilities including 2 shafts & the main experimental drifts) took
place between 1999 and 2005. During the same period ANDRA produced and issued the “Dossier
2005”, which served as an official support document (the Repository concept) for the passing of the
law (Nuclear act voted in 2006) which is now governing ANDRA’s activities. The reference concept
developed in the “Dossier 2005” [3] for the emplacement and storage of C type (vitrified
waste) canisters in horizontal disposal cells is illustrated in Fig. 2.
A 40-m-long horizontal disposal cell (where the canisters are emplaced) is lined with a carbon steel
casing to support the clay formation wall and to facilitate any future retrieval operation of the C
type package (ANDRA has to take into account the reversibility factor in its repository design development).
The access gallery enables the transport shuttle to reach the cell mouth for docking operations
and the subsequent emplacement of the canisters.
The main difference between this reference concept and the technical programme developed in ES-
DRED is the length of the disposal cell which has been increased from 40 m to 100 m.
260

Figure 2: French reference concept for the disposal of C type waste in horizontal disposal cells
3. New waste canister transport and emplacement technology in Germany
3.1 Characteristics of Waste Canisters in Germany
In order to align and optimise the emplacement technologies for both categories of waste (vitrified
waste and spent fuel), alternative technical approaches were sought in Germany during recent years.
In this context, the borehole emplacement technique for consolidated spent fuel as already foreseen
for high-level reprocessing waste was reconsidered. The new fuel rod canister (called BSK 3 canister)
was designed by the German nuclear industry. It can be filled with the fuel rods of 3 PWR or 9
BWR fuel assemblies. The BSK 3 canister was designed to contain spent fuel rods with a total activity
of up to 0.8E+17 Bq, and to be capable to transfer a maximum heat load of 6 kW. The BSK 3
concept offers the following optimisation possibilities:
• The new steel canister has nearly the same diameter as the standardized canisters for HLW
and compacted technological waste, as delivered from reprocessing abroad.
• The standardized canister diameter provides the possibility to apply the same transfer and
handling technology for both categories of waste (vitrified HLW and spent fuel) and thus to
save money.
• The new BSK 3 canister is tightly closed by welding and designed to withstand the
lithostatic pressure at the emplacement level.
• Thermal calculations verified that the residual heat generation of a canister loaded with fuel
rods burned up to 50 GWd/tHM will enable its emplacement in a salt repository after only
about 3 to 7 years after reactor unloading of the fuel assemblies.
• Compared with the emplacement of POLLUX ® casks the creep process of the host rock
(rock salt) will be accelerated resulting in a faster (earlier) encapsulation of the entire waste
canister. This may reduce the requirements for geotechnical barriers.
Heat-generating vitrified HLW is contained in standard Cogema canisters (CSD-V) while low heatproducing
high-active technical waste, mainly caps and claddings, is contained in Cogema type
CSD-C containers.
261

The main waste canister characteristics, and the numbers of canisters that need to be disposed of
according to the phase-out scenario, are given in Table 1.
Table 1. Characteristics of the waste canisters for disposal of heat-generating waste in Germany
HLW Canister CSD-C BSK Canister
Number of canisters 4,778 8,764 ca. 5,525
Number of boreholes needed 30 55 95
Length mm 1,338 � 1,345 4,980
Diameter mm 430 � 440 � 440
Total mass kg ca. 492 � 850 5,226
Mass HM tHM - - 1.6
Heat generation kW
• at loading 0.02 21,220
• after 10 years 120*) 3,030
• after 30 years 67**) 1,930
*) after 9 years **) after 29 years
The BSK 3 concept, therefore, may provide a common solution for the emplacement of all types of
heat-generating radioactive waste in Germany, thus considerably reducing the necessary effort in
terms of time and costs.
3.2 Approach to developing the BSK 3 emplacement technology
A research programme was launched in order to develop and test the necessary technical components
for the transport and handling of BSK 3 canisters. The main objective was to develop the
components for demonstrating the functionality and reliability of a suitable emplacement technology.
In addition, the results of the tests and investigations are to provide all information required for
the licensing of this new back-end technology, thus meeting the legal requirements for a German
HLW repository.
In the context of the Integrated Project ESDRED [1], the BSK 3 canister transport and emplacement
concept is specific to the German reference concept in salt but it may be applicable to other host
rocks as well. The following objectives for the BSK 3 research and development project were set:
• General objective: - to develop and test the emplacement technology for BSK 3 canisters
on a 1:1 scale,
• Detailed objectives: - to prove the technical feasibility of constructing the single components
as well as of the entire emplacement system for BSK 3 canisters
- to prove the operational safety by corresponding demonstration
tests,
- to derive safety measures for the operation in a repository,
- to investigate the approvability of the emplacement system.
In accordance with this set of objectives, an emplacement system was developed for the handling
and disposal of BSK 3 canisters that comprises a transfer cask, which provides appropriate shielding
during the transport and emplacement process, a transport unit consisting of a mining locomotive
and a transport cart, an emplacement device for handling the transfer cask and lowering the
BSK 3 canister into the emplacement borehole, and a borehole lock which provides radiation pro-
262

tection during the operational phase of the repository. Figure 3 shows the components of the entire
transport and emplacement system in an underground emplacement drift. It was selected out of a
variety of different options. Aboveground, in a hot cell of a conditioning plant, the BSK 3 canister
is inserted into the transfer cask. After shipment to the repository, the transfer cask is transported by
the transport cart through the shaft to the emplacement drift underground. The mining locomotive
drives the transport cart with the transfer cask to the emplacement device. The emplacement device,
previously positioned on top of the emplacement borehole, lifts the transfer cask from the transport
cart, tilts the cask into an upright position and lowers it down onto the top of the borehole lock. The
borehole lock and the lock of the transfer cask are opened simultaneously, and the BSK 3 canister is
lowered down by means of a rope and canister grab.
Figure 3: Sketch of the BSK 3 transport and emplacement system with all necessary components
The BSK 3 emplacement system required the development of the following new components:
• a BSK 3 canister, capable to contain the rods of 3 PWR or 9 BWR fuel assemblies,
• a transfer cask for the safe enclosure and transport of the BSK 3 canister,
• a suitable emplacement device,
• a borehole lock, and
• a transport unit consisting of a transport cart and a battery-operated mining locomotive for
rail-bound transport in the repository.
An early idea to reuse the transport cart, which had been successfully used during the demonstration
tests with POLLUX ® casks in the 1990s, had to be discarded. From an economical point of view it
was less expensive to build a new one than to modify the existing one. However, the batteryoperated
mining locomotive is used again.
3.3 Demonstration programme with BSK 3 canister
Due to the lack of an underground laboratory in salt rock in Germany it was decided to perform the
demonstration tests in a surface facility. For this purpose, a former turbine hall of a power station
owned by E.ON in the village of Landesbergen in the vicinity of Hanover, Lower-Saxony, has been
rented. This building provides the possibility to simulate the emplacement process of a BSK 3 canister
in a vertical borehole. The components of the emplacement system are assembled at a level of
10 m above ground floor (Fig. 4), while a 10-m-long vertical steel metal casing simulates the em-
263

placement borehole. The BSK 3 canister is lowered down by the grab of the emplacement device
and – unlike in a real repository – removed again for further tests.
The test programme comprises demonstration tests, simulation tests, and tests to resolve operational
disturbances. In total, approx. 500 complete emplacement cycles (500 times filling the borehole
with a canister and 500 times retrieving the same canister) will be simulated in order to obtain information
on the reliability of the entire system and of each component.
Figure 4: Design of BSK 3 test facility
Figure 5 shows a view onto the completely new test platform of the test facility in Landesbergen.
All components necessary for the demonstration and testing of the BSK 3 emplacement system are
positioned on the test platform.
Figure 5: Full-scale demonstration platform at the test site in Landesbergen
3.4. Achievements
All the components have been designed in detail, the drawings and reports evaluated by external
experts confirming the compliance with the requirements of the German mining law and atomic energy
act. The components were manufactured on a full-scale between summer 2007 and spring
2008. The construction work to prepare a suitable test platform was performed between February
and April 2008. The single components (mining locomotive, transport cart, BSK 3 dummy, transfer
264

cask, emplacement device and borehole lock) were delivered to the test site between April and June
2008. After the individual components had been accepted on site (SAT), the demonstration programme
- performed in two shifts - was started and will last until the end of 2008. The intention is
to prove the reliability of the emplacement system by means of a large number of demonstration
tests (approx. 500) and to draw conclusions and give recommendations for the industrial application
in a real repository.
4. French concept for the disposal of heat-generating waste
4.1 Characteristics of waste canisters in France
The C type package (vitrified waste) consists of a primary package, the well known Cogema CSD-
V, contained in what is called an overpack, i.e. a second envelope of 55-mm-thick carbon steel. It is
equipped with 12 ceramic sliding runners (6 mounted at each end) which reduce the friction force
exerted (when the canister is pushed inside the cell lining) and at a later stage prevent corrosion
sticking (between the outside wall of the canister and the inside wall of the liner), and which are to
facilitate any future retrieval operations.
The C type canister weighs approximately 2 tons at a length of 1.6 m and an outer diameter of about
60 cm. (Size & weight of the canisters may vary, according to the variety of C type canisters produced).
There are various existing scenarios concerning the production of C type canisters. It is envisaged
to store between 30 000 & 50 000 of them.
4.2 Approach for developing the Pushing Robot emplacement technology
Within the frame of the European IP ESDRED project, a research & development programme was
launched in order to design, manufacture and test the necessary technical components for the transport
and emplacement of C type canisters into horizontal disposal cells in Argillite. The main objective
was to develop a system for demonstrating the functionality and reliability of a suitable emplacement
technology, called the Pushing Robot. In addition, the results of the tests and investigations
should later provide to the Public at large valuable information on practical, down-to-earth
operations likely to be encountered in the future industrial storage. After testing, the demonstration
model is to be moved and re-erected in the CTe (ANDRA’s technological show-room, now in construction
in Saudron, near the Bure site).
The following objectives for the Pushing Robot research and development project were set:
to develop and test the emplacement technology for the Pushing Robot on a 1:1 scale,
to prove the technical feasibility of constructing the entire emplacement system for the C type
canisters and the single components,
to investigate the robustness of the emplacement system,
to present the Pushing Robot Demonstrator at work in the CTe in spring 2009.
The emplacement system comprises a transfer cask, which provides appropriate shielding during
the transport and emplacement process, a transport unit consisting of a shuttle with hydraulic motor,
a docking table for connecting the transfer cask to the cell mouth, and finally, a Pushing Robot to
move the canister into the disposal cell. Figure 6 shows the components of the entire transport and
emplacement system in an underground emplacement drift mock-up.
265

Figure 6: Sketch of the Pushing Robot transport and emplacement system with all necessary components
in its testing configuration
4.3 Demonstration programme with C type waste package
Due to the lack of space available in the Bure URL, it was decided to perform the demonstration
tests in a surface facility. For this purpose, a former workshop in the vicinity of Saint-Etienne
(Loire Department) has been rented. This building provides the necessary lifting means and length
to simulate the emplacement process of a C type canister into a 100-m-long horizontal disposal cell.
The components of the emplacement system are now being assembled.
The test programme comprises demonstration tests, simulation tests, and tests to resolve operational
disturbances. Once all the test configurations are passed and solved, a total of approx. 500 hours of
operations will be simulated in order to obtain information on the reliability of the entire system and
of each component. Emergency situations, for which remedial devices have been developed, will
also be tested.
4.4 Achievements
All the components have been designed in detail. The manufacturing of all the system components
is also completed. The components are now progressively delivered to the test site between June
2008 and mid August 2008. After the individual components have been accepted on site (SAT) the
demonstration programme will start in September and should last until the end of 2008. Figures 7 &
8 show the status of preparation of the shielding cask and of the test bench respectively.
266

Figure 7: View of the shielding cask during factory commissioning
Figure 8: View of the test bench under erection at site
5. Acknowledgements
The development and demonstration of the German BSK 3 emplacement technology - as part of the
IP ESDRED - is financed by the European Commission, the German Ministry of Economics and
Technology represented by the project management agency Karlsruhe (PTKA) and the German nuclear
industry represented by GNS. The latter in particular provides money for manufacturing the
components of the emplacement system.
The development and demonstration of the French Pushing Robot emplacement technology - as part
of the IP ESDRED - is also co-financed by the European Commission within the frame of the 6 th
Euratom Framework Programme (2002-2006).
References
[1] www.esdred.info,
[2] ENGELMANN, H.-J., et al., 1995, “Systemanalyse Endlagerkonzepte”, Abschlussbericht,
Hauptband, DEAB T 59,
[3] Dossier 2005 – “Architecture and management of a geological disposal system”
267

268

Emplacement of Heavy Canisters into Horizontal Disposal Drifts using Fluid
(Air/Water) Cushion Technology
Jean-Michel Bosgiraud 1 , Wolf K. Seidler 1 , Louis Londe 1 , Erik Thurner 2 , Stig Pettersson 2 ,
Bo Halvarsson 3
1 ANDRA France, 2 SKB Sweden, 3 Vattenfall Power Consultants Co Sweden
Summary
The disposal of certain types of radioactive waste canisters in a deep repository involves handling
and emplacement of very heavy loads. The weight of these particular canisters can be in
the order of 20 to 50 metric tons. They generally have to be handled underground in openings
that are not much larger than the canisters themselves as it is time consuming and expensive
to excavate and backfill large openings in a repository. This therefore calls for the development
of special technology that can meet the requirements for safe operation in an industrial
scale in restrained operating spaces. Air/water cushion lifting systems are used world wide in
the industry for moving heavy loads. However, until now the technology needed for emplacing
heavy cylindrical radioactive waste packages in bored drifts (with narrow annular gaps)
has not been developed or demonstrated previously. This paper describes the related R&D
work carried out by ANDRA (for air cushion technology) and by SKB and Posiva (for water
cushion technology) respectively, mainly within the framework of the European Commission
(EC) funded Integrated Project called ESDRED (6 th European Framework Programme). The
background for both the air and the water cushion applications is presented. The specific characteristics
of the two different emplacement concepts are also elaborated. The various phases
of the Test Programmes (including the Prototype phases) are detailed and illustrated for the
two lifting media. Conclusions are drawn for each system developed and evaluated. Finally,
based on the R&D experience, improvements deemed necessary for an industrial application
are listed. The tests performed so far have shown that the emplacement equipment developed
is operating efficiently. However further tests are required to verify the availability and the reliability
of the equipment over longer periods of time and to identify the modifications that
would be needed for an industrial application in a nuclear and mining environment.
1 Introduction and background
In the industry, air and water cushion systems for lifting and handling heavy loads (up to several
hundred tons) are being used world wide. The application of such a technology is however generally
limited to situations where the air or water cushions are acting on flat and smooth “sliding” surfaces.
Applications where the air/water cushions act on cylindrical surfaces (i.e. either on the surface
of the load to be moved or on the surface on which the cushions “slide”) are not so common.
The specific applications considered within the scope of the ESDRED work [1] are most likely
without precedent, since the emplacement concepts relate to moving and placing cylindrical waste
canisters inside horizontally bored disposal drifts (cells) whose cylindrical walls have a rough surface.
At the same time the annular gap between the outside diameter (OD) of the canister and the
inside diameter (ID) of the disposal drift wall is limited to a few cm. Working underground with
radioprotection constraints, requiring remote control (i.e. no direct access or vision of the moving
load) adds a further level of complexity. The two emplacement concepts that were developed and
tested are presented below.
269

1.1 Water Cushion Application
SKB (Sweden) and Posiva (Finland) have since 2001 a joint project called the KBS-3H disposal
concept. The main components in the system are shown in (Figure 6). The Super Container (SC) to
be emplaced consists of a copper canister (containing Spent Fuel), a buffer annulus (made of bentonite
rings) and an outer perforated steel cylinder equipped with short feet. The weight of the SC is
in the order of 45 tons, with an OD diameter of 1.77 m and a length of 5.57 m. It is emplaced inside
a 300 m long horizontal disposal drift excavated in granite by using a water cushion deposition machine.
Figure 6: Main components of the KBS-3H emplacement concept (SKB & Posiva)
The deposition equipment for the KBS-3H concept has become part of the ESDRED Project and
has been partly financed by EC within the 6th Framework Programme since February 2004. However,
between 2001 and 2004, SKB and Posiva had already carried out prototype testing using both
air and water cushions in order to prove the feasibility of this technology for cylindrical objects and
surfaces. Water was selected as the appropriate lifting medium for two main reasons: i) it is a fluid
compatible with the host formation (granite), ii) it avoids the important pressure loss that would be
experienced over a 300 m long air umbilical. As an alternative, an air compressor with a 75 to 100
kW electrical motor mounted on the mobile emplacement system would have created too much heat
and also increased the size and weight of the emplacement machine. Unlike air, water can be continuously
recycled through a small water tank mounted on the deposition machine. In this case, a 7
kW electrical pump is sufficient to provide the necessary water pressure and flow.
270

1.2 Air Cushion Application
ANDRA in France has selected the air cushion technology for emplacement of SF (Spent Fuel) canisters
as well as for the transport and placement of sets of buffer rings. In the ANDRA case (unlike
the KBS-3H concept described above), the sets of buffer rings and the SF canisters are handled
separately and not as one package. The ANDRA system has also been developed within the framework
of the ESDRED Project. The main components in the ANDRA disposal system for emplacement
of SF canister are shown in (Figure 7).
The buffer (bentonite/sand) rings are assembled in sets of four (4). Each set of rings has a weight of
17 tons, with an OD of 2.25 m and a length of 2 m. The weight of the SF canister is 43 tons and it
has an OD 1.25 m and a length of 5.39 m. The disposal cell excavated in clay has a length of approximately
40 m. Only the emplacement of the SF canisters is described in this paper.
Figure 7: Main components of ANDRA’s emplacement concept for a spent fuel canister disposal
cell
Air was selected as the appropriate lifting medium for two main reasons: i) it is a medium compatible
with the host formation (clay), ii) the pressure loss experienced over a 40 m long umbilical is
compatible with the proper functioning of the air cushions.
2 Development of the Demonstrators
2.1 Testing initial prototypes
The main objective of the original test campaigns was to investigate if standard cushions (i.e. “off
the shelf” components) could be efficiently used for cylindrical objects/surfaces with a limited diameter.
As these cushions are designed to operate on flat surfaces, their structure has to be curved
to the appropriate radius. A successful trial was a pre-requisite to the further development of a full
size demonstrator. A second objective was to determine the appropriate operating parameters for a
good performance of the future full scale emplacement system. The cushions are fixed onto a pallet
which supports the pay load. This pallet lifts when the cushions are activated. The actual lifting
height of the pallet depends on the fluid flow, the pressure at cushion inlet and the design of the
cushion.
271

In the case of SKB and Posiva, the test bench used for the prototype had the same diameter as the
KBS-3H disposal drift but the length and the weight of the mock-up canister was reduced to ¼
scale. The load during the tests was thus limited to approximately 12,250 kg (1/4 of the real load)
and only eight (8) pairs of cushions were installed instead of 32 for the full load. As the inclination
tolerance for the real disposal drift is 2 degrees (°) ±1°, an inclination of 3° was therefore accomplished
on the prototype test rig. Preliminary prototype tests were performed successfully at the
SOLVING facility in Jakobstad, Finland. The two first tests were performed with air in April and
July 2003, and the third test was performed with water in March 2004.
In the ANDRA concept, the use of air cushion technology as emplacement means for the spent fuel
(SF) canister had not been tested prior to the ESDRED Project. The feasibility tests carried out by
SKB for the KBS-3H Super Container (before the start-up of the ESDRED Project) could not be
considered as a solid enough basis for confirming the feasibility of ANDRA’s specific application.
This was primarily due to the fact that ANDRA’s canister and disposal cell have a smaller diameter
than the equivalent SKB components and because ANDRA’s SF canister has a higher linear weight.
Therefore, ANDRA decided that it also needed to perform preliminary prototype testing, similar to
that of SKB, with an air cushion supplier (BERTIN), in France. These tests were successfully carried
out from July 2004 to January 2005 and later repeated by BERTIN, on behalf of ME-
CACHIMIE, who had been selected by ANDRA as the final supplier for the full scale deposition
equipment. These tests [2] were carried out using a dummy canister with a 1:1 scale outer diameter
of 1.25 m, a 1:3 scale length of 1.93 m and a 1:3 scale weight of 13.74 tons (instead of 43 tons) as
compared to the real canister. The number of air cushions used was six (6) instead of 18 for the real
case. This preliminary prototype testing confirmed that the air cushions after certain modifications
were working effectively and could subsequently be used even for a heavy cylindrical object with
an outer diameter of only 1.25 m. The main operating parameters were also determined for this specific
application.
2.2 Selection of Contractors
After the prototype testing was completed, it was decided to proceed with the next step of the R&D
programme by implementing the full scale demonstrator phase for each of the two emplacement
systems being considered. The bid and tender process resulted in two separate contracts which were
awarded respectively to CNIM (France) by SKB/Posiva and to MECACHIMIE (France) by AN-
DRA.
The work programme in the two cases started with preliminary and detailed studies, followed by the
manufacturing and erection of the equipment. The systems then underwent the testing campaign per
se. These activities were carried out in line with the main milestones posted in the initial schedule
of work. The present paper focuses on the test campaigns and the related results, taking for granted
that the engineering, supply and manufacturing is of limited interest to the reader [3].
3 Full-Scale Demonstrator by SKB and Posiva
For SKB and Posiva the FAT were carried out at the CNIM factory, at La Seyne-sur-Mer, in
France, in February 2006, before delivery of the equipment to the Äspö HRL in Sweden. However,
during the FAT, all the planned tests could not be performed. Once the equipment was installed in
the real underground conditions and during the initial start-up of the SAT in May 2006 it was discovered
that the unbalance of the SC could not be controlled and derailing occur during the testing.
Preventive actions had to be taken. The deposition machine was therefore retrofitted with a guidance
system intended to prevent the uncontrolled rotation of the SC. At the same time a fork was
272

attached to the electrical cart radioprotection shield to improve the alignment of the load vis-à-vis
the water cushion pallet. Following this retrofitting, the effective SAT at the HRL could start.
An overview of the set-up of the equipment at the Äspö HRL test site is shown in (Figure 8). The
picture is taken from the rear of the chamber in which the emplacement equipment was prepositioned
in front of the mouth of the disposal drift.
Figure 8: Set-up of equipment at the Äspö HRL (level -220 m) test site. The Super Container is inside
the transport tube with the shielding gamma gates open
Two SCs (built with SF copper canister and buffer material mock-ups) and two Distance Blocks
(spacers with dimensions similar to the canister mock-up) were manufactured for the purpose. The
mock-ups were representative of the real payloads, with the correct physical dimensions and
weights. To ensure that the guidance system functioned properly, it soon became evident that the
lifting height of the water cushions had to be reduced. It was therefore decided to replace the original
water cushions with a different brand of cushions that had a reduced lifting height and that also
had less sensitivity to load variations. The pallet was also equipped with four lift sensors for indication
of the lifting height. The SAT was carried out in accordance with a detailed SAT Programme
and included the following check operations detailed in Table 2.
Table 2: Testing Sequence for the KBS-3H
Test designation
Checking of the HMI (Human-Machine Interface) and Control System with power on
Checking of the machine moving parts using the portable controls
Checking of the machine moving parts from the control room
Checking of the water cushion pallet hydraulic circuit
Deposition machine tests without load
273

Deposition machine tests with load
All tests from the SAT were recorded. The average transport speed obtained (cycle time) was 14.9
mm/sec.
The first tests with the machine showed that there was a high risk that the rotation of the container
about the long axis could increase cumulatively each time the container was moved due to the gap
between the guides on the pallet and the slide plate. This gap is 5 mm, which allows the container to
rotate approximately +/- 0.2 – 0.3°. As soon as a rotation of the canister is detected by the sensors, a
compensating system (ballast system) is activated to offset the rotation phenomena.
Another important observation was that if the container, together with the pallet and the slide plate,
was rotated more than 3.5 - 4°, then this movement could create problems for the proper functioning
of the water cushions (due to the uneven load distribution resulting from such a configuration).
As reported previously, the water cushions are sensitive to load variations (the problem that can occur
with a too big rotation is that some of the cushions, which get more loaded than normal, are not
able anymore to lift the container). After considerable effort, the conclusion was that it is impossible
to properly handle an unbalanced SC with the presently developed water cushion system.
During the tests it was observed that the system is sensitive to the alignment between the emplacement
equipment and the drift so that having the best possible initial alignment of the whole set up is
of paramount importance.
The Demonstration test period started immediately after the completion of the SAT. During this test
phase, the SC was repeatedly transported to the far end of the deposition test drift (only 95 m instead
of up to 300 m for a real application) and recovered. According to this endurance demonstration
test programme, the goal was to make one deposition and subsequent recovery per day. The
cumulative travel distance of the deposition equipment to May 2008 was approximately 22 km. The
transportation of the SC was performed in both manual and automatic modes.
The performance requirement for an average deposition speed of 20 mm per second (mm/s) was
finally reached after making some minor adjustments to the water cushion control valves. However,
there continues to be a problem with the water cushion pressure relief valves. They have a tendency
to jam resulting in high cushion pressures that can damage the cushions, which may result in an uneven
lowering of the SC (the uneven lowering will result in a rotation of the SC, which the ballast
system cannot compensate for). The function and reliability of these relief valves is presently being
reviewed.
Besides the problem with the water cushion valves and some initial problems while running the machine
in automatic mode (due to damaged laser sensors on the slide plate), the tests have been performed
without any major problems. The tests have also shown that there is no problem controlling
the container rotation if the set-up is well aligned from the very start-up of operations. However, the
system is more sensitive when moving forward than when reversing.
4 Full-Scale Demonstrator by ANDRA
The test campaign related to the emplacement of the CU1 (SF) canister took place from May 2006
to September 2006. This campaign started with the erection of a complete test bench in the configuration
shown below in (Figure 9).
274

Polycarbonate tube Gamma gates Dummy canister
Supporting frame
Electrical cart
Figure 9: ANDRA - Set-up of the CU1 emplacement system at the HRB test site (SGN-Mécachimie’s
premises in Beaumont-Hague)
The complete test bench was composed of the following main parts:
a supporting frame equipped with adjustable feet for simulating the geometrical defaults
likely to be encountered in a real disposal cell underground or/and the steps/misalignment
between the docked shielding cask and the disposal cell mouth;
a polycarbonate tube (for viewing during demonstrations) with stainless steel sliding track
sections fixed to the full length of its invert. These sections have two (2) guide rails welded
to the upper surface of the sliding track. When the SC canister is set down onto the rails,
there is enough clearance between the bottom of the canister and the top of the sliding track
so that once the air cushions are deflated the slide plate attached to the electrical cart can be
advanced, i.e. the pallet and sliding plate can be moved separately from the SC. The rails
also act as a guide for the slide plate and air cushion pallet, which follow the path of the SC
to its final destination. The ID of the polycarbonate tube is similar to the diameter of the inner
steel sleeve in a real disposal cell;
two gamma gates: one attached to the cell mouth and one attached to the shielding cask.
The shielding cask gate is motorized and it moves the passive cell mouth gate;
an electrical cart (the deposition machine) equipped with a radioprotection shield and an
electrical pushing jack for advancing the SC in 1 m increments (see Figure 10);
a slide plate attached to the body of the electrical cart;
an air cushion pallet attached to the pushing jack;
a control & monitoring console (see Figure 10);
a 43 ton dummy canister (5.4 m long) whose centre of gravity could be adjusted longitudinally
and radially.
275
Air hose winch

Figure 10: Details of electrical pushing jack connected to the spent fuel canister (left)
and Control & Monitoring Console (right)
The primary objectives and challenges in this test programme were as follows:
to show that the emplacement equipment could meet or exceed all the specified technical
performances, including the successive emplacement (and subsequent retrieval) of the
dummy canister in automatic mode, inside the polycarbonate/steel tube , the automatic closing
and opening of the gamma gates and finally the specified average travel speed over a
complete emplacement cycle;
to demonstrate that the emplacement equipment could pass over obstacles such as the recesses
in the door frames created by the shielding gates or over the discontinuities between
two (2) consecutive sections of guide rails. For this purpose, the use of a sliding plate could
not be avoided;
to evaluate the sensitivity of the system to the various construction defaults (steps, misalignments,
etc) likely to be encountered underground and to any off-centre (radial or longitudinal)
location of the centre of gravity of the dummy canister;
to identify the weak points of the system likely to require some re-engineering and/or retrofitting
in the real industrial application;
to identify some potential improvements (mainly in terms of ruggedness and performance.
All tests executed during the FAT & SAT were recorded in a test report. What follows is a condensed
overview of the results with reference to the main functional requirements as well as other
observations noted during the tests.
The commissioning of the emplacement system took place during the months of May and June
2006. PLC programming was a large part of the work during that period. The main difficulties encountered
during this commissioning period (and their solutions) are listed below:
the friction coefficient between the lower face of the slide plate and the steel invert (sliding
track) of the polycarbonate tube turned out to be bigger than anticipated. Consequently, the
pushing force, which had to be exerted by the electrical cart’s pushing jack, exceeded the
capacity of that jack. This problem was solved by attaching a Teflon sheet onto the lower
face of the slide plate;
at the end of each 1 m stroke of the pushing jack (moving the air cushion pallet over the
slide plate), the air cushions had to be deflated to lower and place the canister on the sliding
276

track rails. Subsequently, the sliding plate was advanced by another 1 m. The time needed
for deflating and purging the air from the air cushion system turned out to be too long. Consequently,
the cycle time specified could not be achieved. This problem was solved by the
installation of a quick relief (purge) valve;
the compressed air feeding the air cushions carried considerable moisture. This resulted in
the formation of condensation within the air cushions following the quick pressure drop. As
a result, the rubber part of the air cushion tended to separate from its steel supporting plate.
Replacement cushions were glued with a water resistant compound and the problem was
solved;
the presence of moisture in the air also impacted the operation of the flow control system. A
regular purging of the electro-valves turned out to be necessary on a regular basis, i.e. at the
end of every emplacement cycle;
as originally designed, the air cushions could raise the air cushion pallet higher than the top
of the guide rails inducing a tendency for derailing the system. This problem was solved by
increasing the height of the guide rails by adding a 5 mm band spacer underneath the rail;
the air cushions also turned out to be quite sensitive to individual load variation. This phenomenon
appeared mainly when simulating the longitudinal imbalance of the SF canister. In
the most critical simulation tested (combination of longitudinal imbalance together with a
change of inclination of a tube section), the canister could not be moved.
Despite the issues noted above, the test programme turned out to be a complete success. The specified
emplacement performances were exceeded as the average emplacement speed over one complete
cycle was found to be 1.8 m per minute (m/min) versus the 1.2 m/min specified. In addition,
the SF canister emplacement process turned out to be very smooth, without shocks. The stability of
the canister on the pallet was maintained even in the case of radial load unbalance or of geometrical
defaults in the polycarbonate/steel sleeve.
Issues not fully solved within the framework of the test programme, but that should be addressed in
a future version of this equipment, are listed below:
since the air cushions are sensitive to load variation, a more accurate air flow control is
needed, such that fine tuning of each air cushion is possible;
in order to avoid derailing of the air cushion pallet, the air cushion lifting height must
not only be monitored, but also controlled (see previous point);
since the air cushions are sensitive to moisture content in the air, the compressor should
be equipped with a air-dryer;
the air feed inlet should be modified so that the air cushions can be activated and deactivated
more quickly thus resulting in a reduced overall cycle time;
in automatic mode, the winding and unwinding of the air hose umbilical attached to the
back of the electric cart was not perfect and “needed a hand” from time to time. This
was due, at least in part, to the friction coefficient of the hose on the invert slide track,
which created a parasitic (drag) force. A different hose material might reduce the friction
coefficient (and also the wear on the hose) and consequently reduce the drag force
exerted on the electrical trolley. Finally, a spooler mounted on the hose winch would
improve the winding / unwinding of the hose on the winch drum;
alternately a more powerful electrical motor mounted on the electrical cart could compensate
for the friction force (drag) exerted by the hose;
277

the very heavy weight (43 ton) of the SF canister induced some inertia efforts, which
were a real strain on the electrical pushing jack frame, which occasionally emitted some
“cracking noise”. A stiffer jack frame would reduce the stresses and the bending effects
on the jack;
Finally a slide plate made of composite material (carbon fibre or similar) instead of
stainless steel would help to reduce the friction between the bottom of the slide plate
and the top of the slide track fixed to the invert of the polycarbonate tube.
5 Conclusions
The series of industrial scale tests carried out from May-June 2006 to September 2007 by
SKB/Posiva and by ANDRA on their respective emplacement equipment helped to validate the use
of fluid cushion technology for placing heavy loads in very confined spaces. This work also identified
some of the limitations of the equipment as well as the necessary refinements/modifications
that should be implemented prior to a full scale industrial application in a future deep geological
repository in clay or granite. These tests and their results were compiled in a report [5].
5.1 Conclusions Related to the Testing of the Water Cushion System (SKB/Posiva)
The tests performed have shown that the emplacement equipment designed and fabricated within
the scope of the ESDRED Project can operate effectively for the transport and deposition of Super
Containers with a weight of 45 tons in horizontal drifts excavated in hard rock. Further tests are
however required to verify the availability and the reliability of this equipment over longer time periods.
It has also been observed that the water cushion technique used by SKB/Posiva is sensitive to load
variations. This means that the Super Container to be transported must be well balanced. This requirement
implies that all fuel positions in the canister must be completely filled with fuel elements
or fuel dummies. Finally, the system is also sensitive to the set-up alignment between the transport
tube for the Super Container, the deposition drift and the start tube for the deposition machine.
5.2 Conclusions Related to the Tests of the Air Cushion System (ANDRA)
The tests performed have shown that the emplacement equipment designed and fabricated within
the scope of the ESDRED Project can be operated effectively for the transport and emplacement of
spent fuel containers with a weight of 43 tons in mock-ups of horizontal disposal cells. Further tests
will also need to be conducted in real underground conditions and over a longer period of time to
assess the availability and the reliability of this equipment. It has also been observed that the air
cushion technique used by ANDRA is sensitive to load variations. This means that the air cushions
must be individually monitored and controlled. Finally, an efficient spooling system is considered
necessary for a proper functioning of the air hose winch.
References:
[1] “Input Data and Functional Requirements” (Project Deliverable Module 3 WP1 D1).
[2] “Report on Prototype Test for Spent Fuel Canister CU1” (Project Deliverable Module 3 WP2
D2).
[3] “Detailed Design and Manufacturing of Equipment” (Project Deliverable Module 3 WP3 D3).
[4] “Commissioning Report” (Project Deliverable Module 3 WP4 D5).
[5] “Module 3 Final Report” (Project Deliverable Module 3 WP5 D6).
278

Application of Low pH Concrete in the Construction and the Operation of
Underground Repositories
José Luis García-Siñeriz 1 , Mª Cruz Alonso 2 , Jesús Alonso 3
1 AITEMIN, Spain
2 IETcc-CSIC, Spain
3 Enresa, Spain
Summary
Module 4 of ESDRED IP project deals with the development and demonstration of low-pH
concrete formulations suitable for the construction of underground repositories for the disposal
of high activity wastes. The use of low-pH concrete instead of conventional OPC based
one will avoid the potential physicochemical transformations and changes in the radionuclide
confinement properties of the disposal components due to the hyper alkaline plume. Two
kinds of application have been addressed: the construction of plugs for drifts in crystalline
rock, and rock support in both crystalline and clayey rock. In all these cases the wet shotcrete
method has been applied.
Low-pH shotcrete formulations were developed and tested in Spain and thereafter two shotcrete
plugs have been built, one at Äspö HRL in Sweden and then, a second one at the Grimsel
Test Site in Switzerland. The first plug has been loaded up to failure with a hydraulic pressure
provided by a pump, and thereafter dismantled and analysed. The swelling pressure of a
re-saturated bentonite barrier loads the demonstration plug built at Grimsel. The later test is
going on at present.
1. Introduction
The construction and closure of underground repositories for the disposal of high activity wastes
(high level vitrified waste and spent fuel) will require the use of big amounts (up to thousands of
tons) of cementitious materials for structural support and for the construction of auxiliary structures
needed for the operation of the repository. Besides other applications, most underground repository
concepts consider the use of cementitious materials for the construction of temporary or permanent
plugs and rock wall support. For instance, the use of concrete for rock support will be a key issue
for repository concepts in clayey rock to guarantee the stability of the excavations (shafts, main tunnels
and deposition drifts), and plugs are required for confining backfills in repository tunnels and
shafts in the waste application phase and in the final phase of closing the repository.
OPC based concrete will develop a pore water pH above 12.5 and the propagation of this alkaline
fluid into the clay-based sealing materials (the bentonite) or into the geological medium (clay or
granite host rocks) can last for a very long time (up to thousands of years). This phenomenon,
called the hyper-alkaline plume or high-pH plume, may cause physicochemical transformations and
changes in the radionuclide confinement properties of the disposal components.
The hyper alkaline plume can be avoided if low-pH cements are developed and used for concrete
formulation. Another issue addressed in relation to the construction of concrete plugs is the use of
the shotcreting technique, which is a standard for rock support. This technique provides a very good
contact between concrete and rock, filling all voids and holes, even at the roof part. Although the
utilization and performance of standard shotcrete in conventional construction works is well known,
even in the construction of underground sealing plugs, as detailed by Bárcena et al. (2003) [1], there
279

is no experience in either the workability or the performance of low-pH shotcrete, therefore, testing
of this specific material under realistic conditions is required.
2. Methodology
�
The objectives of Module 4 of ESDRED have been to develop low-pH cementitious materials and
to test them at full scale in the actual underground environment. Two kinds of application have
been addressed: the construction of plugs for drifts in crystalline rock, and rock support in both
crystalline and clayey rock. In all these cases the wet shotcrete method has been applied.
The functional requirements for both applications were established by the national radioactive
waste management agencies involved in the research, ENRESA, NAGRA, SKB, POSIVA and
ANDRA, as summarised in Table 1 and Table 2.
Table 3: Main functional requirements for the
shotcrete plug
Requirement Target
Pore water pH < 11
Hydraulic conductivity K < 10 -10 m/s
Final mechanical properties:
� Young Modulus
� Compressive
strength
< 20 GPa
10 MPa
Workability > 2 h
Pump ability 250m
Peak hydration temperature 40º C
Construction rate 1 m/day
280
Table 2: Main functional requirements for
rock support
Requirement Target (updated)
Pore water pH < 11
Mechanical properties:
� Compressive � 10 MPa (36 hours)
Strength
� 20 MPa (7 days)
� 30 MPa (28 days)
� 40 MPa (90 days)
� Young Modulus � 15 GPa (7 days)
� 20 GPa (28 days)
� Bonding � 0.5 MPa (7 days)
� 0.9 MPa (28 days)
Durability (sulphate resistant)
Workability � 2 hours
Pump ability > 15m
Slump 15 – 20 cm
Use of organic components
(fibres or admixtures)
Compatible with PA,
needs to be studied
Steel fibres Steel (or glass) fibres
compatible with PA,
needs to be studied in
a later phase.

2.1 Low-pH concrete design
The concrete can be seen as a composite material composed of an aggregate skeleton bound by a
paste matrix. The paste itself is composed of the low-pH cement formulation, water and the chemical
admixtures. As most of the physicochemical reactions occur at the paste phase, the compatibility
among different constituents can be assessed in paste evaluations, the aggregate being almost inert.
Thus, the selection of the concrete components was divided in two stages: paste components and
aggregates proportioning.
Paste components
Several low pH cement formulations were developed and eleven of them (seven based on CAC and
four on OPC) were chosen for the shotcrete design process. The selection was made considering
their pore fluid pH at 90 days and their setting time, see more details at García Calvo et al. (2008)
[2]. Besides, the other materials tested for the paste were:
� Two types of super plasticizers: SP-1 (Policarboxilate, pH = 4,3) and SP-2: (Naphthalene
formaldehyde; pH = 7,5)
� Two types of accelerating admixtures: Ac-1 (Liquid formed by special inorganic substances;
pH = 12), and Ac-2 (Liquid, non-alkali, formed by inorganic substances; pH = 3)
Different tests have to be performed successively to determine suitable combinations of low-pH
cement formulation and the admixtures to be used. Calibrated siliceous sand (standard UNE-EN
196-1) was used as aggregate in the mortar samples evaluated in this phase. Four mixes of cement
formulation with suitable admixtures were finally selected. They are compiled in table 3 including
their pore fluid pH and their compressive strength in mortar samples.
Table 3: Paste components selected for basic concrete designs
Cement formulation Accelerator
Super plasticizer
281
w/c
pH in mortar (90
days of curing)
CS (28 days
of curing)
70%CAC-20%SF-10%FA Ac-2 SP-2 0.52 11.1 15 MPa
70%CAC-10%SF-20%FA Ac-2 SP-2 0.49 11.5 17.5 MPa
60%OPC-40%SF Ac-2 SP-2 0.77 11.1 20.6 MPa
35%OPC-35%SF-30%FA Ac-2 SP-2 0.67 10.9 11.4 MPa
CS: Compressive Strength measured on mortars of equivalent consistency.
Aggregate proportioning
Around 70 % of the concrete is made of aggregates and they strongly influence water demand,
workability, pump ability and project ability of the concrete. To select suitable aggregates, two
main considerations arise:
1. The suitability of aggregates in terms of strength, surface hardness, dimensional stability
and the resistance to alkali-aggregate reactions, among others.
2. The aggregate grading, i.e., the distribution of the size of the particles, as described by
Fernández-Luco et al. (2005) [3].
Two types of aggregates were considered in the design of the concrete mixes. For the short plug,
made in Äspö (Sweden), crystalline rock from the excavation was crushed and sieved to produce
both fine and coarse aggregate; the shape of these aggregate was flaky and texture was harsh. For

the case of the long plug elaborated in Grimsel (Switzerland), more suitable aggregates were used,
made of natural siliceous gravel and river sand. To determine the relative proportions of each aggregate
fraction, the reference grading limits of the Sprayed Concrete Association (SCA) were
used, slightly adapted to the actual maximum size of the coarse fraction selected.
Basic low-pH concrete design and main properties
The integration of concrete components was made by means of the absolute volume method using
the aggregates and the paste components selected. A cement content of approximately 300 kg/m 3
was determined and the water was adjusted in trial mixes for a slump in the range 12–17 cm. During
experimental trials, a formulation based on CAC showed variations and instability (strong
thyxotropic behaviour) at the fresh state, and thus it was rejected for further studies. The nominal
compositions of the low pH concretes suitable for pumping and shotcreting are given in Table 4.
Cement formulation
Table 4: Nominal composition of basic concrete types
Short Plug (aggregate from the excavation) Long Plug (convent. aggregates)
70%CAC+20
%SF+10%FA
60%OPC+40
%SF
282
35%OPC+35%S
F+30%FA
60%OPC+40%SF
Water (kg/m 3 ) 262 277 237 230
Binder (kg/m 3 ) 310 307 316 275
Water/binder 0.85 0.9 0.75 0.84
Filler (kg/m 3 ) - - - 70
Gravel (kg/m 3 ) 621 615 635 -
Fine Gravel (kg/m 3 ) 201 200 205 588
Sand (kg/m 3 ) 825 818 843 1045
Super plasticizer
(kg/m 3 )
Air-entraining admixture
(kg/m 3 )
5.58 5.5 5.7 5.7
- 0.6 0.6
The main parameters analyzed in the concretes made using these nominal compositions were:
� In fresh concretes: unit weight (kg/m 3 ), consistency (slump), cohesion and aspect (qualitative
assessment).
� In the hardened state: compressive strength, elastic modulus and pH, determined at different
ages (time of curing).
Results obtained in basic concretes with super plasticizer are given in table 5 and figures 1 & 2.

Properties
Table 5: Properties of basic concretes at the fresh state
70%CAC+20%S
F+10%FA
Short Plug
(aggregate from the excavation)
60%OPC+40%SF 35%OPC+35%SF
+30%FA
Unit weight (t/m 3 ) 2.23 2.23 2.25 2.27
Slump (cm) 17 12 13 15
Cohesion Good Good Good Good
Aspect Good Good Good Good
Comp. Strength (MPa)
40
35
30
25
20
Basic concretes
15
70CAC/20SF/10FA sp
10
60OPC/40SF sp
5
0
35OPC/35SF/30FA sp
60OPC/40SF lp
0 20 40 60 80 100
Curing time (days)
283
Long Plug (conventionalaggregates)
60%OPC+40%SF
Figure 1: Evolution of compressive strength over curing time (sp: short plug; lp: long plug)
pH
12,2
12
11,8
11,6
11,4
11,2
11
10,8
10,6
Basic concretes
70CAC/20SF/10FA sp
60OPC/40SF sp
35OPC/35SF/30FA sp
60OPC/40SF lp
10,4
0 20 40 60 80 100
Curing time (days)
Figure 2: Evolution of pH vs. curing time. (sp: short plug; lp: long plug)
Other properties of the low-pH concrete
The hydraulic conductivity of the low-pH concrete was determined using granitic water from Äspö
in order to simulate the real conditions of an underground repository. The mean value obtained
from cored shotcrete samples was 1.03·10 -10 m/s, which is similar to that of the surrounding rock
(which is in the order of 1·10 -10 m/s) and this value does not increase with time [2]. Besides, the pH

values measured in the percolated waters are never above 9, with a mean pH value of 8.1. The hydraulic
conductivity tests also allowed the evaluation of the resistance of low-pH cementitious materials
to long-term water aggression. A complete characterization of the degradation level of lowpH
concretes due to their interaction with granitic water will be done after 2 years, but preliminary
results after 1 year, indicate that the water aggression increase the total porosity of the low-pH concrete,
but this increase mainly occurred in smaller pores. Moreover, this increase in total porosity
does not generate an increase in the hydraulic conductivity of these low-pH concretes [2].
Shrinkage testing was conducted on mortar and concrete specimens cured at different relative humidity
and compared with reference samples (without mineral additions) of equivalent consistency.
The preliminary results, after 9 moths of testing, show that the high mineral addition contents existing
in the low-pH cements do not generate a significant increase in the shrinkage of the cementitious
materials designed. Figure 3 shows, as an example, the shrinkage values measured in the lowpH
concretes (60%OPC+40%SF formulation) at different relative humidity conditions (98% and

Field testing of low-pH formulations
The shotcreting trials for checking and optimising the low-pH concrete formulations developed
were carried out in Leon (Spain). Shotcreting tests were carried out, pumping the concrete along a
pipeline with elevations, over short and long distances, and spraying both manually and with a
spraying robot, over panel and into a steel reinforced concrete tube resembling the Äspö gallery
(Figure 4).
Figure 4: Testing of shotcrete formulation for plug construction in Spain
The final shotcrete formulation for the construction of the short plug can be found in Table 6.
Table 6: Concrete formulation for short and long plugs
Component Short plug
(kg/m 3 Long plug
) (kg/m 3 )
Water 277 230
Ordinary Portland Cement: CEM I 42.5 R/SR 184 165
Silica Fume 123 110
Coarse aggregate (5-12) 616
Gravel (4-8) 590
Medium aggregate (2-5) 200
Sand (0-4) 1045
Fine aggregate (0-2) 818
Filler (limestone) 70
Super plasticizer “Sikament TN-100” 5.5 2.8
Air entrapper “Sika Aer 5” 0.6 --
Accelerant "Sigunita L-53 AF S" 18.5 16.5
Additional shotcrete trials were carried out to test the shotcrete formulation adapted to the aggregates
from Grimsel for the long plug test (Table 6). The spraying tests were carried out over a panel
resembling the long plug gallery, which measures 3.5 m in diameter, to check the self-supporting
capacity of the fresh shotcrete in such large cross section. Furthermore, a final test was performed at
the VSH Hagerbach Test Gallery in Flums Hochwiese (Switzerland), in order to test the correct behaviour
of the equipment with the same set-up to be used during the plug construction (mixing procedure,
pumping length, spraying section, etc).
Short plug test
The short plug test was designed as a parallel 1 meter long shotcrete plug (without keys in the rock)
constructed in a horizontal drift measuring 1.85 m in diameter, excavated by full face push boring
technique in the -220 m level of the Äspö URL. For the construction of the short low-pH shotcrete
285

plug the concrete was mixed manually by introducing all the dosed components directly into a
mixer truck. The spraying was done manually using a stand-alone concrete pump, over 15 m of distance
and 2 m of elevation (Figure 5).
Figure 5: Construction of short low-pH plug in Äspö URL
After the hardening period, a mechanical pressure was applied at the rear face of the plug up to take
it to failure by injecting and pressurising water into a water chamber left in the back end of the drift.
This water chamber was hydraulically sealed by means of an isolation membrane covering the rock
walls and the rear face of the plug. A number of sensors (extensometers, pressure cells and acoustic
sensors) were installed within the plug and surrounding rock for measuring different parameters related
to the behaviour of the plug during the test.
Increasing the pressure in the water chamber stepwise performed the test. The plug overcame elastic
deformations during the pressure increases, with recovery when pressure dropped. Despite a significant
water leakage that was detected at the bottom of the plug, it was possible to increase the
pressure up to 27 bar, when it was considered that the plug had “failed”, given the sudden increase
in the rate of displacement. After releasing the pressure, two more load tests were done, and the
plug was moved again when reaching a pressure of 25 bar. The plug underwent a total displacement
of 16 mm.
Long plug test
The test consisted of a 4 m long parallel low-pH shotcrete plug constructed at the back end of a 3.5
m diameter horizontal gallery, excavated in granite with a TBM in the Grimsel URL (Switzerland).
The end of the gallery was sealed with 1 m of buffer constructed with blocks of highly compacted
bentonite (Figure 6). The bentonite was provided with geotextyle mats for water injection, working
as an artificial hydration system to accelerate the saturation process. Besides, a number of sensors
were installed to follow the evolution of the test, namely total pressure cells, displacement sensors,
water content sensors and piezometers. Both the tubings from the hydration system and the cables
from the sensors were led, through a pass-through borehole excavated in the rock, to the service
area, where they were connected to the water injection system and the data acquisition and control
system respectively. According to the preliminary mechanical scoping calculations, the maximum
pressure that the plug can support is estimated in 5 MPa.
286

The one meter thick bentonite buffer was built with vertical layers of highly compacted bentonite
blocks, manufactured from compacted powder bentonite “FEBEX” type [4], resulting in a global
dry density of 1.55 g/cm 3 which yields a mean swelling pressure of 4.15 MPa, with a natural variability of ±25 %,
that is, ± 1 MPa approximately.
Figure 6: Long low-pH shotcrete plug test layout
The plug was constructed in 7 curved layers applied during four days in total with a spraying robot
and the concrete mixer and pump were installed at 80 m from the construction point.
Figure 7: Long plug construction and”as built” shape of layers
The buffer saturation phase was delayed due to a water leakage at bottom of the plug but after the
bentonite swelling it was resumed, as detailed by Bárcena et al. (2008) [5], being total pressure and
humidity rising elsewhere after five months of continuous water injection. The objective is to hydrate
the buffer as fast as possible to reach the target buffer pressure (4.5 MPa) before the maximum
planned duration of the project (December 2008).
287

2.3 Testing of low-pH shotcrete for rock support
The work was carried out in steps starting with modification of the concrete formulation for the
plug to meet the specification for rock support and using cement, super plasticizer, accelerator, etc.
from Sweden and local aggregate. The modified formulations were then tested in pilot scale followed
by field tests of the selected formulation at Äspö HRL in Sweden. Both the pilot and field
tests were successful. The formulation developed in Sweden was modified for cement, aggregates,
super plasticizer, accelerator, etc. available in Switzerland. Further pilot tests were successfully
conducted in Switzerland using this modified formulation.
Figure 8: Low-pH shotcrete for rock support: field activities in Sweden (left) and Switzerland
(right)
3. Results & conclusions
The low-pH concretes designed fulfil with the functional requirements established for construction
in underground repositories and conventional wet-stream shotcreting technique proved to be appropriate
for the construction of plugs and rock support with the selected low pH concrete, according
to the results of the different tests made. It was demonstrated that a solution for minimising the effects
of the hyper alkaline plume in the repository is now available at industrial scale
Moreover, the shrinkage evaluation made up to now shows that the high contents of mineral admixtures
do not produce a significant increase in this parameter, which is relevant for the construction
of plugs. Durability tests indicated that low-pH concretes are stable enough and corrosion tests have
showed that the use of low-pH cements with conventional reinforcements is not suitable.
In particular, the use of low-pH shotcrete for plug construction provides the following improvements
for the repository:
� Better compatibility of engineered materials and natural barriers due to an improved sealing
material: the low-pH concrete.
� Improvement of seal/plug designs because concrete plugs could be built with no reinforcement
and with no recesses excavated in the rock for competent formations (granite).
� Improvement of seal/plug construction methods and equipment because the concrete plugs
could be built using the shotcreting emplacement method, which is much faster that the cast
concrete, can be easily automated and could be almost continue due to the low heat release
of the low-pH concrete during hardening.
� Increase of the long-term safety due to a more stable multiple barrier system (natural and
engineered) thanks to the reduction of the hyper alkaline plume effect.
288

4. Acknowledgements
This project has been co-funded by the European Commission and performed as part of the sixth
Euratom Framework Programme for nuclear research and training activities (2002-2006) under contract
FI6W-CT-2004-508851.
References
[1] Bárcena, I., Fuentes-Cantillana, J.L., García-Siñeriz, J.L. (2003). Dismantling of the heater 1
at the FEBEX ‘in-situ’ test. Technical Publication 09/2003 ISSN 1134-380X. ENRESA. Madrid.
[2] García Calvo, J.L., Alonso, M.C., Fernández Luco, L., Hidalgo, A., Sánchez, M., 2008. Implications
of the use of low-pH cementitious materials in high activity radioactive waste repositories.
International Conference: Underground Disposal Unit Design & Emplacement
Processes for a Deep Geological Repository, 16-18 June, Prague.
[3] Fernández-Luco, L., Alonso, M.C., García, J.L., Hidalgo, A., 2005. Shotcrete development
for low-pH cements. R&D on low-pH cement for a geological repository. Workshop 15-16
June, Madrid.
[4] Huertas, F. et al, 2006. Full-scale Engineered Barriers Experiment- Updated Final Report
1994-2004. Technical Publication 05-0/2006 ISSN 1134-380X. ENRESA. Madrid.
[5] Bárcena, I., García-Siñeriz, J.L., 2008. Full scale demonstration of shotcrete sealing under
realistic working conditions. International Conference: Underground Disposal Unit Design &
Emplacement Processes for a Deep Geological Repository, 16-18 June, Prague.
289

290

Summary
ACTINET – A Network of Excellence for Actinide Sciences
Thomas Fanghänel 1 , Klaus Gompper 2 , Horst Geckeis 2 , Pascal Chaix 3
1 European Commission, JRC, ITU, Karlsruhe, Germany
2 Forschungszentrum Karlsruhe, INE, Karlsruhe, Germany
3 CEA Saclay, France
The Network of Excellence for Actinide Sciences ACTINET within the 6 th Framework Programme
of the EC started in March 2004 and its first phase will terminate at the end of
2008. The large actinide laboratories, universities and other research institutions in Europe
have joined this network. The ACTINET objective of stimulating European actinide research,
coordinating it, promoting integration, training young scientists, and maintaining and
enhancing European competence was pursued by using a number of instruments:
As so-called pool facilities, the large European actinide laboratories with their unique experimental
and analytical equipment were made available to scientists from Europe for joint
research projects. Establishment of a Theoretical User Lab was a promising step to make use
of the synergy between theory and experiment in various fields of actinide sciences. Joint
research projects are funded by the network. ACTINET supports the mobility in particular
of young scientists and provides grants. Seminars, workshops, and annual schools contribute
to education and training. ACTINET provides an important contribution to maintaining and
enhancing European competence in actinide sciences in the medium and long term.
ACTINET has become a living network that contributes decisively to the support, coordination,
and integration of European actinide research. Due to the key role of actinides in the
use of nuclear energy, industries and research institutions, operators of nuclear power plants
and nuclear facilities, and licensing and supervisory authorities benefit from these activities.
ACTINET needs the support of all stakeholders to further strengthen the network and to establish
it permanently as an instrument of integrating and coordinating European actinide research.
1 Introduction
Glenn Th. Seaborg postulated in 1944 that the 14 elements following actinium and having the
atomic numbers from 90 to 103 have to be considered as actinides in analogy to the lanthanides.
Similar to lanthanides (4f shell), the 5f shell of the actinides is filled with electrons. While the
chemical behaviour of the lanthanides is mainly governed by their trivalent oxidation state (with a
few well defined and well understood exceptions) the chemical behaviour in particular of the lower
actinides is much more complex with respect to their redox chemistry and other physical chemical
properties.
Actinides, in particular uranium and plutonium, but also the minor actinides neptunium, americium,
and curium play a key role in the use of nuclear energy. This applies to nuclear power plants, where
291

energy is produced by fission of uranium and plutonium as well as in the nuclear fuel cycle, where
fissile elements are separated from spent nuclear fuel for reuse. In the disposal of radioactive waste,
actinides have to be considered in long-term safety assessments due to their long half-lives and radiotoxicities.
Actinides also are in the focus of the “Partitioning and Transmutation” concept, a potential
alternative to the final disposal of long-lived radionuclides, and of the development of nuclear
fuels for advanced reactor systems.
Actinde research is highly complex and expensive. The reasons are high work safety requirements,
as actinides, due to their radiotoxicity, may only be handled in special controlled areas and there in
glove boxes or hot cells exclusively (except natural thorium and uranium). In addition, U-233, U-
235, and Pu-239, for instance, may be used as nuclear fuels and are suitable for nuclear weapons.
They are subject to rather strict licensing and control requirements and may only be handled in secured
and monitored areas. Consequently, there are only few laboratories in Europe that have the
license, the infrastructure and the recourses’ to handle actinides. These laboratories are pursuing
national research programmes and so far have been accessible for external scientists, e.g. from universities,
to a very limited extent only.
In recent years, the number of young scientists working in the field of actinide research decreased
in Europe. Without young scientists, however, Europe may suffer a loss of competence that cannot
be compensated in the short term. Without enthusiastic and ambitious young scientists, it will be
impossible to maintain competence, which does not mean to preserve the current state of science
and technology only, but to accumulate new knowledge in order to improve our understanding of
the various processes and systems. This is a prerequisite to cover the demand of research institutions,
nuclear industry and licensing authorities for excellent young scientists and engineers in the
future.
2. Bundling of European Activities
In view of the high relevance of actinides to the use of nuclear energy and the necessity of maintaining
competence, it appeared reasonable to stimulate European actinide research and to improve
cooperation. For this purpose, a project (ACTINET-5) was launched under the 5 th Framework Programme
of the European Commission with the vision to develop the frame and the terms of reference
for the implementation of a European Network of Excellence on Actinide Research within the
6 th Framework Programme. Objectives of the project included:
Definition and execution of joint transnational research projects.
Better use of the infrastructure available at the actinide laboratories.
Facilitating access of European scientists to the actinide laboratories.
Familiarizing young scientists with actinide research.
Enhancing training and education in the field of actinide research.
Specific promotion of young scientists is aimed at further developing and sustainable maintaining
European competence in the field of actinide sciences.
Within the framework of ACTINET-5, contacts between the large European actinide laboratories
were intensified and potential structures and research topics were discussed with interested parties
from universities and other research institutions. A consortium of 27 partners from 13 European
countries was founded. This consortium comprises a so-called core group that represents the large
European actinide laboratories. These are the French Commissariat à l’Energie Atomique (CEA),
292

the Institute for Nuclear Waste Disposal (INE) of the Forschungszentrum Karlsruhe, the European
Institute for Transuranium Elements (ITU) in Karlsruhe and the Belgian Studiecentrum voor
Kernenergie - Centre d’Étude de l’Énergie Nucléaire (SCK-CEN). On behalf of this consortium and
coordinated by the CEA, a project proposal for a “Network of Excellence (NoE) for Actinide Sciences
ACTINET” was submitted for a period of four years under the 6 th Framework Programme of
the EU. The proposal was approved by the Commission and funds of 6 million EUR were granted.
The project started in March 2004.
3. The NoE ACTINET
The objectives of the NoE ACTINET can be summarized as follows:
Stimulation and enhancement of actinide sciences in Europe
Significant improvement of access of European scientists to the large actinide laboratories
Better utilization of the experimental and analytical infrastructure existing in Europe for actinide
research
Increase in mobility in particular of young scientists from the institutions of the ACTINET
members
Establishment of joint R&D projects in the field of fundamental actinide sciences
Maintaining European excellence by scientific evaluation and selection of joint R&D projects
Transfer of knowledge to young scientists by intensified education and training
Maintenance and extension of European competence in the field of actinide research
Apart from the organization and management of the network, activities of the NoE ACTINET concentrate
on integration, joint research projects as well as on education, training, and transfer of
knowledge.
ACTINET is open to new partners. Several new members have joined the network during the last
four years.
3.1. Organization and Management
The organization structure of ACTINET is similar to that of other networks or integrated projects
under the 6 th Framework Programme of the EU. All members of the consortium are represented in
the governing board. Among others, it is the task of this governing board to define general principles
of the network, to approve and allocate (annually) the budget under the given financial boundary
conditions, and to decide on new members. The scientific advisory committee comprises scientists
who have been selected for their scientific reputation and represent various areas of actinide
sciences. They are supposed to accompany the network from the scientific point of view, to provide
consultancy, and to define new topics or foci, if applicable. The scientific advisory committee
members evaluate the joint projects proposed in the field of research, education, and training. The
executive committee consists of 9 members, four of which represent the core institutions. They support
the coordinator in the management of the network and prepare the decisions of the governing
board. Within the limits of the budget approved, they decide on the funds granted to positively
evaluated projects. The executive committee is supported by the management team that comprises a
number of working groups with defined tasks, such as the support of the three research areas, organization
and coordination of the actinide laboratories, or organization of education and training.
293

3.2. Integration
3.2.1. Pool Facilities
A major step towards integration is improved access of European scientists to the actinide laboratories
within the framework of joint research projects. For this purpose, a European “pool” of institutions
was and is being established under ACTINET, in which experimental and analytical infrastructure
is available for actinide research. Presently, these “pool facilities” include institutions in
France, Belgium, Switzerland, and Germany. In France, the pool facilities are the CEA at Marcoule,
Cadarache, and Saclay. In Belgium, laboratories of SCK-CEN at Mol represent pool facilities. In
Switzerland, the synchrotron radiation source Swiss Light Source (SLS) of the Paul Scherrer Institute
in Villingen is a pool facility. As a European establishment, the Institute for Transuranium
Elements (ITU) is pooling its unique nuclear research facilities. German pool facilities comprise the
Institute for Radiochemistry of the Forschungszentrum Dresden-Rossendorf with its Rossendorf
beamline (ROBL) at the European Synchrotron Radiation Facility (ESRF) and the Institut für Nukleare
Entsorgung (INE) of the Forschungszentrum Karlsruhe (FZK) with its radiochemical laboratories
and the INE Beamline for Actinide Research at the ANKA synchrotron radiation source. AC-
TINET allocates funds for providing access of external scientists to the pooled facilities.
Pool facilities enable a wide range of various experiments with actinides or actinide-containing material.
Such experiments range from the investigation of irradiated materials and selective actinide
partitioning to actinides speciation in the geo- and biosphere, to name just a few. The pool facilities
are equipped with most modern analytical devices to cover the whole spectrum of analytical problems.
The focus is not only on qualitative and quantitative analysis down to the ultra trace range,
but increasingly on the speciation of actinides and their compounds, which is indispensable for understanding
the occurring processes or materials properties. Important examples are nuclear magnetic
resonance spectroscopy as well as laser and X-ray spectroscopy using synchrotron radiation.
An overview of and detailed information on the equipment and activities of the ACTINET pool facilities
are provided on the ACTINET homepage under http://www.actinet-network.org.
As mentioned above, handling of actinides is subject to strict safety requirements and permitted in
controlled areas only. For access to and work in these areas, a personal reliability check and a medical
examination, but also training concerning special work techniques and safety regulations are
required. Due to specific plant-related requirements and national provisions, the time needed to
meet these requirements was varying considerably among the various pool facilities. Within the
framework of ACTINET, a certain level of harmonization was achieved, for instance, by mutual
recognition of training courses and examinations.
3.2.2. Theoretical User Laboratory
In recent years, theoretical chemistry increasingly gained importance for actinide science. To efficiently
use the synergy between experiment and theory, however, it turned out to be necessary to
intensify the not always simple communication between experimentalists and theoreticians. For this
purpose a concept for a platform called “Theoretical User Lab (ThUL) was developed and implemented.
Organising workshops on selected topics, training and education of young scientists but
also the procurement and joint use of special software are among the key issues of ThUL. In October
2005 a brainstorming seminar was organised with about 60 scientists both theoreticians and experimentalist
and the scope of ThUL was developed A ThUL School was established and has already
been organised twice, in May 2006 in Lille (France), and in November 2007 in Cadarache
(France). About 50 students and 18 lecturers discussed various topics e.g. theoretical and experi-
294

mental solid state physics and chemistry in the gas phase and in solution. Apart from a general introduction
to calculation methods and actinide-specific approximation techniques, experimental results
on speciation processes were discussed. The lectures on theory were combined with computer
exercises on small applications.
3.3. Joint Research Projects
3.3.1. Research Areas
The NoE ACTINET covers three main research areas in actinide sciences. The first area “Partitioning
Chemistry of Actinides and Basic Actinide Sciences” includes fundamental chemistry and physics
of actinides in the nuclear fuel cycle, but also issues like partitioning. The research area “Actinides
in the Geological Environment” concerns the behaviour of actinides in waste disposal systems
and, hence, contributes to the long-term safety assessment. The area of “Actinide Materials under
and after Irradiation” focuses on fuel matrices for advanced reactors or transmutation facilities, but
also on aspects of long-term interim storage of spent nuclear fuel. These research areas are coordinated
by working groups, with the responsible scientists coming from ITU (area 1), from FZK-INE
(area 2), and from CEA (area 3). It is the task of the groups to stimulate joint R&D projects, to support
and coordinate the development of project proposals, and to identify and integrate new issues.
3.3.2. Funding of Joint Research Projects
The most important instrument of the NoE ACTINET to integrate European actinide research is the
joint definition and execution of research projects using the pool facilities. Calls for proposals are
made by ACTINET two times per year in May and December. The proposal must be submitted by
at least two members of the consortium from different European countries. Funds are not granted
for labour costs, but for integration and mobility, i.e. to cover expenses of travelling and accommodation
to/at other research institutions. In particular, integration and mobility of young scientists is
supported. The first call was published in May 2004 (a couple of months after the implementation
of the Network), and the eighth call had its deadline in December 2007.
Proposals of joint research projects are collected by the executive committee via the coordinator of
ACTINET and handed over to the scientific advisory committee (SAC). The SAC reviews the proposals
from the scientific point of view and ranks them in three classes from scientifically excellent
to unacceptable. A project proposal may also not be accepted, because it is outside the scope and
goal of ACTINET or because it does not fulfil the integration requirements. Based on the SAC
evaluation, the executive committee decides on the acceptance and funding of the proposals.
A total of approximately 150 proposals have been received, and reviewed by the Scientific Advisory
Committee. Among these proposals, 83 research projects have been selected by the Executive
Committee, ranging from instrumentation to quantum chemistry, from solution chemistry to the
physics of irradiated actinide materials. For these projects, access is given to the requested pooled
facilities, and support is given for mobility, accommodation, sample transports. A list of approved
projects and more information on each project may be found on the internet (http://www.actinetnetwork.org/joint_projects.)
A new instrument to enhance integration and mobility that has been available since the 5th call for
proposals for ACTINET: specific funding of excellent young scientists by grants. These grants are
given within the framework of joint research projects and are subject to prior scientific evaluation.
The grants have one year duration. 13 fellowships have been granted.
295

3.4. Education and Training
Apart from the support of joint research projects, support of education and training represents a major
step towards integration. An instrument is the annual “Actinide Summer School” that is organized
alternately by CEA and ITU. While ITU covers fundamental aspects of the chemistry and
physics of actinides, CEA focuses on selected applied areas. The number of participants is in the
order of about 70 to 80 students. The “Theoretical User Lab School” mentioned above is another
instrument that also takes place annually in spring and deals with selected topics.
In analogy to joint research projects, education and training projects may be proposed. These may
be seminars or courses, etc. that will be funded by ACTINET in case of a positive scientific evaluation.
The first five calls for proposals resulted in 19 proposals and 10 of them were funded. The
contents cover a variety of topics and reflect all research areas of ACTINET. The seminars and
courses are supposed to familiarize young scientists with special aspects of actinide sciences and
modern analytical or speciation methods. Again, funding is limited to the mobility of the participants,
i.e. travelling, accommodation, and course registration expenses may be funded.
ACTINET may also financially support conferences and seminars that contribute to the integration
of European actinide research. As a rule, a positive scientific evaluation is the prerequisite.
3.5. Use and Dissemination of Knowledge
The transfer and use of knowledge are of high relevance to European integration in actinide research.
Results of joint research projects are published in about 135 papers and, thus, made available
to the scientific community. Workshops and seminars are aimed at conveying knowledge relating
to selected topics, but also at critically scrutinizing it. Course events, such as the Actinide Summer
School or the ThUL School, are aimed at training young scientists.
ACTINET representatives give presentations both oral and posters at relevant conferences. Close
contact exists with the integrated projects under the 6 th Framework Programme, which are related to
actinide sciences, such as EUROPART, EUROTRANS, NF-PRO or FUNMIG.
All relevant information on the Network of Excellence for Actinide Sciences ACTINET on the consortium
members, core institutions, pool facilities, the Theoretical User Lab, current and completed
joint research projects are available on our web site http://www.actinet-network.org. Activities in
the field of education and training are announced and course material, for example, presentations,
are provided.
4. What’s next
The first phase of ACTINET will terminate at the end of 2008. The next phase has been prepared by
the members of the network, and will be supported by the European Commission as an Integrated
Infrastructure Initiative (ACTINET-I3).
All the major features of the network will be preserved, while the new formal structure is designed
to reduce the administrative burden for both the coordinator and the partners. For example the consortium
will be restricted to the major organisations providing most of the support to the network
(pooled facilities).
296

As a new feature, a “Stakeholder Group” will be introduced with representative of the industry, the
waste management organisations and NGO’s. The ”Stakeholder Group” will provide their views on
what are the major relevant challenges, and will actively support specific activities within the network.
5. Acknowledgments
This project has been co-funded by the European Commission and performed as part of the 6 th
Euratom Framework Programme for nuclear research and training activities (2002-2006) under contract
ACTINET-6 FI6W-CT-2004-508836.
297

298

IP FUNMIG: The FP6 Far-Field Project
Gunnar Buckau 1)* , Lara Duro 2) , Bernhard Kienzler 1) and Anne Delos 2)
1 Forschungszentrum Karlsruhe, Institut für Nukleare Entsorgung (INE),
Hermann-von-Helmholtz Platz 1, 76433 Eggenstein-Leopoldshafen, Germany
Summary
2 Amphos21, Passeig de Rubi 29-31 E08197 Valldoreix - Barcelona Spain
Integrated Projects are instruments introduced within the European Commission’s 6 th Framework
Programme. Within the EURATOM program, the Integrated Project “Fundamental
processes of radionuclide migration” (FUNMIG) was launched January 2005 and ends after
four years December 2008. It has an R&D program dealing with all aspects of radionuclide
migration from a high level waste repository to the biosphere, covering basic generally applicable
processes, the different topics related specifically to the three host-rock types presently
under consideration in Europe (clay, crystalline and salt), and the application of the results to
the disposal Safety Case. In addition to the broad R&D program, all other aspects of FP projects,
and especially the different aspects of the Integrated Projects are reflected in the project
structure, program and activities.
The project is close to its end and the different project activities are under finalization. This
includes closing up the scientific-technical work program, holding the final training course
and final annual workshop, and moving towards final reporting. The systematic approach to
reporting is based on publication of a comprehensive final EUR report, a number of institutional
reports, followed by a broad set of various types of open publications. In addition, scientific
highlights are published in the form of a Special Issue in Applied Geochemistry.
Among the institutional reports, one key document is a forthcoming Nagra report where the
overall outcome of the scientific-technical achievements is documented together with assessment
and conclusions concerning the application of these achievements for the disposal Safety
Case. This latter document is expected to become a reference document for the community.
1. Introduction and overview of the project
Within the FP6 EURATOM program, the Integrated Project “Fundamental processes of radionuclide
migration” (IP FUNMIG) has the largest number of groups and European countries involved
(Fig. 1.1). With 51 Contractors and 29 Associated Groups from 18 EURATOM signatory states (+
Russia, Korea and Canada), the project responds to the challenge of European integration. All the
other elements that can be found in FP projects are also found in and dealt with in IP FUNMIG
(Fig. 1.2). The IP started January 2005, and will end after four years in December 2008.
299

Fig. 1.1: EU Member States involved in IP FUNMIG are shown in dark grey (number of Contractors + Associated
Groups in brackets). EU Member States that are not involved are shown in light grey. Switzerland
is shown in dark grey, as a EURATOM signatory state involved in the project, but not as EU Member State.
The project is centered around the R&D program on radionuclide migration processes in the farfield
of a nuclear waste repository,
(i) covering basic processes applicable to all types of host-rock types and disposal concepts,
(ii) conducting investigations specifically addressing key issues for the three host-rock types
presently under investigation in Europe (clay, crystalline and salt (with respect to salt; the
overburden)), and
(iii) from the R&D outcome, creating tools for PA and elaborating upon the application to the
disposal Safety Case.
The fundamental processes studied are those involved in the potential migration of radionuclides
from a deep geological repository for high level nuclear waste in either type of host-rocks. It focuses
on the radionuclide-host rock interactions that govern the barrier function between radioactive
waste and the biosphere.
300

Fig. 1.2: FP R&D Projects with the different elements providing value to the project.
Documentation and Dissemination of results builds on a series of measures. The basic approach to
documentation of the project outcome is publication of results in open literature. This includes, for
example project own public annual workshop proceedings, scientific journals, various national and
international workshop and conference proceedings, institutional reports and doctoral thesis. The
final reporting is done through a comprehensive EUR report, making reference to the different publications
and institutional reports. In addition, a special issue in Applied Geochemistry will be used
for documentation and dissemination of scientific highlights. The open Annual Project Workshops
are also important for dissemination because of the broad participation, including participants from
outside the project.
With respect to scientific-technical achievements, project internal reporting is only used for very
brief overview documentation. In this very brief internal documentation, reference is made to where
the outcome is published, outside the project. The reasons for this approach are that (i) the outcome
becomes available to a broader community within publication structures with long life-times and
that are easy to trace, (ii) partners avoid duplication by generation of project internal reports with a
short life-time and thus save human resources otherwise inefficiently spent, (iii) immature information
is not published as project reports with no added value except for as transient documentation
for the originator, and (iv) despite the very large project and the broad spectrum of activities, a good
overview is kept over the status of the project and the outcome.
Training and education provide for the future of the field, including maintaining the experience and
knowledge base in view of contingencies for future challenges. Training and education is implemented
through,
(i) specific training courses,
(ii) joint research activities between different organization and individuals, and
(iii) mobility measures, especially for young scientists.
Joint research activities with mobility measures for young researchers are also conducted through
NoE ACTINET. This type of training on the job by joint activities with other institutions/organizations,
including scientific and/or analytical visits proves to be very efficient.
301

Fig. 1.3: Stakeholders and interested/concerned community
The community value through stakeholder representation (Fig. 1.2) requires adequate involvement
of the different players in the concerned community (Fig. 1.3). Implementers, universities, private
companies and research organizations are all represented under the Contractors to the project (7, 24,
3 and 17, respectively). Eight organizations representing regulatory functions have joined in as Associated
Groups, allowing them to participate without compromising their integrity and without allocation
of undue resources, especially where they do not have own R&D activities and programs.
The broader scientific community is addressed via the broad set of publications, presentation of the
project, or parts of it, at different national and international conferences and workshops as well as
through invitation to participate in the open annual workshops. Another important element in interacting
with the broader scientific community is through the many scientific investigations intersecting
with the broad set of disciplines involved. The broader interested community is addressed, especially
through cooperation with COWAM and OBRA, i.e. EURATOM projects dealing specifically
with this aspect.
2. Objectives of the project
IP FUNMIG aims at:
- Providing tools for scientifically sound performance assessment for radionuclide migration
from near-field to the hydrosphere/biosphere;
- Covering the variability of different radioactive waste disposal approaches and host-rock
types under investigation in Europe;
- Ensuring optimized use of resources and communication on this issue between Member
States with large programs and high competence levels;
- Providing for knowledge transfer in order to foster a common competence level among all
European countries;
- Providing communication with national regulatory bodies responsible for the fulfillment of
compliance with safety standards;
- Ensuring applicability of results for different radioactive waste disposal options and national
needs.
For the overall project, the spatial scales reach from nanometres to kilometres and the time-scales
from laboratory to geological systems. Integration of processes and abstraction to PA are key issues.
The scientific and technological knowledge gained in the project improve the state of the art
of PA abstraction and visualization methodologies. The knowledge acquired throughout the project
302

is brought forward to the general scientific community and broader stakeholder communities by active
training, and dissemination and transfer of knowledge.
3. Project structure
The IP FUNMIG consists of several substructures (Research and Technology Development Components
(RTDC)) and one component on training and knowledge transfer:
RTDC 1: Well established processes.
RTDC 2: Less established processes (ill defined).
RTDC 3: Radionuclide migration in clay-rich host rock.
RTDC 4: Processes and transport studies relevant for crystalline rock disposal concepts.
RTDC 5: Processes and transport studies relevant for salt rock disposal concepts.
RTDC 6: Integration of processes and abstraction to performance assessment.
Component 7: Training, knowledge management and dissemination of knowledge.
The six RTDCs cover basic processes common to all the disposal concepts under consideration in
Europe, through to studies of host-rock type specific processes and application of the results to the
disposal Safety Case. RTDC 1 investigates the conceptually well established radionuclide migration
processes with the aim to fill in critical data gaps while RTDC2 is aimed at deepening the understanding
of conceptually less understood fundamental processes driving radionuclide migration in
the geosphere. The RTDCs 3 to 5 focus on investigations of specific rock types currently under discussion
in Europe for hosting High Level Nuclear Waste repositories (clay, crystalline and salt).
RTDC 6 provides a forum for documenting the general aspects of performance assessment tool development
and bringing the outcome of the other RTDCs under one umbrella.
Management and dissemination of knowledge is conducted under Component 7, including operation
of public web-space and project internal portal. Training workshops are also organized and science
shops are held for communication dissemination with a broader community.
4. Annual workshops
Annual workshops combine a number of different objectives, meetings and activities:
- Clarifying administrative issues and taking necessary decisions;
- Taking decisions related to reporting, dissemination and communication;
- Planning for the following project year;
- Holding work meetings of the individual RTDC’s;
- Providing for across RTDC scientific exchange, especially by poster sessions; and
- Holding topical session sequentially treating the project key scientific-technical topics.
Amongst the different meetings held in association with the annual workshops are Executive Committee,
Governing Board and General Assembly. Clustering of these administrative meetings with
the R&D meetings and activities provides for optimum use of time and travel resources.
4.1 Topical Sessions
The aim of the Topical Sessions is to provide for in-depth treatment of selected topics and document
their status in the annual workshop proceedings. Each topical session consists of a series of
talks given by experts in the field. For the different processes treated in FUNMIG, in view of the
overall objectives of the project, the key questions were as follows:
303

- Present state of scientific level for the given process.
- Present state of technical implementation of this scientific knowledge into PA.
- If the level of technical implementation in PA is lower that the level of scientific knowledge,
why is this the case (for example, tools not yet available, scientific knowledge not yet sufficiently
well rationalized or further development of technical implementation of that process
is irrelevant)?
- How can FUNMIG contribute to improvement in PA with respect to this process, if desired?
1 st Topical Session
The selected topics were:
- Scientific knowledge base of processes/topics and its implementation in PA from the point
of view of Waste Management Organizations, and
- Diffusion/retention in compacted clayey materials (Radionuclide migration in clay-rich host
formations).
The topics selected were at comparably advanced state of development with a sound scientific basis
in good progress. It therefore served as a good example for discussion of treatment in PA and the
expected contribution from scientific progress also in other areas.
2 nd Topical Session
Topics around disposal in crystalline rock were treated (RTDC4):
- Role of biogeochemical processes on radionuclide migration.
- Characterization of geochemical conditions in crystalline rock/ Process identification and
verification by real system analysis.
- Fluid flow system characterization in crystalline rock (Effects of the heterogeneity and upscaling)
Presentations within these topics covered a wide variety of different issues:
i Role of biogeochemical processes on radionuclide migration,
ii Effects of micro-organisms upon radionuclide migration,
iii Interactions of microbes with actinides,
iv Characterization of geochemical conditions in crystalline rock/ process identification and
verification by real system analysis,
v Results from the groundwater hydro-geochemical investigation program in Sweden,
vi Geochemical fluxes in the geosphere: quantitative understanding by identification and verification
of processes,
vii Fluid flow system characterization in crystalline rock (Effects of the heterogeneity and upscaling),
viii Effect of heterogeneities in transport and upscaling, and
ix Diffusion process within crystalline matrices: from petrography to solute residence time distribution.
3 rd Topical Session
The third topical session covered the “Influence of organics on radionuclide migration processes”.
It was co-organized by the RTDC 1 and 2 leaders. The following topics were presented:
- General overview on FUNMIG relevant organics with respect to mobile – immobile, natural
– waste derived organics, common FEPs concerning organics, and on starting point, investigations
and achievements on organics within FUNMIG.
304

- Radionuclide complexation and kinetics.
- New advanced analytical methods: Expectations and achievements.
- Organics and radionuclide redox states, including if the humics determine or buffer the system
redox state.
- Organics on mineral surfaces
- Natural organics: Application to PA and Safety Case.
4 th Topical Session
The forthcoming fourths and final Topical Session will deal with the application of knowledge to
PA and the Safety Case. A panel with representatives from implementers, regulators and research
organizations will contribute to formulation of the overall outcome of the project with respect to its
value for different types of end-users. The outcome of this Topical Session will be a key contributor
to the overall conclusions from the project and the final reporting.
4.2 Proceedings
The scientific-technical output of the respective project year, and the topical session of the concerned
annual workshop, is published as workshop proceedings. The 1 st AWS was held at Saclay,
France in November 2005, the 2 nd AWS in Stockholm, Sweden and the 3 rd AWS in Edinburgh, UK
in 2007. The final AWS will be held in November 2008 in Karlsruhe, Germany. The proceedings of
the 1 st and 2 nd AWS are published as CEA-R-6122 (2006) and SKB-TR-07-05 (2007) reports, respectively.
The 3 rd AWS proceedings will be published by NDA, UK during 2008. The final workshop
proceedings will be available as a FZKA report in 2009. The scientific-technical papers in the
proceedings include contributions from project external participants. These scientific-technical papers
are subject to review by external experts from USA, Canada and Germany. The proceedings
contain:
(i) Overview summaries of the scientific-technical progress within the respective RTDC’s,
(ii) Reviewed scientific-technical contributions, and
(iii) The outcome of the Topical Session held.
The proceedings thus form corner stones in ongoing documentation of the results and are key elements
in the overall project reporting. They are also important instruments for dissemination of the
project outcome to a broader interested community.
5. Application of the results to the Safety Case
The application of the results to the disposal Safety Case is evaluated and documented under the
auspice of Nagra. A working document has been established where the tasks conducted within
FUNMIG are evaluated in view of the application of scientific data and knowledge to the disposal
Safety Case. The basis for the assessment is relevant FEP’s for the different host-rock types and
how the scientific results feed into their safety assessment. The final document will be published as
a Nagra report and form a very important element in the final reporting. With respect to the potential
migration of radionuclides from a repository, it is expected that this report will be a reference
document for the regard of scientific knowledge in the Safety Case.
305

6. Training courses
Key events of the training activities are three training courses. The first training course (“Fundamentals
of Radionuclide Migration”) was held in Barcelona, November 2005. The second training
course (“The use of scientific results in site characterization”) was held also in Barcelona, November
2006. The 3 rd FUNMIG Training Course covers “The transfer of scientific results to Performance
Assessment“. It will be held at Barcelona, 6-7 October 2008. The main objective of this course
is to provide contextual training on the transfer of scientific results into Performance Assessment. It
addresses young professionals as well as scientists entering the research areas of FUNMIG as well
as implementers, regulators, scientists and students from outside the project (for registration, see
www.funmig.com). In addition, a half-day seminar entitled “The Use of Scientific Data in PA” was
held the day before to the 2 nd Annual Workshop.
7. Final reporting system
The final reporting of the scientific-technical results is based on a combination of (Fig. 7.1):
i. institutional reports
ii. A comprehensive EUR report, and
iii. Applied Geochemistry special issue.
The comprehensive EUR report gives overviews of the scientific-technical achievements and application
to the Safety Case. For details, it relies on relevant institutional external citable reports attached
on a CD as well as a broad set of publications. The institutional reports carrying the main
information from the project are the annual workshop proceedings, an ENRESA report on boundary
conditions and a NAGRA report on the application to the Safety Case (cf. above). Scientific highlights
are also published in an Applied Geochemistry special issue. This special issue is scheduled
for publication around mid-2010.
1 st AWS Proceedings
(CEA Report)
2
SKB
nd AWS Proceedings
(SKB Report)
3 rd AWS Proceedings
(NDA Report)
4 th AWS Proceedings
(FZKA Report)
Final S+T Report (Comprehensive EUR)
CD
Various documents that
are referred to in the
Final S+T Report
306
Special Issue
Applied Geochemistry
Boundary Conditions
(ENRESA Report)
Application to Safety Case
(NAGRA Report)
Fig. 7.7: Final reporting system for scientific-technical reporting

8. Further information and key events
Final FUNMIG Workshop is hosted by Forschungszentrum Karlsruhe. It will take place at
Karlsruhe, Germany, November 24 – 27, 2008 (see homepage).
Training Course: “The transfer of scientific results to Performance Assessment“, Barcelona, 6-7
October 2008.
Information about FUNMIG and about these events (registration…) can be found in the internet:
www.funmig.com.
307

308

Radionuclide migration in clay-rich host formations: Process understanding,
integration and up-scaling for safety case use
Scott Altmann 1 , Christophe Tournassat 2 , Florence Goutelard 3 , Jean-Claude Parneix 4 ,
Thomas Gimmi 5 , Norbert Maes 6 , Pascal Reiller 7
Summary
1 ANDRA, Châtenay-Malabry, France
2 BRGM, Orleans-La Source, France
3,7 CEA, Saclay, France
4 ERM, Poitiers, France
5 PSI, Villigen, Switzerland
6 SCK•CEN, Mol, Belgium
The FUNMIG RTDC3 work programme studied the major phenomena likely to govern radionuclide
(RN) migration in clayrock geological barrier systems. Research carried out on anionic
and cationic RN diffusion and retention in compact clayrock over a wide range of spacetime
scales resulted in significant improvements in understanding of the distribution of total
RN mass between dissolved and sorbed species, consolidated the conceptual models describing
diffusion-driven transport of anionic RN in clayrocks and produced credible strategies / methods
for carrying out the up-scaling needed to obtain representative parameter values usable for
performance assessment simulations of a clayrock geological barrier system, in particular taking
into account the effects of spatial heterogeneity of rock physical-chemical properties. Results
show that continued research is needed regarding phenomena affecting migration of
highly-sorbing RN.
1 Introduction
Deep underground disposal in low permeability ‘clayrock’ formations has been put forward by Belgium,
Switzerland and France as the most appropriate solution for managing the high and intermediate
level, long half life radioactive wastes generated by their respective nuclear energy programs.
These concepts rely on various favourable characteristics of the host rock formation to insure that
release of radio nuclides (RN) to the biosphere will always remain at levels well below those capable
of potentially affecting human health. Stakeholder confidence in these projects is based in large
part on the capacity of the corresponding waste management organizations (WMO: i.e.
ONDRAF/NIRAS (BE), Nagra (CH), Andra (FR)) to demonstrate that the models used to predict
radionuclide migration through the respective clayrock formations (Boom clay, Opalinus clay,
Callovo-Oxfordien (COx)) are based on a scientifically sound understanding of all contributing
phenomena. WMO publish ‘Safety Case’ (SC) reports (1, 2, 3) describing, among many other aspects,
the state of knowledge regarding RN migration phenomena in the geological barrier (GB, i.e.
the host formation) and the results of Performance Assessment (PA) simulations of RN migration
towards the biosphere. While each of these SC has its own specificities, they reveal many common
points concerning the state of knowledge, and remaining questions, regarding understanding and
modelling RN migration in the respective GBS. RTDC3 was conceived based on the state of
knowledge in 2004 regarding RN migration in clayrocks and analogous compacted clay mineral
309

materials, the essential aspects of which are presented in the three SC. This state-of-knowledge led
to formulation of several ‘key questions’:
Do we have a sound theoretical basis for describing RN speciation in the porosity of highlycompacted
clay materials and clayrocks, in particular the distribution of total RN mass
between dissolved and sorbed species?
Do we have a coherent conceptual model describing diffusion-driven transport of anionic
and cationic RN in clayrocks?
Do we have credible strategies / methods for carrying out the up-scaling needed to obtain
representative parameter values usable for performance assessment simulations of a
clayrock geological barrier system, in particular taking into account the effects of spatial
heterogeneity of rock physical-chemical properties.
The goal of RTDC3 was to improve our capacity to provide positive answers to these questions for
use in upcoming clayrock safety cases.
At the most simplistic level, most of the research carried out in RTDC3 can be structured around
conceptual models based (i) on Fick’s first and second laws for diffusive driven transport, adapted
to take into account the effect of reversible sorption of RN on clayrock surfaces, and (ii) consideration
of whether or not the same set of Fick’s law parameter values can be used to represent RN migration
at all space-time scales considered in clayrock safety cases (< 10 -3 m to > 10 2 m).
Fick’s first law (for steady state RN flux): with where J:
flux (mol·s -1 ·m -2 ); De, D0, Dp: respectively ‘effective’, ‘free solution’, ‘pore’ diffusion
coefficients
(m 2 ·s -1 ); C: concentration (mol·m -3 ); x: distance (m); /
310
2 : term representing the effects of
pore space geometry on RN diffusion (dimensionless); : porosity accessible for RN
diffusion (dimensionless).
Fick’s second law (time dependence of RN mass transfer): with
where Da: ‘apparent’ diffusion coefficient (m 2 ·s -1 ); : rock density (kg·m -3 )
and Kd: coefficient representing the partitioning of total RN mass present at position x
between mobile dissolved species and immobilized sorbed species (m 3 ·kg).
Most of the work carried out in RTDC3 was focused on improving conceptual models for diffusion
driven transport for two classes of RN of key importance for SC:
Non or very weakly sorbing RN, principally those which are anions ( 36 Cl - , 129 I - …). Here the
objective was to improve understanding of terms in Fick’s 1 st law (porosity organization,
anion exclusion and mobility in clay domains, effects of mineral composition on
porosity…). These aspects are discussed below in §2 and §3;
Moderately and highly sorbing RN, principally in cationic form ( 135 Cs + , actinides, analogue
elements,…). Here the focus was on the partitioning term in Fick’s 2 nd law, but interesting
information concerning mass transport was also obtained. Sections §2 and §4 describe these
aspects.
RTDC3 involved the collaborative and complementary efforts of research teams from 24 different
organizations (ANDRA(FR), ARMINES(FR), BRGM(FR), CEA(FR), CIEMAT(ES), ERM(FR),
FZK-INE(DE), GRS(DE), II-CRC(HU), NAGRA(CH), ONDRAF/NIRAS(BE), PSI(CH),
SCK•CEN(BE), UDC(ES), UNIV-BERNE(CH), UJF(FR), LPEC(FR), LMM(FR), AIED(FR)) in-

cluding research institutes, laboratories, SME, national radwaste management agencies and four
Associated Groups (Hydr’asa(FR), La Trobe University(AUS), CEREGE(FR), UnivAvignon(FR)).
2 Characterizing and understanding clayrock properties influencing RN migration
Clayrock composition and structure largely govern the migration characteristics for any given RN
species. This rather blunt affirmation is in fact the working hypothesis for a significant part of the
research carried out in RTDC3, which is why it merits explanation, in particular in relation to the
two situations studied in detail in RTDC3: diffusion-driven transport of anionic RN species and retardation-by-sorption
of cationic RN species. First, what do we mean by composition and structure?
Composition includes both the mineralogical (and organic) phases making up the rock solid matrix
and, most importantly, the speciation of the pore solution and contacting mineral surfaces. Pore solution
and mineral surface speciation will largely determine RN dissolved speciation (e.g. predominance
of anionic or cationic forms), solid-solution partitioning of RN mass (Kd) and the intensity of
electrostatic field effects on the solution volume accessible to anionic RN. As for structure, it refers
to the organization, geometry and dimensions of the connected porosity of a given clayrock, and its
relation to contacting minerals. Porosity structure, taken together with the electrostatic field effects
on anions mentioned above, will determine the accessible porosity and diffusion path ‘tortuosity’
for cationic and anionic RN species. In addition, since current knowledge indicates that the permanently
negatively charged swelling clay minerals present in clayrocks play a key role in determining
RN migration behaviour, the structure and composition of the porosity associated with the clay
mineral fraction is expected to be of prime importance. In addition, clayrock formation databases
(1, 2, 3) generally show that values measured for a given composition or structure parameter vary
for rock samples taken from different positions within the formation, i.e. the formation is not an
homogeneous entity as regards parameters which might affect RN migration.
Research carried out in RTDC3 focused on enhancing understanding of clayrock structure and composition
at the various scales and spatial resolutions which will be needed for interpreting, integrating
and up-scaling the results of studies on RN diffusion and sorption described in subsequent sections,
i.e.
at the formation, i.e. Geological Barrier System (GBS), scale (~10 2 m) with a resolution of
~10 -1 m,
at the ‘macroscopic’ scale (10 -1 to 10 -2 m), typical of that associated with lab and in-situ
determinations of RN Kd (or R) and De parameters, with resolutions ranging down to ~ 10 -5
m,
at the ‘mesoscopic’ scale (~ 10 -3 to 10 -4 m), characteristic of that of diffusion profiles for
high Kd RN (cf. §4.2), with resolutions down to 10 -6 m.
All scales were studied on the COx formation in order to provide a common ‘safety case’ context
for inter-relating and up-scaling study results. Clayrock samples from all four clayrock formations
(Opalinus clay, Boom clay, Callovo-Oxfordian, Boda claystone (HU)) were also characterized in
terms of mineralogy, structure, porewater composition and water states at the macroscopic scale in
order to identify key common characteristics and essential differences likely to impact RN migration
(CIEMAT, ERM).
311

2.1 Geological formation scale (10 2 – 10 3 m) results
Clayrock safety cases generally present a detailed geological model of the host formation emphasizing
its stratigraphic organization and corresponding (vertical) variability in mineralogy. On the
other hand, the PA calculations carried out in these same safety cases generally assume that the entire
formation has uniform characteristics as regards RN migration, i.e. single values for De, Kd, etc.
selected based on the results of measurements on many rock samples taken throughout the formation.
While this process is robust from a PA standpoint, as demonstrated by an analysis carried out
by SCK•CEN (cf. §3.4), safety case confidence could be enhanced if a method (tool) existed for
evaluating the effect of formation-scale geological variability on the GBS representation in Performance
Assessment. The first step in this process was achieved by research carried out by Andra
which used signal profiles obtained by high resolution electrical logging of three boreholes traversing
the COx to generate profiles of rock carbonate mineral concentrations with sub-cm scale resolution.
Geostatistical methods were then used to divide the formation profile into three mineralogical
classes based on carbonate content (which is inversely correlated with clay mineral content). This
information was subsequently used to define the meshing of a RN transport model, each class being
assigned different values for RN migration parameters (De, Kd) based on the results of measurements
on representative samples of differing carbonate content (cf. §3.4 and §4.2 for the rest of the
story). Another RTDC3 action along this same line, i.e. being able to link information concerning
the spatial variability of rock characteristics at the formation scale with possible effects on RN migration,
was development of a database for the COx formation containing all data regarding parameters
likely to influence directly or indirectly RN migration (BRGM).
2.2 Macroscopic scale (mm-dm) results
The vast majority of the RN migration-related data (rock composition, structure, Kd, De, etc.) presented
in safety cases are based on measurements made on cm-dm scale rock volumes. There are
excellent reasons for this, among which are practical upper limitations on the sample dimensions
which can be accommodated in the space-time framework laboratory experiments and lower limits
imposed by the need to make measurements on volumes of rock of sufficient size to guarantee that
the values measured are sufficiently representative of ‘real rock’ complexity to be credible for
safety case use. Several actions in RTDC3 were devoted to enhancing understanding of clayrock
composition and structure at this important scale:
A detailed analysis (CIEMAT) of porewater composition and water states in the above
mentioned four different clayrocks led to development of a common conceptual model for
the distribution and composition of the different types of water (external and internal water)
present in highly compacted clayrocks, main inputs needed for constructing models for
water-rock interaction, RN speciation and solute transport. The resulting comparative
database clearly illustrates the main commonalities and differences of the four rock types.
Investigation of processes controlling the redox state of COx clayrock pore waters (BRGM,
LPEC, La Trobe University) show that the upper limit of dissolved Fe(II) in pore water must
be less than 1/100 of the Ca concentration and that clay-associated Fe(II) is a highly
reactive, redox determining component.
Macroscopic-scale clayrock volumes can exhibit significant internal variability in terms of
structure, mineral composition and porosity, all of which are capable of affecting RN
migration. Samples from the four clayrock formations were characterized (ERM, Hydr’asa,
CEA) using a wide range of methods in order to visualize and map (2D, 3D) the spatial
distribution of porosity, mineralogy and structural discontinuities (pyrite inclusions…) with
resolutions reaching down to the m scale. The results of the most complete
characterization, that carried out on a single dm-sized COx clayrock sample taken from a
312

diffusion experiment carried out in the Bure URL, were used along with results of studies at
the formation scale, mesoscopic and microscopic scales, as the basis for the conceptual
model used to integrate and up-scale many of the research results of RTDC3.
2.3 Mesoscopic scale (< mm) results
An important objective of RTDC3 research was to improve understanding of how clayrock composition
and structure influence RN diffusion (mainly for anions) and retention by sorption (mainly
for cations). If we set aside the possible effects of discontinuities (pyrite inclusions, fissures…) on
RN transport measurements carried out at the macroscopic scale, and if we assume the clay mineral
fraction and its associated porosity to be of prime importance, it seems reasonable to expect that a
detailed understanding of the latter’s organization and connectivity could help in reaching this objective.
A major effort was consecrated by Hydr’asa, ERM, Andra and CEA on developing and applying
methods for quantifying and analyzing the form and organization of clay, quartz, carbonate
and other mineral grains in sub-mm volumes of clayrock with sub m resolution.
Two main results were achieved:
The results of measurements and statistical analysis of the form factors (length to width
ratio) and orientations relative to the sedimentation plane of non porous quartz and
carbonate mineral grains which show that both grain populations have elongated form
factors, are preferentially oriented parallel to the sedimentation plane and that adjacent
grains are always separated by the clay matrix. Taken together, these results show that the
COx clayrock exhibits two domains of mineral particle organization: 1) the spatial
arrangement of clay particles, at the m scale, inside the clay matrix and 2) the spatial
organization of the contiguous clay matrix, at the

porosity outside the clay interlayer volume, and it is expected that anion repulsion will also affect
the amount and ‘geometry’ of this ‘external’ porosity accessible for anion diffusion. Cations and
neutral species (HTO), on the other hand, are able to access and diffuse in, all of the pore volume.
RTDC3 consecrated a significant effort toward increasing understanding and modelling equilibrium
mass distribution (accessible porosities) and mobility (diffusion) of anions, HTO (and cations) in
clay mineral domains at the microscopic-scale (~ m), and testing the models against experimental
data measured on a compacted pure clay mineral (montmorillonite) synthesized and characterized
specifically for project needs (LMPC).
From a theoretical and modelling perspective, the challenge is to describe and model, in a scientifically
rigorous fashion, diffusion of anions, cations and HTO molecules in compacted clay materials
as a function of material density (which affects the pore size distribution) and solution composition
(cation charge, ionic strength…). The main outcomes and advances in understanding along this line
are summarized below:
Results of molecular dynamics simulations of a montmorillonite in contact with a NaCl
solution (BRGM, AIED) were used to estimate reasonable bounds for the anion exclusion
volume, distance of water structuring and cation partitioning between the diffuse layer and
the sorbed plane in the system. The H2O data were consistent with 2 H-NMR measurements
(LAIEM) on the synthetic clay mineral. The calculated ion distributions were then used to
‘calibrate’ the electrical double layer parameters of a surface complexation model contained
in a code capable of coupling geochemical speciation and diffusion (PHREEQC2 v2.14).
This code was then used to calculate diffusion of HTO anions and cations through
compacted montmorillonite under different conditions (density, solution composition).
Comparison of model results with existing data sets show that it is able to simulate the
principal observed characteristics. The results of comparison with the experimental data
obtained in FUNMIG (see below) are not yet available.
Two new theoretical (and corresponding numerical) models of anion, cation and HTO
diffusion in compacted clay domains were developed respectively by Armines and CEA. The
distinctive feature of the Armine model is that it proposes that the hydration water associated
with cations present in clay interlayers be treated as part of the solid phase, i.e. not as a
constitutive part of overall porosity. This paradigm change allows a single value to be used
for the porosity accessible for anion, cation and HTO diffusion, with only the latter two
molecules being able to exchange mass with the pools of cation and hydration water present
in the interlayer (solid). This model, which has many similarities with that developed by
BRGM, is able to satisfactorily model data sets of anion, cation and HTO diffusion over a
wide range of clay densities.
The model developed by CEA takes a completely different approach, representing the compacted
clay in terms of an ordered arrangement of charged, non porous, several nanometrethick
rectangular entities, immersed in a continuous dielectric medium (the pore solution).
The rectangles represent the external surfaces (basal and edge), i.e. the interlayer volume is
not considered, of real clay particles. Changes in pore size distribution as a function of density
is taken into consideration by changing the particle population spacing. It is worth noting
that this is the only model approach, at this scale, which yields different values for De
perpendicular or parallel to particle orientation, i.e. this model allows introduction of diffusion
anisotropy in the clay domains. Model simulation results are generally coherent with
experimental observations of diffusion in compacted clays; e.g. De and accessible porosity
values for anions decrease with ionic strength, De for the alkaline elements increase from Na
to Cs.
314

Experimental data sets, for comparison with the blind predictions made using the theoretical
models described above, are being generated by carrying out diffusion experiments with an
anion ( 36 Cl), HTO and mono and divalent cations ( 22 Na, 45 Ca) on compacted synthetic
montmorillonite samples, as a function of ionic strength (CEA). While these measurements
are not completely finished at this time, initial results show for example that, as expected, (i)
anion exclusion increases with decreasing ionic strength and there is no impact of ionic
strength on HTO diffusion and (ii) 45 Ca diffusion is enhanced by and strongly depends on
ionic strength (due to reduced Ca 2+ sorption due to competition of Mg 2+ for ion exchange
sites).
3.2 Diffusion in mesoscopic scale (~mm) clayrock volumes
Considerable effort was invested in improving understanding of how clayrock mineral-porosity organization
can affect diffusion of mobile (non sorbing) RN since this seems to be a highly promising
approach for establishing links between diffusion properties and observed variations in rock
mineralogy. The working hypothesis, based on the observations presented in §2.3, was that the spatial
organization of the contiguous clay matrix porosity could affect (i) the value of the apparent diffusion
coefficient (Da) for non sorbing tracers by modifying diffusion path tortuosity and (ii) the
anisotropy of Da values measured in directions perpendicular or parallel to the sedimentation
planes. The study, carried out by Hydr’asa/Andra, was based on simulations of HTO diffusion in
numerical models of the 2D and 3D mineral-porosity distributions quantified in §2.3. These simulations
were carried out using the Time Domain Diffusion (TDD) method which simulates diffusion
by tracking the ‘random walk’ of anion particles in the 2D or 3D pixel grids based on the digitized
images of mineral grain organization. Each grid pixel is characterized by its porosity (a constant
value for all clay pixels, null for all others) and an isotropic Da value (for the clay pixels). The effects
of grain organization were quantified by performing simulations of diffusion, in rock volumes
having different compositions, in directions parallel and perpendicular to the sedimentation surface
plane. The results show (i) that Da perpendicular to the sedimentation surface decreases with increasing
fraction of non porous minerals and (ii) that the elongated shape of carbonate and quartz
grains and their orientation relative to the sedimentation surface introduce geometrical anisotropy in
the organization of the connected porosity at the mesoscopic scale, which in turn induces diffusion
anisotropy at a larger scale. The anisotropy of diffusion, which is observed experimentally, probably
has two components: inside and outside the clay matrix. The global diffusion coefficient is related
to the clay matrix diffusion coefficient by a geometric factor Gm which is specific to the clay
matrix geometry (this work). The clay matrix diffusion coefficient is itself related to free diffusion
of solute and a geometric factor Gcp related to clay particles (as described by the models presented
in §3.1).
3.3 Diffusion at the macroscopic scale (cm-dm)
Several premises are behind diffusion measurements made at the ~cm scale, among the most important
being (i) that they are made on samples representative of the ‘average’ properties (mineralogical
composition, porosity characteristics, etc.) of the rock unit from which they were taken and (ii)
that the measured parameter values (De, Da(perpendicular to bedding); De, Da(parallel), accessible
porosity, mineralogy, etc.) integrate, in a representative fashion, the effects on these parameters of
local variations in rock properties at smaller scales. One of the working hypotheses guiding the
RTDC3 experimental program at this scale was that the major characteristics of anion diffusion
should be coherent with, and explainable by, phenomena which were studied and modelled at the
smaller, mesoscopic scale (§3.2), in particular the role of non porous mineral grain organization in
determining diffusion anisotropy, i.e. Da(bedding parallel) > Da(bedding perpendicular), and the
315

eduction in De with increasing proportion of non porous minerals. RTDC3 efforts were therefore
focused (i) on improving the capacity to measure, model and quantify the effects of rock heterogeneity
and bedding on diffusion at the cm scale and (ii) on evaluating the effect of rock mineralogy
on diffusion, in part by comparing diffusion in samples from different clayrock formations.
Detailed analyses of HTO diffusion in dm-scale volumes of Opalinus and Callovo-Oxfordian clayrock
were carried out by CIEMAT using a novel technique consisting of placing a solid source of
radioactive anionic and HTO tracers at the centre of a pluri-dm sized clayrock cylinder. 3D tracer
distribution maps were obtained at the end of the experiment by coring. Numerical modelling by
UDC using a code capable of considering bedding plane relative anisotropy was used (i) to carry
out sensitivity analyses to identify relevant diffusion and retention parameters and (ii) to determine
best estimates for parameter values by solving the inverse problem. Results for the Callovo-
Oxfordian give lowest error values of 4·10 -11 m 2 ·s -1 and 2.23 10 -11 m 2 ·s -1 respectively for
De(bedding parallel) and De(bedding perpendicular), which leads to a diffusion anisotropy ratio of
1.8. This value is of the same order, but greater than, that calculated by the TDD method (§3.2)
which could be due to the fact that the TDD model does not consider possible anisotropy within the
clay domains, i.e. at the scale of the models presented in §3.1. Results for HTO diffusion in the
Opalinus clay give much higher anisotropy ratios, of the order of 10, which might indicate a higher
degree of preferential orientation of clay particles than for the Callovo-Oxfordian.
Effective diffusion coefficients ‘bedding perpendicular’ were measured (CEA) for Cl - and HTO on a
set of Callovo-Oxfordian rock samples having carbonate contents covering the entire observed
range in the formation. Rocks having extreme, and relatively rare, high carbonate contents were of
particular interest since they will necessarily have very low fractions of the clay minerals governing
RN diffusion (and sorption, cf. §4.2). The results for De(Cl) show a ‘threshold’ effect, with De(Cl)
remaining in the normal range of values for the formation (5*10 -12 m 2 .s -1 ) for carbonate fractions
below ~35%, then falling off progressively to roughly 40% of this value as carbonate increases to
70%. This tendency is similar to that predicted by TDD modelling of the effect of increasing the
fraction of non porous grains at the mesoscopic scale (§3.2).
One RTDC3 goal was to develop a comparative database of diffusion properties of different clayrock
formations. The characteristics of 99 TcO4 - , HTO and H 14 CO3 - in clayrock samples originating
from two depths in the Boda claystone formation were determined by II-HAS. While the two depths
show significant differences in mineralogy (e.g. absence or presence of analcime), the De values
measured for the 99 TcO4 - anion and HTO were generally coherent with results on other clayrocks,
i.e. De(TcO4): 4.2·10 -12 m 2 ·s -1 < De(HTO): 1.4·10 -11 m 2 ·s -1 .
3.4 Diffusion at geological and Safety Case (SC) time-space scales
Performance assessment calculations in existing SC (cf. 1,2,3) show that non sorbing RN diffuse
across the entire GBS thickness (roughly 50 meters thick) during typical PA timeframes (10 6 years).
In all of these SC, single values for diffusion-determining parameters (De, Da, ) were used to represent
RN diffusion throughout the entire GBS volume. The parameter values chosen for both base
case (most probable) and sensitivity (pessimistic) calculations are demonstrably valid and robust,
being based on statistical evaluation of measurements made on a representative population of cmdm
scale samples. One of the objectives of RTDC3 was to provide information and methods for
supporting homogeneous representations of GBS diffusion properties. Three complementary approaches
were taken along this line:
A theoretical and statistical analysis of the effects of scaling on RN transport (SCK•CEN),
316

An up-scaling methodology linking (i) diffusion parameter value variation as a function of
rock mineralogy measured at the cm scale and (ii) rock mineralogy measured at the GBS
scale, followed by comparative diffusion modelling with a homogeneous model (Andra,
CEA),
Natural tracer based studies (UniBerne, GRS).
A comprehensive evaluation of the potential effects of observed (or induced) spatial variability in
the Boom clay formation properties on RN transport was carried out by SCK•CEN. The study
shows that, from a theoretical standpoint, microscopic flow and transport processes can be upscaled
to the scale of the formation ‘layers’ so long as parameter values are associated with rock
volumes equal to or exceeding that of a representative volume element (RVE) for the Boom clay
(pluri-mm to cm). Consideration of two other types of information, statistical analysis of parameter
values measured on cm scale samples from throughout the formation and in situ measurements integrating
large rock volumes (hydraulic tests, diffusion experiments), allows a strong case to be
made for using parameter values measured on small samples as a basis for determining a representative
value and associated uncertainty applicable to the entire geological formation at the repository
site.
A similar conclusion was reached regarding anion diffusion through the Callovo-Oxfordian formation
using a method developed by Andra and CEA. The approach is based on a logical extension of
the working hypothesis which guided studies at the mesoscopic (§3.2) and macroscopic (§3.3)
scales, i.e. that non sorbing tracer diffusion should be determined largely by effects of rock mineral
composition on porosity organization. It therefore constitutes the last step in up-scaling process understanding
gained at small scales to the parameterization of model for non sorbing RN migration in
the GBS for PA purposes. It consists of three main steps:
Determination of the relationship between diffusion parameter values (De, ) for Cl - and
carbonate mineral content in Callovo-Oxfordian rock samples (cf. §3.3). This relationship
was used to define three rock classes having statistically different De values.
Signal treatment and geo-statistical methods were used (i) to obtain vertical profiles of rock
carbonate content with cm scale resolution from high resolution borehole log data and (ii)
to assign formation intervals (10 cm thick) to one of the three De(Cl) rock classes,
Modelling of diffusion, using the above as a basis for meshing and with an anion source
term located in the centre.
The calculated anion flux vs. time curve at the top of the formation was compared to that calculated
for the same system but with a single De value used for the entire formation, i.e. the configuration
used for PA calculations. The difference is insignificant.
In certain geological contexts information obtained by measuring and modelling spatial distributions
of conservative (non sorbing) natural tracers can provide a powerful argument for supporting
use of Fick’s law representations of non sorbing RN transport at the GBS scale, based on parameter
values measured at laboratory space-time scales. Such an approach requires measurement of natural
tracer concentration profiles within a geological formation (including boundary formations) followed
by model interpretation to extract plausible Fick’s law parameters 4 . UniBerne applied this
approach to studying natural tracers (Cl, Br, He and water isotopes) in the Opalinus Clay and adjacent
formations in the Mont Russelin anticline (Jura, CH). The Cl - distribution shows a regular,
4
The OECD Nuclear Energy Agency ‘CLAYTRAC Project: Natural Tracer Profiles Across Argillaceous Formations’
(2008) treats this aspect in detail.
317

well-defined profile, with the highest values being found in the centre of the Liassic clayrock anticline,
close to the contact with Opalinus Clay. The data were modelled using a 2D geometry and the
Dp(Cl) value determined experimentally on Mont Terri samples. The results show that the observed
Cl tracer distributions are consistent with diffusion as the dominating transport process, assuming
that the groundwater flow system in the overlying Dogger aquifer developed about 4 My ago,
which is coherent with the geological model.
GRS modeled tracer profiles measured at the Mont Terri URL with the added objective of testing
the benefits and disadvantages of using models of differing complexity to represent RN diffusion
driven transport at geological (and GBS) space-time scales. The results show that, while both complex
and simple models are able to represent the data satisfactorily, a higher degree in complexity
does not improve the agreement between the simulations and the experimental data. The most probable
reason for this is considered to be that the complexity of the model is too high and does not
correspond with the level of detail and the quality of the input data.
4 Understanding migration of sorbing RN in clayrocks
The principal method used for generating databases of RN sorption behaviour on clayrocks is by
‘Kd’ measurements in batch systems, i.e. using crushed clayrock material. Kd values obtained for
many RN (e.g. actinides, 139 Cs) are sufficiently high such that, when used in PA modelling, these
RN are entirely confined within the GBS over the simulation time frame. Three important questions
of interest to future clayrock SC can be raised regarding this approach:
i. Can the Kd dataset obtained for a given RN / dispersed clayrock system be interpreted in
terms of a chemically plausible thermodynamic model, i.e. mass action laws for adsorbed
RN species, surface site types and concentrations, other reactions…?
ii. Are the migration characteristics observed for a given RN in an intact clayrock coherent
with the predicted behaviour based on batch sorption behaviour, i.e. either by direct use of
Rd values or using the thermodynamic model?
iii. If not, can means be provided for selecting a Kd for PA use which compensates for the
discrepancy, ideally in terms of justifiable (and robust) adaptations of parameter values in
the thermodynamic model (i.e. mass action laws, activity correction model, extensive
parameter values)?
Research carried out in FUNMIG focused mainly on enhancing understanding regarding the first
and second questions, mostly for strongly sorbing RN or analogue elements, but also provided some
insights into the latter aspect.
4.1 Fundamentals of RN sorption reactions on clays and clayrocks
The majority of the work on this subject was carried out within the framework of RTDC1, with two
main types of approaches being used: spectroscopy-based measurements to determine RN surface
species characteristics at the molecular level and thermodynamic modelling of RN isotherm data
(batch systems). The latter is of the greatest utility for clayrock SC since it responds directly to the
second question posed above. The highlights of these actions are briefly mentioned hereafter.
Regarding surface speciation and redox reactions:
The results of a study (UJF-LGIT) of the nature (hydration/hydrolysis state, inner/outer
sphere complex) of Sm 3+ species sorbed on synthetic montmorillonite samples using variety
of analytical methods (neutron diffraction (ND), EXAFS, Quasi-elastic Neutron Scattering
(QENS)) show somewhat contradictory results. ND measurements indicate that Sm 3+ is
bound to the clay surface and is probably partially hydrolyzed while EXAFS measurements
318

indicate that Sm 3+ is present as an outer-sphere complex with nine water molecules
surrounding the cation. Comparative QENS measurements of the mobility of the hydration
water associated with Ni and Sm sorbed on fluorated hectorite show similar behaviour, with
mobility in Sm-hectorite being somewhat greater than in the Ni-hectorite.
The results of EXAFS-based studies of Y, Lu and U(VI) sorption on clay minerals
(montmorillonite, hectorite) by CEA are coherent with formation of inner-sphere complexes,
but the location of these complexes at clay layer edges could be proposed only by analogy
with other sorbate cations. Evidence of differences in U(VI) sorption mechanisms was also
observed for the two minerals. TRLIF data for Eu sorption on both clays show increasing
formation of inner sphere complexes as pH increases.
The redox reactivity under anoxic conditions of Se(IV) with Fe(II) adsorbed on synthetic,
structural Fe-free montmorillonite was studied by UJF, in collaboration with BRGM, LPEC,
LMPC/LMM. The results show slow reduction of Se and formation of a nano-particulate
Se(0) solid phase when selenite is added to a montmorillonite previously equilibrated with
Fe 2+ solution; this was not observed in Fe-free systems. These, and other, results clearly
suggest that the Se and Fe redox reactions are not directly coupled, leading to the hypothesis
that electrons produced in the absence of Se by oxidation of sorbed Fe(II) are stored, for
example by formation of surface H2 species, and are then available for the later Se(IV)
reduction.
Regarding thermodynamic modelling of sorption:
The results of batch sorption studies (CIEMAT) of Sr, Pu, selenite and europium sorption
onto Na-smectite, Na-illite and mixed systems showed for selenite that (i) sorption was
higher in smectite than in illite and, in both clays, was independent of ionic strength and
decreased with pH, (ii) linear sorption isotherms over a broad concentration range (1•10 -10
to 1•10 -4 M), and (iii) that data could be satisfactorily modelled (from pH 3 to 8) considering
the formation of surface complexes at the edge sites of the clay and using a one site, non
electrostatic model. Regarding Eu (III), results show that (i) ionic exchange is important at
pH < 4, (ii) surface complexation becomes increasingly important as pH increases to 10, and
(iii) that data could be represented satisfactorily using a model incorporating nonelectrostatic,
two site surface complexation and cation exchange.
Measurements and modelling of Ni(II), Co(II) and U(VI) sorption on Opalinus clay, and of
Co(II) on illite were carried out by PSI, with the results for U(VI) being particularly
illustrative. Here, predictive modelling of sorption was carried out using the 2 site
protolysis, non electrostatic surface complexation and cation exchange sorption model used
in previous studies to represent U(VI) sorption on purified Na-illite assuming that (i) illite is
the main sorbing phase in the Opalinus clay and (ii) only the UO2 2+ and the hydrolyzed
species sorb. In the case of the clayrock, it was found that the U(VI) sorption isotherm could
only be modelled if the neutral Ca2UO2(CO3)3(aq) complex was included in the calculations
and assumed to be non-sorbing.
Sorption data for Cs, Sr, Am and Th on dispersed Boom Clay were gathered and interpreted
(SCK•CEN) in terms of surface complexation models, and sorption experiments on
compacted samples (clay disks) were performed for Cs and Sr to check the impact of
compaction (decreased accessibility to sorption sites) on Kd, with no significant differences
being observed. Cs sorption could be modelled using the same 3 site, cation-exchange
model used by PSI for modelling sorption on Opa. SCK•CEN also studied the influence of
the natural organic matter (NOM) present in the porewater on sorption of Am(III) and
Th(IV) onto Boom Clay, including the potential role of colloids. Data on Am(III) and
319

Th(IV) sorption on montmorillonite and illite was used to calibrate a predictive sorption
model for Boom clay and Tipping's Humic ion binding model VI was used to account for
NOM interactions. The NOM model has been found to adequately simulate Eu-NOM
interactions and Eu sorption to illite in presence of NOM.
Armines adapted the retention and porosity model developed for bentonite to COx clayrock
by assuming additivity of the various mineralogical contributions, based principally on the
content of interstratified illite/smectite (I/S) minerals and the illite to smectite ratio in the I/S
minerals. The approach led to a reasonable quantitative representation of the CEC and
surface complexation site densities as a function of mineralogical composition. Sorption
isotherms calculated for Cs and Ni were in good agreement with measured Kd data.
4.2 Migration of (strongly) sorbing RN in intact clayrock
The most convincing (from a safety case stand point) approach for quantifying migration of highly
sorbing RN in clayrock is direct measurement of Da values (cf. Fick’s second law, §1) in intact
samples under conditions representative of the GBS. While this approach has been used in SC, it is
much more frequent to estimate Da using based on (i) assumed De values for cationic
RN and (ii) Kd datasets measured on crushed rock samples. The reason for this is evident from
Fick’s second law which tells us that the rate of propagation of a sorbing RN in a clayrock (i.e. the
distance travelled by the migration front away from the source in a given period of time) will be
inversely proportional to its Kd value. For example, for a RN having a Kd = 10 m 3 kg -1 (order of
magnitude for actinides) and De = 10 -11 m 2 s -1 (order of magnitude for HTO), RN mass will have
penetrated roughly 0.5 or 1.5 mm into a typical clayrock after respectively 1 or 10 years of contact.
From a practical standpoint, this means that for a one year experimental time frame, one must be
able to quantify RN mass distribution within a sub-mm thickness of rock in order to determine Da.
Despite such a challenge, RTDC3 considered that it was sufficiently important for safety case confidence
building to take it on.
The work program was guided by two main objectives: (i) determining if Kd values measured in
batch systems under varying conditions could be used to estimate RN sorption behaviour in the intact
rock and (ii) determining if De values for strongly sorbing RN could be estimated based on values
measured for non or weakly sorbing tracers. Studies were carried out on three different clayrocks
(Opalinus clay, Boom clay, Callovo-Oxfordian) using elements representing the range of
sorption behaviour for cationic RN: strongly-complexing (Am, Eu, Pu, U), moderately-sorbing (Cs
Co, Cu), weakly-sorbing (e.g. Sr, Na). The main results concern:
Development and application of new analytical methods for carrying out in-diffusion
experiments and quantifying high Kd tracer migration in intact rock;
Comparison of sorption equilibrium model parameter values determined in batch and
compact rock systems;
Comparison of De + sorption parameter value sets obtained by modelling data from indiffusion
experiments with predictions made using an assumed De value and batch sorption
parameters.
320

Analytical method development
It was necessary to develop a variety of innovative analytical methods during FUNMIG in order to
be able to quantify the migration of highly-sorbing tracers in intact clayrock samples. The ‘hiresolution
abrasive peeling’ method developed by PSI consists of diffusing tracer into a flat clayrock
surface using a specially developed cell, for a given period of time (e.g. ~170 days for Eu(III)),
and then abrasively removing roughly 10 m thick layers of rock for total tracer concentration determination.
With this method, tracer concentration profiles extending less than 200 m into the
rock could be determined with roughly 10 m resolution. Along this same line, CIEMAT demonstrated
that nuclear ion beam Rutherford Backscattering Spectroscopy (RBS) could be successfully
used to quantify Eu(III) depth profiles extending to ~1.5 m (~50 nm resolution) below a polished
clayrock surface, after different in-diffusion times. FZK-INE developed a special cell for carrying
out actinide in-diffusion experiments on clayrock samples maintained at in-situ confining pressures
and an autoradiography technique for quantifying tracer distribution relative to the input surface. A
different approach was taken by the CEA driven by the need to associate the spatial distribution of
tracer mass after an in-diffusion experiment with the corresponding rock mineral-porosity organization
for carrying out diffusion modelling by the TDD method (cf. §3.2). The developed method,
based on hi-resolution Laser Induced Breakdown Spectroscopy (LIBS), allows the simultaneous
determination of tracer and rock mineral element distribution away from the tracer input surface
with a ~3 m spatial resolution. CEA also developed column-based method for determining Da parameters
for sorbing RN diffusion into and out of ~ 2 mm thick clayrock ‘plates’, with the number
of plates and flow rate being varied depending on RN the sorption and diffusion characteristics.
Results of comparison of RN sorption equilibrium in batch and intact rock systems
The question addressed here is quite simple – for a given mass of rock equilibrated with a given activity
of a sorbing RN under otherwise identical conditions, does one measure the same total sorbed
mass of RN if the rock is present as ground particles or as a compact solid? While this question can
be answered fairly readily for weakly and moderately sorbing RN as will be shown below, it turns
out to be quite difficult in the case of strongly sorbing RN. The key objective here is to reach an
equilibrium state in systems containing compact rock samples, i.e. kinetics related to RN mass
transport (diffusion) into the sample have reached insignificant levels. When this is the case, results
can be interpreted using only a chemical equilibrium model, i.e. without diffusion. The results of
measurements carried out by PSI show that sorption equilibrium reached for Na + , Sr ++ and Cs + on
compacted Opalinus clayrock are comparable (within a factor of 2) to corresponding states measured
on crushed samples. Similar results were obtained by SCK-CEN for Sr and Cs on Boom clay.
The Rd values measured on compacted samples tend to be higher than for crushed rock which
might be due to the longer equilibration times used for the compact material studies. Measurements
of Co +2 sorption, while not completely attaining equilibrium within the nearly 700 day experimental
time frame, indicate the same result. Taken together, these results tend to support the following conclusions:
Crushed and whole rock samples have similar sorption site populations per unit mass
(sorption site types and corresponding total concentrations),
Similar mass action laws apply for sorption in intact and crushed materials,
Sorption-induced retardation of RN mass transport (i.e. very low Da value) results in very
long time frames for reaching equilibrium for highly sorbing RN in compact rock systems.
This limits the capacity to directly determine sorption equilibrium for actinides in compact
rock.
321

RN migration experiments and model interpretation
A major part of the research carried out in RTDC3 was concentrated on measuring mass transfer of
sorbing RN in compact clayrock-containing systems and interpreting the resulting datasets using
numerical models coupling the effects of diffusion and sorption equilibrium. A wide variety of experimental
techniques were used to study RN migration characteristics in rock volumes ranging
from mm (cf. techniques mentioned above) to dm (laboratory and in-situ experiment) scales. Generally
speaking, two types of information were sought (i) time-dependent evolution of RN tracer
concentration in the source term and, in certain cases, in a ‘sink’ reservoir, and (ii) tracer mass distribution
in the rock volume after a given time(s). Numerical modelling was then used to seek plausible
sets of values for diffusion (e.g. Da, De) and sorption (Kd, complexation model) parameters,
and in certain cases the degree of spatial variability within the rock, leading to the best possible representation
of the experimental dataset. The model results were then compared with those expected
based on initial hypotheses concerning RN migration (e.g. Kd(intact rock) = Kd(batch), De(cations)
= De(HTO), homogeneous porous medium), and conclusions drawn regarding the applicability or
non applicability of the reference model for representing RN migration. The following sections
briefly summarize the principal results of research carried out on each of the three clayrocks:
Opalinus clay, Callovo-Oxfordien and Boom clay.
Opalinus clay
Migration in Opalinus clayrock of a wide range of weakly to strongly sorbing cations (Na + , Sr ++ ,
Co ++ , Cs + , U(VI), Eu(III), Pu(IV)) was studied using a number of differing experimental configurations.
The results of Cs + in-diffusion and through-diffusion experiments carried out by PSI tend to
show that RN migration behaviour is generally coherent with the expected sorption model but is
significantly affected by relatively complex mass transport processes. One hypothesis is that the
porosity is made up of well-connected and ‘dead-end’ pores, which would necessitate considering
at least two Da terms in the diffusion equation, with that for the ‘dead end’ pores representing both
the kinetics of RN mass transport into this volume and RN retention on the sorption sites in contact
with this porosity. In-diffusion studies of Co(II) and Eu(III), with profiles extending to less than 0.5
mm after roughly 400 days for the later, also show evidence of mass transport complexity (dual
path). The general conclusion is that mass transport of sorbing RN in clayrocks is still not well understood,
especially given the difficulties associated with making measurements at such small spatial
scales (modifications of surface layer properties….).
Somewhat different results regarding the transferability of batch Kd to RN migration were obtained
from in-diffusion studies carried out by CIEMAT using three techniques: diffusion from a solid
source term in large-scale (dm) clayrock blocks (Sr), using a ‘filter source sandwich’ configuration
(Sr, Co, Eu, U(VI), Cs) and RBS detection (Eu). Generally speaking, the Kd values estimated by
modelling migration experiment data were found to be significantly smaller (10% or smaller for Co
and Cs) than those measured in batch experiments. Whether this difference is due to real differences
in sorption equilibrium between compact and crushed material, or includes the effects of mass
transport complexities (e.g. kinetic effects related to restricted access to some sorption sites) similar
to those observed by PSI, is not yet clear. On the other hand, the Da range obtained for Cs (3 –
12*10 -14 m 2 .s -1 ) is similar to the value obtained by PSI (~ 6*10 -14 m 2 .s -1 ) and Da values measured
for Co and Eu are intermediate between the Da values for fast and slow diffusion path values fit by
PSI. This suggests that some of the differences between Kd and observed effects on Da might be attributable
to differences in the modelling approach used for taking sorption into account.
Migration of most of these tracers is also being studied by means of the DR in-situ experiment being
carried out by Nagra/PSI at the Mont Terri URL, the first step of which was predictive modelling
(PSI, UDC, GRS) of tracer mass loss from the source solution using a variety of numerical
codes. While preliminary data on tracer loss from the injection interval tend to confirm the general
322

ehaviour expected based on measurements on rock samples, data modelling also shows that for
strongly sorbing tracers, complexities in mass transfer (borehole mixing, diffusion in filters, etc.)
strongly influence experimental data, making clear-cut estimations of values for diffusion-retention
parameters in the undisturbed clayrock difficult. Some of these ambiguities will certainly be removed,
at least for the more mobile tracers, when information on tracer distribution in the surrounding
rock becomes available (post-FUNMIG) for providing further constraints on model interpretations.
Finally, Pu(V) diffusion into Opalinus clayrock samples kept under in-situ confining pressures was
measured by FZK-INE, with results from both batch and in-diffusion experiments showing that Pu
is reduced to the Pu(IV), probably by Fe(II) contained in rock minerals (probably chlorite), and is
retained (sorption or other process?) on preferential sites.
Callovo-Oxfordian
The Callovo-Oxfordian research program is, by design, quite similar to that described for the
Opalinus clayrock since one of the objectives of RTDC3 was to generate comparable data sets for
different clayrocks in order to identify common characteristics and eventual significant differences
in RN migration properties. The principal difference between the two is in the working hypothesis,
and consequent experimental approach, taken by a consortium of French partners (CEA, ERM,
Hydr’asa, Andra) for quantifying and modelling migration of highly-sorbing RN in clayrock. The
guiding assumption here was that, given the very small spatial scales covered by RN migration during
the time frames of in-diffusion experiments owing to the preponderant effect of sorption, it
would be important to be able to quantify how tracer mass present in the diffusion profile was distributed
relative to rock mineral constituents and associated porosity. The approach taken involved
carrying out, on a single, oriented cm-scale volume of clayrock:
Characterization of the mineral-pore space organization and construction of the corresponding
2-3D numerical models, including analysis of grain organization (cf. §2.2), followed by TDD
modelling of diffusion of non sorbing tracers (cf. §3.2);
In-diffusing a highly-sorbing tracer (Eu, Cu) into the rock (gradient perpendicular to sedimentation
plane) for a given period, then sectioning the rock perpendicular to bedding;
Simultaneous mapping of tracer mass and mineral grain 2D spatial distributions, relative to the
in-diffusion surface, using LIBS and Electron Probe Micro Analyser (calculation of the average
diffusion profile), followed by construction of numerical models of the in-diffusion zone;
Inverse modelling of tracer diffusion (source term, spatial distribution) to determine Da values
for the porous mineral zones (clay matrix + disseminated pyrite), followed by comparison with
batch Kd.
The results for Cu(II) (the most complete dataset available) are rich in information concerning both
tracer migration phenomena and potential experimental artefacts which can make measurements on
such small rock volumes difficult to interpret. Regarding the former, it was found (i) that Cu does
not penetrate into, nor sorb significantly onto, the carbonate and quartz grains and associates preferentially
with the pyrite minerals dispersed in the clay matrix, (ii) Cu diffusion profiles (clay and
pyrite) developed over a distance of roughly 2 mm into the rock and (iii) that evaluation of the entire
dataset (reservoir + profile) yields values for De (2.5 * 10 -10 m 2 .s -1 ) and Rd (3500 mg/L) which
are consistent both with results showing that De for cations in the COx are generally significantly
higher than the value for HTO (~ 2.5*10 -11 m 2 s -1 ) and measurements of Kd(Cu) which give values
in the 3000 to 9000 range (note that Rd could be controlled by the redox reactivity of this tracer).
On the other hand, detailed analysis also shows that the roughly 50 m layer of rock in contact
with the source solution is significantly perturbed, both in terms of its mineral-porosity composi-
323

tion/organization and its Cu(II) retention characteristics - the Da in this zone is roughly two orders
of magnitude less than the rest of the profile. These observations have, by the way, quite a number
of similarities with those for Co(II) in-diffusion in the Opalinus clayrock.
CIEMAT carried out a program of ‘filter sandwich’ and block-scale diffusion measurements similar
to that described for the Opalinus clayrock, and obtained generally similar results, i.e. Kd extracted
from fitted Da values are significantly smaller than those observed in comparable batch experiments.
Note finally that FZK/INE carried out similar measurements of Pu(V) migration in COx samples
and obtained similar results as for Opalinus.
Kd values for Cs were determined (CEA) on the same set of ‘variable carbonate content’ COx samples
studied in §3.2. As for De(Cl), the results exhibit a ‘threshold’ effect, with Kd remaining in the
normal range of values for the formation for carbonate fractions below ~70%, then falling off drastically
to roughly 10% of this value. For completeness it can be noted that, when this data is used to
parameterize the distribution of Kd(Cs) values at the formation scale (cf. §3.4), the calculated Cs
flux vs. time curve at the top of the formation is, as expected, identical to that calculated using a
constant Kd for the entire formation.
Boom clay
SCK•CEN carried out in- and through-diffusion experiments with Cs and Sr in Boom clay with the
objective of extracting ‘diffusion-operant Kd’ values for comparison with the Kd values measured in
batch and compacted systems (cf. §4.1). The Da values obtained for both Cs and Sr, respectively
~1.4·10 -13 m²·s -1 and ~9·10 -12 m²·s -1 , match values obtained previously using a variety of experimental
techniques. However, in both cases, it was not possible to extract unambiguous, reliable values
for the ( + Kd) term. Because of this, comparison with measured Kd values was not possible. On
the other hand, it could be shown that batch Kd values tend to overestimate values for the ( + Kd)
term. A coupled sorption/transport simulation implementing the 3 site, cation-exchange model for
Cs+ sorption (cf. §4.1) was used to carry out a sensitivity analysis on the effect of increased pore
diffusion coefficients (related to "surface diffusion" effects) and decreasing available sorption sites.
Results show that good fits to the migration profile required either (i) that the total sorption site
concentration needed to be decreased to a fraction (5%) of that used to model Cs sorption data obtained
on dispersed and compact Boom Clay, or (ii) that the Dp value needed to be raised by a factor
of 16 compared to the Dp of HTO (Dp(HTO)=2.3·10 -10 m²·s -1 ). The extreme nature of both of these
results suggests, irrespective of the underlying hypothesis, that in a compacted system not all sorption
sites seem to be available, a conclusion which tends to go in the same direction as results seen
for the other two clayrocks.
5 Main RTDC3 messages for Clayrock Safety Cases
The overall results of RTDC3 can be summarized by returning to the three questions posed in the
introduction:
Do we have a sound theoretical basis for describing RN speciation in the porosity of highlycompacted
clay materials and clayrocks, in particular the distribution of total RN mass
between dissolved and sorbed species?
A fairly positive response seems justified based on two main results. The first is the observation
that similar equilibrium sorption states (Kd) are observed in dispersed and compacted
materials for moderately sorbing cations (Sr ++ , Cs + , Co ++ ) for all clayrocks. This
implies that the same sorption site populations are accessible under both conditions and
that the corresponding mass action laws are valid. It was not possible to demonstrate this
for highly sorbing RN (actinides…) because of the extremely long times needed to reach
324

equilibrium conditions (associated with other problems), but there does not seem to be any
clear reason why they should not have a similar behaviour. The work on pure clay systems
provides a sound basis for partitioning the mass of both anionic and cationic RN between
different porosity volumes (anionic exclusion, interlayer, EDL, bulk).
Do we have a coherent conceptual model describing diffusion-driven transport of anionic
and cationic RN in clayrocks?
Here the answer is clearly mixed. For anions, yes. The results of the studies carried out at
scales ranging from molecular/microscopic, to mesoscopic, to macroscopic, and geological
formation scales offer a sound scientific basis for explaining and modelling migration of
anionic RN. As for cations, the picture is not so clear, with all of the results tending to
show that coupled diffusion-sorption migration is much more complex than expected, leading
generally to greater mobility than that predicted by coupling Fick and batch Kd. Several
hypotheses have been advanced for this, perhaps the most plausible being that cationic RN
diffuse along more than one type of ‘pathway’ (or porosity) in a clayrock, each having a
corresponding Dp value and sorption site population. In this case, mass transport kinetics
could limit access to the sites in the lower Dp porosity. It should also not be forgotten that
these studies are necessarily carried out on very small rock volumes, with the accompanying
possibility that effects of mineral-porosity heterogeneity existing at this scale might
also have an influence. It is not impossible that the reduced effect of sorption retardation
observed at these mm scales becomes less important when migration over larger space (and
time) scales are considered. In any case, more research is indicated in this area.
Do we have credible strategies / methods for carrying out the up-scaling needed to obtain
representative parameter values usable for performance assessment simulations of a
clayrock geological barrier system, in particular taking into account the effects of spatial
heterogeneity of rock physical-chemical properties.
Here the answer is an unqualified yes, backed up by the multiple lines of argument and
demonstration provided by the theoretical, experimental, up-scaling and natural tracer studies
presented above.
6 Acknowledgements
This project has been co-funded by the European Commission and performed as part of the sixth
Euratom Framework Programme for nuclear research and training activities (2002-2006) under contract
FI6W-CT-2004-516514.
References
[1] ONDRAF/NIRAS (2001), Safety assessment and feasibility Interim Report 2 (Safir 2), Nirond
2001-05
[2] Nagra (2002), Project Opalinus Clay, Demonstration of disposal feasibility for spent fuel,
vitrified high-level waste and long-lived intermediate-level waste, Safety Report, Technical
report 02-05
[3] Andra (2005), Dossier 2005 Argile
325

326

Laboratory and In Situ Investigations on Radionuclide Migration in Crystalline
Host Rock: FUNMIG Project
Tiziana Missana 1 , Paloma Gomez 1 , Andrés Pérez- Estaún 2 , Horst Geckeis 3 , Javier Samper 4 , Marcus
Laaksoharju 5 , Marco Dentz 6 , Ursula Alonso 1 , Belen Buíl 1 , Marja Siitari-Kauppi 7 , Modesto Montoto
8 , Jesús Suso 9 , Gustavo Carretero 9 .
1 CIEMAT, Spain; 2 CSIC, Spain; 3 FZK-INE, Germany; 4 Universidad de la Coruña, Spain;
5 GEOPOINT, Sweden; 6 UPC, Spain; 7 Helsinki University, Finland; 8 Universidad de Oviedo, Spain;
9 AITEMIN, Spain
Summary
The overall aim of the component 4 of the FUNMIG project (RTDC4) is the investigation,
both at laboratory and in-situ scale, of specific processes influencing radionuclide migration
in crystalline rock formations. Processes considered fundamental in performance assessment
(PA) as fluid-flow system, matrix diffusion and sorption, were studied focusing on those aspects
not yet completely understood (i.e. effects of the heterogeneities, interaction mechanisms,
transport model selection and up-scaling methodologies).
The performed studies comprised basic investigations on the chemistry of the crystalline
groundwater, also accounting for the potential consequences caused by the presence of the
bentonite barrier as, for instance, the formation of bentonite colloids. A great effort was made
to elucidate the role of colloids on radionuclide transport in crystalline rocks, because of their
possible high relevance, although colloids are not directly included in PA.
Part of these studies benefit from supporting data from real sites and in-situ data from the
FEBEX gallery (NAGRA’s Grimsel Test Site, GTS, Switzerland). At the GTS, an experiment
simulating at real scale a high-level waste repository in granite was installed more than 10
years ago. This is a unique opportunity to study migration processes from the bentonite barrier
to granite in a realistic environment.
1. Introduction
Crystalline rocks present suitable properties to host a deep geological repository (DGR) of high
level radioactive waste and, therefore, radionuclide migration in these systems has been studied for
many years. In performance assessment (PA) of a deep geological repository of radioactive waste in
crystalline rocks, different processes are considered fundamental to describe radionuclide migration
(i.e. distribution of the groundwater flow, matrix diffusion and sorption). However, several aspects
dealing with those processes are not yet completely understood and most of them were mentioned
in the EU RETROCK Project report [1]. Any improvement on the knowledge of these most important
processes is helpful for the formulation of advanced models and their simplification for PA applications.
To the crystalline rock component, RTDC4, 17 organisations from 7 different countries and 5 associated
groups are participating. RTDC 4 is structured in 6 work packages corresponding to the main
investigations identified as necessary to address within the FUNMIG-IP. The WPs are the following:
WP 4.1 “Characterisation of geochemical conditions in crystalline host rocks”; WP 4.2 “Fluid
flow system characterisation”; WP 4.3 “Generation, quantification, characterisation, stability and
mobility of groundwater colloids”; WP 4.4 “RN transport studies, including the effects of inor-
327

ganic/organic colloids”; WP 4.5 “Process identification and verification by real system analysis”
and, finally, WP 4.6 “Up-scaling of processes”.
In RTDC4, a large experimental work-plan was settled to gather data under well-defined conditions
for assessing the effects of the rock heterogeneities on the distribution of the groundwater flow, matrix
diffusion and sorption. Additional aims of the studies performed were to evaluate the adequacy
of models to describe transport in this complex environment and to develop up-scaling methodologies.
Reactive transport models, at present, do not take into account the complex dynamics inherent to a
heterogeneous reactive transport system increasing the uncertainties on radionuclide migration prediction.
In RTDC 4, efforts were made to quantify effective reactions and transport in heterogeneous
media for a more realistic large scale modelling and to implement concepts and modelling
frameworks in numerical PA tools (WP 4.6) [2]. The dependence of transport parameters on the
scale was analysed and different methodologies were compared. Advantages and limitations of upscaling
methodologies in the context of underground radioactive repositories were evaluated (WP
4.6) [3].
The characterisation of water flow-paths in crystalline rocks is essential for the analysis of the main
migration and/or retention processes: novel and complementary experimental approaches were proposed
to analyse the rock matrix and fractures structure at different scales. The effects of physical
and chemical rock heterogeneity on the main parameters needed for PA calculations (i.e. distribution
coefficients, diffusion coefficients or porosity) were analysed within different work packages.
Even when the processes are well understood, for example in the case of matrix diffusion, it is
sometimes recognised the difficulty to obtain relevant experimental data as input for transport models.
The development of new experimental methodologies may overcome these difficulties and
represents additionally a scientific challenge.
At present, some processes are neglected in PA because they present high degree of uncertainty or
because of data obtained in realistic conditions are too scarce. In crystalline rocks, colloids are neglected
because of their low concentration in the far-field. However, recent studies showed that
bentonite colloids can be generated from the engineered barriers [4] and that they could be particularly
relevant for the migration of high sorbing elements as tri- or tetravalent actinides [5], thus, in
RTDC4, they were thoroughly studied. The complete description of radionuclide transport in the
presence of colloids needs to understand colloidal behaviour in crystalline groundwater (generation,
stability, colloid-RN interactions and rock-colloid interactions). All these aspects were dealt with in
different WPs of RTDC 4.
Lastly, since models developed on the basis of results obtained from laboratory tests might not be
directly applied to field conditions, field studies play an important role. Apart from providing sitespecific
data, they allow investigating the migration processes at a larger spatial scale and under
almost “real” conditions. Successful research programmes have been often based on a continuous
feedback from laboratory to in-situ data.
As an example, in the frame of WP4.5, hydrological and geochemical data from the Swedish Forsmark
and Simpevarp sites are analysed as a support of the safety case for nuclear waste disposal in
fractured crystalline rock in Sweden. In-situ studies were also carried out at the FEBEX site (WP
4.1 and 4.2). The FEBEX experiment reproduces at a real scale a high-level waste repository in
granite and it was installed more than 10 years ago. At moment, it represents the only realistic environment
where the processes affecting RN migration from the bentonite to granite can be studied.
In summary, the main aim of the work carried out within RTDC4 is to obtain (realistic) data, both
from laboratory and in-situ for the validation of existing models, and to improve the knowledge on
less known processes to facilitate their inclusion in PA models (e.g. colloid behaviour). In this paper,
a short description of the main processes studied in RTDC4, including a summary of the main
328

esults obtained so far. Most of the bibliographic references included refer to studies carried out
during the project.
2. Transport in crystalline rocks: PA relevant processes
Groundwater chemistry
RNs mainly migrate dissolved in the groundwater: the chemistry of the water (salinity, pH, Eh,
complexing agents) is one of the main parameters controlling the aqueous speciation of RNs, their
solubility and their retention in the medium. Chemistry influences several other processes, colloid
formation and stability amongst others. When studying a crystalline medium, first key points are to
understand what chemical reactions and sorption processes occur in the host rock (also in the presence
of the engineered barrier) and what their effects on radionuclide mobility are (WP 4.1). Geochemical
characterization includes the identification of water types and presence of elements affecting
the water chemistry and its evolution (water/rock interactions, water paths, residence times,
flow/mixing region, redox condition, effects of micro-organisms or colloids). Furthermore, to know
if present geochemical models are able to account for these processes and if they are adequately
calibrated in front of real systems is necessary.
Site specific studies, as those carried out in the Swedish sites Forsmark and Simpevarp (WP4.5),
represent a direct support for PA [6]. The evaluation of field geochemical data (Eh, pH, TOC, colloids
presence, Mg-Ca, etc…) is needed to verify if they met safety criteria for site selection. On the
other hand, these studies are fundamental to understand phenomena that cannot be obtained from
laboratory studies. The understanding of the hydro-geochemical conditions of the past and present
is the basis to predict future evolutions. Therefore, the complete characterization of a site allows
gaining capability and confidence for the extrapolation of data when the information is scarce or not
available. In certain environments as the Fennoscandian shield, the climate may have also very important
effects. For this reason, different methodologies were implemented to understand the effects
of glacial melt water intrusion [7].
In-situ studies carried out at the FEBEX site serve to quantify mass transfer processes from the bentonite
to granite, to perform in-depth analysis of the effects of the bentonite on the water chemistry
and to analyse the presence and stability of bentonite colloids in realistic conditions. Attention is
focused on hydro-geological structure (fractures, fault regions, lamprophyre...) previously identified
[8]. Within FUNMIG, at the end of 2005, two investigation boreholes FU05.001 and FU05.002
were drilled quasi parallel to the FEBEX gallery, and relatively near to the bentonite surface (30
and 60 cm respectively) [9]. To characterize the crystalline rock, three short boreholes were additionally
drilled for the geophysical experiments (FU05.003, -4 and -5) [8]. Several other boreholes
around the gallery already existed (19 radial boreholes, FBX, Bous, etc.) that represent additional
source of information both for hydro-geochemistry and hydrogeology. A schematic of FEBEX tunnel
with the location of the new boreholes and main hydro-geological structures is presented in Figure
1.
Several water sampling campaigns were carried out for both water chemistry and colloid analyses.
The waters from the new boreholes are slightly alkaline and with low electrical conductivity being
very favourable conditions for bentonite colloid stability [4], so that this place is very adequate to
study colloid generation and mobility processes. On the other hand, the migration through granite of
conservative ions like I - and ReO4 - (installed in 1996 with the FEBEX experiment, Figure 1) and
other “natural” tracers from the bentonite (Cl - , Na + ..) is being studied. I - was observed both in
FU05.001 (interval 4, 0.46 ppm) and in FU05.002 (interval 3, 0.85 ppm) in correspondence to the
initial location of the I - impregnated filter papers (Figure 1). The fact that only iodine and not rhenium
was detected must be related to the reducing conditions present in Grimsel waters. The mobility
of Re would be greatly affected if reduction to the oxidation state (IV) occurred.
329

Figure 1: Schematic of the FEBEX drift with the new boreholes, the packed-intervals, the main
fractures and the locations of the Re and I tracers in the FEBEX experiment.
Most intervals of new boreholes, closer to the bentonite, showed higher concentration of the main
ions than the old radial ones. An increase of Na + and Cl - was observed in all the intervals of
FU05.001, particularly relevant in the packed-off section isolating a small lamprophyre dyke (interval
4, Figure 1). Based on the data obtained in these in-situ studies a mass-transfer conceptual
model was developed [9, 10]. Additionally, these studied allowed to determine the mean effective
diffusion coefficient for Cl - (De = 5.0E-11 m 2 /s). This result will be compared with Cl - diffusion
coefficients obtained in the large scale migration experiment that simulates the in-situ FEBEX configuration
(bentonite+ large block granite) (WP4.4) [11].
Distribution of the groundwater flow: characterisation and modelling
In crystalline rocks water flow takes place in the fractures which are the main conducting paths being
advection the dominant transport mechanism. The water flow in the porous rock matrix, with
low hydraulic conductivity, is negligible and the main transport mechanism here is diffusion. The
predominance of advection over diffusion depends on the characteristics of the fluid flow system.
Therefore, the characterisation of the fluid flow system is a key point for evaluating which paths are
actually available for RN transport and retention. Fracture network can be very complex and the
pore space can be connected or not. It is recognised that small scale features may have an important
influence on the overall transport behaviour so that a study of the rock pore space from m to the
dm scale was one of the objectives of WP 4.2.
At the in-situ scale (dm-m) the characterization of the granite in the FEBEX tunnel was carried out
with geophysical experiments, including Natural Gamma, Borehole Ground Penetrating Radar
(GPR) and Cross-hole Ultrasonic Tomography. The main objective of the work was to visualize the
geometry of the network of fractures in the region between the main boreholes in a quasi 3D-shape.
Different fractures cut both FU05.001 and FU05.002 boreholes, all showing low transmissivity
(1·10 -11 -1·10 -12 m 2 /s) with exception of the interval 1 of FU05.001 (6-8·10 -10 m 2 /s), at the back of
the gallery. The hydro-geological conceptual model of the FEBEX site (10s of meters) was updated
on a smaller scale (close to the granite-bentonite interface). Geophysical studies allowed identifying
three different fractured regions, slightly parallel to the gallery, and validating indirect visualization
methods for the determination of fracture network in a crystalline rock [12].
At a laboratory scale (WP4.2) several techniques were used for the characterisation of the pore
structure of different granite samples (Grimsel, Äspö, Olkiluoto, selected cores from the FEBEX
site). Different rock matrix characterisation methods (PMMA method, X-ray tomography, confocal
laser microscopy) were compared to highlight the applicability and limits of each technique [13].
330

The links between pore apertures and mineralogy were studied combining PMMA method and
autoradiography with electron microscopy (FESEM/EDAX). Heterogeneity, anisotropy and connective
pore-network and heterogeneities were identified as a support of transport experiments with
solutes and/or colloids (WP 4.3 and WP 4.4).
Positron emission tomography (PET) studies were performed to analyse the water flow distribution
and colloid transport in a crystalline rock core from Äspö [14]. This non - destructive technique is
applied for the direct 3D visualisation of solute transport using PET tracers ( 18 F, 124 I...). The transport
paths through the fracture were observed to be modulated by the flow rate and, at localised
sites, matrix diffusion was observed [15]. The applicability of PET measurements for investigations
of the spatial distribution of transport processes of both dissolved components and colloids in granite
was demonstrated.
Geometrical description (3D) of the Äspö fractured core was also obtained by X-ray computer micro-tomography
(XTC); using this information advanced simulation of the fluid flow can be done.
A new experimental approach for non-destructive spatially resolved studies of mapping the mineral
composition and the volumetric microstructure of the host rock samples was developed, combining
3D dual-energy cone-beam tomography and X-ray fluorescence [16].
Matrix diffusion.
Matrix diffusion (MD) is considered a very important retardation process in crystalline rock above
all for not sorbing elements. The state of the art of this process was well described in RETROCK
and its theoretical bases in [17]. The effectiveness of matrix diffusion as retardation mechanism depends
on the penetration depth into the rock from the water conducting zones, and it is very dependent
on the porosity of the rock, the flow rate, RN diffusivity as well as on the flow-wetted surface.
To assess its relevance as retention mechanism it is necessary to assess the role of diffusion
against advection, if the extension of the RN diffusion within the matrix is limited or not and if the
pore system is stable over time.
One of the main recognised problems related to diffusion studies is the difficulty of obtaining experimental
data partly because of disturbances and artefacts that may exist in laboratory samples.
Additionally, diffusion lengths are extremely short (in experimental time span of months - years)
above all for highly sorbing radionuclide; thus available data are very scarce. Another process of
interest is the diffusion of anions, which can be affected by anionic exclusion, because PA calculations
showed that doses are mainly controlled by anionic species 129 I - and 36 Cl - . To demonstrate the
role of MD on the retention of these non-sorbing radionuclides is a key point.
One of the final goals of RTDC4 studies is to analyse the role of heterogeneities for the matrix diffusion
process and to develop models that explicitly take into account the heterogeneity of the rock
matrix. In most of the models, matrix diffusion is assumed to be Fickian and homogeneous, but
some authors suggested that these assumptions may not be valid [18]; the uncertainties in the migration
pathways for contaminants may make inappropriate the deterministic treatment of transport
and therefore stochastic methods have been also developed. Indeed, simple models do not include
the effect of the heterogeneity in rock properties. First attempts modelling the diffusion in a rock
matrix consisting of heterogeneous porosity patterns exist [19, 20, 21]. To validate these models, it
is necessary to obtain diffusion coefficients and to correlate diffusion profiles with the physical (and
mineralogical) properties of the rock matrix [22].
Thus, a microscale approach to matrix diffusion was started by combining diffusion studies and
characterization matrix porosity at a mineral scale [23]. The Rutherford backscattering spectrometry
(RBS) is a nuclear ion beam technique that allows measuring concentration profiles in a micrometric
scale with a resolution that allows measurements within a single mineral. Apparent diffusion coefficients
of uranium could be determined in three different granite types (Grimsel, El-Berrocal and
Los Ratones, both from Spain) in different minerals [23]. Differences observed are related to the
nature of the mineral grains, particularly grain porosity. The measured Da values for the U diffusion
331

within feldspars in the three granites were comparable but the most interesting difference was found
between the various quartz grains. In some granite where quartz minerals showed no accessible
PMMA porosity, uranium access was subject to the existence of micro-cracks or intergranular fissures.
Figure 2 shows the RBS spectra of Grimsel granite obtained on quartz, feldspar and biotite.
In the Table below, the values of apparent diffusion coefficients obtained for each mineral type and
its porosity are shown. The RBS technique was also used to quantify Cs diffusion on single minerals
in Czech granites [24].
Energy (MeV)
Normalized Yield
7
6
5
4
3
2
1
1.4 1.6 1.8 2.0
Grimsel Quartz + U 1 day
Grimsel Feldspar + U 1 day
Grimsel biotite + U 1 day
Simulations
0
500 600 700
Channel
800 900
Sorption
“Sorption” is the general term used to define an unknown retention mechanism at a solid surface.
RN sorption may take place at the fracture walls, but also on the materials filling the fractures. In
PA, sorption is handled as a reversible attachment of dissolved species to surfaces using the “Kd
approach”. The Kd is experimentally derived, generally from static “batch” experiments under sitespecific
conditions. Pore surface of the rock matrix is considered to dominate sorption, while the
sorption on fracture or infills is minor and neglected in PA. Limitations of the Kd concept are fully
recognised: in particular, the Kd-approach do not take into account the chemistry of the pore solution
and its variability. Besides, other relevant processes as precipitation /co-precipitation and solid
solution formation may be hidden in Kd values. Mechanistic approach to sorption and retention
processes is widely treated in RTDC 1. In RTDC 4, two main problems related to Kd values are being
evaluated: Kd are not obtained directly on intact rocks [1] and, furthermore, the effect of the
heterogeneities is totally neglected.
Different “visualisation” techniques are available to observe the regions in which radionuclides interacts
(e.g. modern autoradiography method [24]), and to perform sorption studies on intact rocks.
The main challenge is to quantify RN retention at a mineral level, therefore sorption experiments
were carried out with small rock pieces using the particle induced X-ray emission technique
( PIXE). A mapping of the single elements on the solid surface allows identifying both the main
minerals present and the reactive areas where the RN is sorbed. The quantification of RN retention
in single minerals can be done only by specific analyses of the individual PIXE spectra in each
scanned point within 2x2 mm 2 areas and this methodology was developed in RTDC4. Small regions
within single minerals can be selected as shown in Figure 3 where uranium is preferentially sorbed
in a biotite. The variability of the surface distribution coefficient (Ka) was analysed as the studied
areas increased. This investigation tried to understand how the distribution coefficients must be upscaled
for consideration of the mineralogical heterogeneity found in any natural system.
332
Mineral Da (m2/s) Porosity
Feldspar (1.5 0.5) E-13 0.5 %
Quartz (1.1 0.5) E-13 0.5 %
Dark minerals (5.2 0.5) E-13 > 1.4 %
Figure 2: RBS spectra of Grimsel granite main
minerals after contact with U solution 1 day: (�)
Quartz, (�) Feldspar and (�) Biotite. Simulations
are plotted as continuous lines. In the table,
the values of apparent diffusion coefficients obtained
for each mineral type and its porosity are
shown.

Q
B
Si
P
K
-
333
Fe
Figure 3: Elemental distribution maps (Si, K, Fe and U) obtained by PIXE on a 2x2 mm 2 granite
area after the contact with uranium (10 days). Red squares refer to the areas selected to obtain the
individual PIXE spectra for a quantitative analysis, identified here as (Q) quartz; (B) biotite; (P)
plagioclase.
3. Transport in crystalline rocks: Effects of colloids.
To play a role in RN migration, colloids must exist in a non-negligible concentration, be mobile,
stable and be able to adsorb radionuclides in irreversible form [25]. These conditions must be verified
investigating scenarios, geochemistry, hydrogeology, and other physical factors as well as possible
artefacts that could bias data interpretation. In poorly mineralized waters, such as those present
at the GTS, bentonite colloids may fulfil several of the above-mentioned conditions, so that the
determination of the effects on radionuclide migration has to be studied in depth. The hydration and
loss of density of the bentonite backfill are necessary conditions for the colloids to be formed at the
bentonite/granite interface [4, 26] but the quantification, in realistic conditions, of the colloid source
term from the engineered barriers is still an open issue [1, 27].
In-situ transport studies at the GTS in the CRR project [5, 28] demonstrated that bentonite colloid
migration was not retarded with respect to the water flow; the colloid recovery depended on not
very well identified filtration processes taking place along the flow path as size exclusion,
rock/colloid interactions and diffusion in the rock matrix. The recovery of bentonite colloids and
highly sorbing tri- and tetravalent elements was very high [29] but water flow conditions were not
fully representative of those expected in a geological repository. Thus, it was considered necessary
to perform more experimental studies at a laboratory scale, under constrained conditions as similar
as possible to the real ones.
In RTDC4, different processes related to bentonite colloids were analysed: generation [30], stability
and effects on RN speciation (31), recovery under different flow rates [32], diffusion in the rock
matrix [33]. Other studies on transport and rock/colloid interactions related with colloid properties
(size or surface charge) were carried out also with model colloids (Au, quantum dots composed of
CdSe/ZnS).
The generation of bentonite colloids from compacted clay in contact with stagnant water was analysed.
Figure 3 (left) shows that colloid concentration initially increases but a steady state is
reached in a relatively short time. The higher the clay dry density, the higher the quantity of colloids
generated. Water chemistry controls the stability, size and concentration of the generated colloids:
the concentration of colloids increase, when the salinity of the water decreases (Figure 3,
right) and when pH increases.
+
U

Colloid Concentration (ppm)
160
140
120
100
80
60
40
20
0
BACKppm
1.2DES
1.4DES
1.6DES
2.21
0.75
0.24
1.6 g/cm 3
Background
1.4 g/cm 3
0 100 200 300 400 500
Time (days)
1.2 g/cm 3
334
Colloid concentration (ppm)
160
140
120
100
80
60
40
20
0
Deionised water
Grismel water
NaCl 10 -2 M
1.2 1.4 1.6
Compaction density (g/cm 3 )
Figure 3: Left: Generation of bentonite colloids from bentonite plugs at different initial densities in
deionised water. Right: Dependence on the colloid concentration with the compaction density in
three different waters (deionised, Grimsel groundwater and NaCl 0.01 M. As received FEBEX clay.
Ca-homoionised clay did not form colloid in appreciable concentration, but the presence of Na in
the exchange complex (20 %) completely changes the generation behaviour (for example, asreceived
FEBEX bentonite). Finally, the surface exposed to hydration (and the consequent existence
extrusion paths) also affects colloid generation. Generated bentonite colloids are stable over
month in low mineralised and alkaline pH and their stability may increase in the presence of humic
acids.
Neretnieks and Liu [34] suggested a ‘zero order model’ to describe the colloid generation from
compacted bentonite for PA purposes. The criterion of colloid release is the critical coagulation
concentration corresponding to 1 mmol/L Ca 2+ . The studies performed within FUNMIG clearly
show that taking a certain Ca concentration as a colloid generation criterion does not describe the
real behaviour. Calculated colloid release rates according to the ‘zero order model’ in general are
by far higher than measured in experiments. The new experimental data can be used to develop an
improved colloid generation model.
In most of the transport studies performed in fractured cores, the breakthrough curves of colloids
always presented an elution peak in a position very similar to that of conservative tracers but that
their recovery critically depended on the colloid concentration and on the water flow rate (Figure
4).
In the case of bentonite colloids, significant colloid filtration was observed in spite of the existence
of “unfavourable” electrostatic conditions for colloid-rock attachment. In the Grimsel case, the geochemical
conditions favour high colloidal stability, nonetheless, the filtration of bentonite colloid
increased significantly when the hydrodynamic conditions approached the ones expected in a repository
(low water flow rates) and when the roughness of the fracture surface increased [35].

Colloid Recovery (%)
100
80
60
40
20
0
Flow rate 3-6 ml/h
Flow rate 6-10 ml/h
Flow rate > 10 ml/h
0 20 40 60 80 100 120 140 160 180
Colloid concentration (ppm)
Flow rate

form. Clay colloids were detected in some intervals of the borehole FU05.001 (30 cm far from the
bentonite) and compared to those obtained in the laboratory studies of bentonite colloid generation.
The similarity in both microstructure and composition was shown. The quantification of bentonite
colloids is still a difficult issue even it is clear that, at approximately 30 cm from the bentonite, the
quantity of bentonite colloids cannot be higher than 1 ppm. Higher colloid concentrations were
measured by PCS, showing that artifacts, possibly introduced during the excavation of the new
boreholes, exist. The analysis of these artifacts for a better quantification of the “source term” is a
very important issue at moment [40].
4. Modelling of fluid and solute transport: up-scaling methodologies
Many classical papers exist in the literature, describing the basic concepts of transport in fractured
porous media. Modelling approaches different from the classical advection-dispersion model have
been developed trying to overcome problems related to this “classical” description of solute transport
behaviour at large scales. In fact, the classical formulation of solute transport may significantly
underestimate the late-time behaviour of concentration breakthrough curves in highly heterogeneous
media. A fundamental issue is to validate the application of these models to real sites and to
evaluate their advantages/limitations [41, 42]. A methodology proposed in RTDC4, to up-scale
transport is the dual-domain model. A new random walk particle tracking methodology to efficiently
simulate dual-domain solute transport was developed [43]. This methodology seems more
adequate than standard macro-dispersive transport model but still misspredict chemical reactions
occurring within the system.
From a modelling point of view, to combine flow and transport modelling, taking into account the
possible effects of heterogeneities at different scales is a challenge because flow and transport homogenise
at some spatial and temporal scales. Furthermore, the transported species can react
amongst each other and with the host medium which, together with advective and diffusive mass
transfer mechanism, can lead to a behaviour that is very different from the one observed in a homogeneous
batch environment. The need to incorporate multi-component reactive transport has been
stressed by the RETROCK project [1].
Reaction rates can be successfully obtained from mixing ratios and disequilibrium within the medium
can be often controlled by mixing, therefore these processes have to be analysed. The main
difficulties arise when the effects of space and time variability have additionally to be included:
spatial heterogeneities in the physical and chemical properties of the medium and temporal fluctuations
of the state variables lead to non-trivial scale dependence of transport and reaction phenomena.
In RTDC4, a methodology for the analytical solution for reaction rates for multi-species equilibrium
and non-equilibrium reactive transport in heterogeneous media was developed [44, 45],
quantifying the impact of heterogeneity on reaction rates [46] as well as the effect of heterogeneities
on mixing and spreading. New indicators for mixing were derived [47, 48].
An additional aim was to develop an efficient numerical code for reactive transport modelling for
PA. This includes specifically flexible and transportable reactive transport module for the simulation
of complex geochemical process, and the implementation of effective heterogeneity induced
transport dynamics and coupling with the reactive transport modules (CHEPROO, [2]).
Based on the theoretical and numerical studies on effective reactive transport outlined above, an
effective large scale hybrid transport model is implemented. It is an hybrid model in so far, as large
scale heterogeneity (obtained by hydraulic test, georadar, tracer tests, etc.) is represented explicitly,
while the impact of small scale heterogeneity (difficult to determine deterministically) on transport
is taken into account implicitly using a multirate mass transfer model. The resulting conservative
transport code is coupled with the chemistry module CHEPROO.
336

The multirate mass transfer (MRMT) approach for effective transport modelling in heterogeneous
media has proven to represent well the effective behaviour of conservative elements. MRMT seems
to be a good methodology to integrate the impact of spatial heterogeneity on effective mass transfer
and solute transport (Willmann et al, 2008 under review). An effective reactive transport model
based on MRMT transport mechanisms and coupling with CHEPROO is in development stage and
a research code exist (Willmann et al, 2008 in preparation).
5. Conclusions
The EC-FUNMIG project gave the opportunity to study more in depth processes affecting radionuclide
migration in crystalline media (RTDC 4). Fluid-flow system, matrix diffusion and sorption,
were studied focusing on different aspects that are not yet completely understood, in particular on
the effects of the physico-chemical heterogeneities affecting the crystalline media.
The use of new characterisation techniques of the fluid flow distribution is useful for a better description
of the rock and the evaluation of hydraulic parameters, reducing the uncertainty in PA.
The studies on matrix diffusion were focused on the development of experimental and modelling
methodology to account for the physical and mineralogical rock heterogeneity. Results obtained so
far are promising to describe this mechanism more adequately. In a similar way, the quantification
of sorption by the determination of Kd in the intact rock, proposed in RTDC4, provides more realistic
data based on more in-depth knowledge of retention processes at the rock surface. The research
on this issue should continue on these premises.
Successful studies allowed the development of up-scaling modelling methodologies. Additional
studies focused on the combination of fluid and reactive transport in heterogeneous media. However,
the application of these models to real conditions is still missing and this work needs further
efforts.
A significant part of the RTDC4 work was devoted to study the relevance of colloids coming from
the compacted bentonite barrier on radionuclide migration in crystalline media. The main parameters
affecting the generation and stability of colloid were identified; laboratory experiments provided
data to update existing models on colloid generation and to quantify the “source term” more
precisely.
Both strongly and slight sorbing elements showed, in the presence of colloids, a breakthrough peak
related to colloid-driven transport. However, dynamic experiments demonstrated that bentonite colloid
recovery decreases significantly as approaching the hydrodynamic conditions expected in a repository
(i.e. very low water flow rates) even in conditions unfavourable to colloid attachment to
rock surface. The retention of colloid on the rock surface was analysed and quantification studies
were started, including the determination of bentonite colloids diffusion coefficients.
Results obtained so far, indicated that future experiments should be focused on the study of irreversibility
of RN-colloids and rock-colloid interactions in dynamic conditions.
All the studies carried out in real sites always provided valuable inputs. The recently started studies
at the FEBEX gallery would improve as far as the effects of disturbance due to new borehole excavation
decrease, in particular regarding to the evaluation of the presence of bentonite colloids. To
continue these studies in such a realistic environment is of great interest.
6. Acknowledgements
This project has been co-funded by the European Commission and performed as part of the sixth
Euratom Framework Programme for nuclear research and training activities (2002-2006) under contract
FI6W-CT-2004-516514.
337

References
[1] Retrock (2005). Treatment of radionuclide transport in geosphere within safety assessment.
Final Report EUR 21230 EN.
[2] S. Bea, J. Carrera, C. Ayora, F. Batlle and M.W. Saaltink (2008). CHEPROO: A Fortran 90
Object-Oriented module to solve chemical processes in earth sciences models. Computers &
Geosciences, under review, 2008.
[3] E.F. Cassiraga, D. Fernàndez-Garcia, J.J. Gómez-Hernández, (2005), Performance assessment
of solute transport upscaling methods in the context of nuclear waste disposal, International
Journal of Rock Mechanics and Mining Sciences (ISSN 0148-9062), 42, 756-764, 2005.
[4] T. Missana, U. Alonso, M.J. Turrero (2003) Generation and stability of bentonite colloids at
the bentonite/granite interface of a deep geological radioactive waste repository, J. Contaminant
Hydrology, 61(1-4), 17-31.
[5] H.Geckeis, T. Schäfer, W. Hauser, Th. Rabung, T. Missana, C. Degueldre, A Möri, J.
Eikenberg, Th. Fierz, W.R. Alexander (2004). Results of the Colloid and Radionuclide Retention
experiment (CRR) at the Grimsel Test Site (GTS), Switzerland.- Impact of reaction kinetics
and speciation on radionuclide migration. Radiochimica Acta, 92(9-11), 1-10.
[6] M. Laaksoharju, J. Smellie, E.L. Tullborg, M. Gimeno, J.B. Gómez, L.F. Auqué, J. Molinero,
I Gurban, L. Hallbeck, G. Buckau (2007): Hydrogeochemical methodology for constructing
site descriptive models within the Swedish site investigation programme. Contibution to the
WRI-12 conference in Kunming, China 2007.
[7] J. Suksi, K. Rasilainen, P. Pitkänen, (2006). Variations in 234 U/ 238 U activity ratios in groundwater
– a key to flow system characterisation? Physics and Chemistry of the Earth 31, 556-
571.
[8] Perez-Estaún, A., Carbonell, R., Martinez, L., Dentz, M., Suso, J., Carretero, G., Bueno, J.,
Buil, B., Garralon, A., Gomez, P., Arcos, D., Hernan, P., (2007), New boreholes to investigate
the bentonite/crystalline rock interface in the FEBEX gallery (Grimsel). Rapport CEA-R-
6122, 157-162.
[9] P. Gómez, B. Buil, A. Garralón, M.J. Turrero, L. Sánchez, and J.M. Durán, (2007). Geochemical
gradients at the near-far field interface at the FEBEX experimental gallery (Grimsel
Test Site). Water rock interaction WRI 12, T.D. Bullen and Y. Wang editors, Taylor & Francis
Group, (ISBN 0415451361), Vol 2, pp. 583-587. Paloma et al
[10] B. Buil, P. Gómez,, J. Peña, A. Garralón, M.J. Turrero, L. Sánchez, J.M. Durán, (2008)
Modelization of the mass transfer generated by bentonite into the granite at the FEBEX drift.
Chemical Geology (in press).
[11] J. Samper, Q. Jang, M. García-Gutiérrez, T. Missana, M. Mingarro (2007). Scoping calculations
for the design of a laboratory mock-up experiment of bentonite granite interface. 3d Annual
workshop proceedings EC-FP Funmig Edinburgh
[12] R.Carbonell, Perez-Estaun, A., (2008). High Resolution Geophysical Characterization of
Fractures with a Granitic Pluton, Near Surface Geophysics, submitted
[13] T. Lähdemäki, M. Kelokaski, M. Siitari-Kauppi, M. voutilanen, M. Myllys, T. Timonen, F.
Mateos, M. Montoto (2007). Characterising Low-permeable granitic rocks from micrometer
to centimiter scale: X-ray microcomputed Tomography, Confocal laser scanning microscopy
and 14C-PMMA method: Scientific basis for nuclear waste management XXX,
Mat.Res.Soc.Symp. Proc. 985, 587-592.
[14] M. Gründig, M. Richter, J. Kulenkampff, A. Seese (2007): Studies of the Spatial Water Flow
Distribution and Colloid Transport in a Crystalline Rock Core from Äspö with Positron
Emission Tomography, SKB-Report TR-07-05, 223-229.
338

[15] J. Kulenkampff, M. Grundig, M. Richter, F. Enzmann, Evaluation of positron –emission
tomography for visualisation of migration processes in geomaterials. Physics and Chemistry
of the Earth, in press doi:10.1016/j.pce.2008.05.005
[16] I. Tiseanu et al, Combined 3D tomography and X-ray fluorescence for crystalline host rock,
presented at IP FUNMIG 2nd Annual Workshop, Stockholm, 21st-23rd November, 2006
[17] [1] U. Frick, An evaluation of diffusion in the groundwater of crystalline rocks, Interner
Bericht 92-92E Nagra Report (1996).
[18] Haggerty S. (2002): Matrix diffusion: heavy-tailed residence time distributions and their influence
on radionuclide retention. In Radionuclide retention in Geologic media. Vº GEO-
TRAP Workshop proceedings. AEN-NEA.
[19] P. Sardini, F. Delay, H.K. Hellmuth, G. Porel and E. Oila (2003): interpretation of out diffusion
experiments on crystalline rocks using random walk modelling. J. Contaminant Hydrology,
61:339-350.
[20] P. Sardini, J.C. Robinet, M. Sittari-Kauppi, F. Delay, K.H. Hellmut (2007), Direct simulation
of heterogeneous diffusion and inversion procedure applied to an out diffusion experiment.
Test case of Palmottu granite, Journal of Contaminant Hydrology, 93 21-37.
[21] A. Poteri (2006) Spatial up-scaling of the retention by matrix diffusion, Posiva Work Report
2006-101.
[22] Leskinen, L. Pettinen, M. Siitari-Kauppi, U. Alonso, M. García-Gutierrez, T. Missana A.
Patelli (2007). Determination of Granites' Mineral Specific Porosities by PMMA Method and
FESEM/EDAX in Scientific Basis for Nuclear Waste Management XXX, edited by D.S.
Dunn, C. Poinssot, B. Begg (Mater. Res. Soc. Symp. Proc. 985, Warrendale, PA, 0985-NN11-
20. ISBN 1-55899-752-0.
[23] M. Siitari-Kauppi, A. Leskinen, M. Kelokaski, L. Togneri, U. Alonso, T. Missana, M. Garcia-
Gutérrez, A. Patelli. Physical Matrix charactrisation: studies of crystalline rocks and consolidated
clays by PMMA Method and electron microscopy as a support of diffusion analyses. Informe
Técnicos CIEMAT 1127 Diciembre 2007 (2007).
[24] V. Havlová et al. (2007): Multimethod approach to determination of sorption and retention
within granitic rock. S+T publication, 3rd FUNMIG annual meeting, Nov. 26 – 29, Edinburgh,
2007 (being submitted)
[25] J.N. Ryan, M. Elimelech (1996). Review: Colloid mobilization and transport in groundwater.
Colloids Surfaces A 107, 1-56.
[26] M. Baik, W. Cho, P. Hahn (2007). Erosion of bentonite particles at the interface of a compacted
bentonite and a fractured granite. Engin. Geol. 91, 229-239.
[27] U. Alonso, T. Missana, H. Geckeis, M. Garcia-Gutierrez, M.J. Turrero, A. Möri, Th. Schäfer,
A. Patelli, V. Rigato (2006). Role of inorganic colloids generated in a high-level deep geological
repository in the migration of radionuclides: Open questions. J. Iberian Geol. 32, 79-
94.
[28] A.Möri, W.R. Alexander, H. Geckeis, W. Hauser, Th. Schäfer, J. Eikenberg, T. Fierz, C. Degueldre,
T. Missana (2003). The colloid and radionuclide retardation experiment at the Grimsel
Test Site: influence of bentonite colloids on radionuclide migration in fractured rock. Colloids
Surfaces A, 217, 33-47
[29] W.Hauser, H. Geckeis, J.I. Kim, T. Fierz (2002). A mobile laser-induced breakdown detection
system and its application for the in-situ monitoring of colloid migration. Colloids Surfaces
A, 203, 37-45.
[30] U. Alonso, T. Missana, M. Garcia-Gutierrez (2007). Experimental approach to study the colloid
generation from the bentonite barrier to quantify the source term and to assess its rele-
339

vance on radionuclide migration. Mater. Res. Soc. Symp. Proc. 985: Scientific basis for nuclear
waste management XXX: 605-610
[31] M. Bouby, H. Geckeis, Th. Schäfer, J. Lützenkirchen, H. Seher, A. Bauer, M. Plaschke, W.
Hauser, B. Kienzler (2007) Laboratory study on colloid stability and radionuclide-colloid interaction
under Äspö groundwater conditions. 3 rd Annual Meeting FUNMIG Proceedings.
NDA Repport, UK (In press).
[32] T. Missana, U. Alonso, M. García-Gutiérrez, M. Mingarro (2008).. Role of colloids on the
migration of Eu and Pu in a granite fracture. Applied Geochemistry 23, 1484-1497
[33] U. Alonso, T. Missana, A. Patelli, V. Rigato (2007). Bentonite colloid diffusion through the
host rock of a deep geological repository, Physics and Chemistry of the Earth, 469-476
[34] J. Liu, I. Neretnieks (2006). Physical and chemical stability of the bentonite buffer. SKB Rapport
R-06-103, SKB, Sweden.
[35] N. Albarran, T. Missana, Ú. Alonso, M. García-Gutiérrez, T. Lopez-Torrubia (2008). Dynamic
study of colloid retention in a granite fracture. Colloid and Surfeces A, Submitted
[36] U.Alonso, T. Missana, A. Patelli, D. Ceccato, N. Albarran, M. García-Gutiérrez, T. Lopez, V.
Rigato (2008). Quantification of Au nanoparticle retention on a heterogeneous rock surface.
Colloid and Surfaces A, Submitted
[37] A. Filby, M. Plaschke, H. Geckeis, Th. Fanghänel (2007) Interaction of carboxylated latex
colloids with mineral surfaces. 11th Internat. Conf. on the Chemistry and Migration Behaviour
of Actinides and Fission Products in the Geosphere (Migration 2007), München
[38] U. Alonso, T. Missana, A, Patelli, V. Rigato, J. Ravagnan (2007). Colloid diffusion in crystalline
rock: experimental methodology to measure diffusion coefficients and evaluate size dependence.
Earth and Planetary Science Letters 259, 372-383.
[39] M. Bouby, H. Geckeis (2007) On the dynamics of tetravalent actinide(th/puiv)-humic acid
interaction”.11th Internat. Conf. on the Chemistry and Migration Behaviour of Actinides and
Fission Products in the Geosphere (Migration 2007), München
[40] T. Missana, U. Alonso, N. Albarran, P. Gómez, B. Buil, Th. Schäfer, W. Hauser, H. Seher, A.
Garralón. Bentonite colloid generation from a deep geological repository in granite: an in-situ
study. Geochimica et Cosmochimica Acta, Vol 72 (12) Suppl S, A635 (2008). Goldschmidt
2008 Conference Vancouver July 2008.
[41] P. Salamon, D. Fernàndez-Garcia, D., J. J. Gómez-Hernández (2006), A review and numerical
assessment of the random walk particle tracking method, Journal of Contaminant Hydrology
(ISSN 01697722), VOL. 86, 277-305, 2006.
[42] D. Fernàndez-Garcia, J. J. Gómez-Hernández (2007), Impact of upscaling on solute transport:
Travel times, scale-dependence of dispersivity and propagation of uncertainty, Water Resources
Research (ISSN 0043-1397), VOL. 43, W02423, doi:10.1029/2005WR004727, 2007.
[43] P. Salamon, D. Fernàndez-Garcia, J. J. Gómez-Hernández (2006), Modeling mass transfer
processes using random walk particle tracking, Water Resources Research (ISSN 0043-1397),
VOL. 42, W11417, doi:10.1029/2006WR004927, 2006.
[44] Sánchez-Vila, X., M. Dentz, L. Donado, Transport controlled reaction rates under local nonequilibrium
conditions, Geophysical Research Letters, 34, L10404, 2007.
[45] De Simoni M., Sanchez-Vila X., Carrera J., and Saaltink M. (2007), A mixing ratios-based
formulation for multicomponent reactive transport, Water Resour. Res. (43), W07419,
doi:10.1029/2006WR005256.
[46] J. Luo, M. Dentz, J. Carrera, and P. Kitanidis, Effective reaction parameters for mixing controlled
reactions in heterogeneous media, Water Resour. Res., 44, W02416, 2008.
[47] M. Dentz, J. Carrera (2007) Mixing and spreading in stratified flow. Phys. Fluid 19, 017107,
2007
340

[48] M. Dentz, D. M. Tartakovsky, Four-point closure for transport in steady random flows, Phys.
Rev. E, 77 066307, 2008
341

342

Investigation of far-field processes in sedimentary formations
at a natural analogue site - Ruprechtov
Ulrich Noseck 1 , Vaclavá Havlová 2 , Radek Cervinka 2 , Juhani Suksi 3 , Melissa Denecke 4 , Wolfgang
Hauser 4
Summary
1 GRS, Germany
2 NRI, Czech Republic
3 University of Helsinki, Finland
4 FZK-INE, Germany
The analogue study at Rurpechtov site aimed at in-depth understanding of the behaviour of
uranium and organic matter in a natural sedimentary system similar to overlying strata of salt
domes but also other host rock formations for radwaste disposal. By application of a set of different
microscopic and macroscopic analytical methods the complex immobilisation mechanism
for uranium and the long-term stability of the immobile U(IV) phases were shown.
Sedimentary organic carbon (SOC) is microbially degraded in the lignite rich layers, but DOC
concentrations are relativley low, since only a very small fraction of SOC seems to be accessible.
Beside increase of process understanding another important contribution of this study to
the Safety Case consists in further development and testing of methods for colloid analysis,
sample characterisation on μ-scale, and evaluation of environmental isotope signatures.
1. Introduction
In component RTDC5 of the integrated EC project FUNMIG the far field of the host rock formation
salt is subject of the investigations. In contrast to the two other host-rock components RTDC3
(clay) and RTDC4 (granite) RTDC5 is a natural analogue study. The study is performed at Ruprechtov
site in Czech Republic and represents an analogue for potential migration processes in a
similarly structured overburden of a salt dome but also in other geological formations, which are
foreseen as potential host rocks for radioactive waste repositories.
Results from site characterisation were already available before this study, e.g. [1]. Within the
FUNMIG project specific questions have been addressed, to receive an in-depth understanding of
the evolution of the natural system and the key processes involved in uranium immobilisation as
well as the behaviour of organic matter. Emphasis was put into characterisation of the immobile
uranium phases, their long-term stability and the processes controlling mobility of uranium in the
system. The second major issue comprised the behaviour of colloids and organic matter in the system,
i.e. to better understand the interrelation between sedimentary organic carbon (SOC) and dissolved
organic carbon (DOC) and its impact on the mobility of uranium.
343

The Ruprechtov site, located in the north-western part of the Czech Republic, is geologically situated
in a Tertiary basin [1]. The study area is characterised by a granitic body, which partly crops
out in the west and in the south (see Figure 1) and is widely kaolinised in varying thicknesses (up to
several tens of metres) on its top in the central part. The horizon of major interest is the so-called
clay/lignite layer, a zone of 1-3 m thickness at the interface of kaolin and overlaying pyroclastic
sediments (in 20 to 50 m depth), with high content of SOC, areas of uranium enrichment and partly
aquiferous. This layer does not represent a continuous aquifer, but rather distinct areas. As indicated
in Figure 1 the general groundwater flow direction in the water bearing zones in the Tertiary is from
southwest to northeast. Infiltration area is supposed to be in the outcropping granite in the western
and south-western part, represented by the boreholes NA10, NA8 and RP1.
NA10
NA8
RP1
NA9
NA13
NA12
NA6 NA5
NA7
RP4
NA14
RP3
NA4
RP5
RP2
HR4
344
Water flow
PR4
100 m
LEGEND
Coal and carbonaceous
clays in pyroclastics
Pyroclastic sediments
(undiff.), argillized
Secondary kaolin
(Kaolinite clays and sands)
Primary kaolin
(kaolinized granite)
Granite
(slightly kaolinized)
Granite
(Krusné hory type)
Figure 1: Geological sketch of Ruprechtov site with location of the boreholes [2].
2. Methodology
As described above the investigations in RTDC5 are a natural analogue study, where specific questions
are addressed, reflected by the following three work packages. The methodology for each
work package is briefly described.
Colloid characterisation: A borehole groundwater sampling system and a mobile laser-induced
breakdown detection equipment (LIBD) for colloid detection, combined with a geo-monitoring unit
has been further developed and applied to characterise the natural background colloid concentration
in groundwaters of the Ruprechtov natural analogue site [3]. To minimise artefacts during groundwater
sampling the contact to atmospheric oxygen has been excluded by use of steel sampling cylinders
opening and closing by remote control. The groundwater samples collected in this way are
transported to the laboratory where they were eluted from the cylinders under original hydrostatic
pressure without contact to oxygen. Behind the LIBD detection cell the system consists of in-series
connected detection cells for pH, Eh, electrical conductivity, and oxygen content.
Characterisation of immobile uranium phases: Different microscopic and macroscopic methods
have been applied to selected samples to characterise the immobile uranium phases. These methods

comprise μ-XRF and μ-XAFS spectroscopy, sequential extraction, U(IV)/U(VI) separation with
234 U/ 238 U ratio determination. Detailed information can be found in [4]. In addition, sorption experiments
have been performed and exchangeable uranium determined by isotope exchange with
233 U, e.g. [9], [13]. μ-XRD and μ-XAFS were applied the first time to natural samples and the
method was further developed within the FUNMIG project.
Real system analyses: This work package comprised the integration of results from both other work
packages and available information from site characterisation. Moreover, analyses of specific environmental
isotopes, in particular � 34 S in dissolved sulphates as well as � 13 C and 14 C in DOC have
been performed. Additional work was carried out on characterisation of DOC and SOC (integration
with RTDC2). DOC was characterised by IR and MALDITOF spectroscopy, whereas extracted
humic substances were characterised using elementary analysis, ash and moisture content, UV-Vis
spectroscopy, FTIR spectroscopy and acidobasic titrations. Details can be found in [5].
3. Results
3.1 Colloids
As described above part of the work was dedicated to method-development and qualification. During
the project a new mobile geomonitoring system including a system for the laser-induced breakdown
detection of colloids (LIBD) was developed [3]. It was successfully applied to the detection
of colloids in natural groundwater samples from Ruprechtov site and other sites in Sweden.
It could be shown that colloid concentrations down to μg/l are detected with the special sampling
technique. By comparison with in-situ probe measurements in several boreholes it was also demonstrated
that reducing conditions with minimal access of atmosphere oxygen are maintained. In-situ
experiments in granite demonstrated that colloid generating processes like redox changes (e.g. in
EDZ) and/or mechanical erosion can increase the natural colloid background by up to several orders
of magnitude.
The LIBD determined natural background colloid concentrations found at Ruprechtov are compared
with data of studies performed in Äspö (Sweden) and Grimsel (Switzerland). A comprehensive representation
of colloid concentrations in different water samples as determined by LIBD as a function
of the respective ionic strength is given in Figure 2 [3]. Increasing ionic strength usually forces
colloid aggregation which is reflected in lower colloid concentration in the respective groundwater.
In the Ruprechtov groundwater samples the ionic strength varies in the range of 2·10 -3 to 1.1·10 -2
mol/l without any significant influence on the measured colloid concentration. The broad bandwidth
of detected colloid concentrations in groundwater of ionic strength

Colloid Concentration / μg/l
10000
1000
100
10
1
0.1
0.01
TGT/1
TGT/3
MZD
Grimsel
Volvic
(PE)
TGT/2
Ruprechtov
GTS
Bad Liebenzeller
(glass)
Volvic
(glass)
Gerolsteiner
(PE)
1 10 100 1000
Ionic Strength / mmol/l
346
BDS
LEU
ZUR
Äspö
NaCl solution
Data from C. Degueldre 1996
Forsmark
Figure 2: Comparison of colloid concentration in different types of natural groundwater, mineral
water and synthetic NaCl-solution versus ionic strength. For details see [3]
3.2 Characterisation of immobile uranium phases
The application of macroscopic and microscopic methods provided detailed insight into the U enrichment
processes at the Ruprechtov site. Confocal μ-XRF and μ-XANES notably contributed to
the identification of uranium immobilisation processes. In good agreement with results from other
spectroscopic methods like ASEM and electron-microprobe μ-XANES identified U as U(IV) [6].
As demonstrated in Figure 3 (left), the shape and intensities show the average valence state of the
sampled volume to be U(IV). All three curves do not show the multiple scattering feature 10-15 eV
above the white line (WL) characteristic for U(VI) nor do they show a significant decrease in the
WL intensity, which would be expected for U(VI) as be seen in the schoepite spectrum.
norm. Absorption
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0 μm
40 μm
90 μm
schoepite
17.12 17.14 17.16 17.18 17.20 17.22 17.24
Energy [keV]
Figure 3: Results from μ-XANES (left) and μ-XRD (right) of a sample from borehole NA4 [6,7].

By μ-XANES it was also shown that As exists in two oxidation states, As(0) and As(V). The analyses
of a number of tomographic cross-sections of elemental distributions recorded over different
sample areas show a strong positive correlation between U and As(V). By further development of
the method, using new planar compound refractive lens (CRL) array at the Fluoro-Topo-Beamline
at the synchroton facility ANKA of the Forschungszentrum Karlsruhe, a higher spatial resolution
(focus beam spot size of 2 x 5 μm 2 (V x H)) was achieved. The high resolution made it possible for
the first time to discern an As-rich boundary layer surrounding Fe(II)-nodules, see Figure 3, right
[7]. This suggests that an arsenopyrite mineral coating of framboidal pyrite nodules is present in the
sediment. Uranium occurs in direct vicinity of the As-rich layers. In conclusion of these results a
driving mechanisms for uranium-enrichment by secondary uranium(IV) minerals in the sediment
was suggested. Mobile, groundwater-dissolved U(VI) was reduced on the arsenopyrite layers to
less-soluble U(IV), which formed U(IV) mineral phases. As(0) was oxidised to As(V). Uranium,
therefore, is associated with As(V).
The results from microscopic methods are supported by cluster analysis of sequential extraction results.
They also indicate that U occurs in the tetravalent state, since major part of uranium is extracted
in the respective steps for U(IV) forms and the residual fraction [4]. By cluster analyses,
performed to identify possible correlations between elements, a strong correlation of U with As and
P was found (see Figure 4), supporting the mechanism postulated above and the existence of uranium
phosphate mineral ningyoite identified by SEM-EDX.
Similarity
Similarity
0
0
-100
-200
-300
NA14: 725 mg/kg U
Na As P U K Al Fe S
Na
As
P
U
K
1 2 3 4 5 6 7 8 9
Al
Fe
S
347
Similarity
0
0
-10
-20
-30
-40
As
NA15: 50 mg/kg U
As U P S K Na Fe Al
U
P
S
K
1 2 3 4 5 6 7 8 9
Figure 4: Cluster analyses for extended SE results of samples from the boreholes NA14 and NA15
In order to separate U(IV) and U(VI), a wet chemical method [9] was applied for the first time to
Ruprechtov samples. A major result is that uranium in all samples consists of both U(IV) and U(VI)
[4]. Results from all analyses are summarised in Table 1. The extraction did not dissolve all uranium.
The content of uranium in this insoluble phase is denoted as U(res). In all phases the
234 U/ 238 U activity ratio was determined, which is denoted as AR. The AR differs significantly in the
U(IV) and U(VI) phases, with ratios 1 in the U(VI) phase in
nearly all samples. The AR of the U(res) phase is, with exception of NA12, similar to that observed
in the U(IV) phase. Different (higher) AR in the NA12 residue may indicate involvement of different
U compounds in the sample material, i.e. U(IV) and insoluble U(res) represent different compounds.
Taking into account the higher stability of U(IV) phases we assume that insoluble uranium
Similarity
Na
Fe
Al

exists as a stable mineral phase in oxidation state IV. Most uranium in all samples is U(IV), with
contents between 50 and 90 wt.%. The highest U(IV) fraction is found in NA6-37.
Table 1. Amount of uranium and 234 U/ 238 U-activity ratios (ARs) in the different phases from uranium
separation [4]
sample U [ppm]
U(IV) U(VI) U(res, IV) U(IV) total
[%]
234 238
U/ U [%]
234 238
U/ U [%]
234 238
U/ U [%]
NA6 35a 356 7 28.7 0.54 0.01 41.9 1.42 0.02 29.5 0.65 0.01 58.1
NA6 35b 468 9 45.9 0.56 0.01 33.3 1.69 0.03 20.7 0.64 0.02 66.7
NA6 35c 369 8 23.3 0.47 0.01 47.4 1.16 0.02 29.3 0.67 0.01 52.6
NA6 37a 37.3 2 73.7 0.79 0.03 15.7 2.66 0.07 10.6 0.73 0.01 84.3
NA6 37b 47.5 1,5 66.2 0.52 0.01 9.0 3.37 0.15 24.8 0.86 0.02 91.0
NA6 37c 35.7 2 51.3 0.58 0.01 19.8 2.56 0.08 28.9 0.71 0.04 80.2
NA11 a 52.5 2,2 4.2 0.44 0.02 34.1 0.94 0.01 61.7 0.55 0.02 65.9
NA11 b 53.6 2,5 6.1 0.39 0.02 49.6 0.94 0.02 44.3 0.67 0.02 50.4
NA12 a 27.2 1,4 3.9 0.26 0.02 47.5 1.31 0.04 48.6 1.22 0.03 52.5
NA12 b 31.4 2,1 4.2 0.13 0.02 57.4 1.25 0.04 38.4 1.04 0.03 42.6
NA13 a 216 7 5.4 0.53 0.02 46.8 1.15 0.02 47.8 0.55 0.01 53.2
NA13 b 230 9 1.6 0.58 0.03 46.2 1.15 0.02 52.2 0.53 0.01 53.8
NA14 b 317 10 13.3 0.87 0.02 68.8 1.28 0.02 18.0 0.44 0.01 31.2
NA14 c 354 11 25.4 0.88 0.02 39.6 1.11 0.01 41.5 0.54 0.01 58.5
That U(res) and U(IV) exhibit in nearly all samples an AR below one is a strong indicator for their
long-term stability. AR values significantly below unity are caused by the preferential release of
234 U, which is facilitated by �-recoil process and subsequent 234 U oxidation. In order to attain low
AR values of approx. 0.2 in the U(IV) phase, it must have been stable for a sufficiently long time,
i.e. no significant release of bulk uranium has occurred during the last million years. This is in good
agreement with the hypothesis that the major uranium input into the clay/lignite horizon occurred
during Tertiary, more than 10 My ago [10].
3.3 Real system analysis
One part of the work comprised the re-evaluation of existing, and the evaluation of new hydrological,
geochemical and environmental isotope data from groundwater wells at Ruprechtov site to
characterise the hydrogeological flow regime and the carbon chemistry in the system. The existing
idea about the flow system in the tertiary sediments was confirmed. Additionally, the complexity of
the system was demonstrated [2]. Differences in stable isotope signatures in the northern part of the
site indicate very local connections of the flow systems with the flow system in the underlying
granite via fault zones.
The chemical conditions of the site are characterised by low mineralised waters with ionic strengths
in the range from 0.003 to 0.02 mol/l. The pH-values vary in a range of 6.2 to 8, the Eh-values from
435 mV to -280 mV. More oxidising conditions with lower pH-values are found in the near-surface
granite waters of the infiltration area. In the clay/lignite horizon the conditions are strongly reducing
with Eh-values as low as -280 mV. The Eh-values measured directly by in-situ probe are significantly
lower than those measured on-site in pumped water. The latter method is more susceptible
to disturbances by the contact with atmosphere, which is probably responsible for these differences.
In order to understand the carbon chemistry and the interrelation between SOC and DOC iso-
348

topes of carbon in DIC, DOC and SOC were evaluated. This work is described in detail in [2]. Here
only few major aspects concerning the formation of DOC in the clay/lignite layer are touched.
In Figure 5 on the left the correlation between biogenic DIC and DOC is plotted, assuming that the
13 C content of the source water is of inorganic origin with -27 ‰ and additional DIC is of organic
origin. Besides data for NA8, which well is probably not strongly connected to the clay/lignite water,
the other data follow a line, with increase of biogenic DIC from infiltration waters NA10 and
RP1 to the waters from the clay/lignite layer. It is one important indication that in this layer DOC is
formed by additional release of DIC, as it is observed at other sites, e.g. 11].
Concerning the mechanism of DOC release important information can be gained from other isotope
signatures. A strong indicator for microbial degradation of organic matter by sulphate reduction is
the � 34 S signal in dissolved sulphates. Since sulphate reducing bacteria had already been detected at
the site, appropriate analyses were performed in groundwater from the infiltration area and from the
clay/lignite layers.
The results show that � 34 S-values in waters from the clay-lignite layer are in a range between
16.4‰ and 24.6 ‰, i.e. strongly increased compared to the values between -8.5 and 3.48 observed
in the wells from the infiltration area (see Figure 5, right). The substantial enrichment of 34 S in
these boreholes is a clear indication that microbial sulphate reduction occurs in the clay/lignite layers.
The microbial sulphate reduction is accompanied by isotope fractionation. The lighter isotope
32 S is preferentially metabolised by the microbes leaving residual sulphate in the solution enriched
in 34 S, which is observed here. With the initial sulphate concentration in the infiltration area and the
concentration after sulphate reduction in the clay lignite an enrichment factor can be calculated by
use of the Rayleigh equation. Based on the data from all wells shown in Figure 5 and the corresponding
sulphate concentrations an enrichment factor of 11‰ was calculated [2]. Compared to
other investigations this value is comparably low, but not unusual for bacterial sulphate reduction.
Enrichment factors in a similar range between of ~10 ‰ have been observed e.g. in [12].
DOC [mg/l]
5
4
3
2
1
NA8
NA10
RP1
NA6
NA4
RP2
NA13
NA7
NA12
0
5 15 25 35 45
DIC biogenic [mg/l]
Figure 5: Correlation of biogenic DIC with DOC (left) and � 34 S values in boreholes from the infiltration
area (red circle) and clay/lignite layers(yellow circle) [2]
In order to better understand the interrelation between SOC and DOC, in particular to understand
the relatively low DOC concentrations below 5 mg C/l in the water from clay/lignite layer com-
349
0.2
-8.5
3.48
20.11
23.5
20.4
16.43
24.63

pared to other sites with SOC-bearing sediments [11], a more detailed analysis of SOC from selected
samples of clay/lignite layers have been performed. This work represents a strong link between
RTDC5 and RTDC2. SOC was characterised in detail by
micropetrographical methods,
application of different extraction schemes,
degradability experiments, and
interaction experiments of humic acids with natural clay samples.
The micropetrographical study showed that the sedimentary organic matter at Ruprechtov site is
generally formed by slightly dispersed matter with low degree of coalification, which only reaches
brown coal or lignite degree. The main components of detritic and xylodetritic coal samples and
clay-lignite samples are mineral admixtures and huminite of the maceral group [5].
According to the results from SOC characterisation it seems that the low concentration of dissolved
organic matter in the Ruprechtov system is mainly caused by the low availability of organic matter
to the processes of degradation. Only a very small fraction of SOC is accessible to the groundwater.
An additional reason could be the strong sorption properties of the clay that fix humic acids on the
sediment matrix. This is indicated in first sorption experiments performed by NRI with standard HA
leonardite on the montmorillonite standard SWy-2 and on low TOC clay samples from borehole
Na11. The results are shown in Figure 6 and indicate significant sorption of HA on the clay samples
with higher sorption values on Ruprechtov samples compared to standard montmorillonite [5].
Adsorbed HA (mg per kg of clay)
8000
7000
6000
5000
4000
3000
2000
1000
0
-1000
-2000
0 50 100 150 200 250
Equilibrium concentration of HA (mg/L)
Figure 6: Sorption isotherm of HA on different clay samples
350
NA11 (465nm)
SWy-2 (465nm)
NA11 (280nm)
The integration of all results showed that organic matter did not play such an important role by direct
interaction with uranium, but SOC contributed and still contributes to maintain reducing conditions
in the clay/lignite layers. It can be concluded that SOC within the sedimentary layers was (and
to some extent still is) microbially degraded. By this process DOC is released, providing protons to
additionally dissolve SIC [2]. Moreover SO4 2- is reduced leading (and has lead in the geological
past) to the formation of iron sulphides, especially pyrite. Reducing conditions, being maintained
amongst others by sulphate reducing bacteria, caused the reduction of As, which sorbed onto pyrite
surfaces, forming thin layers of arsenopyrite. Uranium U(VI), originally being released from the
outcropping/underlying granite, was reduced to U(IV) on the arsenopyrite surfaces. UO2 and uranium
phosphates were formed by reaction of U(IV) with phosphates PO4 3- , released by microbial

SOC degradation. These uranium(IV) minerals have been stable and immobile over geological time
frames.
Geochemical calculations with GWB [14] and revised NEA TDB [15] confirmed that U(IV) is the
preferential oxidation state in the clay/lignite layers. These calculations also indicate that the redox
conditions in the clay/lignite horizon are controlled by the SO4 2- /S 2- couple. Uranium concentrations
in the clay/lignite layer observed today are determined by amorphous UO2 and ningyoite. One important
issue which could not be clarified is, whether the more accessible U fraction found by
U(IV)/U(VI) separation really exists in the hexavalent state or has been oxidised after sampling.
4. Conclusions
This natural analogue study contributed to the Safety Case for the far-field transport in sedimentary
layers in different ways. A very important part was the aspect of method development and testing,
e.g. colloid sampling under undisturbed conditions, first application to natural samples and further
development of μ-XRF and μ-XANES as well as application of modern isotope analyses like � 34 S
signatures to identify relevant processes in the field. All these methods are important for characterisation
of a potential repository site including lab and field experiments.
Although Ruprechtov site turned out to be rather complex with regard to hydrogeology and geological
evolution, the mechanisms for immobilisation of uranium have been identified. By application
of a set of different microscopic and macroscopic analytical methods distinct immobile uranium
phases have been characterised and their long-term stability was shown. The results further
indicate that DOC does not contribute to mobilisation of U because of the relatively low DOC concentration
in the clay/lignite layer. DOC is formed by microbial degradation of SOC in the
clay/lignite layers but only a very small fraction of SOC seems to be accessible.
In general it was shown that sedimentary layers can provide a strong barrier function for uranium,
when specific prerequisites are fulfilled. Under the strongly reducing conditions in the clay/lignite
layers at Ruprechtov site, there are no indications for significant uranium release during the last
million years. The low uranium concentrations in the groundwater of app. 10 -9 mol/l are determined
by amorphous UO2 and ningyoite.
5. Acknowledgements
This project has been co-funded by the European Commission and performed as part of the sixth
Euratom Framework Programme for nuclear research and training activities (2002-2006) under contract
FI6W-CT-2004-516514, by the German Federal Ministry of Economics and Technology
(BMWi) under contract no 02E9995, and by RAWRA and Czech Ministry of Trade and Industry
(Pokrok 1H-PK25).
References
[1] Noseck, U., Brasser, Th., Rajlich, P., Laciok, A., Hercik, M., (2004). Mobility of uranium in
Tertiary argillaceous sediments - a natural analogue study. Radiochim. Acta 92, 797-803.
[2] Noseck, U., Rozanski, K., Dulinski, M., Havlova, V., Sracek, O., Brasser, Th., Hercik, M.,
Buckau, G., (2008). Characterisation of hydrogeology and carbon chemistry by use of natural
isotopes – Ruprechtov site, Czech Republic. Appl. Geochem. (in prep.).
351

[3] Hauser, W. Geckeis, H., Götz, R., Noseck, U., Laciok, A. (2005). Colloid Detection in Natural
Ground Water from Ruprechtov by Laser-Induced Breakdown Detection.
[4] Noseck, U., Brasser, Th., Suksi, J., Havlova, V., Hercik, M., Denecke, M.A., Förster, H.J.,
(2008). Identification of uranium enrichment scenarios by multi-method characterisation of
immobile uranium phases. J. Phys. Chem. Earth, doi:10.1016/j.pce.2008.05.018.
[5] Cervinka, R., Havlova, V.; Noseck, U.; Brasser, Th.; Stamberg, K., (2007). Characterisation
of organic matter and natural humic substances extracted from real clay environment. Annual
Workshop Proceedings of the IP Project FUNMIG”. Edinburgh 26.-29.November 2007.
[6] Denecke, M.A., Janssens, K., Proost, K., Rothe, J., Noseck, U., (2005). Confocal micro-XRF
and micro-XAFS studies of uranium speciation in a Tertiary sediment from a waste disposal
natural analogue site. Environ. Sci. Technol. 39(7), 2049-2058.
[7] Denecke, M.A., Somogyi, A., Janssens, K., Simon, R., Dardenne, K., Noseck, U., (2007). Microanalysis
(micro-XRF, micro-XANES and micro-XRD) of a Tertiary sediment using synchrotron
radiation. Microscopy Microanal. 13(3), 165-172.
[8] Ervanne, H., Suksi, J., (1996). Comparison of Ion-Exchange and Coprecipitation Methods in
Determining Uranium Oxidation States in Solid Phases. Radiochemistry 38, 324-327.
[9] Havlová, V., Laciok, A.; Vopálka, D.; Andrlík, M. (2006): Geochemical Study of Uranium
Mobility in Tertiary Argillaceous System at Ruprechtov Site, Czech Republic. Czechoslovak
Journal of Physics, Vol. 56, Suppl. D, 1-6.
[10] Noseck, U., Brasser, Th., 2006. Radionuclide transport and retention in natural rock formations
– Ruprechtov site. Gesellschaft für Anlagen- und Reaktorsicherheit, GRS-218, Köln.
[11] Buckau, G., Artinger, R., Geyer, S., Wolf, M., Fritz, P., Kim, J.I., (2000). Groundwater in-situ
generation of aquatic humic and fulvic acids and the mineralization of sedimentary organic
carbon. Appl. Geochem. 15, 819-832.
[12] Knöller, K., Fauville, A., Mayer, B., Strauch, G., Friese, K., Veizer, J., (2004). Sulfur cycling
in an acid mining lake and its vicinity in Lusatia, Germany. Chem. Geol. 204, 303-323.
[13] Vopalka, D., Havlova, V., Andrlik, M. (2008): Characterization of U(VI) behaviour in the
Ruprechtov site (CZ). Proc. of the International Conference Uranium Mining and Hydrogeology
V, Freiberg, September, 14.-18. 2008.
[14] Bethke, C.M., (2006): The Geochemist’s Workbench Release 6.0. Hydrogeology Program,
University of Illinois.
[15] Yoshida, Y., Shibata M., (2004): Establishment of Data Base Files of Thermodynamic Data
developed by OECD/NEA Part II - Thermodynamic data of Tc, U, Np, Pu and Am with auxiliary
species. JNC Technical Report, JNC TN8400 2004-025.
352

Radionuclide Migration in the Far-Field:
The Use of Research Results in Safety Cases
Bernhard Schwyn 1 , Jürg Schneider 1 , Jörg Rüedi 1 , Jesus Alonso 2 , Scott Altmann 3 ,
Stephane Brassinnes 4 , José Luis Cormenzana López 2 , Aimo Hautojärvi 5 , Jan Marivoet 6 , Ignasi
Puigdomenech 7 , André Ruebel 8 , Cherry Tweed 9 , Tiziana Missana 10 , Ulrich Noseck 8 , Pascal
Reiller 11 , Thorsten Schäfer 12
Summary
1 NAGRA, Switzerland, 2 ENRESA, Spain 3 ANDRA, France
4 ONDRAF/NIRAS, Belgium, 5 POSIVA, Finland, 6 SCK•CEN, Belgium
7 SKB, Sweden, 8 GRS, Germany, 9 NDA, UK
10 CIEMAT, Spain, 11 CEA, France, 12 FZK, Germany
In the Integrated Project FUNMIG fundamental processes governing radionuclide migration
in the geosphere were investigated. Within the RTD Component 6 the involved Waste Management
Organisations (WMOs) and an Integration Monitoring Group coordinated the linking
of the scientific work performed in the Components 1 to 5 to its use in performance assessment
and monitored the improvements achieved within this project for the safety cases of the
individual radioactive waste disposal programmes in Europe. Boundary conditions for the
near-field of a radioactive waste repository, in particular radionuclide fluxes and concentrations
but also repository induced perturbations of the geosphere, were defined based on the
WMOs’ safety cases and on the outcome of the European Project NF-PRO. A comprehensive
catalogue providing concise information for each task performed in FUNMIG was compiled.
It facilitates the retrieval of knowledge acquired within FUNMIG and was used to map these
tasks to internationally accepted lists of Features, Events and Processes (FEPs) for clay-rich,
crystalline and salt host rocks. The subsequent evaluation revealed issues important for future
safety cases, namely the mechanistic understanding, quantification and up-scaling of radionuclide
diffusion in clay-rich host rocks and, for crystalline host rocks, the influence of colloids
on radionuclide transport and the up-scaling of reactive transport modelling in fractured systems.
Finally, the scientific outcome of FUNMIG was synthesised for each of the three host
rocks by collecting the individual tasks in Super-FEPs. The transferability of knowledge between
the host rocks is limited due to the host rock and site specificity of the research performed
in FUNMIG.
1. Introduction
The scope of the Integrated Project FUNMIG was to improve and supplement the scientific knowledge
on radionuclide migration in host rocks for future safety cases for the various European radioactive
waste disposal programmes. In the compilation of a safety case several groups are involved.
E.g., in developing the safety case for the Project Opalinus Clay [1] four groups, each with welldefined
roles, were distinguished:
The management group, responsible for overall management of repository planning.
353

The science and technology group, responsible for the scientific and technical basis of the
safety case.
The performance assessment group, responsible for the development of a system concept, the
safety concept and the approach to safety assessment, as well as the carrying out of the safety
assessment and the compilation of the safety case.
The bias audit group, responsible for ensuring that the scientific basis for safety assessment is
complete, fully documented and exploited in an unbiased manner in safety assessment.
Interaction between these groups must take place, to keep work focused on project goals, to ensure
that the knowledge and expertise of available personnel are fully exploited, and to ensure that all
personnel understand and support the safety case arguments. At the same time, a degree of independence
must be maintained, to avoid unduly biasing the work of one group by the expectations of
another. The science and technology group, for example, should be aware of the possibility of new
phenomena, even though these may not fit conveniently into existing methods or models used by
the safety assessment group.
Because of the scope of FUNMIG the present paper focuses on the interactions between “Science”
(which can be taken to be a sub-group of the science and technology group) and “PA” (which is
identical to the performance assessment group).
The scientific work of FUNMIG was divided into five components: RTDCs 1 and 2 activities were
focused on conceptually well defined (RTDC 1) and conceptually less established processes (RTDC
2), not oriented towards a specific host-rock type or disposal concept. RTDCs 3 - 5 were focussed
on specific processes of relevance in clay-rich, crystalline and salt host-rock type. In the latter case
this refers to the overburden far-field beyond the salt rock formation.
The task of RTDC 6 was to accomplish the above mentioned interaction between science (RTDCs
1-5) and PA, in particular to monitor the scientific outcome of FUNMIG with respect to its use in
future safety cases and to give PA’s feed back to the researchers. “PA” was represented by the following
European Waste Management Organisations (WMOs):
ANDRA, France (clay-rich host rocks); ENRESA, Spain (clay-rich and crystalline host rocks);
GRS, Germany (salt host rock); NDA, UK (clay-rich and crystalline host rocks); NAGRA, Switzerland
(clay-rich host rocks); ONDRAF/NIRAS and SCK•CEN, Belgium (clay-rich host rocks);
POSIVA, Finland (crystalline host rocks); and SKB, Sweden (crystalline host rocks).
The corresponding disposal concepts were presented during the first Topical Session of FUNMIG
[2].
2. Methodology
2.1 Boundary conditions
In an early stage of the project, boundary conditions to the near-field were documented for various
European disposal concepts, namely for Spain, Switzerland, UK, France and Belgium as a prerequisite
to have information on what radionuclides with which fluxes and in which concentrations may
be released from the near-field of a repository into the surrounding host rock. The results reported
in [3] identified the key radionuclides leaving the near field and confirmed the validity of the usual
354

assumption that the radionuclides from the waste enter the far field groundwater in trace amounts
only.
NF-PRO, a sister project within the sixth Framework Programme, was finished in 2007. It dealt
with near-field processes in the European disposal concepts. Results of NF-PRO concerning the interface
of the near field with the geosphere and therefore relevant to FUNMIG were extracted from
the final NF-PRO reports. Hence, not only radionuclide fluxes, but generally potential influences of
the near-field on the geosphere are discussed [3]. Potential disturbances include gas generated by
corrosion processes, transport of corrosion products and bentonite colloids and intrusion of oxygen,
organics and nitrate from the near-field into the geosphere.
2.2 Evaluation of FUNMIG tasks
To organise the interaction between scientists and performance assessors (WMOs) an Integration
Monitoring Group (IMG) was established, composed of one representative for each RTDC. In their
role to coordinate the linking of scientific results to their application in future safety cases performance
assessors and IMG developed a tool to evaluate FUNMIG results in this respect:
A thorough catalogue was compiled by the scientists consisting of Task Abstract Forms, i.e., concise
information on one page for each task performed in FUNMIG. Based on this information the
tasks were mapped on internationally accepted views for clay-rich and crystalline host rocks. FEP-
CAT [4] is a catalogue of Features, Events and Processes (FEPs) for argillaceous rocks. RETROCK
[5] was a European project on the treatment of radionuclide migration in fractured rocks within
safety assessment. A list of processes identified by the participants as safety relevant was used as a
quasi FEP catalogue. For salt host rocks the FEP list recently developed within the German research
project ISIBEL was used.
For this FEP mapping the structure of the project was adopted. Three Task Evaluation Tables
(TETs) were prepared according to the three host rocks addressed, namely clay-rich rocks in RTDC
3, crystalline rocks in RTDC 4 and salt rocks in RTDC 5. The work of the two non host rock specific
components RTDC 1 and RTDC 2 was included in each of the three host rock specific TETs.
Figure 1 (upper part) depicts the structure of the TETs. Based on the information in the above mentioned
Task Abstract Forms the individual tasks were evaluated by the WMOs considering the importance
of the investigated process for the different safety cases on one hand, and considering the
progress made within FUNMIG. Thereto, the WMOs used the experience with their safety cases
and connected suitable sensitivity analyses. As a counterpart in the interaction between the performance
assessors and the scientists, the latter were asked to appraise the benefit of their work for future
safety cases.
355

FEPCAT
RTDC 1 RTDC 2
RTDC 3
Clay-rich rocks
Clay-rich rocks
Super-FEPs
Figure 1: Structure of task mapping and evaluation
The host rock specific Task Evaluation Tables contain the task titles, the FEP numbers the tasks are
mapped to, affected safety assessment parameters and, most importantly, the standardised evaluation
information consisting of a researcher’s view and a WMO’s view. The TETs are included in
the final RTDC 6 report on “FUNMIG topics and processes and their treatment in the safety case”
[6].
Having completed the the TETs the synthesis depicted in Figure 1 was carried out by lumping together
the individual FEPs (to which FUNMIG tasks were mapped) to “Super-FEPs” such as
“Transport mechanisms” or “Retardation”.
3. Results
3.1 Evaluation of individual Tasks
RETROCK Salt-specific
FEP list
RTDC 1 RTDC 2
RTDC 4
Crystalline rocks
Crystalline rocks
Super-FEPs
Mutually applicable knowledge, mutual benefits
Because the outcome of FUNMIG was evaluated on a task level the reporting of the comprehensive
results is beyond the scope of the present paper due to the extensiveness; in this regard we refer to
the Task Evaluation Tables in the final RTDC 6 report on “FUNMIG topics and processes and their
treatment in the safety case” [6]. Although the programme and site specificity of the evaluation results
is evident, there is a considerable agreement between the performance assessors within a host
rock type. Topics, judged by the performance assessors to be particularly important, are listed below.
For clay-rich host rocks topics of high relevance for future safety cases with important progress
within FUNMIG are
The understanding of diffusive transport in clays, namely
o Differences between anions, cations and neutral species,
356
RTDC 1 RTDC 2
RTDC 5
Salt rocks
Salt rocks
Super-FEPs
Sensitivity analysis Sensitivity analysis Sensitivity analysis

o Strongly sorbing radionuclides incl. compatibility of diffusivities with batch sorption
measurements,
o Evaluation of field diffusion experiments,
The structure and surfaces of compacted clays,
The evaluation of natural tracer measurement and up-scaling,
The influence of carbonate and natural organic matter on the sorption of safety relevant radionuclides,
incl. actinides, on clay minerals,
The retention behaviour of the relatively mobile and therefore often dose relevant Selenium,
namely its interaction with pyrite and the sorption on clay minerals incl. the dependence on redox
conditions, compaction degree of sorbing clay and the presence of natural organic matter.
Correspondingly, particularly important topics for crystalline host rocks are
Colloids, incl. bentonite particles from the backfill and humics, and their interaction with radionuclides,
Modelling the up-scaling of reactive transport in heterogeneous fractured systems.
Since hermetical enclosure properties are attributed to the rock salt formation the overburden, to
which work within FUNMIG is focussed on, is considered as supplemental barrier. Migration processes
are therefore of lower relevance.
3.2 Super-FEPS
For clay-rich host rocks the individual FEPs, to which FUNMIG tasks were mapped, were collected
in Super-FEPs. The translation into the FEPCAT terminology showed that virtually all tasks originating
from RTDC 1, 2 and 3 could be mapped onto the FEP group concerning transport mechanisms
(FEPCAT no. A1) and the FEP group concerning retardation mechanisms (FEPCAT no. A2).
Figure 2 depicts the two FEP groups or Super-FEPs . Concerning diffusion related FEPs the two
groups overlap; the group “retardation mechanisms” includes in addition to diffusion related FEPs
also sorption related FEPs according to the broad FEPCAT definition which comprises sorption and
dissolution / precipitation of solids and solid solutions. A few FUNMIG tasks were mapped to FEPs
outside the above mentioned Super-FEPS (cf. Figure 2).
The corresponding parameters typically used in safety assessment are sorption coefficient, solubility,
effective diffusion coefficient and accessible porosity as depicted in Figure 2.
The synthesis is carried out for all three host rock types and is complemented and illustrated by sensitivity
analyses carried out in context with recent safety cases for various European disposal programmes
[6].
4. Discussion
As stated in the introduction of the present paper an interaction between the involved groups is essential
to develop a safety case successfully. In FUNMIG the dialogue between the “supplier
group”, the scientists, and the “customer group”, the performance assessors, was required. In establishing
a co-work between representatives of European Waste Management Organisations as performance
assessors and the Integration Monitoring group (IMG) launched for this purpose, this
challenge was successfully met. The IMG members as representatives of the scientists communicated
the outcome of the research performed within FUNMIG to the performance assessors on one
hand and transmitted the PA feedback to the scientists on the other hand.
357

A1: Transport mechanisms
Diffusivity
Connected matrix porosity
Ion exclusion
Surface diffusion
SA Parameter
Kd
Csol De A3.2/C1.1.2: Evolution of pore-fluid (water and gas) chemistry and
mineralogy in the host rock formation and in embedding units
Sorption coefficient
Maximum solubility
Effective diffusion coefficient
Accessible porosity
B1.2.2: Organics from waste and their effect on transport
properties of the host rock
A2: Retardation
Sorption (broad definition)
Figure 2: FEP groups and single FEPs (taken from FEPCAT) to which the FUNMIG tasks from
RTDC 1, 2 and 3 were mapped, and affected safety assessment parameters for the clay-rich host
rock case.
A very valuable result of this co-work is the comprehensive catalogue of the tasks performed within
FUNMIG providing concise information on aim, type of work, involved research groups, results
(abstracts) and bibliography for each task. It will be attached to the final RTDC 6 report on “FUN-
MIG topics and processes and their treatment in the safety case” [6] and will facilitate the retrieval
of information acquired within FUNMIG and relevant for future safety cases.
In developing the Task Evaluation Tables (TETs) a tool was made available for the information exchange
between the performance assessors and the scientists on the level of the single tasks. The
TETs allowed the performance assessors to address their feedback directly to the scientists concerned
(WMO’s view) and the scientists to appraise their work with respect to achievements for
future safety cases (researcher’s view). The results of the task evaluation by the performance assessors
(WMOs) indicate that the scope of investigated processes in FUNMIG was quite well set and
that substantial progress was made in research fields important for the development of future safety
cases for the different radioactive waste disposal programmes.
To optimize the research programme it would have been desirable to have the relevance of the chosen
research topics for future safety cases available at the beginning of the project. In this respect a
TET can be a helpful tool for planning future Integrated Projects.
The synthesis by collecting the FEPs in Super-FEPs did not reveal ground-breaking findings; it
rather confirmed the WMO’s experience with FEPs related to the barrier function of the host rock
for the radionuclides.
358
B7: Microbiological perturbations
Dissolution / precipitation of solid phases
Solid solutions / co-precipitation
Ion exchange
Surface complexation

Also confirmed was the experience of some of the WMOs that the transferability of information
between host rock types (cf. lower part of figure 1) is restricted to generic research topics. During
the development of a disposal concept the needed information is turning successively more site specific.
Indeed, the majority of the topics dealt with in FUNMIG was host rock oriented.
5. Conclusions
To ensure research oriented towards its use in the safety case an interaction of the researchers and
the performance assessment groups is essential. In FUNMIG this interaction was achieved by the
co-work between the Integration Monitoring Group representing the scientists and the representatives
of the Waste Management Organisations representing the performance assessor groups.
The assignment of FUNMIG research tasks to FEPs proved to be a powerful procedure to evaluate
the relevance of a research topic and the results achieved with respect to their use in a safety case.
This procedure may be used in the planning of projects to optimise research programmes.
The catalogue, compiled as a basis for the above mentioned evaluation procedure with concise information
on each FUNMIG task and results, will in addition facilitate the retrieval of information
acquired within FUNMIG for future safety cases.
6. Acknowledgements
The present paper is based on the outcome of a co-work between the scientists, the Integration
Monitoring Group and the representatives of the Waste Management Organisations participating in
FUNMIG. The Integrated Project FUNMIG was co-financed by the European Commission and
performed as part of the sixth Euratom Framework Programme for nuclear research and training
activities (2002-2006) under contract FI6W-CT-2004-516514.
References
[1] National Cooperative for the Disposal of Radioactive Waste: Project Opalinus Clay: Safety
Report: Demonstration of disposal feasibility for spent fuel, vitrified high-level waste and
long-lived intermediate-level waste (Entsorgungsnachweis). Nagra Technical Report 02-05.
Nagra, Wettingen, 2002.
[2] 1 st Annual workshop proceedings of integrated project fundamental processes of radionuclide
migration, IP FUNMIG, CEA report R-6122, Saclay, France
[3] Boundary Conditions to the near-field, ENRESA Report (in preparation)
[4] Mazurek, M., Pearson, F.J., Volkaert, G. & Bock, H. (2003): Features, events and processes
evaluation catalogue for argillaceous media. OECD / NEA, Paris, 2003.
[5] Treatment of radionuclide transport in geosphere wihin safety assessments (Retrock). Final
Report, EUR 21230 EN. European Commission, Community Research, Brussels, 2005.
[6] Fundamental Processes of Radionuclide Migration: Topics and processes dealt with in the IP
FUNMIG and their treatment in the Safety Case of geologic repositories for radioactive waste.
Nagra Technical Report 09-01. Nagra, Wettingen, (in preparation).
359

360

Summary of the Panel Discussion on the Topic:
"Consensus Views on Key Remaining Issues in Far-Field Processes"
Panel members:
Jörg Hadermann (Chair), PSI, Switzerland
Gunnar Buckau, FZK-INE, Germany
Pierre Toulhoat, University of Lyon, CNRS Lyon, and INERIS, France
Scott Altmann, ANDRA, France
The objective of the panel discussion was to reach consensus views on the key remaining issues in
the area of far-field processes that are of importance for the geological safety case and could be
used as a contribution to identifying research topics for FP7. This goal was reached as far as can be
expected in a 50 minutes panel discussion.
The questions put forward for discussion were:
� What were the major scientific highlights for far-field considerations in the safety case in the
last five years?
� Relevant issues tend to be the same over many years. Where is science expected to make the
largest progress and impact in the next five years?
� Transferability between host rocks is limited. Will commonalities further decrease, and where
would be fruitful interactions?
� Site characterisation is an important step in all projects. What of general scientific interest, in
terms of tools and data, do we expect from site investigations?
In a short introductory remark, the chairman reminded the audience that the remaining issues are
not putting into doubt the safety of repositories, be they in clay, in crystalline or in salt. The main
work in the future will be in the fields of site characterisation and monitoring of the far-field. But
scientific investigations will continue, as is the case in every ripe field. The goal of such investigations
is to enhance the understanding of far-field processes. As an example the competition of species
in the process of diffusion through clays was mentioned. The investigations will help to reduce
uncertainties, contribute to optimisation, and eventually help reducing costs.
The panellists then presented their statements before the word was given to the floor.
Gunnar Buckau emphasized that the remaining issues are no showstoppers but rather objectives,
strategies and issues that deserve attention. He distinguished between different characters of issues:
Those responding to problems emerging from specific safety cases and those related to confidence
building and underpinning the safety statements with basic process understanding. Both, he saw in
the frame of implementation oriented research. As a consequence he called for improving general
basic data and knowledge (which he considered less interesting for advanced programmes), and for
361

maintaining and develop competence networks (which he considered of interest for all and being a
consequence of advanced programs).
As an example of on-going FP7, implementation oriented research he mentioned the Collaborative
Project "Redox Phenomena Controlling Systems" (ReCosy). He specified the various issues dealt
within this project, and especially mentioned methods of redox determination, redox responses of
repository systems upon environmental changes, and redox reactions of selected elements such as
actinides, technetium iodine and selenium. Other issues mentioned were a better understanding of
sorption on crystalline rock, sorption on the overburden of salt host rocks, and especially the interpretation
of data resulting from existing projects.
Pierre Toulhoat put his emphasis on geochemistry of elements and species relevant for disposal in
the geosphere. He advocated more use of specific underground laboratories to investigate geochemical
processes, especially for the in situ validation of speciation predictions. He saw the need
to improve our understanding on conditions when thermodynamic equilibrium dominates, the possibility
of the continuation on the long term of non-equilibrium situations and more knowledge on
competition between various reactions. On the side of direct and obvious impact in the safety case
he mentioned investigations on isotopic dilution and isotopic exchange.
Scott Altmann directly and in detail addressed the questions listed above. He saw the major scientific
progress in our present understanding of pore water solutions in clays. He also mentioned the
progress in sorption modelling and especially the use of thermodynamic models putting on a firmer
basis the retardation considerations within the safety case. However, for some safety relevant elements
such as iodide and selenite the understanding of retardation processes need improvement. In
line with such requirements he advocated a better understanding of mineral reactions and the radionuclide
interaction in this context. When scientific fields develop they start to differentiate. In
the context of this general fact, he did essentially not see much commonality between host rocks
and very fruitful interactions – limited to the far-field, of course - between repository projects in
different host rocks.
After these statements by the panellists the floor was open for questions and comments by the participants.
Jordi Bruno emphasised the increased credibility of safety cases resulting from such large projects
as NF-PRO and FUNMIG. In the future he saw more utilisation of micro-approaches in experimental
investigations.
Gérald Ouzounian did not fully agree and reminded the audience that developing new analysis
methods is not the main issue in the safety case.
From the floor came the question where one could find guidance on what to study. Scott Altmann
recommended looking into the literature and especially existing performance assessments.
Jon Harrington asked for clarification on the importance of coupling processes. Scott Altmann
responded that the strength is dependent on the radionuclide in question. In performance assessment
this coupling is less of importance. However, the sensitivity analysis is all-important. This is
also a result from the variability of the far-field properties. It is hard to correlate variability of the
various properties; and after all, the question is which variability the nuclides are seeing.
Wernt Brewitz reminded the audience that each safety case is unique and, as a matter of fact, cannot
be verified by experimentation.
362

From the panel discussion several possible issues for future research emerged:
- General issue: The host rock specificity must be taken into account in the definition of projects.
Commonalities are negligible.
1- Sorption on crystalline rock and, although less important, on overburden of salt formations,
2- In depth interpretation of existing data, above all from FUNMIG,
3- Geochemical in-situ experiments, micro analytics,
4- Isotopic dilution and exchange,
5- Competition of diffusing species,
6- Retardation processes for iodine and selenium,
7- Mineral reactions and nuclides.
Below is an elaboration on topic 1, 2, 3, 4, 6 and 7. Topic 2 intersects with several of the other
above listed topics and is also relevant to the outcome of the other FP6 IP's.
1. Sorption of radionuclides in crystalline rock
The transport of radionuclides with non-negligible sorption in the far-field of crystalline rock is
strongly dependent on the magnitude of process causing the retention. The present situation is that
sorption data (Kd values) scatter over several orders of magnitude. This broad range of sorption
data reflect on one hand operational differences from underlying experiments, and on the other hand
differences in the composition of the mobile phase and the composition and properties of the stationary
phase. Attempt to relate the differences in sorption values to such variations in the system
properties, however, have not yet given acceptable results. As a consequence, evaluation strategies
are applied in order to select data.
The data selection strategies chosen vary, depending on the objectives, such as selecting values that
are sufficiently conservative for a specific purpose, between the probably most representative ones
and the most representative ones of the set of data available. This may be a necessary intermediate
step in order to proceed with development of the Safety Assessment as part of the disposal Safety
Case (SC). Moving towards a more advanced stage of the disposal SC, however, data should be
used providing for a justified case. Such a justified case needs to build on demonstration of an acceptable
level of process understanding in order to ensure that the data used (i) are void of unacceptable
experimental artefacts, (ii) actually represent the assumed system, and (iii) can be used for
reliable predictions for up-scaling with respect to time and size. The latter is especially important
because the SC builds on a system evolution with changing conditions. Any prediction into the future
will rest on prediction also of the retention behaviour under the different conditions that will be
found in different repository evolution phases. Consequently, even a very elaborate determination
of the best values for the present conditions will not solve the problem of future system evolution.
The salt-dome overburden may also be investigated, however, with less emphasis than that on the
crystalline far-field. In the past, a very broad set of sorption data has been generated for salt-dome
overburden. The problem with these data is comparable with that of crystalline rock, namely difficulties
with justified and trustworthy evaluation, extrapolation and application of the data. Studying
few such systems will also support development, testing and comparison of analytical methods and
evaluation approaches.
363

A project would be justified to determine the underlying sorption processes. It would reflect that
national programmes are reaching a state of development where coming closer to implementation
requires generation of a more acceptable base for sorption assessment. Such a project would build
on the very broad existing competence. It would also benefit from the use of advanced analytical
techniques, especially recent developments where the spatial resolution is rapidly improving for
determination of elemental composition and chemical states on different solids/minerals involved.
Finally, the access to very elaborate site-specific data available through recent site characterization
programmes provides for direct natural analogue / real system comparison and verification.
2. In-depth interpretation and integration of existing data from the large FP6 Disposal
IP’s in view of the application to the Safety Case.
FP6 disposal investigations build around the waste emplacement technology, near-field and farfield
processes, and Performance Assessment (PA)/ the Safety Case. These IP’s complementary to
each other are ESDRED, NF-PRO, FUNMIG and PAMINA. Each of these Integrated Projects
deals with SC Topics, either with reference to the technology or science areas dealt with or certain
PA/SC aspects. The outcome is implemented in present advanced national programmes. An added
value may be expected from synthesizing the outcome of these four IP’s with respect to providing
(i) an overall view of the outcome in terms of the advances in the state of knowledge for European
national programmes at their different levels of development, (ii) an evaluation and prioritisation of
the open issues identified in the list of suggestions for further work and (iii) conclusions and recommendations
for future Euratom disposal programme activities. The latter two aspects are important
elements also in building a disposal technology platform, presently under development.
3. Geochemical in-situ experiments, micro analytics
In order to increase confidence in the performance assessment of nuclear waste disposal, verification
of the speciation of radionuclides through in situ measurements is an important issue. Indeed,
the computation of the migration of radionuclides in the near-field and far-field rely on thermodynamic
databases, based on compilations and surface laboratory experiments. Direct or indirect evidence
of in situ speciation have seldom been gathered. Taking advantage of existing underground
laboratory facilities, it is a real challenge to be able to confront measurements and predictions. Recent
progress in analytical chemistry (spectroscopy, electrochemistry) renders this goal achievable.
4. Isotopic dilution and exchange
For mobile fission products, which mostly contribute to the impact of nuclear waste disposal, it is
relevant to increase the safety margins by looking at processes that should be able to induce a significant
retardation effect. Isotopic dilution and exchange are possible processes and need to be
confirmed. This is especially true for 14 C, 36 Cl and 129 I. Theoretical studies, together with laboratory
studies and field studies are necessary to document such processes.
Retardation processes for mobile fission and activation products ( 79 Se, 129 I, 14 C…) including
mineral reactions and nuclides
The principal radionuclides migrating through the geological barrier in recent Safety Cases (SC)
treating disposal concepts in clayrock formations are 129 I, 36 Cl, 79 Se, 41 Ca (Andra, 2005), 129 I, 36 Cl,
79 Se, 14 C organic (Nagra, 2002) and 129 I, 79 Se, 99 Tc (NIROND, 2001). In all cases except 41 Ca, this
is a result of attributing them retardation values of 1 (Kd = 0). On the other hand, for certain of
these radionuclides ( 79 Se, 129 I and 14 C organic) it is possible to hypothesize mechanisms which
364

might result in a slight, but potentially significant, retardation. Certain of these were brought out
during the International Workshop on Fission and Activation Products in Nuclear Waste Disposal
(16-19/1/2007, La Baule, FR) among which were Se immobilization by ferrous iron sorbed on clay,
Se(II) sorption onto pyrite, exchange of 129 I with natural iodine pool… While existing SC show
that the corresponding radioactive waste disposal concepts remain well within the required safety
limits, additional safety margins, and increased confidence in the safety demonstration would be
gained should research be able to show the capacity of such processes to retard migration of these
radionuclides.
365

366

SESSION VIII: Performance assessment studies – coordination of RD&D for waste disposal
Chairman: Dr Alan J Hooper, NDA, United Kingdom
367

368

Performance Assessment Methodologies in Application to Guide
the Development of the Safety Case
Summary
Jörg Mönig
Gesellschaft für Anlagen- und Reaktorsicherheit (GRS) mbH, Germany
The Integrated Project PAMINA (Performance Assessment Methodologies in Application to
Guide the Development of the Safety Case) in the European Commission’s FP6 involves 27
contractors from all major radioactive waste producing countries in the EU and Switzerland.
The main objective of IP PAMINA is to improve and harmonise integrated performance assessment
(PA) methodologies and tools for various disposal concepts of long-lived radioactive
waste and spent nuclear fuel in different deep geological environments.
The research work is conducted in four RTD components, each of which having different
work packages and tasks. Being a three-year project PAMINA will continue until September
2009. Owing to the structure of the work, at the time being only preliminary results are available
in most work packages. In this paper the programme of work is outlined and some results
are presented for three of the four RTD components.
1. Introduction
The main objective of IP PAMINA is to improve and harmonise integrated performance assessment
(PA) methodologies and tools for various disposal concepts of long-lived radioactive waste and
spent nuclear fuel in different deep geological environments. The IP PAMINA aims at providing a
sound methodological and scientific basis for demonstrating the safety of engineered geological disposal
of spent nuclear fuel and long-lived radioactive waste. A central purpose is the development
of a common understanding of the different approaches used in the different national waste management
programmes.
The Integrated Project PAMINA in the European Commission’s FP6 involves 27 contractors from
all major radioactive waste producing countries in the EU and Switzerland. From their different
roles within their national programmes these members bring in complementary view points and experiences
into the project which allows exploitation of the project results by both national waste
management organisations and regulators alike.
The research work is conducted in four RTD components, each of which having different work
packages and tasks. Being a three-year project PAMINA will continue until September 2009. Owing
to the structure of the work, at the time being only preliminary results are available in most
work packages. This paper will address the programme of work and it will present some results,
where possible, for three of the four RTD components.
369

The main goal of RTDC-1 is to provide a current and comprehensive overview of safety assessment
methodologies, tools and experiences. A handbook of the state of the art of safety assessment methods
will be prepared in a stepwise manner, which includes the experiences of organisations directly
involved in preparing safety assessments as well as of regulators and other organisations using such
results. The principle objective of RTDC-2 is to develop a common understanding of different approaches
to the treatment of different types of uncertainty in performance assessments in the context
of the development of a post-closure safety case. The objective of RTDC-3 is oriented towards
methodological advancements concerning specific aspects that are of or may be of importance for
the safety case, including various aspects of scenario developments for different host rocks, improvements
of methods and codes regarding process understanding and conceptualization, and the
use of safety and function indicators for assessing the repository system or subsystems, respectively,
at various timescales. Within RTDC-4 the needs for implementing more sophisticated approaches
in PA will be identified based on a quantitative comparison of different models that feature
different complexities of underlying assumptions and process conceptualizations. There are
many links between the various work packages in these four RTDCs.
The work in RTD component 2 concerning methods for the treatment of uncertainty in PA will be
presented in a separate paper by D. Galson and R. Wilmot. [1] and specific aspects concerning sensitivity
analysis will be presented in the paper by R. Bolado-Lavin et al. [2].
2. Review Work in RTDC1
In RTD component 1 the state of the art of methodologies and approaches needed for assessing the
safety of deep geological disposal is evaluated on the basis of a comprehensive review of safety assessment
methodologies and tools used in their implementation, which includes the identification of
any deficiencies in both methods and tools, and in the quality of data required. Important objectives
are to develop a better understanding of international methodologies and to distil the lessons
learned from the rich experience accumulated in their development and application, thus identifying
where approaches and terminology can be harmonised across the European Union. However, harmonisation
in this context does not mean uniformity precluding differentiated national features and
practices.
The review is conducted in 11 topics, each topic being addressed from the implementer’s perspective
as well as from the regulator’s perspective. For each topic a common structure is developed for
the contributions, in order to improve their utilisation and the future integration in a Handbook of
the state of the art of PA, which will be one of the key deliverables from IP PAMINA.
The work is structured in three sets as follows:
First set (first project year)
Safety indicators
Sefatey functions
Definitiona and assessment of scenarios
Uncertainty management and uncertainty analysis
Second set (second project year)
Assessment strategy – safety approach
Evolution of the repository system
Modelling strategy
Sensitivity analysis
370

Third set (third project year)
Biosphere
Human intrusion
Criteria for input and data selection
For each review topic, a three-step approach is adopted. The first step is the target definition, in
which the key questions to be addressed in the review are identified and the scope and outstanding
issues for each topic are clearly delineated and described in written guidelines. Implementers and
regulators analyse the same topic from different perspectives to provide a comprehensive understanding
for each of the relevant issues associated to the topic. The target definition is performed
jointly by all contributors to the review.
In the second step an overview of methods and approaches is generated. In this step the participants
prepare individual contributions in a specifically structured format, where the approaches and methods
applied within their respective organisations are explained with appropriated references to the
national and international contexts. It is the aim to include also pertinent information from non-
European countries. In general, this is done by inviting relevant institutions in the US, in Canada or
in Japan to provide the required information.
In the third step, a synthesis is made in order to formulate conclusions on the strong and weak
points perceived in the methods and approaches. This is based on thorough discussions of the individual
contributions on the topics in a workshop.
Work on the first four topics is finished and results are available [3]. Some specific results are outlined
in the following for the topic safety functions. The application of the defence-in-depth principle
led to the introduction of safety functions for geological disposal systems around 1995. Today,
safety functions are intensively used and play an important role in many safety cases since 2000.
Internationally, it is observed that safety demonstration of geological disposal systems is shifting
from a component-based reasoning to a safety-function-based reasoning.
Several definitions of the term safety function can be found in national or international documents,
but they all have similar meanings. However, for the definitions of secondary terms derived from
safety functions (such as the safety function indicators) some homogenisation might be desirable.
The sets of safety functions that are used by most waste management organisations as well as regulators
are very similar. Three main categories of safety functions can be distinguished: these are stability
/isolation, containment (which is called "isolation" by some organisations) and limited and
delayed releases. The importance of a category of safety functions depends on the considered host
formation and repository concept. Methods are being developed to demonstrate that the safety functions
will be available when required. There is general consensus that dilution in aquifers and biosphere
is not considered as a safety function.
Safety functions are already widely used for various applications such as determination of the safety
strategy, development of the repository concept, analysis of the functioning of the repository system,
testing the robustness of the repository system, structuring the safety case, scenario identification,
identification of performance indicators, and communication. There is a clear trend to increase
the use of safety functions within the safety case, as can be seen in recent safety assessment exercises.
371

3. RTDC3
In RTD component 3 methodologies and tools for integrated PA for various geological disposal
concepts will be further developed. The work is separated into four work packages and it includes
the development of scenarios by applying safety functions and stylised scenarios,
the PA approach to gas migration processes with respect to the determination and quantification
of the impact of gas on the engineered and natural barriers, and
the PA approach to radionuclide source term modelling where a more detailed modelling of the
chemical environment and the up-scaling from one canister or disposal cell to a large scale repository
is considered.
In some of the areas, PA representations are known to be less advanced. The last work package in
this RTD component concerns the use of safety and performance indicators with the objective of
testing the applicability of various safety indicators for host rock formations other than granite. First
results are described here.
The work in this work package is based on the review work in RTD component 1 concerning safety
and performance indicators. One of the main findings was that there is international consensus that
a repository safety case can be strengthened by the presentation of a range of safety indicators, to
complement the dose or risk calculations. There are different concepts of assessing repository safety
and performance by means of other indicators. The review also revealed that there is still a large
variety of different views on the terminology used for safety indicators and performance and function
indicators, respectively.
In the work in RTD component 3 an important goal is to develop a common understanding as follows.
Every safety indicator that is applied for the safety case must be defined by a safety statement,
based on a previously defined safety aspect. These safety aspects are based on a specific
safety concern, such as “protection of human life” and “protection of environment”. A safety statement
could, for example, be that all biological effects to a human individual, i.e. the incorporation
of radionuclides released from a repository by humans via different exposition paths, remain so
small that they have no impact on human health. The corresponding safety aspect is the human
health. For a complete safety statement a numerical measure is required, by which the biological
effects due to incorporation of radionuclides can be calculated, as well as a reference value in order
to define a safe level. A common measure for this type of safety statement is the individual effective
dose rate. In the case of the effective dose rate, the national legal limit could be used as reference
value. In general, reference values are either global or can be derived on a local scale. The use
of the latter must be justified thoroughly.
In order to employ a safety indicator in a safety case, it is necessary to calculate the outcome of the
considered numerical measure (e.g., the effective dose rate) by means of a PA model for the corresponding
repository system. The PA model allows the calculation of the radionuclide migration in
the repository system based on a reference design and a description of the geological site. By comparing
the results of the PA model with the reference value of the safety indicator, the determination
of the safety indicator is complete and the results can be added to the safety case.
In contrast to safety indicators there can be a close interaction between the PA model and the definition
and calculation of performance indicators, especially between the compartment structure and
the repository design. For every performance indicator a compartment structure has to be defined.
The chosen compartment structure depends on several conditions, e.g. the host rock or the type of
quantity (concentration, flux) that is calculated. If several performance indicators are used within
372

one study, the compartment structures should be identical, or at least as similar as possible, in order
to allow a comparison of the results. Since performance indicators are not based on a safety statement
they do not require any reference value and their use is more flexible. Performance indicators
are very important for the understanding of the modelled processes and they can be used for optimising
the repository system. Finally not every applied performance indicator may be used in the
safety case, but most of them give valuable arguments for increasing the confidence in the safety of
a repository system.
With respect to the work concerning safety and performance indicators several test cases have been
defined and applied to host rocks such as clay and rock salt, for which the use of safety and performance
indicators has hardly been investigated in detail yet. The general approach is based on the
SPIN methodology. Preliminary results suggest that this methodology works well for all kinds of
host rock. Some new indicators have been identified, which are tested within PAMINA with results
still pending. It is evident, that performance indicators give a good insight to the functioning of the
system. In this context it was observed that performance indicators applied to specific radionuclides
are useful for identifying influential processes. Finally it is noted that performance indicators are
dependent on repository systems and host rocks.
4. RTDC4
In RTD component 4 several benchmark exercises are carried out, in which quantitative comparisons
are made of approaches that, on the one hand, rely on simplifying assumptions and models,
and on the other hand, on complex models that take into account a more complete process conceptualization
in space and time. The main objective is to evaluate the added value of using more complex
and more realistic modelling approaches not yet fully accounted for in PA. This RTD component
comprises of three work packages.
The first work package focuses on processes which determine the evolution of the near field of a
repository in salt as host rock. One aspect being investigated is the convergence process of rock
salt, which determines the advective flow of radionuclides in case of a repository completely filled
with solution. For several test cases results are compared from different approaches to model this
complex process which depends on many parameters. Also the brine intrusion process into a backfilled
drift and the radionuclide transport by density driven exchange are investigated for salt as
host rock.
Specific benchmarks have been set up to study the reactive transport of radionuclides for clay as
host rock. One exercise investigates Cs migration, considering all competitive effects on sorption
processes. Both empirical and thermodynamic models are defined for geochemical modelling. Another
exercise focuses on heavy elements behaviour with regard to sorption and precipitation processes.
Also, benchmarks are carried out for sensitivity analysis on “Kd” and “solubility limit” models
/ geochemical transport with respect to radionuclide migration in the near field. In case of granite
as host rock benchmark calculations using the reactive transport code CORE and GoldSim as PA
code aim to draw conclusions on the usefulness and the need of implementing fully reactive transport
in PA models instead of the Kd and solubility limit approach commonly used in PA.
The second work package is to investigate the usefulness of codes dealing with geometric complex
representations of the geosphere in comparison to coarse or simplified 1D representations often
used in PA codes for modelling the transport behaviour of radionuclides in the far field. Part of this
work is described in the following for a repository located in rock salt. Typically, the water in the
aquifer that is in contact with the cap rock of the salt dome is highly saline, while other aquifers at
373

lower depths are less saline. For the purpose of the benchmark exercise a simplified hydrogeological
model for the overburden was developed that features some of the characteristics of the Gorleben
site without being an accurate model of the real situation there. In the model used two aquifers
are located above a salt dome and are separated by a clay layer. This aquiclude has one hydrological
window near the left side of the model region allowing groundwater exchange between the two aquifers.
At the bottom of the model region at one point radionuclides are introduced at a rate which
corresponds to the convergence-driven advective flow from a repository in rock salt which is been
completely filled with solution. This test case refers to a disturbed evolution of the repository system.
The evolution of the system over 1 million years and the transport behaviour of all pertinent radionuclides
in high-level waste have been calculated with the dedicated reactive transport program
r 3 t. In Fig.1 the results of model calculations are shown for the low-sorbing isotope Cl-36 for a
point in time of 10 000 years. The calculated concentrations are given in Becquerel per cubic metre
of water, colour coded on a logarithmic scale ranging from 10 -15 to 1. Note that the numbers given
on the colour bar give the according exponent. It can be clearly seen that there are two distinct preferential
flow paths for this radionuclide in the overburden. The first flow path, called here pathway
1, is from the radionuclide source in direct vertical direction through the lower aquifer, the aquiclude
and the upper aquifer. The second flow path, called pathway 2, is first in horizontal direction
to the left side in the lower aquifer towards the hydrological window in the clay aquiclude and than
through this window into the upper aquifer and to the surface. The two flow paths are indicated as
arrows in Fig. 1.
Fig.1 Cross section of the Cl-36-concentration, given in Bq/m 3 , after 10 000 years and the two
different abstracted 1D transport pathways of the CHET model
It was observed that the general transport behaviour of all radionuclides is similar, but that different
fractions are transported on the two pathways depending on their adsorption behaviour. While a
high fraction of a low-sorbing radionuclide like C-14 is directly transported upwards on pathway 1,
the highly sorbing radionuclides like Th-230 have only a very limited ability to be transported
across the clay layer. For these radionuclides pathway 2 was the predominating transport pathway.
In general, the r 3 t calculations showed that two distinct radionuclide concentration maxima exist at
the model surface, corresponding to pathway 1 and pathway 2, respectively. These two positions
reflect the locations with the maximum potential radiation exposures. The maxima only slightly
change their position with time.
374

Although generally the maximum Cl-36 concentration values are higher resulting from pathway 2,
in the time frame at about 500 000 a the concentrations at the position corresponding to pathway 1
exceed those of pathway 2. This shows that in case of Cl-36 both pathways could potentially contribute
about the same proportion to the radiation exposure of the population by Cl-36 if water is
used from the upper aquifer. However, the transport on pathway 1 is somewhat slower, even for the
low-sorbing Cl-36.
Both pathways have been modelled separately with the one-dimensional PA code CHET to compare
the results with the concentrations calculated with r 3 t. Comparing the results for the various
radionuclides obtained with the two programs reveals that the 1D PA code overestimates the radionuclide
concentrations by one to two orders of magnitude. The reasons for this are manifold.
Firstly, radionuclide diffusion is badly represented in lower dimensional models. Also, the heterogeneity
of the transport velocities and their averaging resulted in too short transport times in the 1D
model. Finally, the fast transport at the end of pathway 2 resulted in insufficient time to bring decay
chains into radioactive equilibrium. This fact is quite common in PA radionuclide transport modelling
and has to be accounted for by considering an additional transport time in the 1D model that
allows time to establish the radioactive equilibrium in the decay chains.
The transport on the faster transport pathway 2 for all radionuclides resulted in higher radionuclide
concentrations than the transport on pathway 1. For radionuclides with a short half-life, this is even
pronounced by the radioactive decay if transported on the slower pathway. In the case presented
here it is therefore always conservative in terms of determining maximum concentrations to regard
only the transport on pathway 2 and its abstraction by a one dimensional model. To determine the
temporal evolution of the radiation exposure it is however necessary to model both transport pathways.
4. Conclusions
The integrated project PAMINA has already made good progress and is expected to continue doing
so. Strong links have been established with the Integration Group for the Safety Case (IGSC) of the
OECD/NEA with the aim to feed PAMINA results directly into the work of the IGSC. All results of
PAMINA are made publicly available through the PAMINA Internet page (www.ip-pamina.eu) as
soon as possible. This includes not only the deliverables, which represent the main results of the
project, but also relevant milestone reports. The Final PAMINA Workshop will be held on Sept. 28
– 30, 2009, in Germany at Schloss Hohenkammer near Munich. In combination with the final workshop,
a training course will be offered which takes place just before the workshop.
5. Acknowledgements
This project is co-funded by the European Commission and it is performed as part of the sixth
EURATOM Framework Programme for nuclear research and training activities (2002-2006) under
contract FI6W-036404.
References
[1] Galson, D. and Wilmot, R.: The Treatment of Uncertainty in PA and the Safety Case. –
EURADWASTE ’08, Luxemburg, this issue
[2] Bolado-Lavin, R., Röhlig, K.-J. and Becker, D.-A.: Sensitivity analysis techniques for the performance
assessment of a radioactive waste repository. – EURADWASTE ’08, Luxemburg,
this issue
375

[3] Marivoet, J., Beuth, T., Alonso, J. and Becker, D.-A.: Task reports for the first group of topics:
Safety Functions, Definition and Assessment of Scenarios, Uncertainty Management and
Uncertainty Analysis, Safety Indicators and Performance/Function Indicators. – PAMINA
DELIVERABLE N°:1.1.1, available via www.ip-pamina.eu.
376

PAMINA – The Treatment of Uncertainty in PA and the Safety Case
Summary
Daniel Galson and Roger Wilmot
Galson Sciences Ltd, Oakham, UK
The European Commission’s PAMINA Project (Performance Assessment Methodologies in
Application to Guide the Development of the Safety Case) has the aim of developing a common
understanding of and improving integrated PA methodologies for the geological disposal
of radioactive wastes. A core component consists of research on methods for the treatment of
uncertainty in PA and the safety case
This paper summarises the overall programme of work on treatment of uncertainty, and the
results of an initial international review of the treatment of uncertainty and a recently completed
task on the regulatory evaluation of uncertainty. A workshop aimed at discussing the
approaches used by regulators in evaluating uncertainties in the safety case was organised by
the Swedish regulators and Galson Sciences Ltd in June 2008. The workshop provided a forum
for discussion of how regulators can assess and compare quantitative and qualitative lines
of reasoning and evidence in safety cases that are subject to uncertainties.
1. Background
Development of a safety case for the management of long-lived radioactive waste involves consideration
of the evolution of the waste and engineered barrier systems, and the interactions between
these and relatively complex, natural systems, such as climate change. The timescales that must be
considered are much longer than the timescales that can be studied in the laboratory or during site
characterisation. These, and other factors, give rise to various types of uncertainty e.g., on scenarios,
models, and parameter values used in modelling, which need to be taken into account when assessing
long-term performance of a geological disposal facility. It is important to follow a clear
strategy for dealing with uncertainties when developing a safety case.
The European Commission’s PAMINA Project (Performance Assessment Methodologies in Application
to Guide the Development of the Safety Case), which has 27 partner organisations and runs
from 2006 to 2009, has the aim of improving and developing a common understanding of integrated
performance assessment methodologies for disposal concepts for spent fuel and other long-lived
radioactive wastes in a range of geological environments. Galson Sciences Ltd is responsible for the
co-ordination and integration of the Research and Technology Development Component “RTDC-2”
of the PAMINA Project.
The objective of RTDC-2 is to allow development of a common understanding of different approaches
to the treatment of uncertainty in PA, and to provide guidance on, and examples of, good
practice on how to treat different types of uncertainty in the context of the development of a postclosure
safety case, both as a whole and in specific areas. Guidance on the development of work in
RTDC-2 has come from an initial review of key drivers and methodologies for the treatment of un-
377

certainty, conducted in RTDC-1 as Work Package 1.2 (WP1.2). This initial review is summarised in
Section 2.
RTDC-2 is organised in three work packages:
WP2.1 is researching key drivers and methodologies for the treatment of uncertainty, addressing
regulatory compliance, the communication of uncertainty, approaches to system PA, and techniques
for sensitivity analysis.
WP2.2 is proceeding in parallel with WP2.1 and is testing and developing the framework outlined
in WP1.2 by undertaking a series of exercises to provide examples of uncertainty treatment
from different European programmes at different stages of development. The work is divided
into tasks that consider the main types of uncertainties (scenario, model, parameter), the
treatment of spatial variability, and the development of probabilistic safety assessment tools.
WP2.3 is a synthesis task pulling together the WP1.2 review, and research on the treatment of
uncertainty under WP2.1 and the testing and development work under WP2.2 to arrive at final
guidance on approaches for the treatment of uncertainty during PA and safety case development
that contains state-of-the-art examples from RTDC-2 for a range of key areas.
Most of this work is still underway, so we focus here on the outcome of the WP1.2 review, a brief
description of the work in progress, and, in Section 4, a longer summary of the one complete task in
RTDC-2 on the regulatory evaluation of uncertainty.
2. WP1.2 Initial Review of the Treatment of Uncertainty
The aim of WP1.2 was to develop a document that synthesises the state-of-the-art at the beginning
of the project, providing examples on approaches to the treatment of different types of uncertainty
at different stages of safety case development and highlighting areas where further development
would be most helpful. Information on the treatment of uncertainties was gathered from PAMINA
participants and several other organisations using a questionnaire, and via a limited wider review of
the literature. The questionnaire responses obtained represent 16 disposal programmes in 13 countries,
including all of the countries with advanced programmes to implement geological disposal,
allowing the review to give wide coverage of global activity. Selected results from the review are
given here. A more complete summary is provided in [1].
2.1 Types of Uncertainties Considered in PA
There is consensus on both how uncertainties considered in PAs should be classified and the nature
of uncertainties, although this is masked by variations in terminology and differences in the way
uncertainties are treated in programmes. Uncertainties in PAs are generally classified as:
1. Uncertainties arising from an incomplete knowledge or lack of understanding of the behaviour
of engineered systems, physical processes, site characteristics and their representation using
simplified models and computer codes. This type of uncertainty is often called “model” uncertainty.
It includes uncertainties that arise from the modelling process, including assumptions associated
with the reduction of complex “process” models to simplified or stylised conceptual
models for PA purposes, assumptions associated with the representation of conceptual models
in mathematical form, and the inexact implementation of mathematical models in numerical
form and in computer codes.
378

2. Uncertainties associated with the values of the parameters that are used in the implemented
models. They are variously termed “parameter,” or “data” uncertainties.
3. Uncertainties associated with the possible occurrence of features, events and processes (FEPs)
external to the disposal system that may impact the natural or engineered parts of the disposal
system over time. These are usually referred to as “scenario” or “system” uncertainties.
All three classes of uncertainty are related to each other, and particular uncertainties can be handled
in different ways, such that they might be dealt with in one class or another for any single iteration
of a PA/safety case, depending on programmatic decisions (e.g., on how to best to implement PA
calculations or to communicate results) and practical limitations (e.g., on funding or timescales).
The classification system for uncertainties given above essentially arises from the way the PA is
implemented, and says little about the nature of the uncertainties. With respect to nature, a useful
distinction can be made between epistemic and aleatory uncertainties. Epistemic uncertainties are
knowledge-based, and therefore, reducible by nature. Aleatory uncertainties, on the other hand, are
random in nature and are irreducible.
All three classes of uncertainty contain elements that are epistemic and aleatory, although it may be
generally true that “scenario” uncertainties contain a larger element of aleatory uncertainty than the
other two groups. To take an example, typically “parameter” uncertainties may arise for the following
reasons:
The parameter values have not been determined exactly. This type of uncertainty is largely epistemic
in quality, and can be reduced with further effort.
The models use single (or spatially averaged) values for parameters, derived from measurements
at discrete locations, whereas in reality there is continuous variation in parameter values
over space - as well as over time. This type of uncertainty is partly aleatory in quality and cannot
be reduced by further effort.
2.2 Dealing with Uncertainty in the Quantitative PA
2.2.1 Parameter uncertainty
Uncertainties associated with model parameter values can be treated conveniently within most computational
schemes. Common approaches to treating parameter value uncertainty include the following:
1. Setting probability distributions functions (PDFs) for parameter values, which are sampled during
the course of a probabilistic assessment.
2. Repeat deterministic calculations where individual parameter values are varied across a range of
likely or possible values, including deterministic calculations using values representing the best
understanding available (“best estimate”) to better understand the system, e.g., with regard to
sensitivities.
3. Deterministic calculations where deliberately pessimistic values of parameters are taken, producing
a “conservative” estimate of the value of receptor quantities in order to demonstrate
compliance with limits.
379

A simplistic summary might place PA programmes in two camps: those that rely primarily on the
probabilistic approach described in (1) and those that primarily use deterministic approaches (2)
and (3). However, this is, increasingly, an over-simplified picture.
Where a probabilistic approach is preferred, it is normally supplemented by deterministic calculations
(e.g., in Germany, the Netherlands, Spain, the UK, the US). Reasons commonly given for this
preference are the ease with which probabilistic calculations are done, the more complete treatment
of parameter uncertainty, and completeness in terms of describing the whole system from source to
receptor.
Programmes in Belgium, Finland, France, Japan, Sweden and Switzerland are in the camp that favours
largely deterministic approaches, but probabilistic approaches have been or are being considered
in these countries to supplement the deterministic calculations. There is a view that a deterministic
approach has advantages where there are very large uncertainties in the PA, and where the
use of deterministic approaches allows a more transparent treatment of uncertainty.
Regulation can play an important role in determining which approaches to PA are adopted for compliance
calculations.
2.2.2 Conceptual model uncertainty
We focus here on conceptual model uncertainty as there are long-standing tools available to treat
uncertainty in mathematical models and computer codes (e.g., verification, benchmarking exercises,
QA). Conceptual model uncertainties, arising from an incomplete knowledge of the behaviour of
engineered systems, physical processes and site characteristics, and their representation by simplified
models, are perhaps the least well covered in PA. They are often not treated explicitly. In fields
of environmental modelling where it has been possible to compare predictions with measurements,
albeit conducted over shorter time spans than those required for geological disposal facility PA, the
impact of conceptual model uncertainties on assessment endpoints can be significant.
One approach to treat conceptual model uncertainty that is used by organisations taking probabilistic
approaches to parameter uncertainty is to use expert judgement to widen parameter PDFs so as
to represent a greater range of uncertainty than that accounted for by uncertainty in the parameter
values themselves. However, in order to use this approach there must be some understanding of the
effects on assessment endpoints of altering individual parameter values, and a feeling for how much
effect conceptual model uncertainty could have on the same assessment endpoints. Organisations
taking deterministic approaches tend to implement and run separate deterministic calculations based
on the use of alternative models.
2.2.3 Scenario uncertainty
Two main types of approach for treating scenario uncertainty may be delineated:
1. A pure probabilistic sampling approach, in which scenario characteristics (e.g., timing, magnitude)
are sampled from a distribution of possibilities during a Monte Carlo calculation, in much
the same way that parameter values are sampled from PDFs in dealing with parameter uncertainty.
2. Evaluation of a limited set of deterministically defined cases for each scenario in which limited
variations in the characteristics of each scenario are explored – in some cases only a single set
380

of characteristics for each scenario will be defined. Although the characteristics of each scenario
are defined deterministically, scenario consequences may then be assessed probabilistically or
deterministically. Probabilistic consequence assessment means a deterministic approach is taken
for largely “irreducible” uncertainties associated with development of the system over time
(scenario uncertainties), and a probabilistic approach for “reducible” uncertainties associated
with knowledge of the system (many parameter and conceptual model uncertainties). The types
of scenarios typically considered in such an approach are:
a) The reference or normal evolution scenario, which may be the scenario with the greatest
probability of occurrence.
b) Altered evolution scenarios, in which the impacts of more unlikely future conditions are
evaluated. They are sometimes implemented using a pessimistic “bounding” approach to
demonstrate compliance with regulations and to build confidence in safety.
c) “Stylised” scenarios for events and processes for which there are large aleatory uncertainties
that cannot be reliably quantified. In particular, stylised scenarios are usually defined to illustrate
the potential impacts associated with future human intrusion.
Some programmes (e.g. Belgium, UK) consider the use of a discrete set of assessment timeframes
(e.g., first few hundred years, thousand years, hundreds of thousands of years) in structuring the assessment,
in developing assessment models, and in communicating the results.
A recent development is the use of safety functions to describe the required performance of engineered
and natural barriers in different time frames, and the concept of safety function impairment
as a means of comprehensively describing possible future states of the system. For the purpose of
regulatory compliance calculations, this approach can reduce the need for identifying and assessing
all possible reasons for poor barrier performance (or barrier longevity), or for ascribing difficult-todefend
probabilities to barrier failure.
2.3 Methods Used to Address Uncertainties and Provide Confidence
The purpose of uncertainty analysis is to give an absolute estimate of uncertainty in assessment
endpoints, such as dose or risk. It is achieved through propagating estimates of uncertainty in assessment
inputs through assessment models to assessment outputs. The analysis produces estimates
of uncertainties in assessment endpoints without necessarily explaining which input quantities the
uncertainties are derived from. The purpose of sensitivity analyses is to understand how the system
works and which parameters have a strong influence on assessment endpoints. Taken together, such
analyses can identify those sources of uncertainty in parameter values or conceptual model implementation
where the most benefit would be gained – in terms of reduction in overall uncertainty or
greater confidence in PA results - from further investigation or modelling. The value of such analyses
is widely appreciated in PA and safety-case development. A variety of methods for conducting
such analyses is available, and these are described in a companion paper in these proceedings.
The identification and management of uncertainties is an iterative process that can lead to a stepwise
reduction of uncertainties in PA. Factors to weigh in decision making to reduce uncertainties
in PA include whether there are (over)conservatisms in the PA, “how reducible” the uncertainties
are with further targeted information gathering, the likely effectiveness of engineered solutions to
reduce uncertainty, and cost and regulatory acceptability. For an operational disposal facility, some
aspects of the design will be frozen, and there is less scope for design modification.
381

There are at least three non-numerical or qualitative methods for dealing with uncertainties in PA
and the safety case:
1. Robust design, where uncertainties are managed by using conservative engineering design
principles. An example of this approach is provided by the Finnish and Swedish programmes,
where the engineered barriers used in the KBS-3 concept are extremely robust, making some
uncertainties associated with the far field and biosphere easier to discount.
2. Qualitative assessment methods: programmes in many countries employ qualitative assessment
methods, and in some situations they are considered to be as important as the quantitative
methods. This is particularly true where an assessment considers events far removed in space
and time from the original emplacement of waste in the disposal facility, and there are very
large uncertainties associated with the quantitative assessments.
3. Quality Assurance (QA) for development of the PA and safety case: implementing appropriate
QA systems for all aspects of disposal facility development programmes plays a part in the
process of building a compelling safety case and obtaining approval from regulators and stakeholders.
Many programmes have applied custom-designed or internationally accredited QA procedures
to their operations.
3 . PAMINA RTDC-2
3.1 WP2.1 Methodological Research for Treatment of Uncertainty
WP2.1 consists of four tasks:
2.1.A Regulatory Compliance. A workshop was held in June 2008 that examined the treatment of
uncertainty in PA and the safety case with respect to regulatory compliance. The outcome of
this workshop is discussed in Section 4.
2.1.B Communication of Uncertainty. This research aims to understand the effectiveness of different
methods for communicating disposal system performance, communicating how it has
been determined, and communicating the uncertainty associated with the determination and
its significance. The results of a workshop designed to test different approaches to communicate
to lay stakeholders are available [2].
2.1.C Approaches to System PA. This research is examining the advantages and disadvantages of
different approaches to the quantification of uncertainties in system-wide PA calculations,
including deterministic scenario-based assessments versus probabilistic assessments, levels
of conservatism and realism in PA, exploration of the potential of hybrid stochasticsubjective
uncertainty treatment, and alternative approaches for presentation of results from
safety analysis / uncertainty analysis in the form of graphical outputs.
2.1.D Techniques for Sensitivity and Uncertainty Analysis. This research aims to compare the
advantages and disadvantages of different methods for applying sensitivity analysis to PA
calculations. The research comprises a review of the use of sensitivity analysis in PA,
evaluation of the use of sensitivity analysis techniques for a range of actual radioactive
waste disposal facility concepts, testing of sensitivity analysis techniques based on use of
complex and simplified generic PA models, and an international benchmark study aimed at
testing a wide range of the sensitivity analysis techniques on two test cases.
382

3.2 WP2.2 Testing Approaches for Treating Uncertainties
WP2.2 consists of five tasks:
2.2.A Parameter Uncertainty. Research in the area of parameter uncertainties is focused on developing
practical recommendations for reliable and defensible derivation of PDFs for key
parameters used in PA calculations. This includes testing the limitations and (dis)advantages
of methods such as statistical analysis, the Bayesian approach, expert judgement, and hybrid
methods.
2.2.B Model Uncertainty. Work in the area of conceptual model uncertainties includes calculations
to assess the consequences of using alternative representations of key processes, including
radionuclide retardation, gas migration and groundwater flow, in a PA system
model. This work is nearly completed and reports are available on the treatment of uncertainties
associated with modelling the gas pathway [3] and hydro-geochemical impacts on
uranium transport through an engineered barrier system [4].
2.2.C Scenario Uncertainty. This task is considering the extent to which the probabilities of different
types of scenarios can be evaluated, methods for amalgamating consequence results into
risk estimates and the associated limitations (e.g., related to statistical convergence and scenario
termination events), the extent to which it is reasonable to account for the uncertain
occurrence of FEPs in “normal evolution” scenarios, and the definition and utility of analysing
“less likely” and “what if” scenarios.
2.2.D Spatial Variability. This task is evaluating approaches to treating uncertainties in PA calculations
that arise from the spatial variability of facies, materials, and material properties inherent
in the geosphere.
2.2.E Probabilistic Safety Assessment. This task is developing and evaluating an integrated, fully
probabilistic safety assessment approach incorporating scenario, model and parameter uncertainty.
3.3 WP2.3 Synthesis and Integration of RTC-2 Activities
WP2.3 is a synthesis task pulling together research on the drivers for treatment of uncertainty under
WP2.1 and the testing and development of approaches to treatment of uncertainty under WP2.2 to
develop the guidance initially prepared under WP1.2. The main deliverable from the work package
will be a report at the end of the project giving guidance to users on approaches for the treatment of
uncertainty in PA and safety case development and containing illustrative examples from work undertaken
in RTDC-2, as well as from the WP1.2 review. In addition, the guidance will include approaches
to the prioritisation and screening of uncertainties, and consideration of the importance of
sound treatment of uncertainties to the safety case and its communication. The report will provide a
key international reference point for performance assessors and safety case developers.
4. Regulatory Evaluation of Uncertainty
4.1 Introduction
Under Task 2.1.A, the Swedish regulators, with assistance from Galson Sciences Ltd (GSL), organised
and hosted a workshop to elicit views on managing uncertainties in a safety case for a geological
disposal facility mainly from a regulatory perspective. The workshop was held in Stockholm on
383

10-11 June 2008. The workshop was attended by 16 participants drawn from regulators and other
organisations with close interests in the management of uncertainties in the safety case for geological
disposal of radioactive waste. The workshop was facilitated by Roger Wilmot of GSL, and a
workshop report is available [5].
The workshop was grouped into three main sessions with contributed presentations, which provided
a stimulus for wider discussion of the issues:
1. Uncertainties in the safety case. This session addressed some of the key issues relating to the
treatment of uncertainty that are faced by regulators, and included summaries of previous work.
2. Regulatory guidance on the treatment of uncertainties. An important means for regulators to
influence the treatment of uncertainties is through guidance. This session described recent experience
in developing regulatory guidance.
3. Regulatory review of uncertainty treatment. Reviews and assessments of safety cases and license
applications allow regulators to determine whether their requirements and expectations
concerning the treatment of uncertainty have been met. This session described some recent review
experience.
A final discussion session that considered points that had been raised throughout the workshop is
summarised in Section 4.2. Key messages from the workshop are summarised in Section 4.3.
4.2 Discussion
Discussion of the regulatory review process highlighted some general points. First, the regulator
does not need to replicate the full safety assessment produced by a developer (a situation that pertained
earlier in the UK). Instead, the regulator is concerned with reviewing the research, development
and demonstration (RD&D) programme or safety case submitted by a developer, and in doing
so it will use its own capabilities to assess and evaluate key processes and uncertainties. Following
a review, a regulator is in a position to require the developer to carry out what it considers to be
necessary further research, site characterisation or assessment.
There is the question of when should the regulator request a developer to do a piece of research
rather than doing it itself. Research pursued by a regulator or regulatory support organisation is
likely to be focused on obtaining improvements in scientific and technical knowledge as a basis for
effective reviews and for maintaining and developing regulatory competence. In addition, a regulator
may carry out some ‘seed’ research in order to demonstrate to the developer that a research area
should be investigated in more detail. The Swiss regulator places importance on its experts staying
at the forefront of science and performing quality ‘independent’ research.
Although a developer has the primary responsibility to verify its codes, the question of how much
involvement a regulator should have in code verification was posed for discussion. During the
compliance certification (authorisation process) of the Waste Isolation Pilot Plant (WIPP) in the
US, the regulator requested the developer to re-run the assessment using regulator-provided parameter
values. Although this was not a code verification exercise, it circumvented the problem that
the regulator had no code to carry out its own independent assessment calculations. On the other
hand, the French, German, Swedish and Swiss regulators develop and maintain their own codes and
an independent capability for modelling radionuclide transport and PA. This is considered important
for verifying the developer’s results, for consideration of the assumptions buried in the codes,
384

and for independently investigating the possible impact of alternative PA parameter values and assumptions.
The Finnish regulator plans to use a simple code to verify what the developer has modelled.
The UK environmental regulators rely on consultants to develop and run codes only as necessary.
The question was discussed of how distant or close the regulator should be to the developer during
the development of the safety case to avoid compromising its review process during a licence application
stage. Although the onus for developing the safety assessment, safety case and licence application
rests with the developer, it is the regulator that grants the licence or authorisation, and it is
the regulator, not the developer, which has to defend and justify to the public and other stakeholders
the decision to dispose of waste. Stakeholders must recognise that although a regulator must not be
compromised in any way and should have freedom to make decisions once a licence application is
submitted, it nevertheless needs detailed knowledge of the safety assessment and safety case in order
to review the application and defend its decision to grant or recommend a licence or authorisation.
Ideally, regulatory decisions should not be bound by any commitments made to the developer
prior to receipt of the application. Basic scientific knowledge can be jointly gained and commonly
understood, but it is used differently by the developer and regulator.
A pre-application review procedure means, by necessity, that a regulator will advise or require a
developer at various intervals to do specific pieces of research or investigation, and this appearance
of working together needs to be explained to stakeholders in order to avoid misunderstanding during
the licensing process. A regulator might pose questions to, and place requirements on, the developer
via regulatory review, but the regulator should not provide the answers.
A formal process of staged authorisation has been outlined in the UK in new regulatory guidance on
requirements for authorisation for geological disposal facilities, with a series of formal hold-points
to be decided by the regulator [6]. If adopted, this regulatory process will formalise the need for
dialogue between the regulator and the developer.
4.3 Main Messages
Some key messages arising from the workshop were:
Although international harmonisation of dose and risk constraints would be ideal for communication
with the public, the practicalities of national contexts mitigate against this being achieved.
There is now less emphasis than before being placed in the safety case on the traditional comparison
between safety assessment calculation results and dose/risk criteria set by the regulator.
Optimisation and safety functions are increasingly being used as alternative safety indicators or
additional arguments in a safety case in support of compliance with the regulatory dose/risk criteria
and to build confidence in long-term safety. These changes in the emphasis of safety cases
may require additional regulatory research and capabilities.
Most regulators want to match the level of scientific understanding and knowledge of the developer/implementer
in order to be capable of performing meaningful reviews of RD&D programmes,
safety cases and licence applications.
Many regulators have taken steps to have modelling capabilities independent of the developers’
capabilities in order to be able to verify the results of the developers’ assessment calculations
and to investigate alternative conceptual or models.
385

Close dialogue between a regulator and a developer is beneficial to the development of a safety
case and a licence application, but the dialogue must be controlled and documented and not compromise
a regulator’s freedom to make decisions.
5. Conclusions
There is a high level of awareness of the importance of treating uncertainties in PA and the safety
case, and treatments of varying degrees of sophistication have been implemented in all national programmes.
The PAMINA project is making a significant contribution and the final guidance report
from RTDC-2 on the treatment of uncertainty, to be available in late 2009, will summarise this
work and will form a key international reference for both performance assessors and regulators.
6. Acknowledgements
The authors acknowledge the support of the EC under contract FP6-036404, and co-funding received
from ONDRAF/NIRAS (Belgium), NDA (UK), Nagra (Switzerland), and SSM (Sweden).
References
[1] Galson, D.A. and Khursheed, A. (2007). The treatment of uncertainty in performance assessment
and safety case development. In: Proceedings of 11 th Int. Conf. on Env. Remed. and
Radioactive Waste Management ICEM2007 (2-6 September 2007, Bruges). ASME.
[2] Hooker, P.J. and Greulich-Smith, T. (2008). Report on the PAMINA Stakeholder Workshop:
Communicating Safety Issues for a Geological Repository. PAMINA Deliverable D2.1.B.1.
Galson Sciences Limited, UK [http://www.ip-pamina.eu/publications/reports/index.html].
[3] Norris, S. (2008). Uncertainties associated with Modelling the Consequences of Gas. PA-
MINA Deliverable D2.2.B.2. Nuclear Decommissioning Authority, UK [ibid].
[4] Luukkonen, A. and Hordman, H. (2008). A hydro-geochemical change in an engineered barrier
system – two model responses to uranium transport. PAMINA Deliverable D2.2.B.3.
Technical Research Centre of Finland [ibid].
[5] Hooker, P, and Wilmot, R. (2008). Report on the PAMINA Workshop on the Regulatory
Role in Managing Uncertainties in the Safety Case for Geological Disposal of Radioactive
Wastes. PAMINA Deliverable D2.1.A.1. Galson Sciences Limited, UK [ibid].
[6] Environment Agency and Northern Ireland Environment Agency (2009). Geological Disposal
Facilities on Land for Solid Radioactive Wastes: Guidance on Requirements for Authorisation.
386

Sensitivity Analysis Techniques for the Performance Assessment of a RadioactiveWaste
Repository
Ricardo Bolado Lavín 1 , Klaus-Jürgen Röhlig 2 , Dirk-Alexander Becker 3
1 Institute for Energy, European Commission DG-JRC, Petten, The Netherlands
2 Technical University of Clausthal, Germany
3 GRS-Braunschweig, Germany
Summary�
Sensitivity Analysis (SA) is a key element in the Performance Assessment (PA) of a Radioactive
High Level Waste (HLW) Repository. It helps gaining information about the system, understanding
the relation between input parameters and output variables and steering new experimental
and theoretical research to increase the degree of knowledge about the system. The
Integrated Project (IP) PAMINA is devoting a large effort to the research in this area and the
dissemination of results among its partners. Among the activities under development are a review
of SA methods, a benchmark of SA techniques and the application of different SA techniques
to a PA results coming from several national programs. After the end of this activity,
PAMINA partners will get a better understanding of the rationale behind every available technique
and about their capabilities and shortcomings.
1. Introduction
SA methods may be divided into three broad types: Local methods, screening methods and global
methods. Local methods focus on the study of the system model behaviour under very specific system
conditions (the vicinity of an input space point), while screening methods focus on the functional
relation between inputs and outputs disregarding input parameter distributions, and global
methods focus on how the whole input space (taking into account input distributions) maps into the
output space. Though all of them are important and provide relevant information about the system
model, screening methods and global methods fit better within the structure of a PA, and that is the
reason to focus all efforts on them.
Three main activities are being developed in this area: the review of SA methods available in the
scientific literature, the development of a benchmark on SA techniques and the practical application
of SA techniques to results of PA studies produced by different partners. The objective of the review
of SA techniques is to provide a snapshot of most useful SA techniques and to provide guidance
about merits and shortcomings of each one. Screening techniques are useful to identify irrelevant
input parameters that can be set to their nominal value not losing information. Global methods
may be classified as Monte Carlo based methods, variance based methods, and graphical methods.
The SA benchmark has been designed as a two-step process. The first step is dedicated to analyse a
set of mathematical functions most of whose sensitivity indices are well known. The targets in this
step are to debug SA computational tools used, to get skills in their use and to get progressively in
contact with specific features of mathematical models such as (lack of) linearity, (lack of) monot-
387

ony, interactions, etc., and to check the importance of sample size. The second step consists in analysing
a simplified, though representative, PA model. The complex input-output relation, characterised
by strong interactions among input parameters, makes it a challenging model to test SA techniques.
In this case, the target is twofold: firstly to compare different options within a given SA
techniques (to study the added value of using more complex versions of a given technique – classical
FAST versus extended FAST, first order regressions versus higher order regressions, etc.-), and
secondly to cross-compare the results obtained using different techniques. Finally, different SA
techniques will be used to analyse PA results obtained by different partners (number of parameters
ranging from a few to some a hundred).
The structure of this paper is as follows. In section 2 we review most important screening and
global methods and discuss about their advantages and disadvantages. Section 3 is dedicated to describe
the Sensitivity Analysis Benchmark developed under PAMINA, the work under development
and expected results. The last section contains discussion and conclusions
2. Review of sensitivity analysis methods
A PA model typically involves several hundred input parameters, an important fraction of whom
are uncertain, thus a joint probability density function is needed to characterise their uncertainty and
the possible dependence structure among them. If all inputs are independent, the individual (marginal)
probability density functions (pdfs) are enough in order to characterise such uncertainty. The
use of the Monte Carlo method allows mapping the input space onto the output variable space and
estimate consequences.
In addition to characterising as accurately as possible the consequences associated to a repository,
which is the target of an uncertainty analysis, identifying the most relevant input parameters, whatever
this means, is a key task in a PA. A real problem arises when we ask for ‘relevant’ or ‘important’
input parameters: the interpretation of these words; what means ‘relevant’, ‘important’ and
similar words? An input parameter can be considered important with respect to a given output variable
if a strong correlation exists between both (linear relation), but it could also be considered so if
the output takes remarkably high values when the input takes values in a given region, or if that input
contributes a large fraction of the output variance (considered as a measure of uncertainty).
Another issue that arises in the SA area is the study of interactions. We say that two input parameters
interact when the joint effect of both is different from the addition of their individual effects
(interactions of order 2). This concept is naturally extended to any number of input factors. In general,
main effects (individual effects of each input parameter) are more important than second order
interactions, second are more important than third order interactions and so on, though this is not
always true. Interactions deserve to be studied in order to know the true structure of the system
model under study. Not all SA techniques are able to study interactions and in some cases, though
they are able, the study could be impractical due to different reasons (extreme computational cost,
too large diversity of possibilities, etc.)
The existence of different interpretations of importance have triggered the development, over the
last twenty-five to thirty years, of a variety of SA methods designed to study the model from different
points of view, each one developed according to each given interpretation. Nowadays a large
corpus is available to the SA practitioner, who may choose appropriate methods to perform a specific
type of SA attending to his/her interests and needs. PAMINA pays special attention to screening
and global methods, which are further explained in the next sections.
388

2.1 Screening methods
The target of screening methods is to identify at a low computational cost (hopefully) non-relevant
input parameters. This allows model simplification and all the benefits that this provides (conceptually
simpler models, less computational cost, less input parameter characterisation efforts, etc.).
Screening methods focus on the functional relation between inputs and outputs disregarding input
parameter distributions, but paying attention to the ranges of each input.
The origin of screening methods for computer codes is in the area of Design of Experiments, which
is the branch of Statistics dedicated to study the structure of systems (natural, industrial, commercial,
etc.) whose performance depend on a set of controlled factors and another set of not known
factors, which introduce variability (noise) in the response, see Box and Draper (1987). The oldest
and most simple method consists in fixing one point (typically a middle point, though not necessarily)
and modifying one by one each input, recording the results. This method is quite inefficient
when the number of important factors is small and is completely useless to detect interactions,
which needs additionally the simultaneous variation of both input parameters, unless something else
is done.
Typically, several levels may be defined for each input, those levels are crossed in all possible ways
and the output is computed for each possible crossing (factorial design). This strategy is completely
useless for models with many input parameters due to the huge sample size needed. A possibility is
to reduce the number of levels to two (low and high levels respectively) per input parameter. This
strategy (full two-level factorial design) is still too expensive (2 k runs needed) when the number (k)
of input parameters is large (1048576 runs needed to analyse a 20-input parameter model). A strategy
to reduce the computational cost, paying the price of not being able to estimate some interactions
or even main effects, is to take only a convenient fraction of the full design (fractional factorial
design or 2 k-p design). Saturated designs allow to study a number of input parameters with
number of runs = number of parameters+1, but they do not exist for any number of input parameters.
Supersaturated designs need fewer runs than the number of parameters. Among them the most
used ones are the Iterated Fractional Factorial Design, and the Sequential Bifurcations, both are
group-screening methods.
Morris Factorial Sampling method, see Morris (1991), is becoming more and more popular among
SA practitioners as a screening tool. This is a One-At-a-Time design (OAT). It consists in taking a
number of levels per input factor and a number of trajectories randomly generated. Trajectories start
at different points chosen at random and are built by successively selecting at random one of the
inputs and moving it to one of its possible next levels. These trajectories are used to estimate the
mean value and the standard deviation of each main effect. A high estimated mean main effect indicates
that the input parameter is important; a high estimated standard deviation indicates important
interactions of that input parameter.
2.2 Global methods
Global methods pay attention to how the whole input space maps onto the output space, taking into
account the input distributions. We may classify global methods as Monte Carlo based methods,
variance based methods and graphical methods. Monte Carlo based methods do not need a specific
sampling plan to be applied, a normal sample used to propagate uncertainties in a PA obtained via
Simple Random Sampling (SRS) or Latin Hypercube Sampling (LHS) may be used to compute the
corresponding sensitivity indices. Variance based methods study the contribution of each input parameter
and its related interactions to the output variance. Variance based methods usually need a
389

specific sampling strategy, with the only exception of the simplest method (correlation ratios - CR).
Graphical methods provide complementary visual information that helps understanding the meaning
of numerical sensitivity indices and the global structure of the system model.
2.2.1 Monte Carlo based methods
Monte Carlo based methods may be divided in three types: regression/correlation based methods,
Monte Carlo Filtering and gridding methods. Regression/correlation methods assume that inputs
and outputs can have a linear relation. The simplest method is Pearson’s correlation coefficient.
This is a measure of linear relation that takes values between –1 and +1. Positive values indicate
joint linear increase or decrease, negative values indicate that when the input increases the output
decreases and vice versa. The closer the value is to +1 or to –1, the stronger the relation is, thus the
more important the input parameter is. An alternative related measure of sensitivity is the slope or
regression coefficient of the output versus the input (after standardisation of the sampled values,
subtracting the corresponding sample mean and dividing by the corresponding standard deviation,
in order to avoid scale effects in the sensitivity indices). When the importance of several inputs is
analysed at the same time, the tool used is multiple regression together with input and output standardisation.
In this case, the indices used are the Partial Correlation Coefficients (PCC) and the
Standardised Regression Coefficients (SRC). These methods become really important when the inputs
are correlated, otherwise they are respectively exactly the same as Pearson’s correlation coefficient
and regression coefficients in simple regression. The importance of PCCs and SRCs comes
from the fact that they measure respectively the correlation and the regression coefficient between
one input and one output after removing the influence of all the other inputs. Unfortunately, by default,
these indices are used to analyse only main effects, interactions are hardly ever considered in
the analysis. Additionally, hardly ever practitioners study possible transformations of inputs and
outputs to get a more appropriate regression model.
In many cases, instead of linear, the relation between inputs and outputs is monotonic, in those
cases it is convenient to transform inputs and outputs into their ranks, the largest sampled value is
transformed into n (sample size), the second largest one into n-1,…, the smallest into 1. This way,
the same tools (renamed as Partial Rank Regression Coefficients -PRCC- and Standardised Rank
Regression Coefficients –SRRC-) may be used to assess the importance of each input parameter.
The reliability of the results obtained via linear regression depends on the Coefficient of Determination
(R 2 ) of the regression model obtained. If R 2 is close to 1 the results are very reliable, if it is
close to 0, it means that this sensitivity method is not appropriate to study the system model at hand.
Monte Carlo Filtering (MCF) is based on dividing the output sample in two or more subsets according
to some criterion (achievement of a given condition, exceeding a threshold, etc.) and testing if
the inputs associated to those subsets are different or not. As an example, we could divide the output
sample in two parts, the one that exceeds a safety limit and the rest. We could wonder if points
in both subsamples are related to different regions of a given input or if they may be related to any
region of that input. In the first case knowing the value of that input parameter would be important
in order to be able to predict if the safety limit will be exceeded or not, while in the second case it
would not be. The tools used to provide adequate answers to this type of questions are a set of parametric
and non-parametric statistics and their associated tests, as for example the two-sample t
test, the two-sample F test, the two-sample Smirnov test, the k-sample Smirnov test, the Cramervon
Mises test, the Wilcoxon test (also known as Mann-Whitney test) and the Kruskal-Wallis test,
see Conover (1980) for details about each specific statistic and test. In general, non-parametric tests
should be preferred for fewer restrictions are imposed on the samples used. The idea behind Gridding
and the tests used are similar to Monte Carlo Filtering. The only real difference is that the cri-
390

teria to divide a sample in two or more parts are set on the input space and the test is performed using
the corresponding points in the output space.
2.2.2 Variance based methods
The variance, or equivalently the standard deviation, and the entropy, are the main measures of uncertainty
in the theory of Probability. The larger the variance of a random variable is, the less accurate
our knowledge about it is. Decreasing the variance of a given output variable is quite an attractive
target, that may be achieved sometimes by decreasing the variance of input parameters (this is
not always true, remember the possibility of risk dilution). This is what makes so attractive methods
that try to find out what fraction of the output uncertainty (variance) may be attributed to the uncertainty
(variance) in each input.
Variance based methods find their theoretical support in Sobol’s decomposition of any integrable
function in the unit reference hypercube into 2 k orthogonal summands of different dimension: the
mean value of the function, k functions which depend each one only on one input parameter, k(k-
1)/2 functions that depend only on two input parameters, k(k-1)(k-2)/6 that depend only on three
input parameters and so on. Replacing any output variable of the system model (our function) by its
Sobol’s decomposition in the integral used to compute its variance produces in a straightforward
manner the decomposition of the variance in its components. The quotient between each component
of the variance and the total variance provides the fraction of the variance attributed to each single
input parameter (main effects), each combination of only two input parameters (second order interactions)
and so on. These are called Sobol’s sensitivity indices; see Sobol (1993). It is important to
remark that Sobol’s decomposition is equivalent to the classical Analysis of Variance (ANOVA)
used in Statistics.
Several algorithms have been proposed to compute Sobol’s indices, the first one by himself. The
main problem is related to the efficiency of the method. It needs one specific sample to compute
each sensitivity index. Since its development, huge efforts have been done to improve the strategies
(algorithms) to compute Sobol’s indices, see for example Saltelli (2002) and Tarantola et al. (2006).
It remains as a powerful but expensive method (in terms of computational cost).
Independently, and quite before the development of Sobol’s decomposition and Sobol’s indices, a
method had been developed to compute first order sensitivity indices (equivalent to first order
Sobol’s sensitivity indices): the Fourier Amplitude Sensitivity Test (FAST), see Cukier et al.
(1973), Schaibly and Shuler (1973) and Cukier et al. (1975). In order to compute sensitivity indices,
these authors create a search curve that covers reasonably well the input space. Each input parameter
is assigned an integer frequency. Varying simultaneously all input parameters according to that
set of frequencies generates the search curve. Equally spaced points are sampled from the search
curve and used to perform a Fourier analysis. The coefficients corresponding to the frequency (and
its harmonics) assigned to each input parameter are used to compute the corresponding sensitivity
index. Saltelli et al. (1999) did further improvements of the method, among them the possibility of
computing total sensitivity indices for a given input parameter (the fraction of the variance due to it
and all its interactions of any order). FAST remains unable to compute sensitivity indices for interactions.
Correlation Ratios are an alternative to Sobol’s method and FAST to compute first order sensitivity
indices using a normal sample (SRS, LHS, etc.). So, though a method used to compute variance
based sensitivity indices, it could also be considered Monte Carlo based.
391

2.2.3 Graphical methods
Let us call X=(X1,X2,…,XK) the vector of input parameters and Y to a given scalar output variable.
For a given input Xi, the scatter-plot is the projection of the sample points (X,Y) on the (Xi,Y)
plane. This representation allows the examination of the dependence between Y and Xi. Scatterplots
are very helpful to identify linear relations, monotonic relations and the existence of thresholds
among other potential trends. A wise use of transformations (e.g. logarithmic, ranks, etc.) may
also provide a lot of information about input/output relations. They may be used as supporting material
to explain the results obtained by means of numeric sensitivity techniques, but also to prevent
the use of inadequate techniques.
Three-dimensional scatter-plots or XYZ plots show the projection of the sample points (X,Y) on the
(Xi,Xj,Y) space. The information they are able to provide is also valuable. The extraction of such
information is limited, though challenging, due to obvious interpretation problems when a 3-D figure
is shown on a 2-D display. Software packages that allow changing the angle of the view may
enhance and broaden their applicability.
Extensions of scatter-plots are matrices of scatter-plots and overlay scatter-plots. Matrices of scatter-plots
show simultaneously, under a matrix format, the scatter-plots of different pairs of input
parameters/output variables. They allow identifying quite quickly the pairs with most remarkable
relations, but they are also affected by the loss of accuracy due to including several plots in a reduced
space, typically a fraction of a page. Overlay scatter-plots allow showing on the same plot the
scatter-plot of one output and several inputs. In order to distinguish the points corresponding to different
inputs, different symbols (dots, circles, crosses, diamonds, etc.) and different colours are
used. Frequently only a few inputs may be represented due to either the different scales used in the
plot or to the difficulties to interpret correctly so many overlapped different symbols.
Cobweb plots have been designed to show multidimensional samples in a two-dimensional graph,
see Cooke and Van Noortwijk (1999). Vertical parallel lines separated by equal distances are used
to represent the sampled values of a given number of inputs/outputs, usually not more than ten or
twelve, in order to keep the plot sufficiently clear. Each vertical line is used for a different input/output
and either the raw values or the ranks may be represented (either raw values or ranks in
all lines, never mixed). Sampled values are marked in each vertical line and jagged lines connect
the values corresponding to the same run. Coloured lines can be used to display the different regions
of any input parameter or output variable. Moreover, flexible conditioning capabilities enable
an extensive insight into particular regions of the mapping. The cobweb plots are usually provided
with ‘cross densities’ showing the density of line crossings midway between the vertical axes.
Therefore, an informed and careful analysis of cobweb plots enables the characterisation of dependence
and conditional dependence.
The Contribution to the sample mean plot (CSM plot) represents the fraction of a given output variable
mean that is due to any given fraction of smallest values of any input. This is obtained by putting
on the x axis the values of the Empirical Cumulative Distribution Function (ECDF) of any input
parameter (this values are 1/n, 2/n,…, 1 for any sample of size n of any continuous random
variable), and putting on the y axis the fraction of the output variable sample mean corresponding to
the smallest value of the input parameter, to the two smallest values, and so on. This way, a monotonic
non-decreasing curve is obtained. Plotting the ECDF in the x axis means that equal lengths
represent approximately regions of equal probability of the input parameter. The more the plot deviates
from the diagonal in a given region, the more (or the less) that region of the input variable
contributes to the sample mean or the sample variance. In fact, non-important input variables pro-
392

duce plots close to the diagonal, since large and small output values can be equally found in any of
their regions. Additionally, since the values represented on the x axis are independent of the input
parameter, many such curves may be plotted in the same graphic corresponding to one output variable
and many input parameters. Additionally, a test has been developed to check if deviations from
the diagonal are statistically significant or come from pure statistical randomness, see Bolado et al.
(2008). Figure 1 is an example of a CSM plot for one output and nine inputs (model used in step 2
of the benchmark on SA techniques). Only inputs W and V (1) show statistically significant deviations
from the diagonal. Small values of W are related to large values of the output while the largest
values of the output are related to intermediate values of V (1) . For all the other inputs deviations
from the diagonal are not statistically significant (large and small values of the output may be obtained
in any region of those inputs).
Figure 1: Contribution to the sample mean plot for the dose due to 129 I at 9·10 4 y (output variable)
and all the input parameters in the model used in the second step of the benchmark on SA techniques.
Other graphic tools, as for example the radar plots or the tornado bars, may be used to visualize, in
a comparative manner, the value of a given sensitivity index (for example the Pearson correlation
coefficient) for a given output and all or a fraction of the input parameters. But these are completely
different tools for they show sensitivity indices instead of a representation of the sampled values.
393

3. The sensitivity analysis benchmark
Many PAMINA partners were interested in getting in depth knowledge and experience in the use of
SA techniques. This was the reason to set a benchmark on SA techniques that could help them getting
that objective. The SA benchmark has been designed as a two-step process. The first step is
dedicated to analyse a set of mathematical functions most of whose sensitivity indices are well
known. The second step consists in analysing a simplified, though representative, PA model. Finally,
though not included as a part of the benchmark but as other tasks committed within PA-
MINA, several partners will study their respective system models, or parts of them, using SA techniques.
The first part of the benchmark consists in studying twelve mathematical models (seven mandatory,
five voluntary) with different SA techniques (the techniques are chosen by each participant, attending
to their respective interests). The first target is to provide the right simple framework to test and
debug their respective SA tools. The second one is to start with rather simple models (three input
parameter linear model) and to continue analysing models with some added complexity: increase
the number of parameters, add nonlinearities, consider non-monotonic models, include periodicity,
consider continuous models whose derivative does not exist at some given points, consider models
with interactions, check the different capability to estimate accurately large and small sensitivity
indices, etc. An additional target is to see the importance of the sample size in the accuracy of the
sensitivity indices.
The model under study in the second step of the benchmark reproduces the behaviour of a radioactive
HLW repository and the contaminant disposed of. Only four radionuclides are considered in
this model, 129 I and the decay chain 237 Np, 233 U and 229 Th. The repository is considered without any
geometric complexity, just a point. Engineered barriers are modelled through a containment time
during which there is no release. After such containment period, the contaminant starts releasing at
a fractional constant rate (one rate for Iodine, a different one for the chain members). The contaminant
is carried by groundwater through two consecutive geosphere layers to the biosphere, where it
gets into a water stream from which exposed population take drinking water. This model has thirtythree
inputs, twelve of which are affected by uncertainty. These model inputs are the initial inventory
of each considered radionuclide, their decay rates (�), their dose conversion factors (�), and all
the other inputs that characterise the physical-chemical properties of the near field, both geosphere
layers and the biosphere. The complex input-output relation, characterised by strong interactions
among input parameters, makes it a challenging model to test SA techniques.
In this case, the target is twofold: firstly to compare different options within a given SA techniques
(to study the added value of using more complex versions of a given technique – classical FAST
versus extended FAST, first order regressions versus higher order regressions, the benefits of using
transformations of inputs and outputs in the SA, the added value of and the problems that arise
when estimating the effect of interactions and total sensitivity indices, etc.); secondly to crosscompare
the results obtained using different techniques. This cross-comparison may be studied
from two points of view: what differences arise when using techniques that target different objectives
and why; i.e.: PCC and FAST, what differences arise when using techniques that target equal
objectives; i.e.: FAST and Sobol’s indices. Additionally, the effect of the sample size is also in the
focus of the second step of the benchmark.
Many PAMINA partners are interested in bringing to their PA studies all the knowledge and experience
acquired during the benchmark. PAMINA RTDC2 and RTDC4 offer a very good opportunity
to test all those techniques in real PA models proposed by several participants. Table 1 shows a
394

snapshot of the PA models and SA methods to be tested as a final step in the SA related activities to
be performed. This table shows that a very good coverage of code properties and techniques is going
to be achieved (modelling dimension 1D-2D-3D, number of input parameters from a few to approximately
100, almost all SA techniques tested).
Table 1. Partners, PA code properties and SA methods to be tested under PAMINA RTDCs 2 and 4.
Partner PA case PA model/ computational
cost
GRS-B/ Rock salt EMOS-LOPOS
JRC-Petten dome TSS 30 min/run
Indurated
clay
GRS-K Iron ore
mine
EMOS-CLAYPOS
TSS
30 min/run
NAMMU-2D. Geosphere
hydrology
(saturated conditions)
ENRESA/ Granite GoldSim TSS
JRC-Petten
500-1000 run/day
Plastic clay GoldSim TSS
500-1000 run/day
NRG/ Salt EMOS-ECN
JRC-Petten
A few seconds/run
Soft clay EMOS-ECN
A few seconds/run
Facilia Granite KBS-3 (focused on
biosphere)
ANDRA/ Indurated Alliance platform,
JRC-Petten clay 2-D
Indurated
clay
Alliance platform,
3-D
N. of input SA methods SA soft-
parameters
ware
6 Regression based SimLab
All inputs (linear & rank (JRC-Ispra)
independent based)
JRC-Petten
FAST/EFAST
Smirnov
software

4. Discussion and conclusions
Three main activities are currently developed under PAMINA in the area of SA: A review of SA
methods, a benchmark on SA methods and their application to analyse several PA models proposed
by different participants. The review of SA techniques consists in an exhaustive search of methods
of interest, screening and global methods, collecting information about its theoretical foundations,
and about the interpretation of ‘sensitivity’ behind them. Special attention is paid to the expected
benefits of using them and their shortcomings. PAMINA partners have shown more interest on
global methods than on screening methods. Among global methods, Monte Carlo based methods,
and more specifically regression based methods (linear and rank based) are among the most popular;
although not very powerful, they do not need specific sampling and may be used under the most
popular Monte Carlo techniques (SRS and LHS) used in a normal uncertainty analysis. Most powerful
techniques, such as Sobol indices or FAST, need specific samples and are quite expensive in
terms of number of runs needed. Most current research efforts are devoted to develop less expensive
computing algorithms.
Finally, the benchmark on SA methods is giving our group a good opportunity to test most interesting
methods and learning about them.
References�
[1] Bolado, R., Castaings, W. and Tarantola, S. (2008). Contribution to the Sample Mean Plot
for Graphical and Numerical Sensitivity Analysis. Submitted to Reliability Engineering and
System Safety.
[2] Box, G.E.P. and Draper, N.R. (1987). Empirical Model Building and Response Surfaces.
Wiley Series in probability and Mathematical Statistics. John Wiley & Sons, Inc.
[3] Conover, W.J. (1980). Practical Nonparametric Statistics. Second edition. Applied Probability
and Statistics. John Wiley and Sons, Inc.
[4] Cooke, R.M. and Van Noortwijk, J.M. (2000). Graphical Methods. In ‘Sensitivity analysis’,
Saltelli, A., Chan, K. and Scott, E.M. (Editors). Wiley Series in probability and statistics.
John Wiley & Sons Ltd.
[5] Cukier. R.I., Fortuin, C.M., Shuler, K.E., Petschek, A.G. and Schaibly, J.K. (1973). Study of
the Sensitivity of Coupled Reaction systems to Uncertainties in Rate Coefficients. I. Theory.
Journal of Chemical Physics, Vol. 59 (8), 3873-3878.
[6] Cukier. R.I., Schaibly, J.K. and Shuler, K.E. (1975). Study of the Sensitivity of Coupled Reaction
systems to Uncertainties in Rate Coefficients. III. Analysis of the Approximations.
Journal of Chemical Physics, Vol. 63 (3), 1140-1149.
[7] Helton J.C., Johnson, J.D., Sallaberry, C.J. and Storlie, C.B. (2006). Survey of Samplingbased
Methods for Uncertainty and Sensitivity Analysis. Reliability Engineering and System
Safety, Vol. 91, 1175-1209.
[8] Morris, M.D. (1991). Factorial Sampling Plans for Preliminary Computational Experiments.
Technometrics, Vol. 33, Number 2, 161-174.
[9] Saltelli, A., Tarantola, S. and Chan, K. (1999). A Quantitative, Model Independent Method
for Global Sensitivity Analysis of Model Output. Technometrics, Vol. 41, Number 1, 39-56.
[10] Saltelli, A. (2002). Making Best Use of Model Evaluations to compute Sensitivity Indices.
Computer Physics Communications, Vol. 145, 280-297.
[11] Schaibly, J.K. and Shuler, K.E. (1975). Study of the Sensitivity of Coupled Reaction systems
to Uncertainties in Rate Coefficients. II. Applications. Journal of Chemical Physics, Vol. 59
(8), 3879-3888.
396

[12] Sobol, I.M. (1993). Sensitivity estimates for Nonlinear Mathematical Models. MMCE, Vol.
1, No. 4, 407-414.
[13] Tarantola, S. Gatelli, D. and Mara, T.A. (2006). Random Balance Designs for the Estimation
of First Order Global Sensitivity Indices. Reliability Engineering and System Safety,
Vol. 91, 717-727.
397

398

CARD – Proposed European Technology Platform For The Co-ordination Of
RD&D For Geological Disposal
Alan Hooper 1 , Julio Astudillo 2 , Wernt Brewitz 3 , Monica Hammarstrom 4 , Lawrence Johnson 5 ,
Metka Kralj 6 , Philippe Lalieux 7 , Patrick Landais 8 , Irena Mele 6 , Gerald Ouzounian 8 , Marjatta
Palmu 9 , Frantisek Woller 10 , Juhani Vira 9
Summary
1 NDA-RWMD, United Kingdom; 2 ENRESA, Spain; 3 GRS, Germany
4 SKB, Sweden; 5 Nagra, Switzerland; 6 ARAO, Slovenia
7 ONDRAF/NIRAS, Belgium; 8 Andra, France; 9 POSIVA Oy, Finland
10 RAWRA, Czech Republic
The aim of the CARD project was to assess the feasibility of a European Technology Platform
(TP) that would provide a framework for networking and co-operation in the field of RD&D
for geological disposal of radioactive waste in the EU. The project sought inputs, principally
through a questionnaire and a subsequent open workshop, from radioactive waste management
organisations (geological disposal implementers), research organisations, regulators, local
communities and other stakeholders as prospective participants in a TP. An analysis of the
responses enabled the identification of the general benefits or objectives required by prospective
TP participants.
The proposed structure and working methods of a TP that would deliver these benefits and objectives
have been developed and consulted upon. A step-by-step implementation plan has
been outlined where the next step is the drafting of a Vision Document for the TP that will
provide the basis for organisations to commit to participation. Once the participants are established,
they will generate a Strategic Research Agenda defining the priorities for RD&D to
support the successful implementation of geological disposal in EU Member States.
1. Introduction
The aim of the CARD Project was to assess the feasibility of a Technology Platform (TP) 5 that
would provide a European framework for networking and co-operation in the field of RD&D for
geological disposal of radioactive waste in the EU, see reference [1]. Under the EC contract, the
study collected inputs from radioactive waste management organisations (geological disposal implementers)
and other potential participants in a TP, comprising research organisations, regulators,
local communities and other stakeholders. The project partners then analysed these inputs and, finding
there is a sufficient level of support (meaning coherent support for a common proposal), developed
the basis for a proposal for a TP.
5 The ‘Technology Platform’ is an instrument devised by the European Commission to provide a framework for coordination
of R&D activities in key technical areas with a view to assisting Europe to compete efficiently in the
development of advanced and complex technologies, e.g. see http://cordis.europa.eu/technologylatforms/home_en.html.
399

2. Methodology
The project participants met and discussed the objectives, structure and working methods of
a TP on two occasions (November 2006 and May 2007). They prepared a preliminary vision
for a TP and a detailed questionnaire on that vision. The questionnaire was responded to by
82 national organisations, including Formal National Appointees 6 , research providers, regulatory
bodies, safety authorities and other stakeholders.
The project participants judged that responses to the questionnaire did indeed demonstrate a
sufficient level of support for a European TP in the field of RD&D for geological disposal of
radioactive waste. Therefore they developed a draft proposal for the TP on the basis of their
analysis of the responses to the questionnaire.
The organisations that had been invited to respond to the questionnaire were invited to an
open workshop held in Brussels in March 2008 to share the findings of the Project and to
give feedback on the proposals for the structure and operation of the TP. The workshop was
attended by 54 participants from 11 countries. The high level of support for the proposed TP
was confirmed and a high proportion of the participants contributed ideas on how the TP
could be established and operated to meet the overall objective (of more efficient implementation
of geological disposal in EU member states) and the associated needs of national programmes
and of individual organisations.
The Discussion section of this document provides a proposal for a TP, based on the preliminary vision
developed by participants in the CARD project that was further developed to take account of
responses from the questionnaire and of feedback from the open workshop.
3. Results
All 10 organisations that have participated as partners in the CARD Project support the concept of a
European TP in the field of RD&D for geological disposal of radioactive waste and believe it could
have important benefits. Primarily, the benefits are:
Increased confidence in the scientific and technical basis of geological disposal as a safe and
feasible solution – provided by a coherent scientific and technical effort, through collaboration
of WMOs and research providers (both TSOs and non-TSOs), and possibilities for other
stakeholders to witness and influence that effort.
Economic – through sharing costs of projects that address common RD&D goals and/or
through better co-ordination of existing and future projects.
Reservations of the WMOs participating as partners in the project are:
Cost and staff resource based – that participating in a TP must not impose a significant additional
administrative load on organisations or their contracted research providers.
6 “Formal National Appointee” is a term used by the CARD project. It means an organisation that has been formally
appointed by government, often under national legislation, or otherwise entrusted with the responsibility either for
managing the development and/or implementation of deep geological repositories for radioactive waste in a given
country (the WMO), or for providing technical support including RD&D and/or safety assessment capability (a
Technical Support Organisation or TSO).
400

Direction and control based – that shared projects are aimed at common goals, and that key
RD&D resources directed to solving immediate issues within national programmes should
remain focused on those immediate issues.
Responses from the questionnaire showed as follows.
All 13 WMOs that responded expect to participate. One gave a specific reservation related
to control of key RD&D resources.
All 7 TSOs that responded also expect to participate. One gave a specific reservation related
to dependence on the attitude of its national WMO.
The 38 non-TSO research providers responding gave mainly positive responses with at least
65% expecting to participate and only one direct no. The main reservations were concerning
the resource or support to participate.
Among the 23 “other stakeholders” (i.e. regulators, government ministries and local community
organisations) there was general encouragement for a TP as positive to confidence
building. Many of the organisations, however, consider themselves “not competent” or interested
to participate in a primarily technical forum. Only 9 (43%) judged it was possible to
likely that they would participate. Regulators mentioned resources and independence. Social
stakeholders mentioned lack of social dimension.
Thus, while support for a TP is high in all three groups, direct participation from other stakeholders
– mainly ministry departments, regulators and social stakeholders – may be limited.
Feedback from the open workshop showed that there is a high level of support and interest with respect
to the proposed TP. In line with the analysis of the responses to the questionnaire there was
little participation from representatives of Government ministries (except in the case where the ministries
themselves hold the responsibility for implementing the national RD&D programme) and of
social stakeholders. However, the interests of social stakeholders were represented by participants
in national and international initiatives concerning, for example, waste governance and education
and training. Regulators and TSOs made proposals for the operation of the TP in a way that would
meet their needs and objectives while not compromising their independence.
A number of key points emerged in the course of the workshop and these are summarised as follows:
Greater clarity is required on the scope of the TP, i.e. what is included in geological disposal
RD&D.
The TP will represent a valuable source of guidance to the EC on the topics that should be included
in its Framework Programmes.
The TP represents a vehicle for co-ordinating education and training with respect to radioactive
waste management; interface arrangements are required with related initiatives in the nuclear
field. As such the TP represents an opportunity to ensure continuity in the expertise and
knowledge over the extended periods of time needed in the development and operation of a
disposal facility.
Knowledge management should be a highly prioritised activity for the TP, involving the commissioning
of books and reports on the state-of-the-art of relevant topics, effectively “handbooks”
for radioactive waste management.
401

Closely related to this proposal on knowledge management, the involvement of a wide range
of stakeholders, including social stakeholders, will enhance the value of knowledge management
initiatives and inform their objectives.
The value of a reference group for national programmes involved in defining or reviewing policy
was emphasised and the TP was proposed as a means of providing the necessary expertise,
when required in this role; the associated need for a method of communicating with the TP
was recognised.
The interfaces with other similar organisations, in particular the Sustainable Nuclear Energy
Technology Platform (SNETP), need to be defined clearly: in the case of the SNETP, it will be
important to specify the respective remits with respect to waste processing and packaging (i.e.
conditioning).
The TP will have an important strategic role to play, including the provision of inputs and
feedback to the European Parliament.
The scientific and technical community requires a network of personal contacts to operate effectively
in achieving the co-operation envisaged: the TP must facilitate such networks.
4. Discussion
Conditions for success of a European TP for networking and co-operation in the field of RD&D
for geological disposal have been identified and tested to the extent possible at this development
stage. They are as follows:
A shared vision of the TP by those participants having national programme responsibilities and
a willingness to support a common strategic research agenda (SRA), i.e. an agreed set of goals
for the RD&D most suitable for collaboration and needed to develop geological disposal to the
level of practical implementation, and agreed time scales for their accomplishment;
Sufficient authority and willingness of the WMO participants needed to commit resources to
projects;
Active and constructive support of all participants including a range of stakeholders;
Appropriate structure and working methods to realise the general objectives and specific project
goals efficiently.
Relevant aspects of the TP proposal are discussed in the following sub-sections.
4.1 General ground rules
The Technology Platform will be established and directed by the organisations that have national
programme responsibilities for commissioning and applying RD&D in implementing or planning
geological disposal, or in formulation of disposal policy (typically this will be a WMO) and be to
serve their needs. The EC will take an interest as an observer, offer advice in relation to its experience
of similar ventures and provide some support for coordination activities of the platform. Organisations
will decide for themselves whether or not to participate, and at what level of commitment,
depending on the benefits they see in participation and on their own resources.
The Technology Platform can begin as an information exchange and discussion forum and is expected
to develop as a vehicle for practical co-operation in specific RD&D projects. It is not intended
to duplicate existing discussion fora (e.g. as provided by the NEA and IAEA) or existing
402

multi-national or bilateral research agreements. It is expected that these latter agreements will be
built upon for the benefit of the TP and that the TP will also benefit from the structured dialogues
that essentially will continue to occur at the national level between WMOs and research organisations
with responsibility for each national programme. Rather, the TP is to help identify RD&D
needs that are common to at least some of the participants to offer practical solutions by which
interested participants can co-operate in meeting those needs, and to provide a platform for open
discussion and exchange of RD&D results.
It is not expected that participants will surrender control of their RD&D resources, rather, where
there is a joint benefit, they will pool parts of their resources with others for the purposes of specific
projects with joint agreed goals and timescales.
Questionnaire responses confirm that only a few per cent of funding for RD&D related to radioactive
waste management comes via the EC and the bulk is committed directly by the WMOs for
their national research programmes. An important proposed aim of a TP is to co-ordinate shared
objectives and projects in the work programmes at the command of WMOs in topics and areas
where a joint benefit of co-operation is seen. A subsidiary element in promoting such co-operation
is that views expressed in the SRA and the direction of projects within the Technology Platform
would be a valuable source of guidance to the EC in setting priorities in its Framework Programmes.
This information coming out of the TP will be of value to focus support to the implementation
of geological disposal in EU–member countries by identifying areas of highest added
value to implementation by European cooperation
4.2 Benefits and objectives
The following general benefits or objectives have been identified, which a TP should seek to realise:
Gaining understanding of who is doing what RD&D and for what reasons, and thus to learn
each others’ planning strategy and underlying structure for planning RD&D activities and organising
information (e.g. requirements management, knowledge management, strategic resource
management).
For advanced national programmes, supporting the implementation process (and strengthening
the foundation of repository safety cases) through discussion on key issues and formulation of
focused and efficient RD&D responses, also taking account of views from regulators and other
stakeholders. There is also the prospect of sharing resources to tackle issues that may not be
keys to implementing geological disposal but nonetheless need to be handled in national programmes
(e.g. ‘exotic’ wastes).
For less advanced national programmes, giving advance insight on future requirements
through the same processes and giving the opportunity to allocate resources to encourage early
solutions and follow developments.
Enhancing public acceptance and confidence through demonstrated openness of discussing
problems and the RD&D requirements, and developing broadly-based technical consensus on
the state-of-the-art of science and technology (as a part of knowledge management), allowing
objective identification of the uncertainties that still remain.
This demonstrates that establishing a forum and mechanisms for sharing of RD&D information and
results is the most highly regarded objective across all organisations. Establishing a forum for discussion
of RD&D issues and priorities amongst RD&D funders, managers and other stakeholders,
and establishing mechanisms for co-ordinating RD&D on topics of shared interest between programmes
is highly regarded by the formal national appointees.
403

4.3 Structure
The TP structure must allow a level of access to all committed participants – to allow open discussion
and exchange – but also provide a formal structure committed for efficient planning, management
and reporting of projects or activities.
The basic structure proposed for a TP includes:
A forum for exchange of information and discussion of RD&D needs, as well as results, in relation
to implementation of geological disposal.
A working programme controlled by an executive group that is supported by a secretariat.
Within the working programme would be:
Working groups with specified mandates related to the TP (e.g. development of the SRA, development
of supporting activities such as education and training).
Collaborative projects and activities following agreed work plans and objectives.
The structure must accommodate the needs and constraints of:
Organisations that are in charge of implementing disposal facilities and/or entrusted by their
Government with developing radioactive waste disposal solutions.
Research providers, with an interest in scientific co-operation as a means of providing input to
and gaining information from research programmes;
Other stakeholders with technical interests and concerns, for example, regulatory bodies, government
ministries and involved municipalities, with an interest in information from, and influencing,
European research programmes.
The proposed structure developed by the CARD participants is illustrated in Figure 1. This is a
simplified outline developed to promote feedback from interested parties at the open workshop.
The workshop participants were broadly satisfied that this is an adequate structure to act as a guide
for initial development of the TP. An early task for the executive group and secretariat would be to
review and add detail to this structure.
Fig. 1: Structure for the Technology Platform for RD&D for geological disposal.
A key issue in developing this structure is to determine the inputs that are required from different
types of organisations or organisations with different responsibilities, and how best these can be
made.
404

4.4 Working methods
The Strategic Research Agenda
A key vehicle of other European TPs is the “Strategic Research Agenda” (SRA). For geological disposal
RD&D this will be a document, arrived at by technical analysis and discussion between
WMOs and TSOs and other research providers, taking account of the views of other stakeholders. It
will lay out RD&D goals within the Technology Platform and time scales for their accomplishment.
It will form a focus for ongoing discussion and will be subject to review and on-going development.
In line with the planning that typifies national implementation programmes, it is suggested that this
could be structured around short-term, medium-term and long-term objectives that take account of
the planned implementation of geological disposal in a number of member states around 2020.
Once the SRA is agreed, it will form the high-level guidance for development of proposals and detailed
plans of work within the TP. In the context of a TP in the area of RD&D for geological disposal,
the SRA represents a shared view on the RD&D that is required in support of implementation
of geological disposal in Europe and where international co-operation will enable or improve its
quality or timeliness of delivery.
Dialogue and control
Dialogue would be generated primarily within the Exchange Forum, whereas control and monitoring
of activities is achieved under the supervision of the TP executive group and secretariat, see
Figure 1.
The Exchange Forum would use a range of methods to promote dialogue. These could include:
Operation of a website with information on the TP programme, access to results, and proposals
for review and comment (a pilot website www.cardproject.eu was developed within the CARD
project).
Meetings to discuss RD&D priorities, the SRA and the TP programme;
Workshops on specific RD&D topics or functions and support activities.
For management efficiency, and to ensure that the organisations with national programme responsibilities
for deploying research budgets (WMOs or their equivalent), retain control of their RD&D
resources, the implementation of projects within the TP would be controlled by an executive group
appointed by these organisations as a key part of establishing the TP. The appointment of the executive
group would be on the basis of technical competence, covering the complete spectrum of
their needs, and strategic-level management competence. This executive group, meeting regularly,
would commission working groups to develop both the overall SRA and to assess and make technical
plans for RD&D projects or development of TP functions. It would formally open RD&D projects
and activities, monitor their performance, and close projects and activities on reaching their
goals in accord with the SRA. It would develop reports on the activities and outcomes of the TP
primarily as an efficient means of providing information to its Exchange Forum and stakeholders. It
would actively seek views, and respond to views, developed by stakeholders, in particular within
the Exchange Forum.
The executive group would be supported in its duties by a Secretariat, which would also provide
support to activities of the Exchange Forum.
Individual RD&D projects and activities would be managed by management groups and methods
suited to their structure and objectives, and under the control of participants in the individual
RD&D projects and activities.
405

4.5 Implementation
It is proposed that the Technology Platform is implemented in a step-by-step manner.
Attention is first on obtaining the commitment of organisations to participate in the TP. In order to
achieve this it is proposed that a TP “Vision Document” is drawn for distribution to prospective
participating organisations for their consideration and signature.
The drafting of this document is an activity that will be undertaken on a free-will basis by organisations
that are already strongly committed to the prospective TP. The Swedish Nuclear Fuel and Radioactive
Waste Management Company, SKB, has expressed a willingness to lead this initiative,
two other organisations involved in the CARD Project, Posiva and GRS, have formally expressed a
willingness to support SKB, and a number of the participants in the open workshop notified that
their organisations could be approached for support also.
In order to maintain the momentum that has been developed with the CARD Project and to build
upon the good will and support shown by a number of organisations and stakeholders, it is agreed
among the CARD partners that a target of November 2008 should be in mind for completing the
Vision Document and obtaining a critical mass of organisations willing to commit to it.
Scope
Consistent feedback from the open workshop concerned the need for clarity on the scope of the TP
in the Vision Document. In the light of discussions on this matter within the Project, it is proposed
that the overall goal of the TP should be declared as practical implementation of member states’
policies on the geological disposal of radioactive wastes. Such a policy is in place in a number of
member states in respect of high activity, long-lived wastes and this is expected to be the focus for
the TP.
The term geological disposal is taken to mean disposal at depth in suitable geological formations
where the geology will contribute to the long-term isolation and containment of long-lived radionuclides.
Therefore disposal of low-level wastes and short-lived intermediate-level wastes at or near
the ground surface is not included in the scope of the TP. All types of potential host rock are to be
included, in particular the classical categories of crystalline rocks, argillaceous rocks (including
both indurated claystones and plastic clays) and evaporites (in particular rock salt)
The TP should include in its scope consideration of the conditioning of wastes to make them suitable
for disposal. The Vision Document will need to comment explicitly on the interface that will
be required with the SNETP to ensure that there is neither duplication nor significant omission of
important activities in this area between the two TPs.
Content of the Vision Document
The feedback from the open workshop confirmed that much of the information required to be presented
in the Vision Document is available in the material collected and analysed in the CARD Project.
In particular the strongly supported benefits and objectives of the TP are clearly identified.
The vision of the Technology Platform is to establish:
a forum for discussion of RD&D issues and priorities;
a means for sharing RD&D information and results, including information and experience on
RD&D planning and management;
a mechanism for co-ordinating RD&D on topics of shared interest between programmes and
group of organisations.
406

The Vision Document should contain proposals for achievement of these benefits and objectives. In
addition to the identified strategic benefits and objectives, the development of books and reports on
the state-of-the-art of relevant topics merits attention following the comments from the open workshop.
There appears to be general agreement on the principles underpinning the structure that has been
proposed for the TP and the experience in the CARD Project has been that it is not helpful to attempt
to be more prescriptive than is necessary for organisations to see how they would best participate
to meet their own objectives. It will be important to establish methods of working and the
development of personal networks and the capability to function as a reference group were specific
points supported at the open workshop.
As noted by participants in the open workshop, it will be important to state clearly the proposed interactions
of the TP respectively with national programmes, the EC, international organisations and
other TPs and European initiatives (e.g. concerning education and training in the nuclear field).
Some of the information proposed to be included in the future Vision Document will guide the
scope of a subsequent Strategic Research Agenda (SRA) and it is suggested that a proposed scope
of the SRA could be included in the Vision Document. However, it will be for the organisations
participating in the resulting TP to subsequently review and revise that scope and then develop the
detailed content of the SRA.
Planned Actions
Once the Vision Document has been finalised and a critical mass of organisations have signified
their commitment to supporting it, the TP should be launched at a workshop. This should be
planned and designed to attract the participation of key decision makers and senior managers to
emphasise the importance of the step that is being taken and the strategic aspects of the future operation
of the TP
It will of course be for the organisations participating in the TP to set priorities, but a clear early
priority will be to develop the SRA for review and subsequent agreement.
The EC has signalled willingness in principle to support the provision of a secretariat at the early
stage of operation of a TP. It is a prerequisite that this possibility should be pursued but it is also
considered that the TP should develop a resource plan to ensure its sustainable operation over the
long term. This should be done as soon as possible once the benefits of participation are apparent.
Taking the Sustainable Nuclear Energy Technology Platform (SNE-TP) as an example such initial
secretariat support may be the equivalent of 2 person-years
As implied by the term “Vision Document”, initially participating organisations will be committing
to a vision of what the TP will achieve. However, co-operation in projects will demand commitments
of resources and considerations of issues such as intellectual property rights and liabilities.
The type of consortium agreement that is often associated with EC Framework Programme projects
is recommended as a tried and tested model for use by the participants in specific co-operation projects,
not least because the legal departments of most of the likely participants are already familiar
with its use.
5. Conclusions
407

5.1 The CARD Project has shown that a Technology Platform is a feasible method of providing a
framework for networking and co-operation in the field of RD&D for geological disposal in the EU.
In particular the proposed structure and methods of working can meet the identified requirements
for networking and co-operation of those organisations that are central to implementation of geological
disposal in Member States.
5.2 The CARD Project has established and tested the prioritised needs and objectives of potential
participants in the Technology Platform. The resulting database of information provides the basis
for production of a Vision Document for the Technology Platform.
5.3 There is a high level of support and good-will for the establishment of a Technology Platform
and momentum should be maintained by moving as quickly as possible to its launch.
6. Acknowledgement
This project has been co-funded by the European Commission and performed as part of the sixth
EURATOM Framework Programme for nuclear research and training activities (2002 – 2006) under
contract FI6W-CT-2006-036496.
Reference
[1] Sixth Framework Programme Co-ordination Action, Proposal 036496, Co-ordination of research,
development and demonstration (RD&D) priorities and strategies for geological disposal,
Annex 1 – Description of Work, 11 May 2006
408

Summary of the Panel Discussion on the Topic:
"Coordination of RD&D for Waste Disposal in Europe"
Panel members:
Alan Hooper (Chair), NDA-RWMD, United Kingdom
Jordi Bruno, Amphos XXI, Spain
Vit�zslav Duda, RAWRA, Czech Republic
Édouard Scott-de Martinville, IRSN, France
Peter Wikberg, SKB, Sweden
All five of the questions posed in the Conference Programme were addressed by the panel as follows:
� In the context of the shared vision on implementation of geological disposal in Europe, which
areas of RD&D in particular would benefit most from a “technology platform” approach (i.e.
close coordination between all key stakeholders, agreed strategic research agenda and deployment
strategy)?
One strongly held view was that the area that would benefit most would be research into physical
and chemical processes. It was argued that stakeholder acceptance would be improved significantly
if there was greater visibility of a shared view between programmes on what are the important processes
and how to address them. Continued research into such processes provides a valuable framework
for training and development of the early career professionals that will necessarily come into
the field of geological disposal if implementation is to be sustained over the relevant long timescales.
The existing bilateral cooperation between Sweden and Finland provides good examples of
both such types of benefits.
It was pointed out that there is significant interest in the benefits derived from coordination of technology
development in the FP6 ESDRED Integrated Project and that technology development projects
can provide helpful evidence of the practicality of establishing barrier systems to stakeholders.
� Do coordinated RD&D programmes produce some types of results more readily than standalone
national programmes, in particular in respect of regulatory confidence and technical
quality?
It was concluded that confidence is improved when there is evidence of overlap of key issues and
the means of addressing them between different programmes and that the technology platform approach
should improve visibility of such overlaps, in particular to the regulators.
It was pointed out that in many countries the level of acceptance is significantly improved in the
light of evidence of meaningful international cooperation on relevant RD&D topics. For this reason,
a number of regulatory bodies formally request information on their national programme’s international
cooperation programme.
409

� What steps should be taken to maximise the associated benefits from coordinated RD&D such
as training or information exchange?
The view was expressed that the experience gained from recent EC Framework Programmes in the
field of radioactive waste management science and technology has been very beneficial to training
and information exchange in European programmes. Therefore there is a successful model already
in existence that should be built upon. The main development that was urged was the integration of
engineering into the strategic research agenda of a future technology platform. This resonated with
the discussion in response to the first question.
� How can the benefits of coordinated RD&D be maximised for national programmes with relatively
small scientific and technical resources?
There was a clear statement on behalf of such national programmes that coordinated RD&D, such
as would be conducted in the framework of a technology platform, is vital to their success. Effective
communication of the motivation of the RD&D programme and of its outputs, to make this information
accessible to a wide range of stakeholders, will help to maximise the benefits.
� Are there reservations or constraints to bear in mind when considering coordination of strategic
RD&D?
The main issue identified was the importance of an RD&D activity to a key decision point in a national
programme. Whereas an activity might be highly suitable for inclusion in a coordinated programme
in a technical sense, the implementing organisation might legitimately require managing
the risk to the national programme by retaining total control over the activity to ensure delivery of
the required information on the required timescale. However, successful operation of a technology
platform in its early years might well give confidence that such programme risks could be managed
equally effectively in the resulting coordinated international programme and that the foreseen benefits
of involvement in the coordinated programme would outweigh the negligible risk from relinquishing
individual control.
410

Posters based on projects performed as part of Euratom FP6, ISTC and supported
by the EC-DG Energy and Transport (TREN)
411

412

Partitioning and Transmutation
413

414

Development of the Methods for Immobilization of Long-lived
Radioactive Waste in Carbon Matrices for Storage and Transmutation
Murat Abdulakhatov 1 , Sergey Bartenev 1 , Mikhail Goikhman 2 , Alexander Gribanov 2 , Valery
Guselnikov 3 , Nikolai Firsin 1 , John Krasznai 4 , Yuri Novikov 3 , Yuri Sazanov 2 , Mikhail Zykov 1
1 V.G. Khlopin Radium Institute (KRI), Saint-Petersburg, Russia
2 Institute of Macromolecular Compounds (IMC), Saint-Petersburg, Russia
3 Petersburg Nuclear Physics Institute (PNPI), Gatchina, Russia
4 Kinectrics Inc., Toronto, Canada
Summary
Conditions for immobilization of long-lived radioactive waste in carbon matrices were investigated.
Stable isotopes of rhenium, iodine and europium were used as chemical analogues of
low-lived radionucledes: 99 Tc, 129 I and 241 Am, respectively. It is shown that the carbon matrices
incorporating the above elements can be produced by carbonization of composites with
ITA-31 polyimide binder, chemical analogues of investigated radionuclides and carbon fabric
as reinforcing component. The elements under investigation were used both in the form of
salts or oxides and in the form of their complexes with ion-exchange resins. The produced
composites were carbonized in inert gas (argon) or in vacuum. The physical-chemical properties
of the samples were studied. It was revealed that the resultant matrices meet the requirements
imposed on waste storage, disposal and transmutation.
1. Introduction
Further development of nuclear power over the world depends to a great extent on solving two main
problems: safety of nuclear power and industrial facilities as well as ecologically reliable and economic
disposal of radioactive waste (RAW) containing long-lived alpha-, beta- and gamma radionuclides.
Hence, the long-term isolation of RAW from biosphere is an important problem responsible
for further development of nuclear power. Of prime attention is the question on management
of long-lived radionuclides because of their serious hazard if they escape into the environment.
The point is that the radioactivity of fission products ( 99 Tc, 129 I) and minor actinides such as
237 241
Np and Am remains practically at the same level over a period of thousand hundreds years.
Two variants are best suited to handle these long-lived radionuclides. One of them consists in
elaborating a method for immobilization of radionuclides into matrices of high thermal, chemical
and radiation resistance for subsequent long-term storage or final disposal [1]. Another variant is
transmutation of long-lived radionuclides into short-lived or stable isotopes in special nuclear reactors
or Accelerator Driven Systems (ADS) [2-3]. It should be emphasized that the both variants involve
production of matrices possessing some properties suitable for prolonged storage or disposal
and transmutation of long-lived radionuclides. The different research centres carry out research on
the development of new techniques for production of reliable matrices. One of such techniques consists
in production of carbon matrices. Carbon matrices can be obtained by carbonization of organic
substances. To meet the requirements placed on the properties of carbon matrices being obtained,
some composites on the basis of polyimide binder ITA-31 were developed [4]. By producing the
carbon matrices it is extremely important to bind nuclides being immobilized by ion-exchange resins
(IER) and complexing agents, which may be used then as components in synthesis of polymer
415

materials. For attaining the stated goal, different materials may be used. In this work several commercial
IER with high thermal, chemical and radiation stability were chosen. The conditions for
production of the carbon matrices incorporating rhenium, stable iodine and stable europium as the
chemical analogues of 99 Tc, 129 I and 241 Am, respectively, were investigated.
2. Methodology
2.1 Initial materials and reagents
Polyimide binder ITA-31 with the following composition was used: dianhydride of 3,3 / ,4,4 / -benzo-
phenonetetracarboxylic acid - 50% and tetraacetyl derivative of 4,4 / -diaminodiphenyl ether - 50%.
During composite synthesis the binder was mixed with IER containing rhenium in the form of perrhenate-ion
(ReO4 - ), iodine in the form of iodide-ion (I - ) and europium in the form Eu +3 . The
strongly basic anionites like AV-17, VP-1AP, VPB, Amberlite IRA-400 and Dowex 1×4 were used
for preliminary sorption of rhenium and iodine. For preliminary sorption of europium the Russian
IER SF-5 and KB-10 were applied. The carbon fabric of ELUR type was used in composite synthesis
as reinforcing addition for higher strength of matrices.
2.2 Synthesis of composites and their carbonization
Synthesis of composites was carried out in two stages [5]. At the first stage the composite-1 incorporating
ITA-31 binder and the IER containing a corresponding nuclide at the ratios of ITA-31:IER
8:1 ÷ 1:1 was synthesized. ITA-31 and IER were mixed at 260 – 280°�. Then the carbon fabric
ELUR was added to the resultant composite-1 for its higher strength to afford its mass content from
20 to 60% in end product (composite-2). The process for production and pressing of composite-2
was conducted at 280-320°� and under excess pressure of 20 – 250 atm., while the carbonization
process proceeded at 600 – 650°�.
2.3 Determination of leach rates
The procedures recommended by International standards were used for determination of leach rates
of rhenium, iodine and europium from obtained matrices [6-7]. ICP MS and emission spectral
analysis was used for determining I, Re and Eu leach rates.
3. Results
3.1 Thermal stability of composites
Taking into account that the radioactive waste containing long-lived radionuclides may be heatedup
due to radioactive decay, special attention should be given to thermal resistance of matrices used
for long-term storage or final disposal of RAW. In the previous works conducted by the authors it
was shown that the carbonization process of the composites containing polyimide binders is an efficient
method for production of heat-resistant carbon-plastics and carbon-carbon materials on their
basis [4]. However, it was unknown which properties would be typical for the composites based on
ITA-31 and other components of produced matrices. Such components may involve, for instance,
the IER containing the elements under investigation. In the early stage of investigations some experiments
to measure thermal resistance of initial components and related composites were performed.
In Fig. 1 the data are given of differential thermogravimetric analysis (DTA) of VP-1AP in
different chemical forms, including that containing the stable iodine as I - . For correlation purposes
there are also presented the DTA individual results on the ITA-31 and composite containing ITA-31
and VP-1�P in I - - form.
416

Figure 1 Results of DTA
1 – ITA-31; 2 – VP-1AP (OH –
); 3 – ITA-31+VP-1AP(OH –
); 4 – ITA-31+ VP-1AP(OH –
)+ELUR; 5 – VP-1AP (NO3
417
–
);6 – VP-1AP (I –
)
The thermal experiments have demonstrated that the IER containing chemical analogues of investigated
elements enter into chemical reactions with polyimide binders like ITA-31 and it is possible
in such a manner to produce the heat-resistant matrices for incorporation of the above elements in
the course of subsequent carbonization process. In the range of 600 - 1000°� no significant losses
of rhenium and europium are observed and only some marked losses of iodine occur due to lower
melting point of its compounds.
3.2 Synthesis of composites incorporating ITA-31 binder and elements under study
The technology for production of carbon-plastics elaborated in the previous works [4] involved the
ITA-31 synthesis, the procedure for impregnation of a required number of carbon-fabric layers, the
pressing at 300�� for 1 hour under excess pressure of 0.1 – 0.5 atm. The mass ratio between binder
and carbon fabric was 1:1.
The composite (prepreg) being converted into carbon-plastic state was then subjected to carbonization
in vacuum or inert medium in the temperature range of 20 - 600�� at a rate of 7 �/min. It should
be noted that both prepreg and carbon-plastic produced in accordance with this technology possess
high mechanical and thermal strength. The properties of the carbon-plastics produced by incorporation
of IER (for example VP-1AP) into their composition at various ratios between components are
shown in the Table 1.
As it is seen from the results obtained, the strength values of carbon-plastics remain rather high
over a wide range of ratios between ITA-31 and IER. These values meet the requirements of the
Russian and International standards for RAW incorporating matrices.
However, by using such carbon-plastic molding technique a partial loss of composite occurs due to
its flowing out of press mold. In production practice of carbon-plastics containing no harmful
chemical and radioactive substances these losses can be collected and returned into the process
without any radioactive substances these losses can be collected and returned into the process without
any technical difficulties. However, in the case of producing the carbon-plastics (matrices) containing
radioactive substances the composite flowing from press-mold may lead to non-controlled
contamination of equipment and work places.
Other variants of molding matrices were elaborated in this work, which could prevent the noncontrolled
losses of RAW during molding.
In the first case the preliminarily prepared prepreg containing ITA-31 and ELUR was crushed to
sizes of 5 – 10 mm and mixed on heating with the resin saturated by element being investigated. At

the bottom of the heated press-mold the PM film (polyimide film) was placed and above it the mixture
of crushed prepreg and resin. Pressing was conducted at 5 - 250 atm.
Another variant envisaged mixing of crushed carbon with the resin containing the elements under
study; then, molten binder was added on heating and stirring; the resulting mixture was transferred
into the heated press-mold. Pellets were pressed at 300�� for 1 hour.
It was found that among three variants of matrix production the third one is most feasible; in this
case no prepregs should be preliminarily prepared. This method allows to avoid the non-controlled
losses of investigated elements and to obtain sufficiently strong pellets. Besides, one can introduce
nuclides in a carrier (ion-exchange resin, fullerene soot) as well as in any other solid chemical form,
for example oxide.
Table 1. Properties of composites containing VP-1AP
Ratio of ITA-31: VP-1AP
Properties of �-� composite
Compression strength, MPa Modulus of elasticity, GP� Mass loss on carbonization, %
8 : 1 84 90 12
6 : 1 79 82 14
4 : 1 80 81 16
2 : 1 54 57 17
3.3 Leach rates of investigated elements
Alongside the thermal and strength characteristics the leach rates of immobilized radionuclides or
their chemical analogs are of significance in defining the chemical stability of matrices. In this work
the leach rate was determined in accordance with the recommendations of the International standards.
Distilled water was used as liquid phase; open surface area for different samples was 1.2 –
2.0 cm 2 . It is determined that the leach rates of rhenium from sample prepared on the basis of rhenium-saturated
resin VP-1AP are 6.8×10 -5 – 1.7×10 -7 g/cm 2 ×day upon leaching during 1 – 90 days;
of europium from sample prepared on the basis of SF-5 resin saturated with europium are 2.5×10 -5
– 1.6×10 -7 g/cm 2 ×day on leaching within the same time interval; of iodine from sample prepared on
the basis of AB-17 saturated with iodide-ion are 4.8×10 -5 – 2.7×10 -6 g/cm 2 ×day.
3.4 Radiation stability of matrices.
It is shown that irradiation of the matri�es by neutrons and gamma-irradiation in research reactor
VVR-M in PNPI at neutron-flux density ~ 4·10 n/cm 2 ·s and integral flow ~ 10 18 n does not render
the essential influence on their mechanical properties and leach rates of investigated elements. The
values of these parameters for the samples after irradiation practically such, either as for samples
before irradiation.
4. Conclusions
The chemical analogues of such long-lived radionuclides as 99 Tc, 129 I, and 241 Am were immobilized.
Rhenium, stable iodine and europium were used as the corresponding analogs, respectively. It
was shown that the samples carbonized at temperatures up to 600-650°� possess high thermal stability.
The produced samples exhibit high mechanical strength. Their compression strength varies
418

from 47 to 84 MPa, the modulus of elasticity from 49 to 90 GPa. The leach rates of rhenium, iodine
and europium for different samples range between 2.5×10 -5 – 1.7×10 -7 g/cm 2 ×day. On the basis of
the results obtained in this work strong matrices of high thermal, chemical and radiation stability for
disposal of long-lived radionuclides can be developed.
5. Acknowledgments
The work was carried out under financial support of International Science and Technology Centre
(ISTC), Moscow (Project #2391).
References
[1] «Management of radioactive waste and spent nuclear fuel», MINATOM, Information and
Analytical Collection, Moscow, 2000, issue 2, p. 90.
[2] A.Fernandez et al. "Fuel/Target Concepts for Transmutation of Actinides". Proc.
6 th OECD/NEA Information Exchange Meeting on Actinide and Fission Product P&T
(Madrid, Dec. 11-13, 2000).
[3] Michael W. Cappiello, Dana C. Christensen «The Role of LANSE in the Nuclear Energy Future»,
Los Alamos Science, Number 30, 2006
[4] M.Ya.Goykhman et al. "Study of the mechanism high-temperature curing of polyimide ITA
binders". Acta Montana, Series B , No 7 (1997) 9-19.
[5] «Method of immobilization for long-lived radionuclides», M.K. Abdulakhatov, S.A.
Bartenev, M.Ya. Goikhman et al. Patent application, Priority � 2007105656, G21F 9/16,
G21F 9/23, Russia
[6] ASTM Standard. C 1220-92, Standard Test Method for Static Leaching of Monolitic Waste
Forms for Disposal of Radioactive Waste
[7] ASTM Standard. C 1285-94, Standard Test Method for Determing Chemical Durability of
Nuclear Waste Glasses: The Product Consistency Test (PCT)
419

420

Near-field processes
421

422

Summary
TIMODAZ – Thermal Impact on the Damaged Zone around
a Radioactive Waste Disposal in Clay Host Rocks
Xiangling Li
EURIDICE, Belgium
A proper evaluation of the Damaged Zone (DZ) in the host formation is an important item for
the long-term safety of underground disposal of the spent nuclear fuel and long lived radioactive-waste.
The DZ is first initiated during the repository construction. Its behaviour is a dynamic
problem, dependent on changing conditions that vary from the open-drift period to the
initial closure period and the entire heating-cooling cycle of the decaying waste. The TIMO-
DAZ project focuses on the study of the combined effect of the EDZ and the thermal output
on the repository host rock. The influences of the temperature increase on the evolution of the
EDZ as well as the possible additional damage created by the thermal load will be studied.
Three types of clay are investigated in TIMODAZ project: the Boom Clay, the Opalinus Clay
and the Callovo-Oxfordian argilitte. New laboratory experimental equipments and test protocols
are conceived (WP3) and new In-Situ experiments in small and large scales (WP4) are
also designed to better understand the processes occurring within the clay around a disposal
system. All experimental researches are closely linked with the development and testing of
sound, phenomenology-based models which are essential in meeting the Safety Case requirement
of adequate understanding of the long-term evolution (WP5). Furthermore, all experimental
and modelling developments will be situated in the long-term performance assessment
contexts (WP6).
1. Introduction
Disposal of spent nuclear fuel and long lived radioactive waste in deep clay geological formations
is one of the promising options worldwide. In this concept, the host clay formation is considered as
a principal barrier on which rest the fulfilment of key safety functions. Hence, preventing unnecessary
damage to the host formation is one of the objectives of repository design. A proper evaluation
of the Damaged Zone (DZ) in the host formation is thus an important item for the long-term safety
of underground disposal.
In any case, the excavation process of the geological repository cavities (disposal drifts, transport
galleries and access shafts) and its later operation inevitably lead to the creation of a Damaged Zone
(DZ) within the clay around the engineered part of the disposal system. The role that the so-called
Excavation Damaged Zone (EDZ) may play in the transport of radionuclides following closure of
the repository and degradation of the waste packages has been the research subject of the EC FP5
project SELFRAC (SELFRAC, 2007).
As a side effect of radioactive decay, vitrified high-level wastes and spent fuel release significant
amount of heat, even after several decades of cooling in surface facilities. The TIMODAZ project
(Thermal Impact on the Damaged Zone Around a Radioactive Waste Disposal in Clay Host Rocks)
423

focuses on the study of the combined effect of the EDZ and the thermal output from the waste on
the repository host rock. The influence of the temperature increase on the evolution of the EDZ as
well as the possible additional damage created by the thermal load will be studied. The chemical
evolution as well as its interaction with the THM processes around the underground repository will
be addressed too in the project.
The research activities covered by TIMODAZ calls for multidisciplinary expertise involving both
European radioactive waste management organisations together with the main nuclear research institutes
supported by other research institutions, universities, industrial partners and consultancy
companies (SME’s). The TIMODAZ consortium is composed of 15 participating organisations representing
in total 8 countries: EURIDICE (BE), NAGRA (CH), SCK•CEN (BE), GRS (DE), NRG
(NL), CIMNE (ES), EPFL (CH), ULG (BE), UJF (FR), ENPC (FR), CEG-CTU (CZ), ITASCA
(FR), ASC (UK), ITC (CH) and SOLEXPERTS (CH).
2. Project structure
Three types of clay are investigated in TIMODAZ project: the Boom Clay of Belgium, the Opalinus
Clay of Switzerland and the Callovo-Oxfordian argilitte of France.
Even if the characteristics of these clays are different, the THM processes governing the fracturing
and the sealing present some similarities. Within the EC FP5 project "SELFRAC" (Fractures and
Self-Healing within the Excavation Disturbed Zone In Clays), both laboratory and in situ tests have
demonstrated the self-sealing capacity of both Boom Clay and Opalinus Clay, the former one presents
a faster self-sealing process than the later one [1]. In the TIMODAZ project, new laboratory
experimental equipments and test protocols are conceived to study the temperature effects on the
EDZ evolution (including sealing/healing capacity) and potential additional damage induced by
heating (work-package 3) and new In-Situ experiments in small and large scales (work-package 4)
are also designed to contribute to a better understanding of the processes occurring within the clay
around a disposal system for heat-emitting waste during the thermal transient phase. All experimental
researches are closely linked with the development and testing of sound, phenomenology-based
models which are essential in meeting the Safety Case requirement of adequate understanding of the
long-term evolution (work-package 5). Furthermore, all experimental and modelling developments
will be situated in the long-term performance assessment contexts, with the constant support of
work-package 6 - Significance of TDZ in Safety Case. To ensure an appropriate and continuous link
between the end-user needs and the priorities of the TIMODAZ project, an end-user group has been
constituted (figure 1) [2].
3. Methodology
The specific methodology of the laboratory experimental work packages of TIMODAZ project
consists of
Fundamental Thermo-Hydro-Mechanical behaviour characterising:
Tests in laboratory under well controlled temperature/stresses/pore pressure conditions with
different well defined loading paths will be carried out in order to determine the parameters of
the Thermo-Hydro-Mechanical constitutive models used for the numerical modelling (figure
2). More specifically, the thermal effects on the damaged clay and the possible damage induced
by the thermal loading itself will be investigated. Specific attention will be given to the
possibility of the creation of an irreversible damage. During the tests, different techniques will
be used to evaluate the sealing/healing processes (water/gas permeability measurements, μCT,
424

XRCT, etc.) (figure 3). Tests include also investigation of the temperature effects on the viscosity
of soil. Some tests will be complemented with a radionuclide migration test, in order to
evaluate any possible relict of preferential migration along the sealed fracture. Mineralogical
analyses will be performed to determine the possible thermally induced modifications of clays
mineralogy which is a dominant factor influencing the key properties of the clays and THM
behaviour.
Figure 1: Structure of the TIMODAZ project
Figure 2: an example of the THM paths to be followed in a triaxial test (EPFL test)
D
C
A
B
XRCT Consolidation
Deviatoric
loading
T° loading
3
XRCT
1 2 4
n Permeability measurement
Figure 3: Different techniques used for investigation the impact of the thermal loading on the process
of localization and/or fracturing during a triaxial heating/cooling test (UJF test)
Small – middle scale simulation tests
Simulation tests will be performed on hollow cylinder samples with mechanical and thermal
loadings similar to the evolution that will be encountered around disposal galleries for heat
425

emitting radioactive waste to study in laboratory the fracturing and sealing processes that develop
in the Excavation-Damaged Zone around galleries and the impact of a thermal phase on
their evolution (figure 4).
Pressure
external 3 ext.
internal 3 int.
Pore pressure
outer-drain u ext.
inner-drain u ext.
Temperature
external t ext.
internal t int. int. int. int.
Figure 4: simulation test on hollow cylinder sample, in the course of test, regular CT scan and
permeability measurements to study the damage evolution under different THM conditions
The in situ experiments include small scale tests (at Mont Terri and HADES) to characterise the
influence of the thermal load on the THM behaviour and the sealing capacity of clays and large
scale heater test (Praclay experiment in HADES) to study the effect of a large scale thermal load on
the behaviour of Boom Clay. The thermal load impact on the stability of gallery liner will be specifically
tested at the Underground Educational Facility - UEF Josef.
Small-scale in situ test at Mont Terri aims at detailed understanding of failure processes in EDZ as
well as the underlying sealing process of borehole EDZ under isothermal and non-isothermal conditions.
The main steps of the tests are:
Impregnation of the test interval with a tracer / marker (dyed, fluorescent epoxy resin) in the
boreholes with/without heater
Overcoring of existing borehole and/or sampling of the EDZ around the heater through a deflected
drill hole.
Analysis of the recovered sample and characterisation of the EDZ along the seal section (figure
5).
Figure 5: A recovered core section image (overcoring) allowing better understanding of fracture
mechanism around a borehole at Mont Terri
426

The small scale in situ test at HADES (ATLAS) will be designed specifically to be able to overheat
the Boom clay to study its ultimate temperature limit and possible thermal induced irreversible
damage.
Numerical tools allowing simulation at time and repository scale of the THM process in the clay
host formation will be further developed and improved in the frame of TIMODAZ project. The development
will focus on the:
Constitutive modelling in a continuum framework to incorporate heat changes and transfers,
moisture transfer, mechanical stress – strain evolution (including the development, evolution
of microcracks and the viscous – creep effects), and their interaction, especially in relation
with thermal effects
New damage model for unsaturated porous media
Utilisation of local 2 nd gradient models in order to provide objective descriptions of the behaviour
in the post-localization regime within the coupled thermo-hydro-mechanical coupling
problem (possibly including the case of non saturated media).
Modelling of sealing processes under thermal load based on PFC3D (discontinuum) code.
Coupled THMC analysis to examine the possible role of geochemical variables on the development
and evolution of the DZ.
Finally, all laboratory and in situ experiments as well as the modelling development will be constantly
supported and linked with the performance assessment to assess the significance of the DZ
in the safety case for disposal in clay host rock and to provide direct feedback to repository design
teams especially the thermal limits that the clays could sustain.
4. Conclusions
The knowledge gained within the TIMODAZ project will allow assessing the significance of the
TDZ (Thermal Damaged Zone) in the Safety Case for disposal in clay host rock and providing direct
feedback to repository design teams.
5. Acknowledgements
This project is co-funded by the European Commission (EC) as part of the sixth Euratom research
and training Framework Programme (FP6) on nuclear energy (2002-2006) under contract FI6W-
036449.
References
[1] Bernier F., Li X.L., Bastiaens W., Ortiz L., Van Geet M., Wouters L., Frieg B., Blümling
P.,Desrues J., Viaggiani G., Coll C., Chanchole S., De Greef V., Hamza R., Malinsky L., Vervoort
A., Vanbrabant Y., Debecker B., Verstraelen J., Govaerts A., Wevers M., Labiouse V.,
Escoffier S., Mathier J.-F., Gastaldo L., Bühler Ch. SELFRAC: Fractures And Self-Healing
Within The Excavation Disturbed Zone In Clays- Final report, European Commission,
CORDIS Web Site, EUR 22585, 2006.
[2] http://www.timodaz.eu/
427

428

TIMODAZ – Lining Stability under Thermal Load
Jaroslav Pacovský, Ji�í Svoboda, Radek Vaší�ek
Centre of Experimental Geotechnics, Czech Technical University in Prague, Czech Republic
Summary
The TIMODAZ Project is co-funded by the European Commission (EC) as part of the sixth
EURATOM research and training Framework Programme (FP6) on nuclear energy. TIMO-
DAZ is a four-year Specific Targeted Project (2006-2010) investigating thermal impact on the
damaged zone around a radioactive waste disposal vessel in clay host rocks. The TIMODAZ
consortium is composed of 14 participating organisations representing a total of 8 European
countries (BE, FR, CH, DE, NL, ES, CZ, UK).
The extreme long-term functioning of the lining around the disposal vessel is one of the premises
for the safe removal of spent fuel canisters from the engineered barrier. The long-term effects
of heat could well bring about a severe reduction in the stability of the lining caused either
by deterioration in the strength properties of the lining material or by the occurrence of
deformations resulting in a collapse in the shape of the lining.
Research into lining stability under thermal loading at the Centre of Experimental Geotechnics
(CEG) employs physical modelling.
1. Introduction
As part of the Timodaz Project, two physical models of the lining of disposal tunnels have been
built and put into operation – a laboratory model constructed at CEG itself and an “in situ” model
constructed in the underground laboratory of the Josef UEF. The effects of long-term thermal loading
acting on the lining and its stability are being investigated with the help of these models.
The laboratory experiment is designed to allow the deformation of the circular lining. Consequently,
a low degree of stress is generated within the prefabricated lining blocks but the deformation
of the circular lining itself could be enormous.
The model constructed in the Josef underground laboratory has been designed so that the lining
cannot be deformed in the direction towards the rock mass and, consequently, enormous stress is
generated within the lining material.
The laboratory model studies whether the thermal load acting on the lining might lead to a level of
deformation which could lead to a decrease in the dimensional stability of the lining. The “in situ”
model investigates whether long-term thermal loading could result in a level of stress acting on the
lining material that would exhaust its strength parameters causing cracking and the subsequent collapse
of the lining.
2. Methodology
Physical modelling is one of the tools available to researchers in addressing the problems of radioactive
waste disposal in a deep repository. Since such research is of a highly specialised and interdisciplinary
nature, all the relevant tools at hand must be applied, i.e. laboratory, “in situ” experimental
research and physical and mathematical modelling, Pacovský et. al. (2007) [1].
429

2.1 Laboratory experiment
The disposal tunnel model, constructed in the underground silo of the CEG laboratory, employed a
prefabricated circular lining. The outside radius of the lining measures 198cm with a segment thickness
of 16cm and a width of 50cm. The model itself is composed of 3 rings; individual segments are
connected using dry locks. The space between the lining and the silo walls is filled with handcompacted
sand which partially prevents transverse deformation of the thermally loaded lining (Fig.
1).
Figure 1: Construction of the laboratory physical model
The longitudinal deformation of the lining is prevented by the silo wall on the one side and by a
steel frame on the other. This form of experiment construction models the case in which thermal
loading does not produce a high level of stress within the lining, rather, thermal loading leads to the
deformation of the segments, with displacement occurring in the dry locks. The experiment should
verify whether or not deformation and displacement reach values which would eventually lead to a
loss in the stability of the whole lining.
The thermal loading of the inside surface of the lining is performed by means of the REVEL heating
system (Fig. 2). The heating medium employed is water which is heated to a temperature of
90 o C by electric boilers. The heating system is fitted with thermal insulation to eliminate heat loss.
Figure 2: Heating system
The experiment is fully instrumented in order to measure temperature, deformation and displacement.
Measurement is performed by means of a data logger with a data reading interval of once
every 10 minutes, Pacovský et al. (2007)[2]. The experiment was launched in November 2007.
430

2.2 “In situ” experiment
The disposal tunnel model was built in a short driven, former mining adit. The lining was made up
of segments obtained from the PRACLAY project (SCK-CEN Mol, Belgium). The outside radius of
the lining measures 250cm with a segment thickness of 30cm and a width of 50cm. The model itself
is composed of 4 rings and the space between the lining and the rock consists of concrete (Fig. 3).
The longitudinal deformation of the lining is prevented on the one side by solid rock and on the
other by a steel frame.
Figure 3: Construction of the “in situ” experiment
The “in-situ” experiment environment models the case in which the lining deformation (displacement
of the segments) is prevented and maximum stress arises as a consequence of thermal loading.
The thermal loading of the inside surface of the lining is performed by means of the REVEL heating
system. The heating medium employed is water which is heated to a temperature of 90 o C by
electric boilers. The heating system is fitted with thermal insulation to eliminate heat loss. The experiment
is fully instrumented in order to measure temperature, stress, deformation and displacement.
Measurement is performed by means of a data logger with a data reading interval of once
every 10 minutes. The experiment was launched in October 2008 (Fig. 4).
Figure 4: The “in situ” experiment under operation
431

3. Results
3.1 Laboratory experiment
The laboratory experiment was constructed without using the original lining from the PRACLAY
project (SCK-CEN Mol, Belgium). The lining used differed both in terms of its inner span, the
thickness of the segments and the quality of the concrete used. The principal objective of this experiment
was to gain experience in the construction of this type of model and to test both the
REVEL heating system and the instrumentation, Pacovský et al. (2007)[3].
From the very beginning, the heating system was set at a temperature of 90 o C i.e. the maximum
loading temperature. Due to effective thermal insulation, temperature equilibrium was soon
achieved. Individual deformations and displacements within the individual lining segments measured
to date do not exceed 3mm, which is less than had been assumed. Figure 5 shows a typical result
from the deformation measurement.
Displacement [mm]
1
0.75
0.5
0.25
0
-0.25
01 Nov 07
30 Jan 08
29 Apr 08
28 Jul 08
Date
26 Oct 08
Figure 5: Measurement of deformation of joints between individual segments of the lining
3.2 “In situ” experiment
The “in situ” experiment is being performed in accordance with a precisely specified procedure.
Due to the requirement for the temperature difference between the inside and outside lining surfaces
(temperature gradient) never to exceed 30 o C, thermal loading must be increased on a step-by-step
basis (60 o C, 75 o C, 90 o C).
Instrumentation provides accurate, sometimes very surprising results one of which was the discovery
of a steep increase in stress where the physical model comes into contact with the rock mass
(Fig.6). Measured values are continuously evaluated. The stress measured inside the lining segments
reaches values of up to 10MPa.
432
24 Jan 09
107
108
109
24 Apr 09

Contact stress [MPa]
3.5
3
2.5
2
1.5
1
0.5
0
19 Oct 08
29 Oct 08
08 Nov 08
18 Nov 08
28 Nov 08
HPC 3
Date
08 Dec 08
18 Dec 08
433
28 Dec 08
Figure 6: Contact stress and temperature measurement
4. Conclusion
The performance of physical models built in a real “in situ” environment provides knowledge
which cannot be obtained through any other form of experimental research. The models, however,
must be designed, constructed and operated to a high professional level. In the case of the “Lining
stability under thermal load” experiment, this has been accomplished.
The laboratory experiment served for gaining the knowledge and experience necessary for the construction
of a follow-up experiment in a genuine underground environment. The instrumentation
and the REVEL heating system have been thoroughly tested. The measured values of deformation
and displacement caused by thermal loading are particularly well applicable to the solution of the
long-term stability of the lining.
The knowledge obtained was used to very good effect during the somewhat demanding construction
of the “in situ” experiment. The measurement of all the relevant parameters with an interval of once
per 10 minutes allows the monitoring of the stress state deformation response of the lining and the
rock mass to gradually increasing thermal loading. The high stress measured both within the lining
and on contact with the rock mass before maximum thermal loading was reached justifies the decision
to construct the lining of the disposal tunnel of high-strength concrete.
The performance of these physical models will continue at least until the end of 2010 which will
allow the gathering of further important results applicable to the safe design of the lining of disposal
tunnels.
5. Acknowledgements
The “Thermal Impact on the Damaged Zone Around a Radioactive Waste Disposal (Vessel) in Clay
Host Rocks (TIMODAZ)” project is co-funded by the European Commission and forms part of the
sixth EURATOM research and training Framework Programme on nuclear energy (2002-2006)
under contract No. FI6W-36449.
References
[1] Pacovský J. (2007). The Use of the Mock-up-CZ Physical Model in the Design of Engineered
Barriers. WITpress – Management of Natural Resources, Sustainable Development and Ecological
Hazards, ISBN 13: 978-1-84564-048-4.
[2] Pacovský J., Zapletal, L., and Svoboda, J.(2007). Development of Saturation in a Bentonite
T 3
07 Jan 09
40
35
30
25
20
15
10
5
Temperature [ o C]

Barrier. Clays in Natural & Engineered Barriers for Radioactive Waste Confinement“. Physics
and Chemistry of the Earth, Volume 32, issues 8-14,2007. ISSN 1474-7065.
[3] Pacovský J., Svoboda, J., and Vaší�ek, R. (2007). Construction, Performance and Dismantling
of the Mock-up-CZ Experiment. Proceedings of the 10th Australia New Zealand conference
on geomechanics, 21-24 October 2007, Brisbane. ISBN 978-0-646-47974-3.
434

TIMODAZ – Characterisation of Rock Mass Crack Damage
Using Ultrasonic Surveys
Juan M. Reyes-Montes 1 , William S. Pettitt 1 , Jonathan R. Haycox 1 , Jennifer R. Andrews 1 ,
R. Paul Young 2
Summary
1 Applied Seismology Consultants, UK
2 University of Toronto, Canada
This paper presents a summary of two different approaches to interpret the evolution of
crack density in a rock mass induced by stress changes associated with excavation and heating-cooling
processes. These techniques were initially developed as part of the OMNIBUS
project, part funded by the EC under the 5 th Euratom framework and are being further extended
to study the impact of thermal-induced stress changes on the host rock as part of the
TIMODAZ project (in the 6 th Euratom research framework programme).
Two approaches have been investigated. First, full waveform ultrasonic data from laboratory
compression tests and in-situ experiments were correlated with a suite of tests in finite difference
models (Haycox and Pettitt, 2004)[1] with variable fracture density, size and fluid
content. Second, a non-interactive crack effective medium theory allows the derivation of
velocity anisotropy and splitting of elastic waves from modelled crack density, aspect ratio
and fabric orientation for moderately jointed samples (e.g. Schubnel et al., 2006)[2]. For the
purpose of this investigation, the isotropic case is considered, for which the method was
used to invert elastic wave velocities to infer the evolution of crack density and aspect ratio.
1. Introduction
Disposal in deep underground geological formations is one of the likely options to dispose of medium
and high level radioactive waste. A key element in the evaluation of the long-term safety of an
underground disposal site is the study and control of the Damaged Zone (DZ) surrounding the infrastructure.
The evolution of the DZ is affected among other factors by excavation induced stress
changes, pore pressures and the heating-cooling cycle of the waste.
Ultrasonic monitoring provides a unique means to non-destructively remotely monitor the evolution
of fractures and stress disturbance around the DZ. In particular, the monitoring of changes in waveform
attenuation and propagation velocity can be used as an estimate of the evolution of crack damage.
In this paper, two different approaches are presented that invert the modelled effect of a fracture
network on the transmission of elastic waves so as to interpret the evolution of rock mass properties
related to its capability to transport fluids.
435

2. Experimental data
2.1 Laboratory experiments
A series of laboratory rock loading experiments were performed at Imperial College London using
a true-triaxial loading frame and instrumentation platens (Pettitt and Haycox, 2004)[3] developed
within the EC Euratom project, SAFETI.
During this study P- and S-wave velocity survey data were measured during the different stages of a
cyclic polyaxial loading experiment on Crosland Hill sandstone. The rock was taken through a series
of hydrostatic and deviatoric damage cycles, similar to King (2002)[4], in order to induce a set
of aligned cracks. The rock type was chosen due to the limitations in the maximum stress applied
by the testing frame. Initial analysis of the P- and S-wave velocity data showed the main change to
be a significant decrease in velocity following the deviatoric loading cycle, parallel to the 3 direction,
suggesting an aligned microcrack fabric had been induced and was confirmed by the location
of the monitored Acoustic Emissions.
2.2 Field-scale experiment
Field scale ultrasonic surveys were conducted at the Äspö Hard Rock Laboratory (HRL) in Sweden,
operated by Svensk Kärnbränslehantering AB (SKB) (Pettitt et al., 2001)[5]. The monitored deposition
hole was excavated over a two-week period in eleven 0.8m steps and monitored by a 24tranducer
ultrasonic array providing 8 transmitters and 16 receivers. 41 months after the excavation,
heaters were installed within a simulated waste canister in the deposition hole causing temperature
in the rock wall to increase rapidly to approximately 50° C.
3. Method
Wave propagation studies in finite difference models constructed using Wave 3D have been used to
describe the effects of fracture geometry on ultrasonic signals (Haycox and Pettitt, 2004)[1]. A sensitivity
analysis was performed by correlating amplitude-ratio and phase-difference results for selected
ray paths in an ultrasonic survey with waveforms produced by numerical simulations of the
experiment.
A second forward modelling approach is provided by the Effective Medium Theory (EMT). EMT
allows microstructural crack parameters such as crack density and aspect ratio to be derived from
elastic wave velocity measurements. Using those parameters, the evolution of permeability with
stress can be predicted. In this study we have used a non-interactive EMT approach (Kachanov,
1994 [6]; Schubnel and Gueguen, 2003 [7]) that assumes the possible compensation of interactions
when cracks are distributed randomly or aligned. In the isotropic case, the effective Young and
shear modulii of a rock could be written as:
ρ ⎛ 3 ⎡
⎞
⎜
⎛ ν 0 ⎞ δ ⎤
= 1+
⎟
⎜
1+
⎢⎜1−
⎟ −1⎥
E 3h
⎟
⎝ 5 ⎣⎝
2 ⎠1
+ δ ⎦⎠
E0 c
μ ⎛
⎞
0 ρc
2 ⎡
⎜
⎛ ν 0 ⎞ δ ⎤
= 1+
⎟
⎜
1+
⎢⎜1−
⎟ −1⎥
μ 3h
⎟
⎝ 5 ⎣⎝
2 ⎠1
+ δ ⎦⎠
3(
1−ν
0 / 2)
3πE
0ζ
Where h =
and δ =
2
2
16(
1−ν
0 ) 16K
f ( 1−ν
0 )
The method can be extended to the case of transversely isotropic media in a similar manner by using
elastic stiffnesses (and compliances) instead of E and �, and taking into account orientations.
436

4. Results
A Wave 3D model was created for the geometry of the polyaxial test on a sandstone cube (Haycox
and Pettitt, 2004)[1], modelling fluid filling by allowing the cracks to have a normal stiffness but no
shear stiffness. Models consisted of a 50mm x 50mm x 50mm cube and used approximately 8x10 6
finite difference zones. The cube was modelled as an isotropic, linear elastic material with P- and Swave
speeds of 4800 m/s and 3300 m/s respectively and a density of 2431 kg/m 3 .
Best fit models were obtained for the 6 measurements of amplitude ratio and phase difference on
the P and two orthogonal S-waves. An example of the correlation between measured and modelled
data is shown in Fig. 1. The model that was found to best represent all the data was a set of aligned
cracks with crack length = 4mm, normal stiffness = 1x10 13 Pa/m, and crack density = 0.2.
A similar strategy was followed in the case of the in-situ experiment at SKB’s HRL, for the period
following the excavation of the prototype borehole (Haycox and Pettitt, 2004) [1]. A number of ray
paths were chosen which pass through specific regions around the void, ranging from ray paths
which pass within centimetres of the deposition hole’s edge and exhibit the greatest changes in velocity
(Pettitt and Haycox, 2004)[3] to ray paths passing away from the deposition hole wall where
no significant changes in transmitted P-waves are observed before and after the excavation. The
comparison of the amplitude and phase spectra from the ray paths crossing through a tensile (or
low-compressive-stress) zone induced in the vicinity of the hole’s wall before and after the excavation
show a decrease in amplitude over a wide range of frequencies.
Results from the best-fit analysis confirm interpretations based on the ultrasonic data and acoustic
emissions alone. The ray path that passes through the tensile zone (which experienced acoustic
emissions and large velocity drops over the excavation period) has the largest crack sizes and density.
For simplicity, we restricted the modelling of elastic wave velocities using the EMT in both the
laboratory and the field case studies to non-interacting dry microcracks, a valid approximation for
the case where no unstable crack propagation is observed. For the laboratory case study, only measurements
of wave velocity carried out along the three principal stress axes have been considered.
For the inversion, matrix elastic parameters were used, with Eo= 65 GPa and � = 0.24. Figure 3a
shows the changes in propagation velocity along the three principal axes measured during the different
episodes of hydrostatic (up to 100 MPa) and deviatoric loading. The results of the modelled
velocities and the associated changes in crack density are shown in Fig. 2. A velocity increase is
observed during hydrostatic loading associated with crack closure. During the deviatoric loading
there is a significant decrease in the velocity of waves propagating in the �3 direction. The opening
of macroscopic fractures in the plane normal to �3 was confirmed by the source location of Acoustic
Emission events during the deviatoric loading (Haycox and Pettitt, 2004) [1].
Figure 4 presents the results of the modelled velocity changes using the EMT for a ray path skimming
the surface of the deposition hole at SKB’s HRL. The changes in transmission velocity were
measured during its excavation, showing a sharp decrease of velocities after the start of the opening
of the void followed by an asymptotic decrease to a value ~30 m/s below the original transmission
velocity in the case of vP. The magnitude of the stress changes associated with the excavation and
the nature of the host rock in this case results in significantly smaller induced changes in transmission
velocity and crack density compared with the laboratory case. A reasonably good fit is observed
between the observed and modelled velocity changes. The results for the heating phase, pre-
437

sented in the accompanying poster, show a small increment in transmission velocities associated
with micro-crack closure.
a b
Figure 11: a) Example measured waveform and P-wave amplitude ratio (calculated against a reference
survey on ‘undamaged’ sandstone taken at the highest hydrostatic load) b) Waveform and
amplitude ratio for the Wave 3D model that best fits the survey shown in a).
Velocity (m/s)
5000
4500
4000
3500
3000
2500
2000
0 10 20 30
Stage
40 50 60
Vp1
Vp1 model
Vp2
Vp2 model
Vp3
Vp3 model
Vs1
Vs1 model
Vs2
Vs2 model
Vs3
Vs3 model
Crack density
438
0.1
S1
0.09
S2
0.08
S3
0.07
0.06
0.05
0.04
0.03
0.02
0.01
0
0 10 20 30 40 50 60
Stage
a b
Figure 12: a) Monitored and EMT modelled transmission velocities in the direction of the three
principal stress axes during the cyclic loading of the Crosland Hill sandstone cube. b) Inverted
crack densities in the three directions.
Velocity change (m/s)
5
0
-5
-10
-15
-20
-25
-30
-35
Vp (DZ)
Vp model (DZ)
Vs (DZ)
Vs model (DZ)
Excavation
-40
0
0 5 10
Time (days)
15 20
10
9
8
7
6
5
4
3
2
1
Excavation depth (m)
Crack density
0.2205
0.2155
0.2105
0.2055
0.2005
0.1955
Crack density
Excavation
0.1905
0
0 5 10
Time (days)
15 20
a b
Figure 13: a) Monitored and modelled (EMT) transmission velocities through a ray path skimming
the surface of SKB’s HRL deposition hole during its excavation. b) Inverted crack densities.
10
9
8
7
6
5
4
3
2
1
Excavation depth (m)

5. Acknowledgements
The authors gratefully acknowledge part funding for the SAFETI and OMNIBUS projects from the
European Commission as part of the fifth Euratom framework programme. The TIMODAZ project
is part funded by the European Commission within the sixth Euratom framework programme.
References
[1] Haycox, J.R. and W.S. Pettitt, (2004) Integrated Analysis of Measured and Modelled data for
the OMNIBUS Project, ASC, in OMNIBUS Final Technical Report.
[2] Schubnel, A., Benson, P.M., Thomson, B.D., Hazzard, J.F. and R.P. Young, Quantifying
damage, saturation and anisotropy in cracked rocks by inverting elastic wave velocities.
[3] Pettitt, W.S. and J.R. Haycox (2004) Acoustic emission and ultrasonic monitoring of deposition
hole DA3545G01 during the excavation and heating phases, ASC, in SAFETI Final
Technical Report.
[4] King, M. S.,(2002). Elastic Wave Propagation in and Permeability for Rocks with Multiple
Parallel Fractures, Int. J. Rock Mech. Min. Sci., 39, pp1033-1043.
[5] Pettitt, W.S., Baker, C., Young, R.P., Dahlström, L.-O. and G. Ramqvist, (2001). The assessment
of damage around critical engineering structures using induced seismicity and ultrasonic
techniques. Pure Appl. Geophys ., 159, 179 – 195.
[6] Kachanov, M. (1994), Elastic solids with many cracks and related problems, Adv. Appl.
Mech. 30 259-445.
[7] Schubnel, A. and Y. Guéguen (2003), Dispersion and anisotropy in cracked rocks. J. Geophys.
Res. 108.
439

440

TIMODAZ – Modelling the Excavated Damage Zone around an underground
gallery - Coupling mechanical, thermal and hydraulical aspects
Summary
Robert Charlier 1 , René Chambon 2 , Xiang-Ling Li 3 , Arnaud Dizier 1,4 ,
Yannick Sieffert 2 , Frédéric Collin 1,4 , Séverine Levasseur 1,4
1 ArGEnCo, Université de Liège, Belgium
2 Laboratoire 3S-R, Université Joseph Fourier – Grenoble I, France
3 EURIDICE, Belgium
4 FRIA - FRS-FNRS, Belgium
A zone with significant irreversible deformations and significant changes in flow and transport
properties is expected to be formed in clay around underground excavations. The stress
perturbation around the excavation could lead to a significant increase of the permeability, related
to diffuse and/or localized crack propagation in the material. Further the drainage and
the heating of disposal will modified the size and the structure of the damage zone. The main
objective of the study is to model these processes at small and large scale in order to assess
their impacts on the performance of radioactive waste geological repositories. This paper concerns
more particularly the thermo-hydro-mechanical modelling of a hollow cylinder experiment
performed in Boom Clay and the hydro-mechanical modelling of a long term dilatometer
experiment performed in Opalinus Clay at Mont Terri Rock Laboratory in Switzerland. This
study shows that simple models already permit to reproduce the behaviours observed experimentally.
1. Introduction
The aim of the TIMODAZ project is to investigate the effect of thermal changes on the excavated
damaged zone (EDZ) around nuclear deep disposals in clay host rocks. During the disposal construction,
rock mass around the gallery is damaged and its permeability may be modified. Further
the drainage and later the heating will modified the size and the structure of the damage zone. In the
framework of the TIMODAZ project, it is intended to develop constitutive models and numerical
tools. A series of benchmark modelling with different constitutive laws and numerical tools are performed.
From now, two benchmarks have started. The first one, described in section 2, concerns a
small scale laboratory experiment on a hollow cylinder clay sample submitted to thermo-hydromechanical
loadings. The second one, described in section 3, is an in-situ experiment which studies
the axial water transmissivity evolution through the EDZ.
2. Laboratory experiment
To study the thermal impacts on damaged zone, a small scale laboratory experiments tests is modelled
and submitted to a benchmark exercise. As illustrated on Figure 1, heating/cooling cycles are
applied on a hollow cylinder performed in Boom Clay which is modelled with in 1D radial axisymmetric
conditions. The clay is supposed to be isotropic and homogeneous and is assumed to be
fully saturated. The internal confining pressure in the central hole is decreased from the initial in
441

situ stress to model the gallery excavation. During this step, a damaged (plastic) zone is expected to
develop around the hole. Temperature is increased at the inner wall, resulting in a radial gradient.
For the modelling of this experiment, a frictional elasto-plastic Drücker-Prager model [4] is proposed
as mechanical law and the thermo-mechanical couplings are based on thermo-elasticity with
different constitutive laws [3]. The modelling is decomposed into several phases. The two first
phases concern the decrease of the internal confining pressure from the initial value. The third
phase is related to a heating/cooling cycle applied at the inner radius of the cylinder.
Figure 1: Representation of the hollow cylinder and boundary conditions
Figure 2: Description of the thermo-hydro-mechanical loadings applied during the modelling
Figure 3 shows profiles of the pore pressure with the radial distance at different time of the experiment.
The two first phases is represented in Figure 3 (a). The decrease of the internal pressure induces
a drawdown of the pore water pressure. Figure 3 (b) describes the evolution of the pore water
pressure with heating/cooling cycle at different times. A rise in temperature induces an increase of
pore water pressure caused by the difference between the thermal coefficient expansion of the water
and of the solid. When the temperature decreases, the viscosity increases resulting in a rise in pore
water pressure. Figure 4 exhibits stress paths for different combinations of the Drücker-Prager criterion.
The plasticity appears at a lower stress state according the hardening rules [3]. Considering a
basic thermo-mechanical model, these stress paths show that the thermal aspects are small comparing
to the hydro-mechanical part of this modelling.
3. In situ experiment
To study the hydraulical impacts on damaged zone, a long term dilatometer experiment is performed
in Opalinus clay. This experiment is developed to test the influence of bentonite swelling
pressure on the transmissivity of the Excavation Damage Zone [1,2]. Its concept is the combination
of dilatometer tests and numerous hydraulic tests with multi-packer system in a newly drilled borehole
(Figure 5). The pressure in the dilatometer probe is increased stepwise and hydraulic tests are
442

periodically performed to evaluate axial transmissivity under different dilatometer pressures. It can
be assumed that pressure changes are preferably transmitted through the EDZ along the borehole.
As a consequence, the hydraulic conductivity of the EDZ can be seen as a function of
(a) (b)
Figure 3: Evolution of radial profile of pore water pressure during the decrease of the internal
pressure (a) and during the heating/cooling cycle (b). The chosen Drücker-Prager criterion in this
case is perfectly plastic.
Figure 4: Stress path at the inner radius of the hollow cylinder for different combinations of the
Drücker-Prager criterion
Figure 5: Experiment layout long-term dilatometer experiment [2]
inflation pressure of the dilatometer. The objective of this second benchmark is to model this behaviour.
A finite element model based on a hydro-mechanical approach including an evolution of
the permeability tensor is developed. The evolution of permeability is based on a cubic law. A fic-
443

tive crack network, oriented along principal strains, is associated to each Gauss point of the finite
element model. The crack opening is then linked to the tensile principal strains [5].
Because of an anisotropic strain tensor along a borehole, it results an anisotropic permeability tensor.
Applying to the modelling of the long term dilatometer test, this approach permits to well reproduce
the behaviours of indurated clays. An EDZ of few centimetres behind the borehole can be
identified. In this area, a high fracture density, characterized by permeability increases of up to several
orders of magnitude is observed as shown on Figure 6, which represents the axial permeability
along (a) and on a section perpendicular (b) to the borehole after the borehole drilling and for the
different dilatometer loads. Moreover, when dilatometer pressure increases, the permeability decreases
and no significant water flow modification can be observed in EDZ along the dilatometer as
illustrated on Figure 7, which represents the evolution with time of the overpressure in I2 after hydraulic
tests and for different dilatometer loads. Finally, the comparisons between numerical predictions
and measurements of the pressures in the intervals exhibit a good agreement and confirm that
our model is able to catch the main hydro-mechanical processes occurring within the EDZ in
Opalinus clay.
1.0E-14
1.0E-15
1.0E-16
1.0E-17
Kyy (m²)
Packer
Interval I2
(a) (b)
borehole
Dilatometer
dilatometer
loading
increases from
3 to 5MPa
Interval I1
borehole drilling
y (m)
1.0E-18
3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5
444
Kyy (m²)
1.0E-14
1.0E-15
1.0E-16
1.0E-17
1.0E-18
1.0E-19
Dilatometer
dilatometer
loading increases
from 3 to 5MPa
x (m)
1.0E-20
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Figure 6: Axial permeability along the borehole (a) and on a section perpendicular to the borehole
in mid-length of the dilatometer probe (b) after the borehole drilling and for different dilatometer
loads
0.16
0.14
0.12
0.10
0.08
0.06
0.04
ΔPw (MPa)
I2 time of reaction
increases with
dilatometer loading
phase 4
Pdilato = 3MPa
phase 8
Pdilato = 4MPa
phase 10
Pdilato = 4.5MPa
0.02
0.00
phases 12 -13
Pdilato = 5MPa
log Δt (s)
1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07
Figure 7: Evolution of overpressure in I2 in time after hydraulic tests and for different dilatometer
loads
4. Conclusions
The excavation damage zone is a phenomenon that occurs in the most rock masses as a consequence
of underground excavation. The EDZ appears as an area around the underground openings,
where geotechnical and hydro-geological properties are altered. The numerical model should be

able to predict the evolution of the permeability in the EDZ, in order to permit a correct design of
the geo-structure. First, this paper presents a laboratory test that evaluates the influence of the temperature
coupling with different elasto-plastic constitutive laws. Results show that the effect of the
temperature is negligible considering thermo-elasticity. Second, this paper presents an in situ test
that simulated the influence of bentonite swelling pressure on the axial transmissivity of an excavation
damage zone in the Opalinus clay. A model based on a cubic law evolution of the permeability
tensor is proposed. The comparisons between numerical predictions and measurements of the pressures
in the intervals exhibit a good agreement and confirm that this model is able to catch the main
hydro-mechanical processes occurring within the EDZ.
5. Acknowledgements
The authors would like to thank the FRIA-FRS-FNRS, the national funds of scientific research in
Belgium, and the European project TIMODAZ for their financial support. TIMODAZ is co-funded
by the European Commission (EC) as part of the sixth Euratom research and training Framework
Programme (FP6) on nuclear energy (2002 – 2006).
References
[1] Bernier F., Li X.L., Bastiaens W., Ortiz L., Van Geet M., Wouters L., Frieg B., Blümling P.,
Desrues J., Viaggiani G., Coll C., Chanchole S., De Greef V., Hamza R., Malinsky L., Vervoort
A., Vanbrabant Y., Debecker B., Verstraelen J., Govaerts A., Wevers M., Labiouse V., Escoffier
S., Mathier J.-F., Gastaldo L., Bühler Ch. (2004). Selfrac: fractures and self-healing within
the excavation disturbed zone in clays. Final technical publishable report, 5th EURATOM
Framework Programme.
[2] Bühler Ch. (2005). Selfrac (SE) Experiment: long term dilatometer experiment. Mont Terri Project
- Technical Note, TN 99-03.
[3] Dizier A., Collin F., Charlier R. (2008) TIMODAZ project. Benchmark 1- Phase 2: Hollow cylinder
modelling.
[4] Drücker D.C. and Prager W. (1952). Soil mechanics and plasticity analysis or limit design -
Quarterly Applied Mathematics, 10 (2), 157-165.
[5] Levasseur S., Charlier R., Frieg B., Collin F. (2009). Hydro-mechanical modelling of the Excavation
Damage Zone around underground excavations: application to long term dilatometer experiment
in Mont Terri Rock Laboratory (Selfrac Project), submitted to Engineering Geology.
445

446

TIMODAZ – Large-Scale Heater Experiments In Boom Clay
Jan Verstricht 1 , Wim Bastiaens 1 , Xiang Ling Li 1 , Philippe Van Marcke 1 , Guangjing Chen 1 , Xavier
Sillen 2
Summary
1 EIG EURIDICE, Mol, Belgium
2 SCK•CEN, Mol, Belgium
The thermal impact of heat-dissipating, high-level radioactive waste is one of the main disturbances
that are considered in the framework of the geological disposal of this waste. Within
the Belgian research programme, which focuses on the Boom Clay formation as potential host
rock, this aspect has been the subject of several field tests complementing the desk studies. A
large scale integration of the thermal issue is now being implemented as the PRACLAY
Heater Test, a main constituent of the overall PRACLAY demonstration programme. This
programme intends to simulate at real scale (except length and time) a disposal gallery
through all phases: construction (shaft, main gallery, test gallery including gallery crossing)
using industrial techniques, followed by the backfilling, sealing and saturation of the PRA-
CLAY test gallery, after which the heating of the test gallery will be switched on during 10
years. As a preparation to this large scale heater test, an already existing test set-up ("AT-
LAS") has been extended and re-activated. The experimental data obtained have allowed improving
the characterisation of the Boom Clay, for example by incorporating the anisotropy of
the thermal conductivity. The PRACLAY Heater Test will allow confirming these findings at
large scale. The host rock around the gallery, as well as the gallery itself, are extensively instrumented
to this purpose. The test results will constitute an important field database for the
validation of the numerical tools to be developed in the frame of the EC project TIMODAZ,
as well as for the evaluation of the Safety and Feasibility Case 1 (SFC1).
1. Introduction
The thermal aspect has been considered from the very beginning in the desk studies and small-scale
field tests on the research of the disposal of heat-dissipating high-level radioactive waste (HLW),
The temperature increase of host formation strongly influences the hydraulic and mechanical behaviour.
It has further an effect on the chemical and microbiological conditions in the host rock.
The early transient thermo-hydro-mechanical (THM) perturbation might be the most severe impact
that the disposal system will undergo on a large spatial scale and in a relatively short period of time.
Complementing the desk studies and modelling work, the thermal aspect is therefore also the object
of many experimental set-ups at the lab and in situ. With the current evolution in the R&D to integration
and demonstration, the study of the impact of the heating at large scale becomes essential.
For more than 30 years, Belgium has been actively studying the long-term management of highlevel
and/or long-lived radioactive waste. The Belgian Research Centre SCK•CEN initiated the
R&D programme in 1974 following international recommendations to isolate HLW from the biosphere
by disposing of it in a stable geological formation with appropriate characteristics. For vari-
447

ous reasons, SCK•CEN chose to focus on the Boom Clay as potential host formation. After a few
years (1980), the programme resulted in the start of the construction of the HADES Underground
Research Facility (Fig. 1), demonstrating the feasibility of constructing in this type of clay at a
depth of 223 m below surface. When the newly established Belgian Waste Management Agency
NIRAS/ONDRAF took over the R&D programme, the first experimental results from HADES confirmed
the Boom Clay as the reference formation for carrying out its studies on the long-term management
of HLW. An expert assessment in the late eighties confirmed the NIRAS/ONDRAF conclusions
on the suitability of the clay formation for HLW disposal. The Boom Clay had been found
to have a very low hydraulic conductivity, a plastic character with good self-sealing properties and
a high capacity to fix radionuclides and hence, to delay their migration towards the biosphere.
These encouraging results prompted SCK•CEN and NIRAS/ONDRAF to launch an ambitious demonstration
programme with the name PRACLAY. Its main objectives dealt on one hand with demonstrating
industrial techniques suitable for constructing a real repository, and on the other hand to
operate a dummy disposal gallery through a heater test during at least 10 years. Demonstration of
industrial techniques was an important aspect as the construction of the first part of HADES had an
exploratory nature (feasibility). The (mainly manual) excavation and construction techniques had
further disturbed significantly the host formation. In the perspective of a real repository, demonstration
of applicable mechanised techniques was therefore essential. The construction of the
dummy disposal gallery would then allow simulating all repository operation phases up to the heating.
The first objective has been fulfilled through the construction (Fig. 1) of the second shaft, the
connecting gallery, and the PRACLAY test gallery (including the gallery crossing) in 2007.
Figure 1: Construction history of HADES, with the first part (right) from the 1980's, and (left) the
extension in the frame of the PRACLAY project.
2. Methodology
2.1 Preparing for the large scale test – the ATLAS test
A HADES experimental set-up from 1992, named ATLAS, has been reactivated to get additional
data for thermal modelling purposes. The original set-up had been installed in the frame of the EC
INTERCLAY II project (1990-1994) and consisted of a 8 m-long heater (11 to 19 m deep) with two
parallel observation boreholes – all in the horizontal plane. To get a larger monitored zone and to
448

study the possible anisotropy, two additional instrumented boreholes have been installed (Fig. 2,
left).
Figure 2 (left): The ATLAS heater (8 m long, up to 1800 W) and monitoring boreholes (temperature
and pore water pressures). The 3D model is also shown (right). The heating power applied during
the first campaign.
During the first heating campaign (from April 2007 to April 2008), three different heating power
levels (400 W, 900 W and 1400 W) were applied (Fig. 2, right). The campaign has provided valuable
data on the thermal characteristics of the host rock. After extending the set-up with an additional
inclined borehole, the second heating campaign will investigate these topics more in detail.
2.2 Demonstration at large scale - the PRACLAY Heater and Seal Test
Initially, PRACLAY was about simulating as closely as possible a disposal gallery at real scale in
HADES. The review of the engineered barrier system design led to a reorientation of the PRA-
CLAY programme to a generic experiment, so that the experimental results remain valid in case the
repository design would change. The original scope of the PRACLAY experiment – demonstration
of the reference design for vitrified HLW – was enlarged to characterisation, verification, confirmation
and demonstration of relevant elements of the disposal system and their behaviour by means
of a combination of small surface and large in-situ experiments. Additional opportunities for large
scale surface testing of specific designs have been offered by the ESDRED Integrated Project, with
e.g. the gallery backfilling test.
The Heater Test will start after completion of the Seal that will close off the backfilled (and later
saturated) part of the PRACLAY gallery. The test will mainly consist in heating the host formation
through heater elements installed in the test gallery. The heating conditions have been established
based on the predicted repository conditions through thermal modelling [1]. The intention is that
within the limited time span of 10 years of heating, the largest thermal disturbances that would occur
in the real case are simulated. The resulting heating procedure specifies a constant temperature
of 80°C at the gallery outside during these 10 years. This temperature will be obtained gradually to
avoid excessive stresses in the gallery lining. The heater test will further give insight in the thermal
effect on the man-made structures (in particular the thermal stress in the gallery lining), and the
THM coupling behaviour of Boom Clay. Instrumented boreholes, installed from the connecting gal-
449

lery, are already monitoring temperatures, pore water and total pressures, and displacements, in the
host rock around the PRACLAY gallery. The monitoring network is currently being extended
through boreholes drilled from the PRACLAY gallery. The gallery lining itself contains rings instrumented
with strain gauges, temperature sensors, pressure and load cells. Inside the PRACLAY
gallery, sensors will characterise the source term of the whole experiment by monitoring the thermal
field, the pore water pressure, and also the global thermal expansion.
Figure 3: PRACLAY Heater Test with observation boreholes
The Seal will isolate the saturated and heated part of the gallery through an annular bentonite ring.
It is the object of a separate test programme, in which we will monitor the Seal performance, as
well as the actual THM evolution inside the bentonite and at its boundaries. A complete instrumentation
programme is currently being prepared to be installed together with the Seal mid-2009.
3. Results
3.1 First heating campaign of the ATLAS Test
Seal
The main purpose of this campaign, that ran from April 2007 to 2008, was to get a more accurate
and extended picture of the temperature field, and of the thermally induced pore water pressures.
The temperature evolution has shown that the thermal conductivity exhibits a strong anisotropy –
the horizontal component (Kh) approaches well the already established value of 1.7 W/m.K, while
the vertical one (Kv) is significantly smaller (about 1.3 W/m.K). The pore water response (Fig. 4)
when changing the heating power further shows an interesting phenomenon (initial drop prior to
increase), which has also been observed at Bure and Mont Terri. It can be explained by the anisotropic
mechanical behaviour of the clay. These test results allow validating the numerical tools that
will be used for the interpretation of the PRACLAY Heater Test observations.
3.2 PRACLAY Heater Test
Heating results are not yet available as the heater test is expected to start in 2010. The current monitoring
allows to establish an extensive and detailed baseline, in which the other phases (construction-related)
are integrated to make it a really integrated test.
450

Figure 4: Pore water pressure changes (in response to the heating) with typical behaviour when
heating power changes.
4. Conclusions
The first heating campaign of the ATLAS heater test has already pointed to some aspects that have
been underexposed until now, such as the anisotropy of the thermal conductivity and of mechanical
properties. The PRACLAY Heater Test will constitute an important field database for the validation
of the numerical tools being developed in the frame of the EC project TIMODAZ. The first year's
test results will further also provide input for the Belgian Safety and Feasibility Case no. 1 (SFC1),
an important milestone in the Belgian programme for HLW disposal.
5. Acknowledgements
The whole PRACLAY project is managed by EIG EURIDICE with the financial contribution from
NIRAS/ONDRAF. Several parts have been and are co-financed by the European Commission as
part of the sixth Euratom research and training Framework Programme (FP6) on nuclear energy
(2002-2006).
References
pressure drop is predicted when mechanical anisotropy is assumed
451

[1] Bernier F., Li X.L., Weetjens E, and Sillen, X. (2004). The Praclay Heater Test and the Praclay
plug test. Proc. Conference & Workshop on the Experimental Programme of the URF
HADES (Jan. 2004). EURIDICE, Mol, Belgium, pp 1-18.
452

NF-PRO – Concrete Degradation and its Influence on the Geochemical
Conditions at the Concrete/Bentonite under Repository Conditions.
Alicia Escribano, Elena Torres, Mª Jesús Turrero, Pedro Luís Martín, Javier Peña, Paloma
Gómez
Summary
CIEMAT, Madrid, Spain
Degradation of concrete can influence the performance of the bentonite barrier, as high-pH
lixiviates could affect to the clay properties.
The experimental studies conducted by CIEMAT are designed to provide experimental evidences
on the physical, chemical and mineralogical changes during the concrete-compacted
bentonite interaction.
Concrete has been analyzed by SEM-EDS, whereas bentonite samples were studied by means
of XRD, FTIR and SEM–EDS. Chemical analysis for exchangeable cations and soluble salts
were also performed in clay samples.
1. Introduction
Deep geological disposal is the most accepted management option for long-lived and highly radioactive
wastes. In the repository, concrete will be used for the reinforcement of the access galleries
and in the final sealing of access routes.
Concrete degradation caused by external agent can affect its durability, resistance and stability
throughout time. Furthermore, concrete degradation generates a diffusive alkaline plume which can
affect the swelling and transport properties of bentonite, as well as the properties of the adjacent
host rock and groundwater flowing through.
The scope of the present study is to provide experimental data about the alteration products at the
concrete/bentonite interface and the related changes of both materials during the interaction.
2. Methodology
2.1 Materials
Bentonite
The tests have been performed with a bentonite (FEBEX) from the Cortijo deArchidona deposit
(Almería, Spain). The conditioning of the bentonite prior its use in the laboratory is detailed in [1].
Its main characteristics are detailed in Villar (2002) [2].
453

Concrete
The cement used is an Ordinary Portland Cement CEM I R/SR 42.5 N. Chemical composition is
showed in Table 1 [3].
Table 1. Chemical Composition of the Cement.
Chemical
composition
(%)
CEM I
R/SR 42.5
N
2.2 Experimental Set up
SiO2 Al2O3 Fe2O3 CaO(total
)
454
MgO SO3 Na2O K2O CaO(free)
19.6 4.43 4.27 64.5 0.95 3.29 0.11 0.28 1.92
It consists in a hermetic cylindrical cell under vacuum, to prevent oxidation (Fig. 1a). A plane
heater on the bottom of the cell (100ºC) and a chamber on the top that allows the circulation of water
at a controlled temperature (around 22ºC), generate a gradient of temperature. A hydration channel
crosses the upper chamber and allows the hydration of the samples under a pressure of 12 bars.
The clay, with its water content at equilibrium with the laboratory conditions, is uniaxially compacted
outside the cell to a dry density of 1.65 g/cm 3 . A concrete slab was placed on the top of the
bentonite. For hydration, a saline solution, whose composition is described in Table 2, was used.
Two sensors were installed to monitor the evolution of water content and temperature with time.
The cell was dismantled after 18 months.
30mm
71.4 mm
Hydration (1.2 MPa)
Saline water
Concrete
25ºC
FEBEX
Bentonite
ρ=1.65 g/cm3 FEBEX
Bentonite
ρ=1.65 g/cm3 FEBEX
Bentonite
ρ=1.65 g/cm3 Heating (100ºC)
50 mm
Tª and RH
sensors
95 mm
Interface section
Section 1
Section 2
Section 3
Section 4
(a) (b)
Figure 1. (a) Scheme of the cell; (b) bentonite after dismantling and sampling scheme.
Table 2. Chemical composition of the saline solution
Chemical species Fe Na K Ca Mg
Concentration
(M)
1.1 E-05 13 E-02 8.2 E-04 1.1 E-02 8.2 E-02
Chemical species Si SO4 2- Cl - HCO3 -
Concentration 2.7 E-04 7.0 E-02 2.3 E-02 1.8 E-03
pH = 7.54
pE = -3.16

3. Results
3.1 Concrete
(M) log P CO2 = -2.65
Carbonatation is one of the most relevant processes in concrete degradation. Presence of calcite is
greater in the hydration area (Fig. 2a); whereas the rest of precipitates are found in the pores close
the interface with the bentonite. Portlandite begins to precipitate in this area until cover all the sample
surface (Fig 2b, 2c).
6 μm
Figure 2. (a) Crystals of Ca CO3 on the sample surface (b) Ca (OH)2 covering concrete surface(c)
detail of crystals
As well, gypsum crystals were observed (Fig.3). They come from the reaction of substitution between
portlandite and sulphate present in the hydration solution.
40 μm 40 μm
Figure 3. Gypsum crystals in a micro fracture.
(a) (b) (c)
3.2 Interface concrete/bentonite
A homogeneous white precipitate is formed at the interface. The thickness of this layer is around 2
μm. The SEM-EDX analysis of a cross-section of this area shows the formation of a continuous
layer of portlandite. Other secondary mineral phases were found at the interface: calcium silicate
hydrates or CSH gels tobermorite-type with Ca/Si molar ratio of 0.6, magnesium hydroxide (brucite)
and ettringite (Aft) (Ca6Al2(SO4)3(OH)12 26(H2O)) (Fig. 4). Also, dissolution of quartz was
detected .This process enables the precipitation of CSH minerals such as tobermorite.
Tobermorite
6�m 6�m
455

Brucite Ettringite
10 �m 8 �m
Figure 4. New phases formed at the concrete-bentonite interface.
FTIR analysis of the white precipitate and bentonite closest to interface show the main features of
clay minerals [4]. A small band that could correspond to the Mg3OH deformation of a smectitic
trioctahedral phase (saponite) was observed at 669 cm -1. In the spectra of the precipitate, three
bands at 1417, 874, 712 cm -1 reveal the occurrence of calcite. A band at 3700 cm -1 , shows the presence
brucite.
Absorbance
Bentonite
Precipitate
3614
3433
3614
3700
4000 3500 3000 2500 2000 1500 1000 500
Wavenumbers (cm -1 )
Figure 5. FTIR spectra of precipitate and bentonite in the interface
X-ray diffraction of the bentonite shows the possible presence of the new phase. It shows a peak at
1.53 Å only in bentonite interface that belongs to a smectitic trioctahedral phase (saponite) (Fig. 6).
Figure 6. XRD pattern of the bentonite interface and deconvolution of the region between 58 and
61º.
3.3 Bentonite
sm
sm
Quartz
K-Feldspar
sm
20 40
2θ(CuKα)
60
Chemical analysis for exchangeable cations do not shows variation in concentration of Ca 2+ and K +
along of the compacted bentonite block. Nevertheless, an increase of Na + and a drop in Mg 2+ is observed
at the interface. These facts are possibly due to the uptake of Na + from saline solution in the
456
1417
1637 1637
1032
1032
Saponite
467
521
521
467

exchangeable positions of bentonite and the release of Mg 2+ from these places to octahedral sites.
Because of this, a new smectitic trioctahedral phase (saponite) just at the interface can be formed.
Analysis of soluble salts shows a saline front moving towards the heater (Fig.7). The temperature
gradient (25ºC section 1 - 100º C section 4) causes an increase of [Cl - ], [Na + ] and [SO4 2- ] in the
section next to the heater.
Concentration (μg/g)
3500
3000
2500
2000
1500
1000
500
0
K
Interface Section 1 Section 2 Section 3 Section 4
+
Ca2+ Mg2+ 500
0
K
Interface Section 1 Section 2 Section 3 Section 4
+
Ca2+ Mg2+ 457
Cl- Cl- Na + Na +
SO 2-
4
Figure 7. Distribution of soluble salts through compacted bentonite
4. Conclusions
Degradation of the concrete was caused mainly by Ca-leaching, carbonatation and precipitation of
secondary phases such as ettringite and gypsum. A variation in the mineralogy and chemistry of the
bentonite in contact with concrete was observed. At the concrete/bentonite interface, different secondary
mineral phases were identified CSH-gel tobermorite-type (Ca/Si = 0.6), portlandite, brucite
and ettringite. Accessory minerals, such as quartz, present some dissolution signs. Sodium replaces
magnesium from the exchange complex in the bentonite close to the interface. Mg 2+ could belong to
a new smectitic trioctahedral phase, saponite, just formed in the first millimetres of bentonite interface.
Concentration of chlorides sulphates and sodium increases near the heater.
5. Acknowledgements
This work has been financially supported by Enresa and the European Commission NF-PRO project
under contract FI6W-CT-2003-02389.
References
[1] ENRESA; AITEMIN; CIEMAT. FEBEX. Full-Scale Engineered Barriers Experiment in
Crystalline Host Rock. Bentonite: Origin, Properties and Fabrication of Blocks. PT-04/98
(1998).
[2] Villar M. V. Thermo-hydro-mechanical characterisation of a bentonite from Cabo de Gata. A
study applied to the use of bentonite as sealing material in high level radioactive waste repositories.
ENRESA PT-04/02 (2002).
[3] A. Hidalgo, S. Petit, C. Domingo, C. Alonso and C. Andrade. Microstructural characterization
of leaching effects in cement pastes due to neutralisation of their alkaline nature. Part I: pportland
cement pastes.Cement and concrete Research Vol. 37 (2007) pp. 63-70
[4] J.Madejová, P. Komadel. Baseline studies of the clay minerals society source clays: Infrared
methods. Clays and Clay Minerals, Vol 49. No 5. 2001. pp. 410-432.

458

NF-PRO – Influence of THM-GCh Behaviour of the Bentonite Barrier
on the Corrosion Processes of the Carbon Steel Canister
Elena Torres 1 , Alicia Escribano 1 , María Jesús Turrero 1 , Pedro Luis Martín 1 , María Victoria Villar 1 ,
Javier Peña 1 , Juan Luis Baldonedo 2 , Paloma Gómez 1
Summary
1 CIEMAT, Madrid, Spain
2 UCM, Madrid, Spain
The effects of the reactions occurring in the canister/compacted bentonite interface should be
understood for assessing the waste isolation. THM parameters and geochemical conditions of
the clay barrier will vary during the performance of the repository. The experiments performed
by CIEMAT within the framework of NF-PRO project are based on a series of short,
medium and long term experiments conceived at different scales. They were designed in order
to provide useful information about the interaction between the clay barrier and the metallic
canister under realistic conditions. Results obtained in short term experiments show the existence
of a saline front moving towards the heater. The analysis of corrosion products show
that the initial precipitation of chloride plays a relevant role in the first stages of the corrosion
process, as it helps to initiate the nucleation of goethite. When bentonite at the interface gets
desiccated, goethite transforms into hematite.
1. Introduction
The effects of the reactions occurring in the canister/compacted bentonite interface should be understood
for assessing the waste isolation Thermo-Hydro-Mechanical (THM) parameters and Geochemical
(GCh) conditions of the clay barrier will vary during the performance of the repository.
Depending on the relative humidity, density and degree of saturation of bentonite will change. The
water content of the bentonite close to the canister and the increase in temperature due to the decay
of high level nuclear wastes will condition the corrosion mechanism of the overpack. The advance
of salts towards the hottest part of the barrier could favour the occurrence of localized corrosion if
anaerobic conditions have not been achieved.
The experimental studies realized by CIEMAT, within the framework of NF-PRO project, are focused
on the study of the carbon steel/bentonite interface and are based on a series of short, medium
and long term experiments conceived at different scales, from conventional laboratory experiments
and experiments in cylindrical cells, to those specifically designed 3D mock up experiments.
The aim of this work is to study the influence that thermo-hydraulic properties of the bentonite has
on the geochemical evolution of clay barrier and how the coupling of both conditions corrosion of
the carbon steel canister.
459

2. Methodology
In the case of cylindrical cells, the experimental set-up consists on hermetic cells (Fig. 1a) where a
compacted block of FEBEX bentonite (1.65 g/cm 3 ) in contact with iron powder at its bottom is hydrated
with reduced granitic water (Grimsel, Switzerland) on the top while a thermal load is applied
from the bottom. The body of the cell is made out of Teflon, although an external steel cylinder prevents
its deformation by swelling. A plane heater (100ºC) constitutes the bottom of the cell while,
on the top of the cell, a chamber allows the circulation of water at a controlled temperature (around
22ºC), so a thermal gradient is established. Two sensors, placed at 25 and 80 mm from the top of
the bentonite block, record the evolution of relative humidity and temperature as the hydration front
advances.
FEBEX bentonite has a content of dioctahedral smectite of the montmorillonite type of 92 3%. It
contains variable quantities of quartz (2 1%), plagioclase, cristobalite (2 1%), potassium feldspar
(traces), calcite (traces) and trydimite (traces). The cation exchange capacity is of 102 meq/100 g,
and the exchangeable cations are Ca (34.77 2.36 meq/100g), Mg (30.96 3.10 meq/100g), Na
(27.12 0.23 meq/100g) and K (2.58 0.4 meq/100g). The water content of the clay at laboratory
conditions is about 13.7 ± 1.3 %.
Measures of dry density and water content were performed on three samples collected at the end of
the tests at different distances from the interface bentonite-Fe powder. Soluble elements were analysed
in aqueous extract solutions at a solid to liquid ratio of 1:8 (5 g of clay in 40 ml of water reacted
for 24 h.). Additional measures of soluble salts were realized at the interface. The sampling of
the bentonite block is shown in figure 1b.
86.8 mm
13 mm mm mm mm mm
Hydration
FEBEX
bentonite
Fe powder
&0 60 μm
Heating
25 mm
Temperature 80 mm
and RH sensors
Hydration – 25ºC
460
Section III
Section II
86.7 mm
Figure 1. a) Scheme of the cylindrical cells used; b) Sampling of the bentonite block.
Iron powder was analysed by means of Transmission Mössbauer Spectroscopy (TMS), Scanning
Electron Microscopy (SEM) coupled to Energy Dispersive X-ray Spectroscopy (EDS), Transmission
Electron Microscopy (TEM) and Scattered Area Electron Diffraction (SAED).
Section I
Interface
Heating – 100ºC

3. Results
3.1. Thermo-hydraulic properties
The distribution of water content along the bentonite column occurs soon after the beginning of the
experiment. As observed in figure 2, Relative Humidity increases in the coldest end of the block,
where hydration is applied. Close to the Fe/bentonite interface, however, where temperature is
about 100ºC, bentonite loses its absorbed water. Water content of samples collected near the interface
(section I) remains below the initial value (13.7%) in both tests.
Apart from the hydraulic gradient, there is also a dry density gradient caused by the different swelling
of bentonite, since the more hydrated sections swelled more. In the zones affected by hydration,
the densities decreased below the initial value (nominal 1.65 g/cm 3 ) due to the expansion caused by
saturation. On the contrary, near the heater, the dry density increased, due to the shrinkage caused
by desiccation.
% Relative Humidity
100
80
60
40
RH% - Hydration
Temperature - Heater
Temperature - Hydration
20
RH% - Heater
0 50 100 150
Time (days)
100
Temperature (ºC)
Relative Humidity %
80
60
40
20
461
100
80
60
40
RH% - Hydration
Temperature - Heater
Temperature - Hydration
RH %- Heater
20
0 100 200 300 400
Time (days)
Figure 2. Evolution of Relative Humidity and Temperature along the bentonite block in the cells
dismantled after a) 6 months; b) 15 months.
3.2. Geochemical evolution
The hydration of bentonite produces the dissolution of the soluble accessory minerals in the bentonite
(sulphates, carbonates and chlorides). These anions can have a certain influence on corrosion
processes. Carbonates can precipitate at the interface as siderite (FeCO3), once the barrier is fully
saturated. Chlorides and sulphates are hygroscopic and its precipitation can favour the starting of
corrosion processes or enhance it, even at low relative humidity. As expected, soluble carbonates
were only detected in saturated areas (934 and 980 mg of CO3 2- per kilogram of dry bentonite, for
the cell dismantled after 6 and 15 months, respectively). Figure 3 shows the advance of the chloride
and sulphate fronts towards the heater. In the cell dismantled after 15 months, it is observed that
most of soluble chloride is concentrated at the interface. So, it seems that after the initial precipitation
of chlorides when heating started, there is a later precipitation that results from the advective
transport of chloride towards the heater[1]. Sulphate concentration is controlled by gypsum solubility.
So, in both tests, the sulphate concentration at the interface is much smaller than that of chloride.
3.3. Corrosion processes
Different corrosion processes were observed in iron powder depending on the distance from the interface.
In both tests, iron oxide accumulations were found at the interface (Fig.4). In the case of the
100
80
60
40
20
0
Temperature (ºC)

cell dismantled after 6 months, room temperature Mössbauer spectra of the bentonite collected at
the interface shows a sextet (quadrupolar splitting: -0,27 mm/s; isomeric shift: 0,25 mm/s; hyperfine
field: 100-350 KOe) typical from goethite. In the FTIR spectra recorded for the iron oxide found at
the interface after 15 months, there is a doublet placed at 570 and 478 cm -1 . These bands are characteristic
from hematite. EDS analysis detected high contents of chlorine (up to 5% at. in some cases),
and traces of common elements present in bentonite either in goethite or in hematite.
Iron powder seems to undergo slight corrosion. In the cell dismantled after 6 months, Fe powder
kept metallic luster and no corrosion products could be identified on it. In the cell dismantled after
15 months, it was observed the existence of corroded and non-corroded areas in iron powder. EDS
analysis of corroded iron powder detected, in most cases, traces of chlorine (0.6% at.) and calcium
(0. 3% at.). Precipitation of chloride was not homogeneous and where it was found, a thin layer of
hematite ( -Fe2O3) (Fig. 5) and maghemite (�-Fe2O3), was grown (below 1 m in all cases). In this
stage, corrosion may continue, however to a lesser extend, as there is no water available at the interface.
mg Cl / Kg of dry bentonite
4000
3000
2000
1000
0
Hydration - 25ºC
6 months
15 months
Heater - 100ºC
Section III Section II Section I Interface
462
mg SO 2-
4 /Kg of dry bentonite
1500
1000
500
0
Hydration - 25ºC
15 months
6 months
Heater - 100ºC
Section III Section II Section I Interface
Figure 3. Movement of chlorides (left) and sulphates (right) along the bentonite block (aqueous extract
solid: liquid 1:8).
Figure 4. (Left) Backscattered SEM image of the Fe/bentonite interface in the six-month test;
(Right) SEM image of hematite found at the interface after 15 months of experiment.
4. Discussion
30 �m �m
Fe 31.4%
O 51.4%
Cl 5.2%
Si 6.8% %
at.
20 �m
Fe 33%
O 57.2%
Cl 3.24%
Ca 0.37% %
at.
The redistribution of water content along the bentonite column occurs soon after the beginning of
the experiment. Desiccation of bentonite causes the precipitation of chlorides and calcite and the
prevalence of anhydrous iron oxides over iron oxyhydroxides[2]. Chlorides and sulphates are hygroscopic
salts. They are known to significantly enhance the corrosion of mild steel at relative hu-

midities as low as 40% in a temperature range between 60 and 90ºC[3]. Aqueous salt solutions
formed after the precipitation of chlorides onto iron powder enable aqueous film electrochemical
corrosion to occur. This could explain the high concentrations of chlorine found in goethite, which
is a typical product of “wet corrosion”, in the cell dismantled after 6 months. At longer times,
hematite is formed at the Fe/bentonite interface by the thermal dehydration of goethite.
Hematite and maghemite found in the iron powder reacted for 15 months seems to have been
formed by the thermal transformation of a hydrated iron oxide (goethite, lepidocrocite, ferrihydrite),
as well. However, in this case, corrosion happens in a much lesser extend and the transformation
may be faster. Chloride traces found in corroded iron powder may help corrosion to start where it
was deposited.
50 �m
50 nm
Figure 5. (Left) TEM image of a hematite suspension obtained from corroded iron powder and
SAED pattern of one of the crystals; (right) EDS analysis of the diffracted crystal.
5. Conclusions
Initial precipitation of chloride plays a relevant role in the first stages of the corrosion process, as it
helps to initiate the nucleation of goethite. When bentonite at the interface gets desiccated, goethite
transforms into hematite. Once, the chloride front reaches the interface and precipitates onto the
iron powder, it seems that corrosion is enhanced again. Results obtained in these tests are preliminary
and will be completed after the dismantling of the four experiments on-going at the moment.
6. Acknowledgements
This is a contribution to the NF-PRO IP (Integrated Project) number FI6W- CT-2003-02389 financed
by the European Union and the CIEMAT/ENRESA association.
References
[1] M.V. Villar, A.M. Fernández, P.L. Martín, J.M. Barcala, R. Gómez-Espina, P. Rivas, Effect
of heating/hydration on compacted bentonite: tests in 60-cm long cells, CIEMAT, Madrid,
2008, p. 76.
[2] J. Majzlan, K.-D. Grevel, A. Navrotsky, Thermodynamics of Fe oxides: Part II. Enthalpies of
formation and relative stability of goethite, lepidocrocite and maghemite, Am. Mineral. 88
(2003) 855-859.
463

[3] G.E. Gdowski, Humid Air Corrosion of YMP Waste Package Candidate Material, CORRO-
SION NACExpo 98, San Diego, CA, 1998.
464

NF-PRO – Experimental and Modelling Studies of the THM Behaviour
of the Clay Barrier
María Victoria Villar 1 , Marcelo Sánchez 2 , Roberto Gómez-Espina 1 , Antonio Gens 3
Summary
1 CIEMAT, Spain
2 University of Strathclyde, UK
3 CIMNE, Spain
A THM mathematical formulation and a computer code to include ‘non-standard’ THM processes
and phenomena have been extended and improved. This upgraded approach has been
used to model some specifically designed laboratory tests. Among the phenomena investigated
are the parameters and processes influencing the hydration kinetics of the clay barrier,
the water flow under low hydraulic gradients similar to those expected towards the end of the
saturation process, the impact of temperature on the hydro-mechanical properties of the bentonite,
and the effect of clay fabric changes on the behaviour of bentonites submitted to heating
and hydration.
1. Introduction
The integration between the basic laboratory research and the development, application and calibration
of numerical models allows to improve the knowledge about the thermal, hydraulic and mechanical
processes (THM) –considered in a coupled way– that control the behaviour and performance
of the clay barrier under the conditions of a Deep Geological Repository (DGR). CIMNE has
extended and improved a THM mathematical formulation and a computer code to include in the
analysis ‘non-standard’ THM processes and phenomena. This upgraded approach has been used to
model some specifically designed laboratory tests performed by CIEMAT with FEBEX bentonite.
The following phenomena have been investigated:
- The parameters and processes influencing the hydration kinetics of the clay barrier, especially
with respect to the effect of thermal gradient. At advanced stages of the system, when the hydraulic
gradient becomes very small, it is possible that coupled phenomena, such as thermoosmotic
flow, could have a noticeable effect on the behaviour of the barrier, causing a trend to
slow down the hydration in the zones close to the containers.
- The water flow under low hydraulic gradients similar to those expected towards the end of the
saturation process. Under low hydraulic gradients Darcy’s simple relationship does not rule the
liquid flow in some soils, probably due to the strong clay-water interactions.
- The impact of temperature on the hydro-mechanical properties of the bentonite, since temperature
changes affect important hydraulic characteristics of compacted clays, whose knowledge is
465

crucial to predict the hydration rate of the barrier, and also the mechanical response of the material,
which has important implications on the design and performance of the repository.
- The effect of clay fabric changes on the behaviour of bentonite submitted to heating and hydration.
2. Results
2.1 Effect of hydraulic gradient on permeability
The hydraulic conductivity of FEBEX clay samples compacted to dry densities between 1.4 and
1.65 g/cm 3 has been measured under low hydraulic gradients (between 200 and 2400) and low injection
pressures (from 200 to 7200 kPa). From the results obtained, the following preliminary remarks
can be made [1]:
- No clear evolution of permeability with time has been observed for more than 900 days.
- The average hydraulic conductivity values obtained are in the order of those obtained for the
same dry densities applying higher hydraulic gradients.
- The permeability values obtained are not dependent on the hydraulic gradient applied, and
there is an overall proportionality between flow and hydraulic gradient. However, the erratic
relationship between flow and hydraulic gradient found for hydraulic gradients lower than
2000, could indicate that the critical gradient for this bentonite would be around this value. The
critical gradient is the hydraulic gradient below which flow occurs but it is not Darcian.
- No measurable flows have been obtained when hydraulic gradients below 200 have been applied
in a sample of dry density 1.4 g/cm 3 . This value could be regarded as a threshold hydraulic
gradient for dry density 1.4 g/cm 3 , since no flow has been obtained below this gradient. For
the dry density 1.65 g/cm 3 the threshold hydraulic gradient would be 1400. However, there is a
dependency of these values also on the injection and backpressures applied: the higher are both,
the lower the threshold hydraulic gradient.
2.2 Evolution of hydration of bentonite with and without thermal gradient
Two infiltration tests performed in cylindrical cells with FEBEX bentonite compacted to nominal
dry density 1.65 g/cm 3 started in the framework of the FEBEX Project and continued in the NF-
PRO Project, thus they have been running for more than six years [1]. One of them has run under
isothermal conditions (test I40), and the other one under thermal gradient (test GT40), in order to
simulate the conditions of the clay barrier in the repository and better understand the hydration
process. The total height of the columns is 40 cm, and relative humidity and temperature sensors are
placed inside the bentonite.
The tests have shown that the permeability to water vapour of dry bentonite is very high and that
the thermal gradient initially set remains constant during the whole test. The initial hydration of
compacted bentonite takes place quicker under thermal gradient than at laboratory temperature.
However this behaviour is reversed as saturation proceeds and later on, the water intake is higher
for the sample tested at room temperature, because the hot zones of the sample tested under thermal
gradient remain desiccated for a long time (Fig. 2.1).
466

These tests have been modelled with a THM coupled formulation based on the macroscopic approach
developed in the context of the continuum theory for porous media by [2] and described in
[3]. The parameters used for the bentonite are given in [3]. The predictions got with this basic
model are shown in Fig. 2.1 as “Base case”. It can be observed that they overestimate the actual hydration
kinetics, especially in the case of the test performed under thermal gradient (GT40). To explain
this discrepancy non-standard processes not previously included in the model have been considered,
among which the existence of a threshold hydraulic gradient and thermo-coupling effects.
The model results obtained considering a threshold gradient of 50, a critical gradient close to 2000,
and a power law for the range of hydraulic gradient with non-Darcian flow, are shown in Fig. 2.1
with solid lines.
Although, at this stage, the new models are quite simplified, the results obtained are interesting. In
this context, this study allows comparing the measurements of actual tests with the computed response
under the hypothetical case in which some of these effects would be present. Each of these
phenomena does not exclude the others and it is possible that an explanation for the whole behaviour
of the barrier would require the combinations of some of them.
Relative Humidity (%)
100
80
60
40
20
TEST
MODEL (TG)
d=0.10m
d=0.20m
d=0.30m
Base case
0
0 10000 20000 30000
Time (hours)
Relative Humidity (%)
467
100
80
60
40
20
MODEL (TG)
d=0.10m
d=0.20m
d=0.30m
0
0 10000 20000 30000
Time (hours)
Figure 2.1. Evolution of relative humidity in three different positions along the bentonite column of
test GT40 (left) and I40 (right): the points correspond to measured vales, the discontinuous line to
the base case model, and the solid lines to the model assuming the existence of threshold gradient
2.3. Effect of temperature on hydro-mechanical properties
The swelling capacity, swelling pressure and permeability of the bentonite compacted to dry density
between 1.5 and 1.7 g/cm 3 have been determined at temperatures between 20 and 80°C [1].
The results show that the effect of temperature on the swelling capacity of the bentonite is smaller
than the effect of the vertical load applied during hydration or the effect of initial dry density. As
shown in Fig. 2.2, in which the final strains obtained in all the tests have been plotted, the effect of
temperature is higher for the higher density and is more noticeable when the load is high. The results
obtained in the swelling pressure tests confirm this trend (Fig. 2.3). On the one hand, all the
dry densities tested, including the lowest one, suffered a decrease of swelling pressure for high
temperature, what would confirm that the effect of temperature is more important as the vertical

load is higher, since the load applied in swelling pressure tests is higher than that applied in swelling
under load tests. On the other hand, the effect of temperature on swelling pressure is less significant
for the lowest density, confirming the trend observed in the swelling under load tests. The
extrapolation of the logarithmic correlation towards higher temperatures would indicate that swelling
pressures higher than 1 MPa would develop even for temperatures of 100°C for the three densities
tested.
Final vertical strain (%)
-25
-20
-15
-10
-5
0
open symbols: 1.5 g/cm 3
filled symbols: 1.6 g/cm 3
5
20 40 60
Temperature (°C)
80 100
468
0.5 MPa
1.5 MPa
3.0 MPa
Figure 2.2. Final vertical strain in swelling under load tests performed with FEBEX bentonite compacted
to nominal dry density 1.6 g/cm 3 (filled symbols and discontinuous lines) and 1.5 g/cm 3
(open symbols and solid lines).
Swelling pressure (MPa)
6
4
2
0
Error bars obtained from values of
tests performed at laboratory
temperature (1.6Mg/m 3 )
Error bars obtained from values
of tests performed at laboratory
temperature (1.5 Mg/m 3 )
Dry density (Mg/m
Test Test
Model Model
3 )
1.6 1.5
20 30 40 50 60 70 80
Temperature (ºC)
Figure 2.3. Swelling pressure as a function of temperature for saturated FEBEX clay compacted to
different nominal dry densities. Experimental and modelling data
The decrease of swelling pressure and swelling capacity of the FEBEX bentonite with temperature
has been explained as a consequence of the transfer of microstructural (interlayer) water to the macrostructure
which is triggered by temperature. This process would be more significant in highdensity
samples, in which the interlayer water predominates initially over the “free”, macroscopic
water.

A double-structure model, based on the general framework for expansive materials proposed by [4]
has been developed to explain the hydro-mechanical behaviour of the bentonite and extended to
consider the effect of temperature on swelling [3]. The existence of two pore structures (macro and
micropores) has been explicitly included in the formulation, what provides the opportunity to take
into account the dominant phenomena that affect the behaviour of each structure in a consistent
way: the swelling features of clay minerals are explicitly considered through a microstructural law,
the relevant effects of the granular-like skeleton are contemplated through the macrostructural law,
and the model also considers the interaction between both structural levels. To account for the temperature
effects, a dependence of the microstructural law on temperature has been suggested (Fig.
2.4), since the clay particles cause the expansive behaviour. In this way, the model is able to capture
the main trends observed in the tests (Fig. 2.3).
(p+s) (MPa)
100.0
10.0
1.0
0.1
Δ T (ºC)
0 20 40 60
1000 1500 2000 2500 3000 3500
K 1 (MPa)
469
τ : 0.12
Figure 2.4. Changes in micro-structural stiffness with temperature
3. Conclusions
The laboratory tests performed, in which the expansive clay and the THM conditions are close to
the ones expected in a DGR, have been very useful to understand the behaviour of the FEBEX clay
and validate the mathematical formulations and computer codes. Most of the constitutive model and
parameters assumed in the analyses are the same ones adopted for the main analysis of large-scale
tests (i.e. FEBEX mock-up and in situ tests). The fact that the model can reproduce the global
trends observed in the tests at different scales with a unique set of parameters points to an appropriate
selection of the processes and parameters.
4. Acknowledgements
Work co-funded by ENRESA and the European Commission (Contract FI6W-CT-2003-02389).
References
[1] Villar, M.V. & Gómez-Espina, R. 2007. NF-PRO. Deliverable 3.2.13. Final report on laboratory
tests performed by CIEMAT for WP3.2. 57 pp.
[2] Olivella, S., Carrera J., Gens, A. & Alonso, E.E. 1994. Non-isothermal multiphase flow of
brine and gas through saline media. Transport in porous media 15: 271-293.
[3] Sánchez, M. & Gens, A. 2007. NF-PRO Project. Deliverable 3.2.14. 93 pp.

[4] Gens, A. & Alonso, E. 1992. A framework for the behaviour of unsaturated expansive clays.
Can. Geotech. J. 29: 1013-1032.
470

Summary
NF-PRO – Mechanical and permeability properties
of highly pre-compacted granular salt bricks
Klaus Salzer, Till Popp, Heinz Böhnel
Institut für Gebirgsmechanik GmbH Leipzig (IfG), Germany
In the salt concept for the disposal of high level radioactive waste in deep geological rock
formations granular salt is the favorite for backfilling of remaining openings because of its
advantageous physical properties. However, its initial sealing capacity is low due to the high
porosity. Besides others, the usage of pre-compacted blocks of crushed salt offers an alternative
sealing concept. Our investigations are aiming to investigate both the mechanical compaction
behaviour and the evolution of transport properties in salt bricks (artificial pressed
granular salt with a porosity of ~8%), as a prerequisite to develop mechanism-based constitutive
models describing these processes in the low porosity region (10 - 1%). A key factor is
the role of water, since it affects in a complex manner the coupled hydraulical/mechanical
properties of granular salt.
1. Introduction
In the salt concept for the disposal of high level radioactive waste in deep geological rock formations
granular salt is the favourite for backfilling of remaining openings because of its advantageous
physical properties because (1) it has good compacting properties, (2) its thermal and mechanical
properties after compaction are similar to the surrounding rock salt, and (3) it is easy available.
Nevertheless a limiting factor for the use of backfill saliferous granulates is their initially high porosity
and permeability, resulting in a low sealing capacity and mechanical integrity just after emplacement.
In the framework of NF-PRO considerable efforts have been obtained regarding the
knowledge about compaction of granular salt, not only with respect to the acting fundamental densification
processes but also to improve its compaction behaviour.
Besides adding bentonite to enhance compaction and to reduce permeability, an alternative concept
is the use of pre-compacted salt stones, e.g. bricks, which are characterized by porosities

2. Experimental Methodology
A salt brick used in our investigations is a cold pressed (p �130 MPa in some few seconds) artificial
rectangular solid (240 mm x 115 mm x 71 mm) of highly compacted crushed salt composed of
nearly pure sodium chloride ( 95% NaCl) with grain sizes, mainly between 0.16 mm and 0.5 mm.
Such specimens of salt bricks have been produced in large series by the company K+S in 1990 in
advance of the planned in situ test dam construction in the Asse salt mine. The fabrication of the
salt bricks is already described in detail in [2].
The laboratory investigations were aiming at a comprehensive data basis regarding compaction and
hydraulical behaviour of the pre-compacted salt bricks (initial porosity

porosity (%)
10
8
6
4
2
208/38
σ 1,3 = 10 MPa
w = 0.64%
208/39
σ 1,3 = 20 MPa
w = 0.46%
0
0 100 200
time (d)
300 400
473
208/36
σ 1;3 = 3 MPa
w = 0.42%
Figure 1. Comparison between measured and calculated long-term creep behaviour for the ZHANG
model, calculations performed with best fit for wetted salt brick samples. Porosity vs. time (symbols
indicate the measured behaviour, continuous lines the calculated one, colours mark the stress conditions
for the creep-test).
a) b)
Figure 2. Hydraulical properties of salt bricks. a) Porosity-permeability data for dry salt brick material
deformed in compression under triaxial loading conditions at room temperature. b) Relationship
between gas threshold pressure and intrinsic permeability for various low permeability rock
formations (e.g. claystones, shales, sandstone) – modified after [4].
Shear tests were performed to investigate contact properties between saltbrick surfaces and the rock
salt. Whereas at dry conditions only some friction occurs, significant strengthening is observed
when moisture is present because of activation of cohesion (Fig. 3). This observation offers a direct

proof of healing in salt rocks. It has to be mentioned that further experiments are necessary to establish
the dependency between the mechanical and hydraulic properties of the interfaces between the
salt blocks among self and between the blocks and the host rock (rock salt).
Figure 3. Shear strength characteristics of lab-dry and moistened brick-brick and brick-rock salt
contacts. Healing effects, as indicated by development of cohesion, are promoted by the presence of
moisture and increase systematically with stationary ageing over 16, 70 and 94 hours at fixed normal
stress (multiple step tests at constant normal stress). – modified after [3].
3.2 Review of appropriate material laws for modelling granular salt compaction
Four published constitutive models (ITASCA, HEIN, ZANG and SPIERS, for details see [1]) describing
crushed salt behaviour have been evaluated and adapted to recalculate the measured longterm
hydrostatic creep behaviour to remaining porosities in the order of some few percent, i.e. in
the order of natural salt.
Actually the ZHANG model [5] seems to be the most suitable constitutive law to describe the compaction
behaviour of the pre-compacted crushed-salt bricks (compare modeling curves for wet salt
in Fig. 1). However an implementation in commercial available codes is necessary allowing prognosis
calculations for real underground situations. Nevertheless, the potential of the SPIERS mechanism
based pressure solution-model (e.g. [6]) for future modeling is obvious but requires more experimental
work.
The new MINKLEY-shear model for contact interfaces, which implies the displacement-dependent
and the velocity-dependent strength softening is a suitable constitutive law to describe the contact
between pre-compacted crushed salt bricks and to the host rock. This shear model is already implemented
in a commercial code allowing prognosis calculation for real underground situations.
The remaining task is to study and to describe the coupled HM-behaviour of such contact interfaces
474

4. Conclusions
Our investigations on artificial salt bricks cover a wide field of relevant rock-mechanical and hydraulical
properties of salt bricks, which demonstrate their favourable properties for usage as precompacted
elements in sealing or back-filling systems. In this context it has to be mentioned that
already practical experiences at relevant in-situ scales are available from the conception of the
“Asse-Damm”- project [2] and the technical realisation of the dam - project “Sondershausen” [1].
However, despite the obvious improvements in knowledge and experimental data base it came out
during the work that some principal challenges in relation to the salt backfill materials still remain,
i.e.
(1) Understanding of physical processes which control the efficiency of granular salt compaction
especially with respect to humidity effects.
(2) Development of generally agreed constitutive models for compaction in granular salt that
can be reliably extrapolated to in-situ conditions.
5. Acknowledgements
The studies presented have been funded as part of the NF-PRO jointly by the European Commission
under the 6 th Euratom Research and Training Framework Programme on Nuclear Energy
(2002-2006), contract FI6W-CT2003-02389, and the German Federal Ministry of Research and
Education under contract 02 E 9904.
References
[1] Salzer, K., Popp, T., Böhnel, H., Naumann, D. & Mühlbauer, J., 2007. Investigation of the
Mechanical Behaviour of Precompacted Crushed Salt in Contact to the Host Rock. Final Activity
Report. 3.5.7. NF-PRO Deliverable 3.5.7: (2007-12-15).
[2] Stockmann, N. (ed.) 1994. Dammbau im Salzgebirge, Abschlussbericht Projektphase II, Berichtszeitraum
vom 01.07.1989 - 31.12.1992, GSF-Bericht 18/94.
[3] Salzer, K., T. Popp & H. Böhnel, (2007): Mechanical and permeability properties of highly
precompacted granular salt bricks. In K.-H. Lux, W. Minkley, M. Wallner, & H.R. Hardy, Jr.
(eds.), Basic and Applied Salt Mechanics; Proc. of the Sixth Conf. on the Mech. Behaviour of
Salt. Hannover 2007. Lisse: Francis & Taylor (Balkema). 239 – 248.
[4] Davies, P.B., 1991. Evaluation of the role of threshold pressure in controlling flow of wastegenerated
gas into bedded salt at the Waste Isolation Pilot Plant (WIPP). Sandia Rep. SAND
90-3246.
[5] Zhang, C. L., Schmidt, M. W., Staupendahl, G., Heemann, U. (1993): Entwicklung eines
Stoffansatzes zur Beschreibung des Kompaktionsverhaltens von Salzgrus. Bericht Nr. 93-73
aus dem Institut für Statik der Technischen Universität Braunschweig.
[6] Spiers, C. J., Peach, C. J., Brzesowsky, R. H., Schutjens, P. M. T. M., Liezenberg, J. L.,
Zwart, H.J. (1989): Long-term rheological and transport properties of dry and wet salt rocks,
Nuclear Science and Technology, EUR 11848 EN, Office for Official Publications of the European
Communities, Luxembourg. 1989.
475

476

Summary
NF-PRO – Impact of Bedding Planes to EDZ-Evolution
and the Coupled HM Properties of Opalinus Clay
Till Popp, Klaus Salzer, Wolfgang Minkley
Institut für Gebirgsmechanik GmbH Leipzig (IfG), Germany
Field observations at the Mont Terri site demonstrate an Excavation Damage Zone (EDZ)
around tunnels consisting of a complex crack network related to the bedding and the existing
stress field. To separate the different impacts of the rock mass and bedding planes laboratory
investigations were performed on Opalinus clay samples including triaxial strength and direct
shear tests. During the deformation tests monitoring of ultrasonic wave velocities, permeability
and volumetric strain measurements facilitates the detection of stress induced onset of
damage. Two stress dependent criteria were estimated, referring to various damage states, i.e.
(1) to initial onset of damage and (2) occurrence of dilatancy. Based on the experimental results
a new modelling approach for a prognosis of the evolution of the EDZ is developed consisting
of two parts: (1) a (visco-)elasto-plastic constitutive model, comprising the hardening/softening
behaviour and dilatancy effects of the rock mass, and (2) a specific friction
model, which described displacement- and velocity-dependent shear strength softening for the
bedding planes. The capability of the new approach is demonstrated by recalculating the spatial
development of EDZ around a drift at the Mont Terri site.
1. Introduction
Besides other host rocks argillaceous clay rock formations are considered for the long term storage
of radioactive waste to exclude a threat to actual and future generations. Argillaceous rocks are inherently
anisotropic due to their sedimentary and tectonical history. In-situ observations demonstrate
that during excavation of underground openings various failure mechanisms and bedding related
phenomena have to be considered (Fig. 3a). Both overlapping effects of brittle matrix behaviour
and bedding plane slip are particularly important during rock stress redistribution in the EDZ.
In addition to their importance as mechanical weakness planes, bedding planes can also act as preferential
flow paths. Since the transport properties of the clay are responsible for the demanded integrity
of a radioactive waste repository, knowledge about the relationship between development of
damage (dilatancy) and hydraulical properties is of utmost importance. In the lab the amount of
damage of rocks is generally described by the parameter dilatancy, i.e. the development of microfractures
depending on the state of stress (stress field geometry and deviator). The determination of
the stress dependent onset of dilatancy is, therefore, of predominant importance for an appraisal of
barrier properties of solid rocks. Focusing on these topics IfG performed in the framework of NF-
PRO a rock-mechanical study divided in a laboratory part and numerical modelling, based on the
477

obtained experimental results [1]. For the investigations Opalinus clay was used as a reference material.
2. Experimental Methodology
Two experimental approaches were used concerning the detection of stress dependent onset of micro-cracking:
Triaxial multi-anvil apparatus: Deformation tests on well oriented sample cubes facilitated a precise
detection of the onset of micro-cracking resp. dilatancy as indicated by seismic velocity
measurements in the three directions (respectively Vp, Vs1 and Vs2 for determination of shear
wave splitting (for details and results see [2]).
Triaxial Kármán cell: Short term triaxial tests with simultaneous monitoring of p- and s-wave
velocities supplemented by permeability and volumetric strain measurements offered information
on coupled HM properties of Opalinus clay loaded and to the bedding.
Focusing on quantification of shear strength properties parallel to the bedding numerous shear tests
were performed on large sample blocks (of approx. 100 mm x 100 mm x 200 mm) parallel to the
bedding in the IfG direct shear apparatus. The obtained results cover a wide range of applied normal
stresses and displacement rates and delivered a reliable estimate of shear properties (i.e.
strength and dilatancy angle) which is an important base of latter rock mechanical modelling.
3. Results
3.1 Experimental results:
Both, seismic monitoring and permeability measurements clearly indicate pre-damage of the investigated
core samples, and, in addition, only partial saturation due to sample disturbances during
sample recovery and preparation. However, increase of confining pressures restores at least partially
the initial sample integrity and saturation conditions as may be inferred e.g. by the observed
permeability decrease respectively seismic velocity increase. The time and stress depending sealing
was found most efficient perpendicular to the bedding.
Differential stress (MPa)
50
40
30
20
II bedding
⊥ bedding
Volumetric strain (%)
0,5
0,4
0,3
0,2
0,1
0,0
-0,1
-0,2
10
-0,3
⊥ bedding 1E-20
0
-0,4
-0,5
II bedding
1E-21
⊥ Bedding
0,0 0,5 1,0 1,5 2,0
0,0 0,5 1,0 1,5 2,0
0,0 0,5 1,0 1,5 2,0
a) Axial strain (%)
b) Axial strain (%) c) Axial strain (%)
478
Permeability (m 2 Permeability (m )
2 )
1E-14
1E-15
1E-16
1E-17
1E-18
1E-19
II Bedding
10 MPa : OPA9
5 MPa : OPA2
3 MPa : OPA4
2 MPa : OPA1
10 MPa : OPA11
3 MPa : OPA Kon1
Figure 1. Summary of triaxial test results on Opalinus Clay in a confining pressure range between
2 and 10 MPa. Note the differences between both groups of loading direction and the well pronounced
effect of increasing confining pressure.

Figure 1 depicts stress strain curves in relation to deformation induced changes of the volumetric
strain and permeability from the various strength tests on Opalinus clay with different loading directions.
Despite the limited number of tests the rock-mechanical test results clearly document that
the strength of Opalinus Clay is sensitive to minimal stress and the stress direction related to the
bedding. In this context it has to be mentioned that detection of initial micro crack-opening in indurated
clay under lab conditions depends mainly on the sensitivity of the measured physical parameter
because during triaxial loading an overall matrix compaction of the clay rock occurs.
Under deviatoric conditions local initial crack opening at 50 – 60% of the failure stress is only indicated
by velocity decrease of radially measured p-waves or s-waves (oscillation direction to crack
planes, compare [2]). Primary at around 90% of the failure stress an increase of volumetric strain
was observed associated with a permeability increase. In consequence, two pronounced stress
boundaries have to be considered representing different stages of damage:
initial damage � 0.5 - 0.6 * peak , respectively dilatancy 0.8 - 0.9 * peak
Rock matrix
(Visco-)elasto-plastic model
it describes as a function of elasto/plastic strain :
60
• hardening
⎛ σMax − σ ⎞ D
50
σ1= σD+
⎜
⎜1
+
⎟ σ 3
σ ϕ + σ ⎟
⎝
3 ⎠
• softening / failure
40
• dilatancy
⋅
(Visco-)elasto-plastic model
it describes as a function of elasto/plastic strain :
60
• hardening
⎛ σMax − σ ⎞ D
50
σ1= σD+
⎜
⎜1
+
⎟ σ 3
σ ϕ + σ ⎟
⎝
3 ⎠
• softening / failure
40
• dilatancy
⋅
Axial stress σ 1 ( MPa )
Volumetric strain V/V ( % )
30
0% plast. strain
0,04% plast. strain
0,1% plast. strain
0,24 plast. strain
20
0,4% plast strain
ε p ( % ) 0,00 0,04 0,10 0,24 0,40
10
σ D ( MPa )
σ φ ( MPa MPa )
10,5
5,0
8,6
5,0
6,0
5,0
2,0
5,0
0,0
5,0
0
σ Max ( MPa ) 55,0 48,0 43,0 38,0 35,0
0 5 10 15 20
2,5
2,0
1,5
1,0
0,5
Confining pressure σ 3 ( MPa MPa )
εp εp ( % ) 0,00 0,04 0,10 0,24 0,4
σ ψ ( MPa ) 3,2 3,1 2,9 2,5 2,0
tan β° 0,15 0,3 0,5 1,0 2,0
0% plast. strain
0,04% plast. strain
0,1% plast. strain
0,24 plast. strain
0,4% plast strain
0,0
0 5 10 15 20
Confining pressure σ 3 ( MPa )
Bedding plane properties
Extended Minkley shear model
it describes the behaviour of weakeness planes
on the basis of: μ K = kinetic friction
Δμ = adhesive friction
τ adhesion = μ K ( 1 + Δ μ ) ⋅ σ N + c
c = cohesion
Numerical modeling of clay rocks as
„discrete material“ using UDEC: The
bedding planes are treated as discontinuities
between blocks (matrix)
479
τ
Shear stress ( MPa )
σ N
Normal stress
σ Ν = 10 MPa
Shear displacement ( mm )
Medium 1
Medium 2
Figure 2. The new modelling approach based on matrix and bedding plane properties. (centre) the
reference case: a specific drift situation at the Mont Terri site with the relevant deformation styles,
i.e. extensional failure and bedding plane slip in the roof respectively brittle failure in the wall. (left
side) the MINKLEY-elasto-plastic constitutive model for the rock matrix (parameter curves of the experimental
data); (right side) the new developed MINKLEY-shear model for describing bedding
plane properties. The inset shows the numerical simulation of a shear test. For details of the used
models see [1].
3.2 Modelling of the EDZ around a specific drift situation at Mont Terri
With respect of a prognosis of the EDZ a new modelling approach has been developed based on the
obtained experimental results (Fig. 2). It consists of two parts, i.e. of the MINKLEY-(visco-)elastoplastic
constitutive model comprising the hardening/softening behaviour and dilatancy effects of the
rock mass and a specific shear friction model, which describes displacement- and velocity-

dependent shear strength softening for the bedding planes (for details see [1]). For the modelling we
applied the commercial UDEC (i.e. Universal Distinct Element Code of Itasca).
Using the site-specific material parameters of matrix properties (derived from the triaxial tests) and
bedding planes (derived from the shear tests), the relevant EDZ phenomena for a local drift situation
in the Mont Terri lab (e.g. tensile fractures at the wall respectively shear slip and tensile fractures
in the roof, could be nicely simulated. Also the spatial extent of the EDZ corresponds fairly
well to the in-situ observations (Fig. 3).
stiffness of bedding planes: 1000 GPa/m
tension forces
16
14
12
10
8
6
4
2
0
1.0
σ v (MPa) (MPa) σ σD D (MPa) (MPa) ε V (%)
σ D (MPa)
a) b) c) c) d)
e) e)
480
2
4
6
8
10
0.1
0.2
0.3 0.3
0.4 0.4
0.5
0.6
0.7 0.7
0.8
0.9
1 GPa/m GPa/m
Figure 3. Simulation of the 2D-situation at the Mont Terri site (Ødrift = 3.5 m) – combined approach
of elasto-plastic description of the rock mass behaviour and the MINKLEY shear model for simulating
the reduced shear strength in the bedding – Note: the influence of bedding plane stiffness (b –
d: bedding plane stiffness = 1000 GPa/m respectively e: 1 GPa/m. a) sketch of the fracture patterns
of the Mont Terri site (modified after Blümling, presented during the NF-PRO second training
course in Cardiff, 2005). b) Vertical stress distribution; c) the distribution of the damage parameter
D; d) the volumetric strain V. and e) the distribution of the damage parameter D.
4. Conclusions
As outcome of our integrated study of experimental and modelling work, it has been confirmed that
in bedded clay formations the evolution of the EDZ mainly depends on rock properties of the indurated
clay and the induced stress state whereby anisotropy effects related to the bedding are obvious.
Deviatoric loading is initially accompanied by a permeability decrease up to 80 - 90% of the failure
stress. Only if localized shear failure takes place a pressure dependent permeability and volume increase
is observed. However, the measured seismic velocities clearly indicate local damage at significant
lower stress conditions. Hence, it has been convincible shown, that due to the overlapping
effect of matrix compaction the detection of dilatancy, i.e. onset of microfracturing, depends on the
sensitivity of the measured physical parameter and the measuring direction. Therefore, the reliability
of the term „dilatancy“, regarding its importance for the EDZ in indurated clays, needs to be dis-
2
4
6
8
10

cussed. In addition, a simple coupling between mechanical (i.e. damage) and hydraulical properties
seems to be unlikely. Although it was demonstrated that the dilatancy concept is also valid in clay
rocks, it‘s application is much more difficult due to the overlapping effects of the porous rock matrix
and anisotropy due to bedding.
The 2D-simulation of the EDZ around a tunnel under the site specific conditions of the Mont Terri
Site demonstrated the capability of the new approach. As came out by the modelling results, the geometry
of the EDZ during excavation is mainly controlled by the instantaneous reaction of the rock
mass on the anisotropic stress regime (elasto-plastic behaviour stress induced tensile failures at
the wall) and by the overlapping bedding plane weakness ( tensile and shear fractures) which
fairly well agrees to in situ observations [3].
So far, we can conclude that high sophisticated rock-mechanical constitutive models are available
to model the development of the EDZ from a mechanical point of view [1, 3]. But looking on the
real in-situ-conditions it becomes obvious that more understanding of the complex effects of humidity
and pore pressures is required for their numerical implementation into the existing constitutive
models. Nevertheless possible consideration of time-dependent effects, i.e. creep, can be easily
done by determining the necessary parameters for the viscous part of the MINKLEY-constitutive
model. As recommendation for future work we have to conclude that more experimental lab and
field investigation are necessary to describe the complex THMC-behaviour of argillaceous clays,
mainly separating effects due to drained and un-drained conditions.
5. Acknowledgements
The studies presented have been funded as part of the NF-PRO jointly by the European Commission
under the 6 th Euratom Research and Training Framework Programme on Nuclear Energy,
(2002-2006), contract FI6W-CT2003-02389, and the German Federal Ministry of Research and
Education under contract 02 E 9874.
References
[1] Popp, T. & Salzer, K., 2007. Impact of bedding planes to coupled HM properties of the damaged
rock – combined rock mechanical laboratory investigations and modelling. NF-PRO Deliverable
4.2.15: Final report (2007-10-01).
[2] Popp, T. & Salzer, K., 2007. Anisotropy of seismic and mechanical properties of Opalinus
clay during triaxial deformation in a multi-anvil apparatus. Physics and Chemistry of the
Earth, Parts A/B/C, 32, 8-14, 879-888.
[3] Blümling, P. & Konietzky, H., 2003. Development of an excavation disturbed zone in claystone
(Opalinus Clay). Geotechnical measurements and modelling: Proceedings of the international
symposium, 23-25 September 2003, Karlsruhe, Germany; Natau O., Fecker, E., Pimentel
E. (eds.); Lisse [et al.]: A.A.Balkema Publishers, 127-132.
481

482

NF-PRO – Studies on Long-Term Stability of Spent Fuel
Vincenzo V. Rondinella, Detlef Wegen, Thierry Wiss, Daniel Serrano-Purroy,
Joaquin Cobos-Sabate, Dimitrios Papaioannou, Jean-Pol Hiernaut, Jean-Paul Glatz
European Commission, Joint Research Centre, Institute for Transuranium Elements,
P.O. Box 2340, 76125 Karlsruhe, Germany
Summary
In the frame of institutional research activities and as contributions to the Integrated Project
NF-PRO, a series of experimental campaigns was carried out at the Institute of Transuranium
Elements (ITU), a Joint Research Centre of the European Commission. Within NF-PRO, the
experiments were included in the WP 1.4 "Evolution of spent fuel prior to water ingress (normal
evolution scenario) and during the transient period (early failure scenario) and impact on
radionuclide release". Source term investigations to determine the grain boundary inventory
of MOX fuel, as an important component of the instant release fraction (IRF), were
performed. For these studies an innovative approach combining Knudsen cell effusion
measurements and leaching experiments was implemented. A further objective of WP1.4 was
to improve knowledge on the microstructural properties of spent fuel and their possible
changes in the scenarios considered. The mechanical stability of spent fuel is important
because it affects the surface area of fuel potentially available for leaching. The effects of
alpha decay damage on the mechanical stability of spent fuel were investigated using alpha
doped UO2 and (U,Pu)O2. The techniques used were SEM/TEM combined with annealing and
DSC measurements. Finally, potential alteration processes of the spent fuel in an early failure
scenario, namely during the transient phase between initial breach of the canister and water
ingress in it, were investigated. This work involved exposing fuel rodlets containing
intentionally manufactured defects in the cladding to a water-saturated atmosphere containing
argon and/or hydrogen to simulate the atmosphere inside the canister emplaced in a geologic
repository. The condition of the fuel inside the cladding was evaluated at the end of the
experiments by SEM/EDS while gas and liquid phase were examined by mass spectrometry.
1. Grain boundary inventory and instant release fractions for MOX
An estimation of the grain boundary inventory of radionuclides belonging to the so called "labile
fraction" was performed for ~39 GWd/t SBR mixed oxide fuel (MOX) (BNFL - now Sellafield Ltd)
[1-2] adopting a combined approach including effusion and leaching tests. Very scarce or no data
were previously available concerning grain boundary leaching/inventory for MOX fuel, after the
work of Gray et al in the’90 in PNNL [3-5], more recently extended on UO2 fuel by Hanson et al.
[6], Roudil et al. [7], Garisto et al. [8], and reviewed by Johnson et al. [9]. The experimental results
obtained in this work indicate that for the species showing fractional release higher than U, i.e. Rb,
Sr, Mo, Cs, Tc, Ba, fractions typically �1 % are released from grain boundary if only the amounts
released after removing pre-existing oxidized surface layers on the sample are considered. By
including also the amounts released during preliminary “washing” cycles, the highest observed
483

elease falls typically between 1 and 2 % for all these species except Ba; the highest value for the
fraction released was determined for Cs (2.3%). These values are generally lower than labile
inventory values reported in literature from tests on UO2 including also gap contributions. They are
in satisfactory agreement with the data reported specifically for grain boundary inventory of low to
medium burnup UO2 [9]. In parallel, measurements of the species effusing from the fuel as a function
of annealing temperature were performed using a Knudsen cell coupled to a mass spectrometer.
The release measured at temperatures below the range where thermally activated transport dominates
the fission gas release was attributed to release from surfaces and grain boundaries. The results
from the two techniques showed good convergence and were in satisfactory agreement with
data reported specifically for grain boundary inventory of low-medium burnup UO2.
2. Alpha decay damage evolution
In order to predict the long term properties evolution of spent fuel, the effects of alpha-decay damage
accumulation on spent fuel properties were investigated under accelerated alpha-decay rate conditions,
using UO2 containing high activity alpha-emitters, the so-called alpha-doped UO2. Macroscopic
and microstructural effects associated with the build-up of defects in the structure were studied.
In particular, the correlation among annealing of defects in the microstructure, release behaviour
of He, and energy stored by the defects in the material were evaluated. The limits of application
of accelerated decay accumulation methods were also evaluated.
The studies on alpha-doped UO2 have been performed on sol-gel producedsamples with a range of
activities: from 10 wt% 233 U-doped UO2 (3.88·10 7 Bq.g -1 , with a damage level of ~ 10 -5 dpa) up to
UO2 doped with 10 wt% 238 PuO2 with 3 dpa of damage, corresponding to standard irradiated UO2
fuel after 10000 years of storage.
Alpha decay damage may lead to long term changes in the fuel microstructure and may further influence
long term fuel leaching. A TEM analysis of the 233 U-doped UO2 has been performed, showing
(Figure 1) numerous non-homogeneously distributed dislocation loops with sizes between 20
and 50 nm, indicating that the UO2 damage rapidly leads to precipitation of dislocation loops.
Figure 1: TEM micrograph of 10 wt% 233 U-doped UO2
A Differential Scanning Calorimetry (DSC) study of 10 wt% 238 Pu-doped UO2 has also been performed
(Figure 2). With this method the energy related to healing of lattice defects could be determined
as a function of temperature, after correcting for the thermal effects due to self-irradiation in
the material.
484

The Cp * (T) curve were analysed in order to identify the different parameters controlling the latent
heat effects of the defects. For each stage, the quantities to be derived are concentration and energy
associated to the annealing of a certain kind of defect i.e. its characteristic mobility (i.e., preexponential
factor and activation energy of the diffusion coefficient).
Four peaks, corresponding to different defect annealing stages were identified from the analysis. A
distinct physical significance was attributed to the various stages with increasing temperature:
oxygen vacancy/interstitial recombination; uranium vacancy/interstitial cluster recombination (5.1
eV); dislocation loop annealing; void growth (the void fractional volume was measured to about 0.3
0.05 %; the resulting defect annealing energy of a vacancy trio is 13.4 eV).
Heat Released, J g -1
0,00
-0,02
-0,04
-0,06
-0,08
-0,10
-0,12
-0,14
400 600 800 1000 1200 1400
Annealing Temperature, K
Figure 2: Analysis of the heat release for (U0.9Pu0.1)O2
3. Evolution of spent fuel in steam
I
II
485
III
IV
Fitting
Experiment
The aim of this study was to investigate the alteration of spent nuclear fuel (SNF) in a disposal vault
under non-oxidising conditions (the behaviour of SNF under oxidizing conditions was investigated
previously [10-13]). Potential alteration processes of the spent fuel during the transient phase corresponding
to initial breach of the canister and first water ingress (early failure scenario) were investigated.
This involved experiments in which fuel rod segments containing intentional defects in the
cladding were exposed to a water-saturated atmosphere containing hydrogen to simulate nonoxidizing
repository conditions. The effect of hydrogen overpressure on the corrosion behaviour of
spent fuel in contact with groundwater was investigated by performing autoclave studies on irradiated
fuel and analogues under conditions relevant for assessing different repository scenarios.
The complete experimental equipment including an autoclave, an oven, a gas-sampling system and
all the electrical connections was installed inside a hot cell during normal cell operation and
adapted to remote handling by tele-manipulators. Three experiments were carried out at 90°C in
humid atmosphere composed of, respectively, 1) pure Ar, 2) mixture of Ar and H2, and 3) pure H2.
SEM investigations showed a significant surface change in the area of the intentionally set defect
only under Ar. On both rodlets exposed to a moist hydrogen-containing atmosphere, no fuel surface
alteration could be seen. Figure 3 shows SEM images of the surface exposed to different
atmospheres.
EDS analysis of the hydrogen-exposed surfaces showed, in addition to uranium, only the presence
of fission products caesium and barium at the outer periphery of the fuel pellet, near the fuel
cladding. Here also zirconium traces were found.

Figure 3: Left: Spent fuel surface exposed 150 days to humid argon. Middle: Fuel surface resulting
from a fresh cut made in the same rodlet. Right: Fuel surface after exposure to humid H2.
In addition to surfaces, also gaseous and aqueous phases were analysed by mass spectrometry. A
gas pressure increase was observed in all cases. It was 0.3 – 0.5 bar during the test under argon
Under hydrogen, the total pressure increase was lower (0-0.2 bar (H2/Ar); 0.1-0.2 bar (H2)). The
main gas component was in all experiments fission gas Xe. Under argon, radiolytically produced H2
was also detected. A possible explanation for the high fission gas release could be the opening of
pathways for release from pressurized gas bubbles in the fuel. The oxygen content in all gas
samples was below the detection limit (

The studies on alpha-doped UO2 have been performed on samples with damage corresponding to
spent UO2 fuel after 10000 years of storage. Regarding the stability of the spent fuel the preliminary
conclusion is that, while it is not expected that overall the fuel will lose its mechanical integrity, the
alpha-damage and the precipitation of helium in regions with low porosity or few pre-existing
trapping sites (bubbles) produced during irradiation might lead to a potential micro-cracking and
local loss of cohesion. This is particularly the case for the high temperature region of irradiated
fuels.
The experiments on spent fuel evolution in steam have produced evidence that the presence of
hydrogen leads to a reduced matrix alteration also under moist atmosphere conditions.
5. Acknowledgements
This project has been co-funded by the European Commission and performed as part of the sixth
Euratom Framework Programme for nuclear research and training activities (2002-2006) under contract
FI6W-CT-2003-02389.
The authors wish to thank Matthew Barker, National Nuclear Laboratory, UK, and Chris Chatwin,
Sellafield Ltd, for supporting the use of the irradiated fuel samples and provision of PIE data.
References
[1] R.J. White, S.B. Fisher, P.M.A. Cook, R. Stratton, C.T. Walker, I.D. Palmer, J. Nucl. Mater.
288 (2001) 43.
[2] S.B. Fisher, R.J. White, P.M.A. Cook, S. Bremier, R.C. Corcoran, R. Stratton, C.T. Walker,
P.K. Ivison, I.D. Palmer, J. Nucl. Mater. 306 (2002) 153.
[3] W.J. Gray, D.M. Strachan, Mat. Res. Soc. Symp. Proc. 212 (1991) 205.
[4] W.J. Gray, L.E. Thomas, Mat. Res. Soc. Symp. Proc. 333 (1994) 391.
[5] W. Gray, Mat. Res. Soc. Symp. Proc. 556 (1999) 487.
[6] B. Hanson, Mat. Res. Soc. Symp. Proc. 824 (2004), 89-94.
[7] D. Roudil, C. Jégou, V. Broudic, B. Muzeau, S. Peuget, X. Deschanels, J. Nucl. Mater. 362
(2007) 411.
[8] N.C. Garisto, L.H. Johnson, W.H. Hocking, 2 nd Int. CANDU Fuel, Oct. 1989, 352.
[9] L. Johnson, C. Ferry, C. Poinssot, P. Lovera, J. Nucl. Mater. 346 (2005) 56.
[10] R.J. Finch, E.C. Buck, P.A. Finn, J.K. Bates, Mat. Res. Soc. Symp. Proc., 556 (1999) 431.
[11] P. Taylor, D.D. Wood and D.G. Owen, J. Nucl. Mater. 223 (1995) 316.
[12] P.C. Burns, R.J. Finch, D.J. Wronkiewicz, Direct Investigations of the Immobilization of Radionuclides
in the Alteration Products of Spent Nuclear Fuel, Final Report, U.S. Department
of Energy, Proj. 73691, 2004.
[13] J.C. Cunnane, CSNF Waste Form Degradation: Summary Abstraction, ANL-EBS-MD-
000015 REV 02, Aug. 2004.
487

488

FUNMIG – Evaluation and Improvement of Numerical THM Modelling
Capabilities for Rock Salt Repositories (THERESA project)
K. Wieczorek 1 , T. Rothfuchs 1 , C.-L. Zhang 1
Th. Spies 2 , U. Heemann 2 , Chr. Lerch 3 , S. Keesmann 3
A. Pudewills 4 , P. Kamlot 5 , J. Grupa 6 , K. Herchen 7 , S. Olivella 8 , Chr. Spiers 9
1 Gesellschaft für Anlagen- u. Reaktorsicherheit (GRS), Germany
2 Federal Institute for Geosciences & Natural Resources (BGR), Germany
3 DBE Technology, Germany, 4 Forschungszentrum Karlsruhe (FZK), Germany
5 Institut für Gebirgsmechanik (IfG), Germany
6 Nuclear Research and Consultancy Group (NRG), The Netherlands
7 Technische Universität Clausthal (TUC), Germany
8 Univ. Politècnica de Catalunya (UPC), Spain, 9 Universiteit Utrecht (UU), The Netherlands
Summary
The objective of the THERESA project is to develop, verify and improve the modelling capabilities
of constitutive models and computer codes for analysis of coupled THMC processes in
geological and engineered barriers for use in Performance Assessment (PA) of the long-term
safety of nuclear waste repositories. Work Package 3 of the project addresses the evaluation
and improvement of numerical modelling capabilities for assessing the performance and
safety of nuclear waste repositories in rock salt, with particular regard to the long-term evolution
of the excavation damaged zone (EDZ), considering thermal-hydraulic-mechanical
(THM) processes. This comprises
Evaluation of the capabilities and/or development needs of the numerical modelling codes
used by the participating teams and compilation of data relevant for model calibration/improvement.
Implementation in the computer codes and testing of the calibrated models.
Definition and benchmark simulation of one test case (TC), with measured data from a
large-scale laboratory test involving coupled THM processes, focusing on code capacities for
realistic system representation, reliable conceptualisation, quantification of uncertainties in
models and results, and capacities for long-term performance assessment predictions.
Expression of the applied process models in terms of a so-called compartment model and
implementation in integrated PA codes to perform long-term performance assessment predictions
for a reference case (RC).
The project partners involved in this work package are the Gesellschaft für Anlagen- und
Reaktorsicherheit (GRS) mbH as WP leader, the Bundesanstalt für Geowissenschaften und
Rohstoffe (BGR), the DBE Technology GmbH, the Forschungszentrum Karlsruhe (FZK)
GmbH, the Institut für Gebirgsmechanik (IfG) GmbH, the Nuclear Research and Consultancy
Group (NRG), the Technische Universität Clausthal (TUC), and the Centre Internacional de
Mètodes Numèrics en Enginyeria (CIMNE).
489

1. Data compilation and evaluation
The development needs are based on an issue evaluation table which has been drafted in the first
months of the project and is regularly updated. An important issue derived from this table is the
model simulation of the evolution and especially the self-sealing of the Excavation Damaged Zone
(EDZ) which evolves as a result of the mechanical response of the rock to excavation of underground
openings. In the region close to the opening the dilatancy boundary will be exceeded, leading
to microfracturing and an associated increase of porosity and permeability of the salt rock. Existing
experimental data relevant for modelling this behaviour and the constitutive models used by
the different partners in their codes were compiled in a deliverable (D5 of the THERESA project,
THERESA 2007) in July 2007.
All constitutive models are able to simulate the elastic and the secondary creep behaviour of the
rock salt, regarding dilatancy and reconsolidation, however, different development states and levels
of confidence had been reached. Although many experimental data are available, reliable laboratory
data including volumetric defor-mation which are essential for dilatancy determination are scarce.
Therefore, GRS and BGR decided to perform additional laboratory tests outside the EC funding to
provide data for model calibration for those partners considering them useful.
2. Model improvement and calibration
While some of the partners (e. g., IfG, TUC) felt the existing data were sufficient for model calibration,
others used the additional lab tests to check their model parameters (FZK) or even modified
their model equations (CIMNE). In the meantime, model calibration has been widely completed. A
respective deliverable is currently under preparation. Figures 1 and 2 show some examples of results
from the GRS tests together with model calculations and Figures 3 and 4 verify the ability of
modelling softening and dilatancy as significant processes in the EDZ. All Figures illustrate the
high level obtained regarding model calibration.
Vertical stress (MPa)
30
-0.016
-0.014
25
-0.012
20
-0.010
-0.008
15
-0.006
-0.004
10
-0.002
5
Vertical stress
Vertical stress (test)
0.000
Volumetric strain
Volumetric strain (test) 0.002
0
0.004
0.00 0.02 0.04 0.06 0.08
Axial Strain (-)
Figure 1: Evolution of axial stress and volumetric strain of an Asse salt sample as a function of axial
strain, measured (GRS) and calculated (CIMNE)
490
Volumetric strain (-)

Permeability (m 2 )
1E-16
1E-17
1E-18
1E-19
1E-20
1E-21
3.5
Model ( k = 3.2E-11. ε ) v
Experiment (sample #3)
1E-22
0 1 2 3 4 5 6 7
Axial strain (%)
Figure 2: Comparison of the calculated (FZK) and measured (GRS) permeability of a salt sample
as a function of axial strain
Differential stress [MPa]
35
30
25
20
15
10
5
Specimen 288/06
Specimen 288/23
Specimen 288/25
Modelling
0
0 2 4 6 8 10 12 14 16 18 20 22 24
Axial deformation [%]
491
Confining pressure 0.2 MPa
Black: 3 tests under deformation rate 5E-06 1/s
Red: numerical simulation
Figure 3: Evolution of differential stress ( 1- 3) versus axial strain, hardening until peak strength
and softening in the dilatant domain, measured and calculated on an Na3 sample from the Asse
mine (IfG)

Volumetric deformation [%]
20
18
16
14
12
10
8
6
4
2
0
Specimen 288/06
Modelling
Specimen 288/25
-2
0 2 4 6 8 10 12 14 16 18 20 22 24
Axial deformation [%]
Confining pressure 0.2 MPa
Black: 3 tests under deformation rate 5E-06 1/s
Red: numerical simulation
492
Specimen 288/23
Figure 4: Evolution of volumetric deformation versus axial strain, strong dilatancy increase while
fracturing and slight increase after generation of shear bands in the dilatant domain, measured and
calculated on Na3 from the Asse mine (IfG)
3. Benchmark test
A laboratory test case has been defined to simulate EDZ formation and self-sealing by a suitable
loading procedure. The TC is performed on a large hollow cylinder (525 mm length, 280 mm diameter,
central borehole diameter 100 mm) of Asse Speisesalz consisting of rather pure halite (see
Figure 5). The axial/radial deformation, borehole convergence, borehole pressure and gas permeability
are continuously measured during the test, while the axial stress, the outer radial stress, and
the temperature are controlled.

hollow
sample
inner
packer
central
heater
outer
jacket
inlet
(gas / water)
axial load
axial load
hollow cylideric sample
outlet
(gas / water)
filter
extensometer
temperature
sensor
outer
heater
pressure /
volume
oil
pump
Figure 5: THM test layout using a large hollow salt cylinder
The TC consists of the following steps:
1. Determination of initial state of the sample
2. Reconsolidation of the sample at increased compressive load
3. Simulation of borehole excavation by reduction of inner packer pressure
4. EDZ extension (if previous step does not lead to a significant permeability increase) by introducing
an outer stress deviator
5. Recompaction at isostatic outer stress and stepwise increase of inner packer pressure
6. Heating of the sample to 70 °C to accelerate self-sealing
7. Cool-down to 30 °C
8. Unloading, dismantling and inspection of the sample
493

An overview of the test procedure with the respective boundary conditions is given in Figure 6.
Temperature (°C)
80
60
40
20
0
temperature axial stress radial stress packer pressure gas pressure
1 2 3 4 5 6 7 8
Δt
22 7
1MPa
2
45MPa
2
10
6
30 C
0 5 10 15 20 25 30 35 40 45 50 55 60 65
10
Time (day)
494
2
10
70 C
5MPa
7 4
Figure 6: Test procedure in terms of boundary conditions to be applied to the sample
4. Outlook
The TC is currently being conducted. After finishing, the actual boundary conditions as maintained
during the experiment will be submitted to the project partners. The measurement results, however,
will only be made available after the modelling teams have finished their simulations of the TC, so
that the modelling results can be regarded as true blind predictions. The calculation results will then
be evaluated in comparison to the actual measurements.
Afterwards, the applied process models will be expressed in terms of a so-called compartment
model and implemented in the integrated PA codes of GRS and NRG to perform long-term performance
assessment predictions for a repository reference case.
Reference
[1] THERESA 2007: THERESA Project, Work Package 3, Deliverable 5, Compilation of existing
constitutive models and experimental field or laboratory data for the thermal-hydraulicmechanical
(THM) modelling of the excavation disturbed zone (EDZ) in rock salt, August
2007.
50
40
30
20
10
0
Stress / pressure (MPa)

Far-field processes
495

496

Summary
FUNMIG – Research on Well-defined Processes
Pascal Reiller
CEA-Saclay, France
The First Research, Technological Development Component (RTDC) from the Integrated Project
FUNMIG was devoted on the assessment of basic thermodynamic data and constants, as
well as basic principles necessary to describe the speciation of radionuclides, their sorption on
minerals, influence of organics and solid-solution formation in the far field of a radioactive
waste disposal site. It includes a large number of European teams that shared and confronted
their particular expertises and views. Common points were identified throughout this project,
which gave the opportunity of common works.
1. Introduction
Main objectives of the integrated project FUNdamental processes of radionuclides MIGration
(FUNMIG) are the fundamental understanding of radionuclide migration processes in the geosphere,
especially focused on tools for application to performance assessment (PA). It is structured
in seven components, two on acquisition of data and validation of models at the laboratory level,
three components are dedicated to host rock specific studies, one is dedicated to the abstraction of
the generated to PA, and last component is devoted to knowledge transfer and training.
The first research, technological development component (RTDC-1) from FUNMIG, regrouping
17 partners from 10 countries, is dedicated to provide fundamental process knowledge and the required
data for processes with comparably well established conceptual understanding. Processes
studied are fundamental and applicable to all types of host rocks and the proper parameters are derived
for the relevant migration systems. The outcome is the fundamental understanding and quantification
of the processes studied. The models are developed at the research model level. It is organised
in four work packages to provide data on (i) ionic species/speciation of radionuclides, processes
determining physico-chemical conditions and generation of missing thermodynamic data, (ii)
ion exchange and surface complexation modelling to described the retention phenomena of radionuclides
by mineral, (iii) the influence of organics, either natural or anthropogenic, on the retention
of radionuclides by minerals, and (iv) formation of solid solutions and secondary phases including
retardation of anions.
RTDC-1 is aiming at proposing solution for the following key scientific questions. Are needed
thermodynamic data available and accurate? Can sorption models calculate sorption isotherms
measured under chemically realistic conditions? Are the modelling simplification justified regarding
spectroscopy? What is the influence of radionuclide complexation on sorption and can this influence
be quantitatively calculated with the sorption models? Is the influence of natural organic
matter on sorption described by thermodynamics? Can we mend the bridge between sorption and
solid solution?
497

The works performed within RTDC-1 permitted to ascertain some of the basic data on actinide
speciation in carbonate, sulphate and phosphate systems, and to tackle the difficult silicate system.
The retention models were compared and benchmarked; an inter-comparison exercise on data acquisition
was also proposed. It is appearing that the modelling of retention of radionuclides by clay
systems is robust as it permits to perform bottom-up modelling of natural samples from laboratory
determination. The modelling on oxides depends largely on the surface state and thermodynamic
stability of the phases. It seems necessary to account for the electrostatic properties of oxides. The
influence of organic molecules on radionuclide retention depends mainly on their origins and interactions
with the minerals. In a general manner, the anthropogenic molecules induce a decrease of
the retention of radionuclides on clays. Synergistic sorption can nevertheless happen at the surface
of oxides. For natural organic matter (NOM), the interaction with mineral has to be weak to apply
linear additive models. Otherwise, a change in the ‘free energy’ of the system must be accounted.
The identification of processes of natural organic layer coating on clays is still not clear. The solid
solution formation has been clearly linked to adsorption. A thermodynamic expression considering
the main component as the solvent can be proposed. A limiting quantity of actinides in natural calcite,
as well as the structure of these solid solutions, has been evidenced. The incorporation of Se to
pyrite is driven by redox processes.
The main results will only be recalled and will not be developed. Readers can refer to the published
works in references. Activity reports are also available in the annual proceedings of the Annual
Workshops [1-3].
2. Results
2.1 Work Package 1: Ionic species/speciation, processes determining physico-chemical conditions
and generation of missing thermodynamic data
The thermodynamic data or thermodynamic constant on complexation of curium (III) in carbonate
solutions at variable temperature, of sulphate complexation of lanthanides (III) and uranium (VI)
[4-6], and phosphate complexation of thorium (IV) was determined [2]. Common work on silicate
complexation of actinides (III) and (IV) were also undergone in the French Commissariat à
l’Énergie atomique and the Swedish Chalmers Technical University.
2.2 Work Package 2 - Ion exchange and surface complexation
The sorption data of radionuclides at different redox state – i.e., Se(IV), Co(II), Ni(II), Eu(III),
Cm(III), Y(III), U(VI), and Th(IV) –, on model oxides and clays – i.e., �-Fe2O3, goethite, -Al2O3,
gibbstite, smectite, illite – [1, 2, 7-9], as well as on actual minerals – i.e., biotite, granodiorite,
montmorillonite, Opalinus clays, Boom clay –, was studied [2]. The different sorption modelling
were used or benchmarked [2, 3]. The combination and confrontation of modelling and spectroscopic
data was also performed [9] as well as the influence of competing reaction [10]
2.3.1 Work Package 3 - Influence of organics on the retention of radionuclides by minerals
This work package is devoted to the influence of organics, either natural or anthropogenic on the
sorption of radionuclides.
The sorption of cellulose degradation product was characterised in far field conditions and was
shown to be very weak compared to natural organics [11-13]. The influence of the main organic
component in nature, i.e. humic substances, was considered [11, 12]. Either a purely additive model
[13-15], or a charge distribution model [16-19] were tested. The influence of the kinetic of humic
complexation was also accounted in transport model [20]. Spectroscopic modification of actinides,
chemical environments in humic-clay systems were also observed [21].
498

2.4 Work Package 4 - Formation of solid solutions and secondary phases, including retardation
of anions
A conceptualization of the mass action law for co-precipitation was proposed [22]. The particular
case of actinide and lanthanide substitution into calcite received particular attention [23, 24].
Amongst anionic radionuclides, the solid-solution formation of Se in iron-sulphur compounds was
particularly studied [1-3].
3. Acknowledgements
M.H. Bradbury is acknowledged for friendly guidance during these four years.
References
[1] P. Reiller, M.H. Bradbury, RTD Component 1. In: 1 st Annual Workshop Proceedings of the
Integrated Project “Fundamental Processes of Radionuclide Migration” – 6th EC FP IP
FUNMIG, CEA-R-6122 (Reiller, P., Buckau, G., Kienzler, B., Duro, L. and Martell, M.
Eds.). CEA, Gif-sur-Yvette (2006) p. 7.
[2] P. Reiller, M.H. Bradbury, RTD Component 1. In: 2nd Annual Workshop Proceedings of the
Integrated Project “Fundamental Processes of Radionuclide Migration” – 6th EC FP IP
FUNMIG (Buckau, G., Kienzler, B., Duro, L. and Montoya, V. Eds.). SKB, Stockholm
(2007) p. 15.
[3] P. Reiller, RTD Component 1. In: 3rd Annual Workshop Proceedings of the Integrated Project
“Fundamental Processes of Radionuclide Migration” – 6th EC FP IP FUNMIG (Buckau,
G., Kienzler, B. and Duro, L. Eds.). NDA, London (in press) p.
[4] T. Vercouter, P. Vitorge, B. Amekraz, E. Giffaut, S. Hubert, C. Moulin, Inorg. Chem. 44
(2005) 5833.
[5] T. Vercouter, B. Amekraz, C. Moulin, E. Giffaut, P. Vitorge, Inorg. Chem. 44 (2005) 7570.
[6] T. Vercouter, P. Vitorge, B. Amekraz, C. Moulin, Inorg. Chem. 47 (2008) 2180.
[7] N. Bryan, R.S. Brar, A.J. Hynes, P. Warwick, N.D.M. Evans, L. Knight, Metal ion sorption
by inorganic oxide surfaces. In: 1st Annual Workshop Proceedings of Integrated Project
“Fundamental processes of Radionuclide Migration” (IP FUNMIG), CEA-R-6122 (Reiller,
P., Buckau, G., Kienzler, B., Duro, L. and Martell, M. Eds.). CEA, Gif-sur-Yvette (2006)
p. 215.
[8] E. Puukko, E. Puhakka, A. Lindberg, M. Olin, M. Hakanen, J. Lehikoinen, Mineral-specific
sorption of Cs, Ni, Eu and Am on granodiorite and mica gneiss. In: 1 st Annual Workshop Proceedings
of the Integrated Project “Fundamental Processes of Radionuclide Migration” – 6th
EC FP IP FUNMIG, CEA-R-6122 (Reiller, P., Buckau, G., Kienzler, B., Duro, L. and
Martell, M. Eds.). CEA, Gif-sur-Yvette (2006) p. 80.
[9] M.H. Bradbury, B. Baeyens, Effect of inorganic carbon on the sorption of Ni(II), U(VI) and
Eu(III) on illite. In: 1st Annual Workshop Proceedings of Integrated Project “Fundamental
processes of Radionuclide Migration” (IP FUNMIG), CEA-R-6122 (Reiller, P., Buckau, G.,
Kienzler, B., Duro, L. and Martell, M. Eds.). CEA, Gif-sur-Yvette (2006) p. 225.
[10] P. Warwick, T. Lewis, N.D.M. Evans, N. Bryan, L. Knight, Sorption of cellulose degradation
products and asoociated components to clay minerals under far field conditions of an intermediate
to loww level nuclear waste repository. In: 2nd Annual Workshop Proceedings of the
499

Integrated Project “Fundamental Processes of Radionuclide Migration” – 6th EC FP IP
FUNMIG (Buckau, G., Kienzler, B., Duro, L. and Montoya, V. Eds.). SKB, Stockholm
(2007) p. 287.
[11] P. Warwick, T. Lewis, N. Evans, N. Bryan, L. Knight, Sorption of selected radionuclides to
clay in the presence of humic acid. In: 1 st Annual Workshop Proceedings of the Integrated
Project “Fundamental Processes of Radionuclide Migration” – 6th EC FP IP FUNMIG, CEA-
R-6122 (Reiller, P., Buckau, G., Kienzler, B., Duro, L. and Martell, M. Eds.). CEA, Gif-sur-
Yvette (2006) p. 184.
[12] N.D.M. Evans, M.H. Khan, P. Warwick, T. Lewis, N. Bryan, L. Knight, Modelling the influence
of humic acid on the sorptionof cadmium to montmorillonite. In: 2nd Annual Workshop
Proceedings of the Integrated Project “Fundamental Processes of Radionuclide Migration” –
6th EC FP IP FUNMIG (Buckau, G., Kienzler, B., Duro, L. and Montoya, V. Eds.). SKB,
Stockholm (2007) p. 293.
[13] L. Weng, W.H. Van Riemsdijk, T. Hiemstra, Improved ligand and charge distribution (LCD)
model. In: 2nd Annual Workshop Proceedings of the Integrated Project “Fundamental Processes
of Radionuclide Migration” – 6th EC FP IP FUNMIG (Buckau, G., Kienzler, B., Duro,
L. and Montoya, V. Eds.). SKB, Stockholm (2007) p. 207.
[14] L.P. Weng, W.H. Van Riemsdijk, T. Hiemstra, Langmuir 22 (2006) 389.
[15] L. Weng, W.H. Van Riemsdijk, T. Hiemstra, J. Colloid Interface Sci. 314 (2007) 107.
[16] D.H. Farelly, L.G. Abrahamsen, A. Pitois, P. Ivanov, B. Siu, N. Li, P. Warwick, N.D.M. Evans,
L. Knight, N. Bryan, Initial kinetic studies of iron oxides and humic acid ternary systems.
In: 2nd Annual Workshop Proceedings of the Integrated Project “Fundamental Processes of
Radionuclide Migration” – 6th EC FP IP FUNMIG (Buckau, G., Kienzler, B., Duro, L. and
Montoya, V. Eds.). SKB, Stockholm (2007) p. 195.
[17] N. Bryan, L.G. Abrahamsen, D.H. Farelly, P. Warwick, N.D.M. Evans, L. Knight, A provisional
humic acid ternary system model. In: 1st Annual Workshop Proceedings of Integrated
Project “Fundamental processes of Radionuclide Migration” (IP FUNMIG), CEA-R-6122
(Reiller, P., Buckau, G., Kienzler, B., Duro, L. and Martell, M. Eds.). CEA, Gif-sur-Yvette
(2006) p. 220.
[18] P. Ivanov, L.G. Abrahamsen, D.H. Farelly, A. Pitois, P. Warwick, N.D.M. Evans, L. Knight,
N. Bryan, A procedure to assess the importance of chemical kinetics in the humic mediated
transport of radionuclides in performance assessment calculation. In: 2nd Annual Workshop
Proceedings of the Integrated Project “Fundamental Processes of Radionuclide Migration” –
6th EC FP IP FUNMIG (Buckau, G., Kienzler, B., Duro, L. and Montoya, V. Eds.). SKB,
Stockholm (2007) p. 187.
[19] L.G. Abrahamsen, D.H. Farelly, A. Pitois, P. Ivanov, P. Warwick, N.D.M. Evans, L. Knight,
N. Bryan, Kinetic studies of the quartz/sand, Eu 3+ and humic acid ternary system. In: 2nd Annual
Workshop Proceedings of the Integrated Project “Fundamental Processes of Radionuclide
Migration” – 6th EC FP IP FUNMIG (Buckau, G., Kienzler, B., Duro, L. and Montoya,
V. Eds.). SKB, Stockholm (2007) p. 179.
[20] A. Krepelova, Influence of Humic Acid on the Sorption of Uranium(VI) and Americium(III)
onto Kaolinite. Ph. D Thesis. Technischen Universität Dresden, Dresden (2007).
[21] T. Arnold, N. Baumann, Spectrochimica Acta A. (in press).
500

[22] P. Vitorge, Law of Mass Action for co-Precipitation. CEA, Report CEA-R-6193, Gif-sur-
Yvette (2008).
[23] S.L.S. Stipp, J.T. Christensen, L.Z. Lakshtanov, J.A. Baker, T.E. Waight, An upper limit estimate
for actinide substitution in calcite: use of total lanthanide concentration as an analogue.
In: 1 st Annual Workshop Proceedings of the Integrated Project “Fundamental Processes of
Radionuclide Migration” – 6th EC FP IP FUNMIG, CEA-R-6122 (Reiller, P., Buckau, G.,
Kienzler, B., Duro, L. and Martell, M. Eds.). CEA, Gif-sur-Yvette (2006) p. 178.
[24] F. Heberling, M.A. Denecke, D. Bosbach, Np(V) co-precipitation with calcite. In: 2nd Annual
Workshop Proceedings of the Integrated Project “Fundamental Processes of Radionuclide
Migration” – 6th EC FP IP FUNMIG (Buckau, G., Kienzler, B., Duro, L. and Montoya, V.
Eds.). SKB, Stockholm (2007) p. 301.
501

502

FUNMIG – Real system analyses of PA relevant processes in sediments:
The Ruprechtov Natural Analogue site
Summary
Vaclava Havlova 1 , Ulrich Noseck 2 , Melissa Denecke 3 , Wolfgang Hauser 3 ,
Juhani Suksi 4 , Kazimierz Rozanski 5
1 NRI Rez plc., Czech Republic
2 GRS, Germany
3 FZK INE, Germany
4 Helsinki University
5 AGH, Poland
Component RTDC 5 of the EC Integrated Project FUNMIG (2004–2008) was aimed to study
real system processes, relevant to performance assessment (PA), in sediment formations that
can form overlying layers of deep geological repository in salt host rocks or layers above any
other potential geological repository host rocks (granite, clay). The study comprised both
laboratory and in-situ experimental work, focused on the Ruprechtov natural analogue site.
The scope of scientific activities, addressed within the frame of FUNMIG RTDC 5 is presented
here, following a “puzzle” approach. Moreover, integration of results from other components
of FUNMIG, namely RTDC 2, into the real system description and vice versa is demonstrated.
1. Introduction
Component RTDC 5 of the EC Integrated Project FUNMIG (2004–2008) was aimed to study real
system processes, relevant to performance assessment (PA), in sediment formations that can form
overlying layers of deep geological repository in salt host rocks or in any potential host rocks (granite,
clay). The activities within the frame of FUNMIG RTDC 5 were aimed to real natural system
understanding, mainly of key processes involved in uranium immobilisation and the role of organic
matter. Unlike the other components, the RTDC 5 allowed a “puzzle approach”: to combine puzzle
pieces of single-track basic investigation to obtain an overall picture of the real system.
The study comprises both laboratory and in-situ work, focused on the Ruprechtov natural analogue
site (Czech Republic). Natural analogues, relevant for a radioactive waste repository safety assessment,
can complement the results from short-term laboratory experiments, since they allow research
of natural systems that have evolved over geological time scales. Moreover, such studies contribute
to model/methodology development and testing, to scientific team building and training as well as
to increase building confidence in communication with the public. The focus of interest at Ruprechtov
site lies in uranium anomalies in distinct organic clay layers (clay/lignite) [1].
The long-term stability of immobile U phases and the processes controlling uranium mobility were
the first target to be studied. A further major task was the investigation of the behaviour of colloids
in the system. The latter one was closely interconnected with FUNMIG RTDC 2 (“Less established
processes”), studying namely natural sedimentary organic matter (SOM) behaviour, quantifying the
SOM fraction that can be released into the groundwater, and its potential complex formation with
uranium [2,3].
503

The Ruprechtov natural analogue site has been studied extensively for several years [1]. The site is
located in the North West part of the Czech Republic. It is underlain by granitic basement, covered
by kaolin layers of various thicknesses. The interface between the kaolin layers and the upper formation
(montmorillonite clays of Tertiary pyroclastic origin) in 20 to 50 m depth is composed by
the so-called clay/lignite layer of few meter thickness with high content of SOC and local uranium
enrichment (up to 600 ppm). Granite of Calsbad type with higher content of U is proposed as a primary
uranium source [1].
2. Methodology and approach
The project scientific activities in RTDC5 followed a scheme that can be assigned as “a puzzle system”
proceeding from investigation of “single puzzle pieces” (e.g. characterisation of natural organic
matter, characterisation of the U immobile phases combining sophisticated analytical methods,
etc.) to a more integral investigation and characterisation of a complex natural system, integrating
results e.g. from single uranium and natural organic matter investigations. Finally a complete
picture puzzle was assembled, i.e. the evaluation of hydrological, geochemical and environmental
data was used to characterize the hydrogeological flow pattern, the geochemical evolution of the
system and in particular the uranium enrichment scenario at the site.
Methodologies and analytical methods used were discribed elsewhere e.g. [2,7,13,14]. Direct combination
of classical methods (sequential extraction), radioanalytical methods (U(IV)/U(VI) separation)
and modern spectroscopical methods was implemented for the first time [4,11,13].
3. Results
The scientific results of the activities that had been performed within RTDC 5 were described in
more detail in [4].
In the first level of the project the single-task scientific topics were dealt with, e.g. sedimentary organic
matter (SOM) behaviour or uranium immobilisation as a smallest puzzle pieces. Organic matter
on the site was found having significant influence neither on U complexation and mobilization
by dissolved organic species in groundwater nor on U direct sorption on organic SOM [2,3,5].
Fe As U
As(V)
As
As(0)
Fe
U
150 μm, 4 μm step
504
150 μm, 2 μm step
Figure 2 : μ-XRF distribution maps for a
150*150 μm 2 area of a thin section of a
sample from NA5. The distribution of the
total As, Fe and U measured with Eexcite =
18 keV and its corresponding Red-Blue-
Green (Fe, As, and U, respectively) image,
as well as the arsenic chemical state
distributions for As(V) and As(0) are
shown. [5]
The application of macroscopic and microscopic methods provided a detailed insight into the U enrichment
processes at the Ruprechtov site. Confocal μ-XRF and μ-XANES identified U in the

sediment as U(IV), being associated with As(V) as a precipitate on arsenopyrite layers, which
formed on pyrite nodules as ningyoite (U phosphate) and uraninite (U oxide) minerals [5- 8]. A
typical μ-XRF elemental distribution map is shown in Fig 2. In order to separate U(IV) and U(VI),
a wet chemical method was applied for the first time to Ruprechtov samples as well, confirming
that the major fraction of immobile uranium occurs in the tetravalent state [9].
In the second step these single task results were put together with results from characterisation of
organic matter and isotope analyses in the system. This gave insight into the role of microbial activity
and organic matter in the uranium immobilization process. It could be shown that sedimentary
organic matter (SOM) contributed and still contributes to maintain reducing conditions in the
clay/lignite layers. [10, 11, 12]. The key processes involved in uranium immobilisation can be
summarised as follows:
- Oxidation of sedimentary organic carbon (SOC) and reduction of oxidising agents (SO4 2- )
- SOC is partly oxidised to inorganic carbon, dissolved in groundwater (DIC) and partly released
as dissolved organic matter (DOC)
- Increase in 34 S in dissolved sulphate in the clay/lignite groundwater by microbial sulphate
reduction [14]
- Formation of framboidal FeS2 by reduction of SO4 2- (see Fig. 2)
- Production of PO4 by degradation of organic matter
- U(VI) reduction on FeAsS sites and oxidation of As(0) to As(V)
- Precipitation of U(IV) phosphate / oxide mineral phases
-
Figure 2: Fromboidal shape of pyrite, typical for microbially driven sulphate reduction
A major result is that uranium in all samples consists of both U(IV) and U(VI), however predominantly
of U(IV) [5, 6, 9]. Activity ratios (AR) below one for U(res) and U(IV) in nearly samples are
a strong indicator for their long-term stability. AR values significantly below unity are caused by
the preferential release of 234 U, which is facilitated by �-recoil process and subsequent 234 U oxidation.
In order to attain low AR values as low as 0.2 in the U(IV) phase, it must have been stable for
a sufficiently long time, i.e. no significant release of bulk uranium has occurred during the last million
years. This is expected under the strongly reducing conditions (-160 mV to –280 mV) in the
clay lignite waters and is in good agreement with the hypothesis that the major uranium input into
the clay/lignite horizon occurred during Tertiary, more than 10 My ago [13].
Finally, the overall picture puzzle shaped up its outlines: After re-evaluation of new hydrological,
geochemical and environmental isotope data from groundwater those were combined with previous
505

esults and integrated to develop a conceptual model for the geological development of the site with
respect to the uranium immobilisation scenario.
The following major steps comprise the uranium imobilization scenario [13, 14]:
o detrital input of U bearing minerals from underlaying granite
o granite kaolinisation
o major volcanic activity deposition of tuffaceous material
o granite alteration + tuff argillitization
o microbial activity in clay/organic rich horizon reduction of dissolved sulphate formation
of pyrite nodules
o sorption of As and subsequent formation of arsenopyrite on pyrite nodules
o CO2-rich water likely initiated U release from accessory minerals in the granite soluble
uranyl-carbonate complexes formed
o U transported to clay/organic rich layers reduced on arsenopyrite layers
o secondary uraninite might have formed at later stages of the geological history at conditions
where phosphate concentration might have decreased.
4. Contribution to the SAFETY CASE
The assembled picture puzzle of the real Ruprechtov system analyses gave a clear message: the
sedimentary rock can provide under certain circumstance a strong barrier function for radionuclide,
in this case for uranium. The key parameter is presence of organic matter that contributed to maintaining
reducing conditions in the clay/lignite layers. Its oxidation provided degradation products
(phosphates) that than took part in uranium mineral formation (ningyoite). Moreover, the results
showed that the presence of organic matter does not naturally mean increased mobilization of radionuclide
due to organic ligand complexation or organic colloid formation. On the contrary, there
are no indications for significant uranium release during the last million years within reducing conditions
of clay/lignite layers.
5. Acknowledgements
This project has been co-funded by the European Commission and performed as part of the sixth
Euratom Framework Programme for nuclear research and training activities (2002-2006) under contract
FI6W-CT-2004-516514, by the German Federal Ministry of Economics and Technology
(BMWi) under contract No. 02 E9995, and by RAWRA and Czech Ministry of Trade and Industry
(Pokrok 1H-PK25).
References
[1] Noseck, U., Brasser, Th., 2006. Radionuclide transport and retention in natural rock formations
– Ruprechtov site. Gesellschaft für Anlagen- und Reaktorsicherheit, GRS-218, Köln.
[2] Cervinka, R., Havlova, V.; Noseck, U.; Brasser, Th.; Stamberg, K., (2007). Characterisation
of organic matter and natural humic substances extracted from real clay environment. Annual
Workshop Proceedings of the IP Project FUNMIG”. Edinburgh 26.-29.November 2007.
506

[3] Cervinka R., Stamberg K., Havlova V. (2008): Report on Uranium complexation by isolated
humic substances from the Ruprechtov site. PID 2.2:20. RTDC 2, EC IP FUNMIG.
[4] Noseck U., Havlova V., Cervinka R., Suksi J., Denecke M, Hauser, W. (2008): Investigation
of far-field processes in sedimentary formations at a natural analogue site – Ruprechtov.
Euradwaste conference, dtto.
[5] Havlová V. et al. (2007): Ruprechtov Site (CZ): Geological Evolution, Uranium Forms, Role
of Organic Matter and Suitability as a Natural Analogue for RN Transport and Retention in
Lignitic Clay. Proc. of Reposafe Conference, Braunschweig Nov. 5 – 9- 2007, submitted.
[6] Denecke M. and Havlova V. (2007): Elemental correlations observed in Ruprechtov Tertiary
sediment. Micr-focus fluorescence mapping and sequential extraction. 2 nd Annual Meeting of
EC Integrated Project FUNMIG, SKB Report TR 07-05.
[7] Denecke, M.A., Janssens, K., Proost, K., Rothe, J., Noseck, U., (2005). Confocal micro-XRF
and micro-XAFS studies of uranium speciation in a Tertiary sediment from a waste disposal
natural analogue site. Environ. Sci. Technol. 39(7), 2049-2058.
[8] Denecke, M.A., Somogyi, A., Janssens, K., Simon, R., Dardenne, K., Noseck, U., (2007).
Microanalysis (micro-XRF, micro-XANES and micro-XRD) of a Tertiary sediment using
synchrotron radiation. Microscopy Microanal. 13(3), 165-172.
[9] Suksi J. et al. (2006): Uranium redox state and 234U/238U ratio analyses in uranium rich
samples from Ruprechtrov site IP FUNMIG 2nd Annual Workshop Proc., SKB TR-07-05.
[10] Noseck U., Suksi J., Havlova V., Brasser T. (2007): Uranium enrichment at Ruprechtov site –
uranium disequilibrium series and Geological development. S+T presentation. 3rd annual
meeting of IP FUNMIG, Edinborough, GB, Nov. 25 – 29, 2007, submitted.
[11] Havlova V., Noseck U., Cervinka R., Brasser T, Denecke M. Hercik M. (2007): Uranium enrichment
at Ruprechtov site - Characterisation of key processes. S+T presentation. 3rd annual
meeting of IP FUNMIG, Edinborough, GB, Nov. 25 – 29, 2007, submitted.
[12] Noseck U., Havlova V., Cervinka R. (2007): Data integration with regard to the behaviour of
organic matter and uranium in the system at Ruprechtov site PID 5.3.2. IP FUNMIG.
[13] Noseck, U., Brasser, Th., Suksi, J., Havlova, V., Hercik, M., Denecke, M.A., Förster, H.J.,
(2008). Identification of uranium enrichment scenarios by multi-method characterisation of
immobile uranium phases. J. Phys. Chem. Earth, doi:10.1016/j.pce.2008.05.018.
[14] Noseck, U., Rozanski, K., Dulinski, M., Havlova, V., Sracek, O., Brasser, Th., Hercik, M.,
Buckau, G., (2008). Characterisation of hydrogeology and carbon chemistry by use of natural
isotopes – Ruprechtov site, Czech Republic. Submitted to Appl. Geochem.
507

508

Training fellowships
509

510

SMARAGD – The Study of Mineral Alterations of Clay Barriers Used for
Radwaste Storage and its Geological Diposal
Summary
Miroslav Honty 1 , Mieke De Craen 1 , Maarten Van Geet 2
1 SCK·CEN, Belgium
2 NIRAS·ONDRAF, Belgium
The SMARAGD was a 2-years postdoctoral fellowship (2006-2007) funded under EURA-
TOM programme of EC in order to study mineral alterations of Boom Clay trigerred by geochemical
perturbations. In Belgium, the Boom Clay is studied as a reference host formation
for research purposes and for the assessment of deep disposal of high-level radioactive waste.
The geochemical perturbations may alter the favourable properties of Boom Clay which are
necessary to fulfill the required safety functions (De Preter, 2007 [1]). Within the SMARAGD
project, two types of geochemical perturbations were considered: oxidation due to ventilation
of underground galleries and the alkaline plume effects due to application of concrete both in
the gallery lining and as a part of the waste package design. The oxidation effects were studied
in situ in the HADES URL facility in Mol, Belgium. The core samples taken from the
Connecting Gallery (ventilated since 2001) and the Test Drift (ventilated since 1987) provided
a unique opportunity to study the oxidation effects at different time scales. The alkaline plume
effects were studied in the laboratory as a batch experiment under well constrained conditions
in terms of temperature, Eh, pH and leachate chemistry evolution. The results of the
SMARAGD project will serve as an input data for modelling long-term performance of the
geological barrier as well as the key parameters to calibrate the existing models. The results
generated during SMARAGD project are complementary to those of ECOCLAY II (2000-
2003) dealing with the effects of cement on clay barrier performance and NF-PRO (2004-
2007), which was investigating the oxidation as one of the key processes affecting the longterm
barrier performance in the near-field.
1. Introduction
The geochemical perturbations occur as a result of construction and operation of the underground
facilities such as URL and/or final high-level waste repository. The geochemical perturbations may
alter the in-situ pH and Eh conditions followed by the change in the pore water chemistry and mineralogy
of Boom Clay. Due to excavation of shafts and galleries, the host rock is inevitably exposed
to atmosphere and Eh will start to increase. The pyrite, one of the most active redox-sensitive mineral
in the Boom Clay, will be oxidized. The side-products of this reaction involve H + generating
acidity (pH decrease), sulphates, thiosulphates, and Fe 3+ precipitates. The reduced sulphur species
are of major concern, because of their highly corrosive effect on the metallic overpack of canisters
with radioactive waste. The acidity might trigger the corrosion of the concrete, moreover, most of the
radionuclides are more mobile under low pH conditions.
The cement, which is present in the concrete gallery lining and also as a principal component in the
current supercontainer design (Wickham, 2005 [2]), is the source of aggressive alkaline fluids when
in the saturated state. The pore fluids released from water-saturated concrete typically range in pH
511

from 12.5 to 13.5, have high ionic strengths, and are dominated by Na and K in the early stage and
by Ca in the later stage. The high pH will change the pore water chemistry which might affect the
mineralogy of the interacting phases.
All the above mentioned processes may induce mineral alterations (dissolution, transformation or
precipitation) in the Boom Clay. The change in mineralogy and geochemical properties may affect
the suitable properties of the Boom Clay as an effective barrier against radionuclide migration. The
objective of the SMARAGD project is to provide high quality data for performance assessment calculations.
Based on these calculations a judgement can be passed on the extent and the importance
of mineralogical alterations on the overall performance of the Boom Clay as a geological barrier
from long-term perspective.
2. Materials and methods
All the samples for the study of the oxidation were taken from the underground research facility
HADES in Mol, Belgium. The clay around the gallery was sampled by means of stainless steel cutting
edges. The cutting edge in the Test Drift was taken between rings 41 and 42 to the East (TD
R41-42E), the cutting edge in the Connecting Gallery was taken between rings 68 and 69 to the
East (CG R68-69E). For the detailed analyses of the effects of oxidation on the clay close to the
gallery lining, both clay cores (TD R41-42E and CG R68-69E) were cut into thin slices. For each
clay core, the first 10 cm of clay was cut into slices of 2 mm. The slicing of the clay core was performed
in an anaerobic glovebox with a controlled CO2 atmosphere. The mineralogy of the solids
was investigated by means of XRD, FTIR and SEM techniques.
In order to study the effects of an alkaline plume on the Boom Clay mineralogy, the bulk rock
Boom Clay powders were interacted with Young Cement Water (YCW) and Evolved Cement Water
(ECW) simulating the early and evolved alkaline fluids released from water-saturated concrete.
The YCW is highly alkaline (pH = 13.5) solution dominated by K (5500 ppm) and Na (1490 ppm),
while ECW is less alkaline (pH = 12.5) with Na and Ca as dominating cations (440 and 409 ppm
respectively). The PE bottles were charged with 10 g of powdered Boom Clay and 121 ml of the
reagent solution. The bottles were put inside the oven and left at a temperature of 60°C. After specific
time intervals (90, 180, 360 and 510 days), the samples were withdrawn, and the solids were
separated from solutions by centrifugation and filtration. The solid leftovers were subject to XRD,
FTIR and SEM investigation. In addition, cation exchange capacity and surface area measurements
were performed on the selected Boom Clay samples at the end of the alkaline plume batch experiment.
3. Results
The mineralogical study of the oxidized samples from HADES URL showed that gypsum
(CaSO4.2H2O) was formed in the clay close to the gallery lining, both in the samples from Test
Drift and the Connecting Gallery based on XRD investigation (Fig. 1). Its presence is limited to the
first ~4.6 cm from the concrete/clay interface in the Test Drift and to the first ~4.2 cm from the
concrete/clay interface in the Connecting Gallery. Further away from the concrete lining, the XRD
patterns approach those of the undisturbed rock. In the Connecting Gallery, jarosite
(KFe3(SO4)2(OH)6) was identified in the clay close to the gallery lining by means of XRD technique.
In contrast, jarosite was absent in the samples from the Test Drift (Fig. 1). Calcite was not
detected in the clay close to the gallery lining in the samples from the Connecting Gallery, while it
was present in the Test Drift. The solid residuums of the Boom Clay samples were analyzed by
XRD after 90, 180, 360 and 510 days of the interaction with YCW and ECW (Fig. 2). The everpresent
change is the progressive decrease of the pyrite intensity in every studied sample. The py-
512

ite reflections completely disappeared after 510 days of the interaction with alkaline solutions. The
most significant non-clay mineral transformation was observed in the region of 26-28 °2theta
CuK , which is typical for Na-plagioclase/K-feldspar identification reflections. Apparently, the Kfeldspar
KAlSi3O8 is formed at the expense of the Na-Ca plagioclase NaCaAlSi3O8. The formation
of gypsum CaSO4.2H2O was observed after long runs in the ECW (Fig. 2 right). The XRD patterns
of the clay fraction indicate lowering of the intensities of mixed-layered illite-smectite, kaolinite,
chlorite as well as the illite reflections in the YCW with the progress of the alkaline attack. The
drop of the intensities is not evident in the case of the ECW, where the diffractograms more or less
aproach the XRD pattern of the undisturbed Boom Clay. The CEC of the undisturbed whole-rock
Boom Clay (26 1.02 meq/100g) decreased to 21 2.66, 22 2.31 and 22 0.65 meq/100g respectively
after 90, 180 and 360 days of the Boom Clay – Young Cement Water interaction. The
initial CEC value of 42 2.47 meq/100g measured in the

elative intensity (cps)
8000
6000
4000
2000
0
mica (9.9 Å)
kaolinite (7.12 Å)
Qtz
Qtz
pyrite (1.63Å)
YCW
Qtz
Qtz
Qtz
pyrite (1.92 Å)
Qtz
pyrite (2.7 Å)
Qtz
0 20 40 60
°2theta (CuKα)
Qtz
8000
6000
4000
2000
0
mica (9.9 Å)
kaolinite (7.12 Å)
Qtz
0 20 40 60
°2theta (CuKα)
514
Qtz
020 gypsum (7.56 Å)
pyrite (1.63Å)
ECW
Qtz
Qtz
Qtz
pyrite (1.92 Å)
Qtz
pyrite (2.7 Å)
Qtz
Qtz
0 days
90 days
180 days
360 days
510 days
Figure 2: The unoriented specimen XRD patterns of the whole rock Boom Clay samples after 90,
180, 360 and 510 days of the interaction with YCW (Young Cement Water) and ECW (Evolved Cement
Water). The uppermost XRD patterns correspond to the initial (undisturbed) Boom Clay sample.
(1) FeS2 + 7/2 O2(aq.) + H2O � Fe 2+ + 2 SO4 2- + 2 H +
(2) CaCO3 + H + � Ca 2+ + HCO3 -
(3) Ca 2+ +SO4 2- + 2H2O � CaSO4. 2H2O
The oxidation of pyrite is accompanied by a release of Fe 2+ , SO4 2- and acidity into the pore water.
The acidity is buffered to a certain extent by the dissolution of calcite as inidicated by the reaction
(2). In fact, the calcite was not detected in the clay close to the gallery lining in the samples from
the Connecting Gallery, while it was present in the Test Drift. On the other hand, the jarosite was
found exclusively in the clay core sampled in the Connecting Gallery. The absence of calcite in the
core from the Connecting Gallery could point to a lesser buffering capacity of the Boom Clay at this
place leading to a precipitation of jarosite, which is only possible at pH

the ECW from 12.5 at the beginning to as low as 5 after 510 days. The most extensive changes occurred
in the mineralogy of clay phases interacted with the YCW, namely mixed-layered illitesmectite,
kaolinite, chlorite and illite. The fact that the position of the XRD reflections of clay minerals
are not changed with time in the alkaline batch experiment suggests that no clay mineral
phase-to-phase transformation occurred, e.g. illitization of smectite, but rather dissolution was a
dominant process. The dissolution is reflected in the decrease of the specific surface area (SSA) and
decrease of the cation-exchange capacity (CEC) parameters.
5. Conclusions
Comparing the two data sets from the Test Drift and the Connecting Gallery (HADES URL), the
general conclusions could be drawn with respect to the degree and the extent of the oxidation at different
times. The mineralogical evidence for the oxidation is traceable within the first ~4.5 cm
ahead from the gallery lining both in the Test Drift and the Connecting Gallery. The gypsum as the
most common oxidation product of pyrite was found in the both data sets, while the jarosite was
found exclusively in the Connecting Gallery. This point to locally different geochemical conditions
concerning Eh and pH in the Test Drift and Connecting Gallery. However, there is no mineralogical
evidence to state that the ventilation could be an important factor affecting the extent of the oxidation
in the the two studied cases. Therefore, the extent of the oxidation is determined by conditions
or processes occuring during or soon after excavation, rather than during the ventilation of the galleries
during the operational phase.
The mineral stability of the Boom Clay depends on the initial base stregth of the applied alkaline
solution. The YCW with the original pH of 13.2 induced more extensive mineral changes than the
ECW having the initial pH of 12.5. The most significant changes in the mineralogy of Boom Clay
caused by the alkaline plume perturbations involve the alteration of Na-Ca plagioclases to Kfeldpsars
in the both studied cases and the dissolution of clay minerals (mainly mixed-layered illitesmectite
phases) in the YCW. The dissolution of clays is accompanied by the decrease in the Cation
Exchange Capacity and the Specific Surface Area parameters. The clay dissolution might increase
the Boom Clay porosity and thus increase the hydraulic conductivity in the repository near-field.
However, important to note is that based on modelling results of Wang et al. (2007 [4]), the extent
of the alkaline plume disturbed zone in Boom Clay is very limited even after 100 ka.
6. Acknowledgements
This project has been funded by the European Commission and performed as part of the sixth Euratom
Framework Programme for nuclear research and training activities (2002-2006) under contract
FI6W-028403.
References
[1] De Preter, P. (2007) The long-term safety functions within the disposal programmes of
ONDRAF/NIRAS. ONDRAF/NIRAS note O/N 207-0526.
[2] Wickham, S.M. (2005) The ONDRAF-NIRAS Supercontainer Concept. Galson Sciences, UK
(2005).
[3] NIROND. "SAFIR-2, Second safety Assessment and Feasibility Interim Report." NIROND,
Brussels (2002).
[4] Wang, L., Jacques, D., and De Cannière, P. (2007): Effects of an alkaline plume on the Boom
Clay as a potential host formation for geological disposal of radioactive waste, SCK•CEN report,
ER-28, first full draft, March 2007.
515

516

Support actions
517

518

Summary
SAPIERR-II – Shared, Regional Repositories:
Developing a practical implementation strategy
Ewoud Verhoef 1 , Charles McCombie, Neil Chapman 2
1 COVRA, Nieuwdorp, Netherlands; 2 ARIUS, Baden, Switzerland
The basic concept within both EC funded SAPIERR I and SAPIERR II projects (FP6) is that
of one or more geological repositories developed in collaboration by two or more European
countries to accept spent nuclear fuel, vitrified high-level waste and other long-lived radioactive
waste from those partner countries. The SAPIERR II project (Strategic Action Plan for
Implementation of Regional European Repositories) examines in detail issues that directly influence
the practicability and acceptability of such facilities. This paper describes the work in
SAPIERR II project (2006-2008) on the development of a possible practical implementation
strategy for shared, regional repositories in Europe.
1. Introduction
This paper reports on new impetus being given to a relatively old idea - multinational waste management
solutions. Soon after the peaceful use of nuclear energy began to spread in the 1960s and
70s there were proposals for multinational solutions to providing full fuel cycle services to power
plant operators. However, for the final steps in the cycle, the management and disposal of spent fuel
or radioactive wastes, it was only reprocessing services that were actually implemented internationally.
These were provided by countries such as France, the UK and Russia. These countries originally
also provided a disposal service since they did not return any reprocessing wastes to their customers.
With time, however, a waste return clause was included in new reprocessing contracts –
mainly as a reaction to public and political pressures in the reprocessing countries.
Interest revived in the late 1990s, driven both by the high costs of geological repository programmes
and also by the security concerns associated with the prospect of fissile material being
widely distributed across the world. Although several initiatives were proposed, none led to success,
partly because the proposed approaches were judged to be premature and too commercial. Accordingly,
in 2002, the not-for-profit organisation, Arius (Association for Regional and International
Underground Storage), was established to help partner organisations from various countries
explore the possibilities of shared disposal facilities. The current growing worldwide interest in initiating
or expanding nuclear power programmes also emphasises the need for all countries to have a
credible disposal strategy. For many, especially new or small programmes multinational cooperation
leading to shared facilities could be an attractive option, since it optimises use of financial and
human resources. For the international community, global environmental and security benefits can
be achieved by having fewer repositories for spent fuel and/or high level wastes.
519

2. SAPIERR I and II
In Europe, the Parliament and the EC have both expressed support for concepts that could lead to
regional shared facilities being implemented in the EU. The EC has funded two projects
that can form the first steps of a staged process towards the implementation of shared regional
or international storage and disposal facilities. In the period 2003 to 2005, the EC
funded the project SAPIERR I (Support Action: Pilot Initiative for European Regional
Repositories), a project devoted to pilot studies on the feasibility of shared regional storage
facilities and geological repositories, for use by European countries. The SAPIERR I
project looked at the basic technical and economic feasibility of implementing regional,
multinational geological repositories in Europe. The studies indicated that shared regional
repositories are feasible and that a first step could be to establish a structured framework
for the future work on regional repositories. The present SAPIERR II project (Strategic
Action Plan for Implementation of Regional European Repositories) examines in more detail
specific organisational, legal, societal economic, safety and security issues that directly
influence the practicability and acceptability of such facilities.
2.1 SAPIERR II work plan
The work plan is designed as a stepwise approach to development of a practical implementation
strategy. The tasks performed in the project are listed and described below. Each task translates into
a work package (WP) within the work plan:
1. Preparation of a management study on the legal and business options for establishing an
European Repository Development Organisation (ERDO) leading to one or more proposed
frameworks (options) for such an organisation.
2. A study on the legal liability issues of international waste transfer within Europe. Even in
national disposal programmes, the issues associated with long-term transfer of liabilities are
complex. For a regional repository, the challenges are still greater. Immediate transfer of all
liabilities and shared responsibilities reaching out to far future times are two extremes that
bracket the possibilities to be considered.
3. A study of the potential economic implications of European regional storage facilities and
repositories. The study analyses the economic implications for potential users of such facilities
and also for host countries. The study examines not only the costs of disposal facilities
but also the benefits, both economic and societal, that a host country and community could
gain.
4. Outline examination of the safety and security impacts of implementing one or two regional
stores or repositories relative to a large number of national facilities. The radiological safety
comparisons are based on existing performance assessments.
5. A review of public and political attitudes in Europe towards the concept of shared regional
repositories. This is based on input from literature studies by representatives of organisations
participating in SAPIERR II, complemented by a review by project team members of
the situation in other European countries and by limited specific questioning of relevant
groups. The work is linked to Work Package 3 since public attitudes can be strongly affected
by local and national benefits.
6. Development of a Strategy and a Project Plan for the work of the EDO. The first tasks of an
ERDO would be agreeing a progressive, staged strategy that would lead to the definition of
potential host countries and eventually, to potential repository sites and definition of a parallel
science and technology programme that could be addressed by the ERDO after its initiation.
520

7. Management and dissemination of information. Contact and consultation with appropriate
national bodies and with EC staff is essential to gather the necessary policy and technical
input for the project and before judging the feasibility of any proposals for future collaboration.
2.2 Results
The most obvious advantages are economic benefits to partner countries. It is estimated that partner
countries could each save of the order of 500 million to 1 billion EUR by sharing development
costs rather than having to implement a national geological repository. If a regional facility is able
to offer disposal services to other European countries after the repository has become operational,
the original partner countries may be able to manage their own current and future wastes with further
significant cost reductions. There will specific economic benefits to the host country and community.
The country and community that hosts a repository will benefit from large initial inwards
flows of capital during the development period and, eventually, of revenues and taxes from operating
the facility over a period of many decades. The sums involved are expected to be of the order of
several billion EUR. A European regional nuclear facility is likely to attract other international,
high-technology activities to the region and can form the basis for a regional economic development
plan.
Figure 1. The benefits to partner countries will be felt at local, national and international level
Most of the problems of developing a shared regional repository are comparable to that of developing
a national repository. In both national and multinational programmes, finding suitable sites remains
the biggest challenge. Since the early 1980s, siting radioactive waste repositories has proved
immensely difficult in every country, but real lessons have been learned in the last decade and a
modern, inclusive process has emerged that is widely accepted today as a model for dealing with
difficult environmental issues. The approach advocated for a European repository will find a site
that is demonstrably environmentally safe and secure. It also aims at working with local communities
that are interested in the project and that may wish to become actively involved in its development.
The approach involves partner countries initially agreeing on excluded areas that are clearly
technically unsuitable for a geological repository. Communities from all other areas would then be
invited to express interest in the project and a community-level and national-level discussion and
evaluation process would begin to find a suitable site. No national declaration of willingness to be a
repository host is necessary to join the exploratory group. Potential host countries will emerge only
521

after extensive interactions have taken place, involving interested communities within the country.
Potential host countries can withdraw from the siting process at any time up to the point where a
final commitment is needed. Shared regional waste management facilities will have to meet the
highest standards of environmental safety. This will be assured by the national regulatory agencies
in the partner countries working closely together. The high profile and level of interest worldwide
in the project indicate that it would be valuable to involve the IAEA and the European High Level
Expert Group, in a wide overview and regulatory capacity.
3. Development of a practical implementation strategy
Over the last four years, the SAPIERR II project has investigated what would be needed to make a
regional approach viable and to identify the benefits that would accrue to partner countries – as well
as realistically acknowledging the challenges faced by repository implementers (see [1] for description
of the work and related reports). The final stage of this preparatory work is to explore with the
governments of potential partner countries their attitudes to, and possible interest in, launching a
new initiative. Based on the organisational, legal, societal economic, safety and security issues carried
out within the SAPIERR II project, a staged, adaptive implementation strategy and organisational
structures for an European Repository Development Organisation (ERDO) is being proposed.
The first step in the strategy is the establishment of a working group of interested countries.
3.1 The ERDO Working Group
The working group will be charged with the task of setting up a not-for-profit ERDO that would,
over the next ten to fifteen years, prepare for a follow on organisation that would implement shared
geological repositories in Europe. Policy-makers in potentially interested European countries have
been invited to participate in the working group. Representatives of the SAPIERR project have
been explaining the technical, economic and societal aspects of shared waste management facilities
to potentially interested countries. This activity will continue until the end of 2008, but already sufficient
support has been gained for establishing a working group in 2009 to define the form, structure
and organisation of the ERDO.
3.2 The ERDO/ERO
If shared repositories are to become a reality, a dedicated organisation will be required that can
work towards the goal on the extended timescales that national disposal programmes have shown to
be necessary. A multinational organisation faces much wider challenges than does a national waste
management body, not least because of the extended range of stakeholders. The figure below gives
an impression of the multiple stakeholders to be managed.
522

Figure 2. Interfaces to be managed by an ERDO (and ERO)
The SAPIERR II project has proposed two types of organisations that may eventually be formed for
performing the work leading to implementation of a regional repository in Europe: A European Repository
Development Organisation (ERDO) and a European Repository Organisation (ERO). The
definitions of ERDO and ERO are as follows:
• ERDO (European Repository Development Organisation): the initiating, non-profit organisation
for a shared geological disposal facilities project. Its objective is to establish the systems, structures
and agreements and carry out all the work necessary for putting in place a shared waste management
solution and geological repository (or repositories). This work would continue through the
investigation of potential sites and up to the point of license application to begin the construction of
a repository. It is assumed that this may take about 10+ years. At this point the ERDO may decide
to transform into or separately establish the ERO.
• ERO (European Repository Organisation): the implementing organisation for waste disposal. The
ERO would be the license holder for the repository and responsible for all subsequent operational
activities in a host country that has agreed to dispose of wastes from other European countries. The
form for the ERO will be chosen at a future date by the members of the ERDO, assuming that they
come to the conclusion that the ERDO organisation needs to be altered. The choice will also be
strongly influenced by the preferences of the country or countries that have been identified as repository
hosts. The ERO could be either non-profit or commercial in structure.
4. Outlook
Even before the founding of the ERDO, there are many important decisions to be taken by the potential
partners. These include the size of the organisation, the legal form, the domicile, the staffing
policy, the budget, etc., etc. All of these decisions require prior date amongst participants to arrive
at a consensus that all can support. Accordingly, the SAPIERR II team has proposed that an interim
step be taken. An ad-hoc ERDO Working Group should consider the key issues for around one year
and then see if there is sufficient agreement to allow formal establishment of the ERDO. A key
point is that participation in the ERDO ad-hoc Working Group does not commit Member States to
523

eing partners if an ERDO is formally established. Invitations to join the ERDO Working Group or
to have discussions on this possibility, have been sent to around twenty different European countries.
The project leaders also had a bilateral discussion with EU Commissioner Piebalgs in which
he expressed his support for multinational initiatives to waste disposal. To date none of the EU
Member States that were approached has declined. The countries that have already indicated that
they will participate in the working group to date are: Bulgaria, Czech Republic, Estonia, Italy, Latvia,
Lithuania, Netherlands, Romania, Slovakia, Slovenia and Spain. The first meeting of the working
group will be held together with the closing seminar of the SAPIERR II project at the end of
January 2009.
5. Acknowledgements
This project has been co-funded by the European Commission and performed as part of the sixth
Euratom Framework Programme for nuclear research and training activities (2002-2006) under contract
number 035958.
References
[1] www.sapierr.net
524

EC-DG TREN
525

526

Improving Financing Schemes for Nuclear Decommissioning and Radioactive
Waste Management in European Member States and at EU Level
Wolfgang Irrek
Wuppertal Institute for Climate, Environment and Energy, Germany
Summary
Member States oversee different regimes for estimating and managing costs and for raising
funding for nuclear decommissioning and radioactive waste management activities, and there
are significant differences in the operation, governance, investment and accessibility of the
existing funds across the EU. Additional actions are needed on the level of the Member States
as well as on the European level in order to ensure that adequate funds for nuclear decommissioning
and waste management are available when necessary. Furthermore, transparency
should be increased by information sharing and reporting. These are the results of a comprehensive
assessment of the financial consequences and risks of the different decommissioning
financing systems from governance, accounting, valuation and investment perspectives in the
course of a study by the Wuppertal Institute and its partners on behalf of the European Commission
[1]. Furthermore, the legal aspects of decommissioning financing in the EU have
been analysed. Based on this, a number of recommendations are made on how to ensure that
adequate funds are available when necessary. These recommendations are made to Member
States and to actions that could be undertaken now on the European level. Furthermore, the
report makes suggestions on how further harmonization of decommissioning financing could
be achieved on the EU level if necessary. Along with these recommendations are suggestions
for information sharing and reporting that should be undertaken across the EU to increase
transparency.
1. Introduction
The European Commission estimates that approximately one third of the 145 power reactors currently
operating in the European Union will need to be shut down by 2025 [2]. This will result in
the need to dismantle, decontaminate and demolish these nuclear facilities as well as to undertake
processing, conditioning and disposal of nuclear waste and spent fuel (‘decommissioning’). It is of
paramount importance that the funding of these decommissioning activities will be adequate and
available when needed in order to avoid negatively affecting the safety of EU citizens. Nuclear operators
are expected to accumulate all the necessary funds during the operating life of facilities.
Member States oversee different regimes for estimating, collecting and managing decommissioning
costs and there are significant differences in the operation, governance, investment
and accessibility of the existing funds across the EU. The research project carried out by Wuppertal
Institute together with seven project partners and five subcontractors on behalf of the European
Commission undertook a comprehensive assessment of the financial consequences and risks of the
different regimes from governance, accounting, valuation and investment perspectives [1].
527

2. Methodology
Based on existing data and information (cf. particularly [3]), a comprehensive and deep analysis
was carried out of the current (and planned) approaches to financing nuclear decommissioning of
nuclear installations in those 16 countries in the EU-27 in which nuclear installations are currently
operated. The following central question guided the analysis: Is decommissioning financing (a) adequate,
(b) available when needed, (c) secure and (d) well-managed in (e) a transparent way to ensure
safe decommissioning based upon the ‘Polluter Pays Principle’, i. e. that the generation and
actors benefiting from the nuclear energy produced will fully pay for safe decommissioning?
Based on the country analyses, the financial consequences and risks of the different decommissioning
financing schemes implemented were analysed. In addition, the legal basis for activities by the
European Commission with respect to decommissioning financing systems in the Member States
was discussed. Finally, recommendations and conclusions were developed.
3. Results of country analyses
The country analyses have shown that differences in current decommissioning financing approaches
in the EU-27 refer to
the decommissioning liabilities, strategies and time schedules,
the approaches to quantifying the decommissioning costs,
the different methods for setting aside and managing funds (cf. Table 1) including the accessibility
of the operators of the nuclear installations to these funds,
how the funding schemes deal with early plant closure or other unforeseen events,
transparency of the schemes to the public, and
stakeholders’ opinion on the funding schemes in their countries.
Table 1: Overview on decommissioning financing systems in the EU-27
Kind of facility Payment from
current budget
Uranium
mine/mill 1
Research reactors
Internal External
Unrestricted Restricted Unrestricted Restricted
e.g., D, CZ e.g., F
e.g., D, E, UK,
IT, B
NPP
Uranium conversion,enrichment
and
fuel fabrication
plants
Reprocessing
plants
UK (NDA) D, B, NL,
IT (SOGIN-
ENEL), CZ
Storage, disposal
e.g., CZ e.g., F, CZ
UK D, NL F
D, UK F
e.g., D, UK e.g., E, F, NL
(COVRA)
528
F, CZ IT (CCSE) FIN, LT, S, UK (NLF:
British Energy), SK, E,
BG, HU, SI
e.g., FIN, S, CZ
This selection is not meant to be exhaustive. There are no provisions for decommissioning in Romania yet.
Source: Wuppertal Institute et al. 2007, 37

The results can be summarised as follows:
The Polluter pays principle for decommissioning is widely accepted. However, only in few
countries it is the basis for granting an operating license.
Costs estimates are subject to high degree of risks and uncertainties.
Differences in reported cost estimates occur due to varying discounting mechanisms and the
timing of dismantling.
Not all Member States require that funds be managed externally and segregated from the
operator.
A number of Member States seem to be moving towards the increased restriction of funds.
This development might be further accelerated by pressure from the financial markets.
In most countries there are only limited rights for the public to access information on decommissioning
costs and funds.
Many operating companies and governments are satisfied with the current situation and
have concerns towards an EU harmonization process of nuclear decommissioning financing.
The discussions on decommissioning funds so far have focused on nuclear power plants.
Decommissioning of other facilities must not be overlooked, in particular for high cost facilities,
such as reprocessing plants or facilities having experienced incidents or accidents.
4. Discussion of financial consequences and risks
What are the financial consequences and risks of the different decommissioning financing systems
from governance, accounting, valuation and investment perspectives?
From a governance perspective, the higher the potential conflict of interests within a particular
decommissioning methodology, the greater the need for additional checks and balances.
Externally managed funds have a lower risk of conflicts of interest.
Using the accounting perspective leads to the conclusion that reliability and comparability
of accounting have to be improved.
The valuation perspective is particularly important to investors. A reliable valuation has to
allow a comprehensive risk assessment. To enable this to happen, transparency is paramount.
The incentive to finance part of future decommissioning costs through a high investment
performance is evident. However, high performance investments can conflict with the prudence
principle, which plays an important role in the field of financial asset management.
5. Conclusions
Member States must ensure that adequate funds will be available when necessary, and that – using
the 'Polluter Pays Principle’ – risks and uncertainties are eliminated as far as possible. These steps
include:
The identification of risks such as the changing of ownership of utilities or the existence of
two or more different decommissioning financing schemes in one market.
Increasing transparency; experience shows that transparency is a key issue for any internal
or external fund. Given this, an operator has to define and establish a procedure which is effective,
clear and transparent.
Assuring a high degree of independence between actors in the governance chain is crucial.
This must include organisational and structural independence of the different organisation as
well as personal independence, particularly with regard to the independence of the fund
manager from the operator and the independence of the licensing authorities. In principle
529

external funds ensuring the independence of decommissioning fund management from the
operator reduce the need for additional checks and balances (cf. also similar conclusion and
recommendations by KPMG [4]). Internal unrestricted financing schemes should be
changed into restricted funds, with a measurable degree of separation.
Introduction of a uniform accounting system, ideally one based on the International Financial
Reporting Standards (IFRSs) for both public and private licensees is necessary.
Additional guarantees to cover unplanned eventualities to ensure that under all circumstances
the polluter pays principle is adhered too should be undertaken (cf. Fig. 1).
140
120
100
80
60
40
20
0
Guarantee I
0 5 10 15 20 25 30 35 40 45 50
Decommissioning activities
during operation (for
simplification not included here)
Years
Figure 1: Guarantee I and Guarantee II covering financial risks related to decommissioning
cost occurring after final shutdown of the plant - Source: Wuppertal Institute et al. 2007, 154
Establishment of investment guidelines for a professional asset & liability management
framework to address the trade-off between high performance and high security of funds
and describing the required qualifications of investment managers.
Action will also be needed on the EU level to increase both transparency and oversight. It is recognised
a number of processes already existing. However, in order to further improve transparency
regular uniform reports should be produced by Member States. The study makes precise recommendations
on three respective reporting levels. The transparency process should be further enhanced
by the establishment of a Council (of trustees) of European Nuclear Decommis-sioning
Funds (CENDF) on the European Level or a European Nuclear Decommissioning Oversight Board
(ENDOB) at a later stage.
According to the experiences with the European Commission’s draft directives of 2003 under Article
31 of the EURATOM Treaty on nuclear safety and radioactive waste management (the “nuclear
package”) and discussions with stakeholders in the course of this project, further legal steps on the
530
Guarantee
II
Decommissioning
(dismantling etc.)
after final shuddown
Provisions covering undiscounted costs
from start of operation
Provisions covering discounted costs from
start of operation + interest accumulated
over lifetime
Provisions as regular undiscounted
(linear) installments over lifetime
Provisions as discounted regular
installments + interest accumulated over
lifetime
Possible real costs of dismantling etc.
after final shutdown

European level should not be envisaged at the moment. However, if the European institutions considered
using the Treaty of the European Communities as a legal base for potential action, further
regulation of decommissioning financing at EU level would be justified. Further harmonisation in
the EU would be achieved by the introduction and implementation of binding legislation by Member
States. Its legal base could focus on the impact of differences in decommissioning financing
schemes on the energy market and/or environmental protection, neither of which are adequately addressed
through the EURATOM Treaty. Such a directive would only be necessary if the current
processes were not fully implemented.
6. Acknowledgements
The study has been carried out on behalf of the European Commission, to which the author of this
paper is extremely grateful for its financial support and co-operation.
A special acknowledgement goes to the members of the consortium who developed and collected
questionnaires whose data and information provided a valuable first basis for this project [2]. Data
and information directly received from operators of nuclear facilities, decommissioning fund managers,
ministries and further stakeholders contacted/interviewed for the purpose of this study added
a lot to this basis. Therefore, the author would particularly like to thank all these stakeholders for
the data and information provided.
Finally, the author would like to thank all the partners who have contributed to the project, in particular:
Kaspar Müller (Ellipson), Dörte Fouquet (Kuhbier Lawfirms), Antony Patrick Froggatt (private
consultant), Abdelkader B. Ameur (AST), Veit Bürger (Öko-Institut), Jozef Krizan, Luba
Kupke-Siposova and Maria Mistrikova (Energia 2000 & partner organisations), Tamás Pázmándi
and Péter Zagyvai (AEKI), Mycle Schneider (Mycle Schneider Consulting), Daiva Semeniene
(AAPC), Ian Smith (private consultant), Stephen Thomas (PSIRU), and, Seppo Vuori (VTT).
References
[1] Wuppertal Institute [Wuppertal Institut für Klima, Umwelt, Energie GmbH]; et al.: Comparison
among different decommissioning funds methodologies for nuclear installations. Service
Contract TREN/05/NUCL/S07.55436 on behalf of the European Commission, Directorate-
General Energy and Transport, H2. Wuppertal, 2007
Download:
http://www.wupperinst.org/en/projects/proj/index.html?&projekt_id=167&bid=130
[2] European Commission: Communication from the Commission to the European Parliament
and the Council, Second Report on the use of financial resources earmarked for the decommissioning
of nuclear installations, spent fuel and radioactive waste, COM(2004) 794 final.
Brussels, 12 Dec 2007
[3] Questionnaires filled-in by most of the EU Member States and candidate countries (except
Bulgaria) in the course of the DG TREN project „Analysis of the factors influencing the selection
of strategies for decommissioning of nuclear installations“ (Contract Number
TREN/04/NUCL/S07.40075) carried out by Colenco and Iberinco.
[4] KPMG [KPMG Corporate Finance N.V.]; NRG (2006): Financiële Zekerheidsstelling Kernenergiewet,
Ministerie van VROM, Amsterdam, Petten.
531

532

LIST OF PARTICIPANTS
533

534

Mr Murat K. ABDULAKHATOV
V.G. Khlopin Radium Institute
2nd Murinsky av., 28
RU – 194021 St. Petersburg
Tel.: +7 812 297 56 41
Fax: +7 813 713 61 67
abdulakhatov@gtn.ru
Mr Bertil ALM
Östhammars Kommun – GMF
Skogsvägen 3
SE – 747 31 Alunda
Tel.: +46 70 922 84 47
Fax: +46 173 175 37
bertil.alm@centerpartiet.se
Dr Florian AMANN
ETH Zürich – Engineering Geology
Wolfgang Pauli Strasse 15
CH – 8093 Zürich
Tel.: +41 44 633 6818
Fax:
florian.amann@erdw.ethz.ch
Mr Peter ANDERSSON
Östhammars Kommun – GMF
SE – 742 21 Östhammar
Tel.: +46 173 864 17
Fax: +46 173 175 37
peter.andersson@osthammar.se
Dr Julio ASTUDILLO PASTOR
ENRESA - Research & Technology Div.
C/Emilio Vargas 7
ES – 28043 Madrid
Tel.: +34 91 566 81 20
Fax: +34 91 566 81 69
jasp@enresa.es
Dr Diana-Maria AVRAM
Politecnico di Milano
Via Ponzio 34/3
IT – 20133 Milano
Tel.: +39 02 2399 6362
Fax: +39 02 2399 6309
diana.avram@mail.polimi.it
LIST OF PARTICIPANTS
535
Dr. phil. Michael ÄBERSOLD
Bundesamt für Energie (BFE)
Disposal of Radioactive Waste Section
Mühlestrasse 4
CH – 3063 Ittingen
Tel.: +41 31 322 56 31 - +41 79 506 50 04
Fax: +41 31 323 25 00
michael.aebersold@bfe.admin.ch
Dr Scott ALTMANN
ANDRA – Scientific Division
Parc de la Croix Blanche – 1-7, rue Jean Monnet
FR – 92298 Châtenay-Malabry Cedex
Tel.: +33 1 46 11 84 81 - +33 1 46 11 81 74
Fax: +33 1 46 11 82 08
scott.altmann@andra.fr
Dr Kjell ANDERSSON
Karita Research AB
P.O. Box 6048
SE – 187 06 Täby
Tel.: +46 8 510 147 55
Fax: +46 8 510 147 55
kjell.andersson@karita.se
Mr Marcel ARNOULD
Carrière Marle 6
FR – 92340 Bourg-La-Reine
Tel.: +33 1 46 63 05 62
Fax: +33 1 46 64 37 65
marcelarnould@yahoo.fr
Ms Jaana AVOLAHTI
Ministry of Employment and the Economy
Aleksanterinkatu 4
FI – 00023 Helsinki – Government
Tel.: +358 9 1606 4836 - +358 50 592 2763
Fax: +358 9 1606 2173
jaana.avolahti@tem.fi
Mr Göran BÄCKBLOM
CONROX AB
Entsavägen 61
SE – 187 35 Täby
Tel.: +46 70 556 26 48
Fax:
goran.backblom@conrox.com

Mr Jozef BALAZ
JAVYS, a.s. - Nuclear Decommissioning
Company, j.s.co.
SK – 919 31 Jaslovské Bohunice
Tel.: +421 910 834 343
Fax: +421 36 639 1107
balaz.jozef@javys.sk
Mr Francesco BASILE
EC – JRC Ispra
Via Enrico Fermi 1 – Office: 84 00/0214
IT – 21020 Ispra (VA)
Tel.: +39 0332 785 170
Fax: +39 0332 785 077
francesco.basile@ec.europa.eu
Dr Dirk-Alexander BECKER
GRS mbH
Long-term safety assessment department
Theodor-Heuss-Strasse 4
DE – 38122 Braunschweig
Tel.: +49 531 8012 269
Fax: +49 531 8012 211
dirk-alexander.becker@grs.de
Mr Gianluca BENAMATI
Italian Parliament
13, Via Migliavacca
IT – 43100 Parma
Tel.: +39 329 831 10 27 - +39 0534 801 423
Fax: +39 0534 801 250
gianluca.benamati@brasimone.enea.it
Dr Anne BERGMANS
Universiteit Antwerpen (UA)
Department of Sociology
St. Jacobstraat 2
BE – 2000 Antwerpen
Tel.: +32 3 275 55 42
Fax: +32 3 275 57 96
anne.bergmans@ua.ac.be
Dr Ved BHATNAGAR
European Commission – DG RTD/J/2
Rue de la Loi 200 – CDMA 1/46
BE – 1049 Bruxelles
Tel.: +32 2 299 58 96
Fax: +32 2 295 49 91
ved.bhatnagar@ec.europa.eu
LIST OF PARTICIPANTS
536
Dr Dalis BALTRUNAS
Institute of Physics
Savanoriu Avenue 231
LT – 2300 Vilnius
Tel.: +370 5 266 1657
Fax: +370 5 260 2317
dalis@ar.fi.lt
Prof. Behrooz BAZARGAN-SABET
École des Mines de Nancy & BRGM
Parc de Saurupt
FR – 54000 Nancy
Tel.: +33 3 83 59 63 29
Fax: +33 3 83 57 18 85
b.bazargan-sabet@brgm.fr
Mr Mahrez BEN BELFADHEL
Nuclear Waste Management Organization (NWMO)
Ontario Power Generation
22 St. Clair Ave. East - 6th Floor
CA – Toronto, Ontario M4T 2S3
Tel.: +1 647 259 3036
Fax: +1 416 934 9526
mbenbelfadhel@nwmo.ca
Ms Mary BERGMAN
Intrasoft
Rue Montoyer 40
BE – 1000 Bruxelles
Tel.: +32 2 238 17 92
Fax:
mary.todd-bergman@intrasoft-intl.com
Ir Frédéric BERNIER
Federal Agency for Nuclear Control (FANC)
Agence fédérale de Contrôle nucléaire (AFCN)
Rue Ravenstein 36
BE – 1000 Bruxelles
Tel.: +32 2 289 20 11 - +32 499 986 966
Fax: +32 2 289 21 12
frederic.bernier@fanc.fgov.be
Ms Ute BLOHM-HIEBER
European Commission – DG TREN/H/2
Rue Robert Stumper 10 – EUFO 4/286
LU – 2557 Luxembourg
Tel.: +352 4301 34151
Fax: '+352 4301 30139
ute.blohm-hieber@ec.europa.eu

Dr Peter BLÜMLING
NAGRA – Science & Technology Division
Hardstrasse 73
CH – 5430 Wettingen
Tel.: +41 56 437 12 93
Fax: +41 56 437 13 45
peter.bluemling@nagra.ch
Dr Ricardo BOLADO LAVÍN
EC – JRC Petten – Institute of Energy
Westerduinweg 3
NL – 1755 LE Petten
Tel.: +31 224 565 349
Fax: +31 224 565 641
ricardo.bolado-lavin@jrc.nl
Mr Jean-Michel BOSGIRAUD
ANDRA
Parc de la Croix Blanche – 1-7, rue Jean Monnet
FR – 92298 Châtenay-Malabry Cedex
Tel.: +33 1 46 11 82 34
Fax: +33 1 46 11 82 08
jean-michel.bosgiraud@andra.fr
Dr Stéphane BOURG
CEA Valrhô – Marcoule
DEN/MAR/DRCP
FR – 30209 Bagnols-sur-Cèze Cedex
Tel.: +33 4 66 79 77 02
Fax: +33 4 66 79 69 80
stephane.bourg@cea.fr
Prof. Dr Wernt BREWITZ
TU Braunschweig – Leichtweiss-Institute
Beethovenstrasse 51A
DE – 38106 Braunschweig
Tel.: +49 531 391 7140 - +49 160 797 6891
Fax: +49 531 391 4584
wernt.brewitz@tu-bs.de
Wernt.Brewitz@t-online.de
Mr Gérard BRUNO
European Commission – DG TREN/H/2
Rue Robert Stumper 10 – EUFO 4/289
LU – 2557 Luxembourg
Tel.: +352 4301 32506
Fax: +352 4301 30139
gerard.bruno@ec.europa.eu
LIST OF PARTICIPANTS
537
Dr Angelika BOHNSTEDT
Forschungszentrum Karlsruhe (FZK) GmbH
Nuklear Safety Research
Weberstrasse 5
DE – 76021 Karlsruhe
Tel.: +49 7247 825 525
Fax: +49 7247 825 508
angelika.bohnstedt@nuklear.fzk.de
Mr Wilhelm BOLLINGERFEHR
DBE TECHNOLOGY GmbH
Research & Development Dept.
Eschenstrasse 55
DE – 31224 Peine
Tel.: +49 5171 43 1525
Fax: +49 5171 43 1506
bollingerfehr@dbe.de
Mr Bernard BOULLIS
CEA Saclay – DEN/DDIN
Bâtiment 125 – Point Courrier n° 7
FR – 91191 Gif-sur-Yvette Cedex
Tel.: +33 1 69 08 64 55
Fax: +33 1 69 08 32 32
bernard.boullis@cea.fr
Dr Vinzenz BRENDLER
Forschungszentrum Dresden-Rossendorf (FZR) GmbH
Institut für Radiochemie
Bautzner Landstrasse 128
DE – 01314 Dresden-Rossendorf
Tel.: +49 351 260 2430
Fax: +49 351 260 3553
v.brendler@fzd.de
Prof. Jordi BRUNO
AMPHOS XXI Consulting S.L.
Waste Management Department
Passeig de Rubi 29-31
ES – 08197 Valldoreix, Barcelona
Tel.: +34 93 583 05 00 - +34 909 32 07 73
Fax: +34 93 589 00 91
jordi.bruno@amphos21.com
Ms Christiane BRUN-YABA
Institut de Radioprotection et de Sûreté nucléaire
(IRSN)
31, av. de la Division Leclerc
FR – 92262 Fontenay-aux-Roses Cedex
Tel.: +33 1 58 35 97 64
Fax: +33 1 58 35 85 09
christine.brun-yaba@irsn.fr

Prof. Gunnar Johan BUCKAU
Forschungszentrum Karlsruhe (FZK) GmbH
Institut für Nukleare Entsorgung (INE)
Hermann-von-Helmholzplatz 1
DE – 76344 Eggenstein-Leopoldshafen
Tel.: +49 7247 824 461 - +49 179 515 02 34
Fax: +49 7247 824 308
buckau@ine.fzk.de
Dr Christoph BUNZMANN
Bundesamt für Strahlenschutz (BfS)
Willy-Brandt-Strasse 5
DE – 38201 Salzgitter
Tel.: +49 30 18333 1511
Fax: +49 30 18333 1535
cbunzmann@bfs.de
Prof. Robert CHARLIER
Université de Liège (ULG)
Chemin des Chevreuils 1 – Bâtiment B52
BE – 4000 Liège 1
Tel.: +32 4 366 93 34
Fax: +32 4 366 95 20
robert.charlier@ulg.ac.be
LIST OF PARTICIPANTS
Dr Chien CHUNG
National Tsing Hua University - Dept. of Biomedical
Engineering & Environmental Sciences
101 Kuang Fu Road, Sec. 2
TW – 300 Hsinchu0
Tel.: +886 3 572 7297
Fax: +886 3 571 8649
cc@mx.nthu.edu.tw
Dr Hans D K CODÉE
COVRA n.v.
Spanjeweg 1
NL – 4380 AE Vlissingen
Tel.: +31 113 616 655
Fax: +31 113 616 650
hans.codee@covra.nl; marianne.cornet@covra.nl
Ms Tammy COTE
Assembly of First Nations
473 Albert Street – 8th floor
CA – Ottawa, Ontario K1R 5B4
Tel.: +1 613 241 6789 ext. 265
Fax: +1 613 241 5808
tacote@afn.ca
538
Ms Maria Crina BUCUR
Institute for Nuclear Research (SCN)
Campului 1
RO – 115400 Pitesti-Mioveni
Tel.: +40 248 213 400
Fax: +40 248 262 499
crina.bucur@nuclear.ro
Dr Pascal CHAIX
CEA Saclay – Nucl. Energy Division – DEN-DSOE
Bâtiment 121
FR – 91191 Gif-sur-Yvette Cedex
Tel.: +33 1 69 08 84 38 - +33 6 85 82 60 26
Fax: +33 1 69 08 15 44
pascal.chaix@cea.fr
Ms Milena CHRISTOSKOVA
DPRAO
51, James Bourcher Blvd., Fl. 18
BG – 1407 Sofia
Tel.: +359 2 962 4948
Fax: +359 2 962 5078
mzh@dprao.bg
Dr Jean COADOU
European Commission – DG TREN.H.2.001
Rue Henry M. Schnadt 1 – EUFO 04/384
LU – 2920 Luxembourg
Tel.: +352 4301 34034
Fax: +352 4301 30139
jean.coadou@ec.europa.eu
Dr Christine COKER
ETL Ltd
3, Nathans' Road
UK – Wembley, Middlesex HA0 3RY
Tel.: +44 208 904 1520
Fax: +44 1280 814195
surfing@dircon.co.uk
Dr Miguel CUÑADO
ENRESA
C/Emilio Vargas 7
ES – 28043 Madrid
Tel.: +34 91 566 81 52
Fax: +34 91 566 81 65
mcup@enresa.es

Ms Gyula DANKO
Golder Associates (Hungary) Kft.
Huvosvolgyi 54
HU – 1021 Budapest
Tel.: +36 1 394 00 05 - +36 30 9621 542
Fax: +36 1 394 00 02
gdanko@golder.hu
Ms Katie DAWSON
Sellafield Ltd
Sellafield
UK – Seascale, Cumbria CA20 1PG
Tel.: +44 19467 85147
Fax:
katie.dawson@sellafieldsites.com
Dr Sylvia DE GRANDIS
ENEA – Località Brasimone
Località Brasimone
IT – 40032 Camugnano
Tel.: +39 05 3480 1243
Fax:
silvia.degrandis@brasimone.enea.it
Dr Peter DE PRETER
EIG EURIDICE ESV – c/o ONDRAF/NIRAS
Boeretang 200
BE – 2400 Mol
Tel.: +32 14 33 27 99
Fax: +32 14 32 37 09
p.depreter@nirond.be
Mr Hugues DESMEDT
European Commission – DG RTD/J/1
Rue de la Loi 200 – CDMA 00/69
BE – 1049 Bruxelles
Tel.: +32 2 299 89 87
Fax:
hugues.desmedt@ec.europa.eu
Mr Gaetano DI BARTOLO
European Commission – DG RTD/J/2
Rue de la Loi 200 – CDMA 1/58
BE – 1049 Bruxelles
Tel.: +32 2 299 03 87 - +32 487 649 399
Fax: +32 2 295 49 91
gaetano.di-bartolo@ec.europa.eu
LIST OF PARTICIPANTS
539
Mr Christophe DAVIES
European Commission – DG RTD/J/2
Rue de la Loi 200 – CDMA 1/61
BE – 1049 Bruxelles
Tel.: +32 2 296 16 70
Fax: +32 2 295 49 91
christophe.davies@ec.europa.eu
Mr Christophe DE BOCK
ONDRAF/NIRAS – R&D Geological Disposal Dept.
Avenue des Arts 14
BE – 1210 Bruxelles
Tel.: +32 2 212 10 07 - +32 485 757 089
Fax: +32 2 218 51 65
c.debock@nirond.be
Mr Aymeric DE LHONEUX
SUEZ-TRACTEBEL S.A. –
TRACTEBEL Engineering
Avenue Ariane 7
BE - 1200 Bruxelles
Tel.: +32 473 696 473
charles-aymeric.delhoneux@tractebel.com
Mr Marc DEMARCHE
ONDRAF/NIRAS
Avenue des Arts 14
BE – 1210 Bruxelles
Tel.: +32 2 212 10 12
Fax:
m.demarche@nirond.be
Mr Michel DETILLEUX
SUEZ-TRACTEBEL S.A. - TRACTEBEL Eng.
Avenue Ariane 7
BE – 1200 Bruxelles
Tel.: +32 2 773 82 25
Fax: +32 2 773 82 00
michel.detilleux@tractebel.com
Dr Daniela DIACONU
Institute for Nuclear Research (SCN)
Campului 1
RO – 115400 Pitesti-Mioveni
Tel.: +40 248 213 400
Fax: +40 248 262 449
daniela.diaconu@nuclear.ro

Dr Mario DIONISI
APAT – Dipartimento Rischio Nucleare e
Radiologico
Via Vitaliano Brancati 48
IT – 00144 Roma
Tel.: +39 06 5007 2303
Fax: +39 06 5007 2941
mario.dionisi@apat.it
LIST OF PARTICIPANTS
Mr Hubert DOUBRE
Commission nationale d'Évaluation (CNE) des
Recherches sur la Gestion des Déchets radioactifs
Tour Mirabeau – 39-43, quai André Citroën
FR – 75015 Paris
Tel.: +33 1 40 58 89 05
Fax: +33 1 40 58 89 38
c.jouvance.cne@wanadoo.fr
Mrs Marie-Claude DUPUIS
ANDRA
Parc de la Croix-Blanche – 1-7, rue Jean Monnet
FR – 92288 Châtenay-Malabry Cedex
Tel.: +33 1 46 11 80 25
Fax: +33 1 46 11 82 25
marie-claude.dupuis@andra.fr
Mr Ladislav EHN
JAVYS, a.s.
Nuclear Decommissioning Company, j.s.co.
SK - 919 31 Jaslovské Bohunice
Tel.: +421 910 834 352
Fax: +421 33 559 15 64
ehn.ladislav@javys.sk
Mr Joaquín FARIAS SEIFERT
AITEMIN
C/Emilio Vargas 7
ES – 28043 Madrid
Tel.: +34 91 566 83 14
Fax: +34 91 566 81 65
jfas@enresa.es
Mr Peter FAROSS
European Commission – DG TREN.H
- EUFO 04/270
LU – 2557 Luxembourg
Tel.: +352 4301 34342
Fax:
peter.faross@ec.europa.eu
540
Dr Andrea DOMINIJANNI
Politecnico di Torino
DITAG
Corso Duca degli Abruzzi, 24
IT – 10129 Torino
Tel.: +39 011 564 7705 -
Fax: +39 011 564 7699
andrea.dominijanni@polito.it
Mr Vit�zslav DUDA
Radioactive Waste Repository Authority (RAWRA)
Dlážd�ná 6
CZ – 110 00 Praha 1
Tel.: +420 221 421 511
Fax: +420 221 421 544
duda@rawra.cz
Dr Sigrid EECKHOUT
EIG EURIDICE ESV
Boeretang 200
BE – 2400 Mol
Tel.: +32 14 33 27 70
Fax: +32 14 32 37 09
seeckhou@sckcen.be
Prof. Dr Thomas FANGHÄNEL
Institute for Transuranium Elements (ITE)
Joint Research Centre (JRC)
Postfach 2340
DE – 76125 Karlsruhe
Tel.: +49 7247 951 483
Fax: +49 7247 951 591
thomas.fanghaenel@ec.europa.eu
Mr Sébastien FARIN
ANDRA
Parc de la Croix-Blanche – 1-7, rue Jean Monnet
FR – 92298 Châtenay-Malabry Cedex
Tel.:
Fax:
sebastien.farin@andra.fr
Mr Régis FARRET
INERIS
20, av. du Général de Gaulle
Parc Technologique Alata
FR – 60550 Verneuil-en-Halatte
Tel.: +33 3 44 55 61 27
Fax: +33 3 44 55 63 17
regis.farret@ineris.fr

LIST OF PARTICIPANTS
Dr Mario Piero FERRANDO
ENEA - National Laboratory for Radioactive Waste
Characterisation
Via Crescentino 41
IT – 13040 Saluggia (VC)
Tel.: +39 0161 483 542 -
Fax: +39 0161 483 381
mario.ferrando@saluggia.enea.it
Mrs Elizabeth (Betsy) FORINASH
OECD - Nuclear Energy Agency (NEA) – AEN/PR
Radioactive Waste Management
1, Square Théodore Judlin
FR – 75015 Paris
Tel.: +33 1 45 24 10 49
Fax: +33 1 44 30 61 11
elizabeth.forinash@oecd.org
Ms Bettina FRANKE
State Authority for Mining, Energy & Geology
Geozentrum Hannover
Stilleweg 2
DE – 30655 Hannover
Tel.: +49 511 643 24 18
Fax: +49 511 643 53-2418
bettina.franke@lbeg.niedersachsen.de
Dr Daniel GALSON
Galson Sciences Ltd
5, Grosvenor House – Melton Road
UK – Oakham, Rutland LE15 6AX
Tel.: +44 1572 770 649
Fax: +44 1572 770 650
dag@galson-sciences.co.uk
Mr Jose Luis GARCÍA SIÑERIZ
AITEMIN – Departamento de Ingenieria y Riesgos
Parque Leganés Tecnológico
ES – 28919 Leganés (Madrid)
Tel.: +34 91 442 49 55
Fax: +34 91 441 78 56
jlg.sineriz@aitemin.es
Dr Francesca GIACOBBO
Politecnico di Milano – Dip. Ingegneria Nucleare
Via Ponzio 34/3
IT – 20133 Milano
Tel.: +39 02 2399 6360
Fax:
francesca.giacobbo@polimi.it
541
Mr Jan FIDLER
Royal Institute of Technology (KTH)
Teknikringen 34
SE – 114 28 Stockholm
Tel.: +46 73 931 62 68
Fax: +46 8 790 50 34
fidler@kth.se
Mr Hans FORSSTRÖM
International Atomic Energy Agency (IAEA)
Nuclear Fuel Cycle and Waste Technology
Wagramer Strasse 5
AT – 1400 Vienna
Tel.: +43 1 2600 25670 - +43 699 1652 5670
Fax: +43 1 26007
h.forsstrom@iaea.org
Ms Julie FROMENT
Ambassade de France
8 B, Boulevard Joseph II
LU – 1840 Luxembourg
Tel.: +33 4 57 27 12 34
Fax:
julie.froment@diplomatie.gouv.fr
Mr Alexander GANETSKY
Crimean Scientific, Techn. & Economic Info Center
Technical and Economic Information Centre
Marshal Zhukov-street 31\51
UA – 95035 Simferopol, Crimea
Tel.:
Fax:
spirit@crimea.com
Ms Sorin-Silviu GHITA
S.N. Nuclearelectrica S.A.
Nuclear Safety Oversight Department
Polona Street 65 – Sector 1
RO – 010494 Bucharest 1
Tel.: +40 21 203 82 50 - +40 755 743 164
Fax: +40 21 316 29 97
sghita@nuclearelectrica.ro
Dr Jean-Paul GLATZ
EC Joint Research Centre
Institute for Transuranium Elements (ITU)
Postfach 2340
DE – 76125 Karlsruhe
Tel.: +49 7247 951 321
Fax: +49 7247 9519 9321
jean-paul.glatz@ec.europa.eu

Dr Paloma GÓMEZ GONZÁLEZ
CIEMAT – DMA - DAG - BIG
Avda. Complutense 22
ES – 28040 Madrid
Tel.: +34 91 346 61 85
Fax: +34 91 346 65 42
paloma.gomez@ciemat.es
Dipl. Ing Rheinhold GRAF
Gesellschaft für Nuklear-Service (GNS) mbH
Hollestrasse 7A
DE – 45127 Essen
Tel.: +49 201 109 1517
Fax: +49 201 109 1134
reinhold.graf@gns.de
Prof. Didier HAAS
EC – JRC ITU BXL – A/6
Rue de la Loi 200 – SDME 10/064
BE – 1049 Bruxelles
Tel.: +32 2 299 26 42
Fax: +32 2 295 01 46
didier.haas@ec.europa.eu
LIST OF PARTICIPANTS
Mrs Monica HAMMARSTRÖM
Swedish Nuclear Fuel & Waste Management Co.
(SKB)
Blekholmstorget 30
SE – 101 24 Stockholm
Tel.: +46 8 459 85 83
Fax: +46 8 661 57 19
monica.hammarstrom@skb.se
Mr Günther HAUPT
Siemens AG
Sachsenstrasse 8
DE – 91050 Erlangen
Tel.: +49 9131 186 379
Fax: +49 9131 181 515307
guenther.haupt@siemens.com
Mr Pavel HEKEL
National Nuclear Fund
Prievozska 30
SK – 821 05 Bratislava
Tel.: +421 2 5828 0421
Fax: +421 2 5828 0420
hekel@njf.sk
542
Dr Enrique GONZÁLEZ ROMERO
CIEMAT
Avda. Complutense 22
ES – 28040 Madrid
Tel.: +34 91 346 61 20 -
Fax: +34 91 346 65 76
enrique.gonzalez@ciemat.es
Mr Philippe GROS-GEAN
ONET Technologies
36, Boulevard des Océans
FR – 13009 Marseille
Tel.: +33 4 91 29 18 74
Fax: +33 4 91 29 18 58
ggoze@onet.fr
Dr Jörg HADERMANN
Paul Scherrer Institut (PSI)
Bündtenstrasse 19c
CH – 5417 Untersiggenthal
Tel.: - +41 56 288 14 68 (privé)
Fax:
joerg.hadermann@psi.ch
Dr Jon HARRINGTON
British Geological Survey (BGS)
Kingsley Durham Centre – Nicker Hill, Keyworth
UK – Nottingham NG12 5GG
Tel.: +44 1159 363 538
Fax: +44 1159 363 261
jfha@bgs.ac.uk
Mrs Vaclava HAVLOVÁ
Nuclear Research Institute Rež (NRI) plc
Dept. of Waste Disposal
Husinec-�ež, cp 130
CZ – 250 68 Rež
Tel.: +420 266 172 405
Fax: +420 266 172 086
hvl@ujv.cz
Mr Wolfgang HILDEN
European Commission – DG TREN/H/2.03
Rue Robert Stumper 10 – EUFO 04/288
LU – 2557 Luxembourg
Tel.: +352 4301 33546
Fax: +352 4301 30139
wolfgang.hilden@ec.europa.eu

Dr Miroslav HONTY
SCK-CEN
Boeretang 200
BE – 2400 Mol
Tel.: +32 14 33 32 32
Fax:
mhonty@sckcen.be
Dr Jana HOSKOVA
European Court of Auditors
Rue Alcide de Gasperi 12
LU – 1615 Luxembourg
Tel.: +352 4398 45985
Fax: +352 4398 46985
jana.hoskova@eca.europa.eu
Dr Stuart HUDSON
Scottish Executive – Environment Group
1-J, North, Victoria Quay
UK – Edinburgh EH6 6QQ
Tel.: +44 131 244 78120
Fax:
stuart.hudson@scotland.gsi.gov.uk
Mr George HUNTER
Scottish Environment Protection Agency
Erskine Court – Castle Business Park
UK – Stirling FK9 4TR
Tel.: +44 1786 452 577
Fax: +44 1786 446 885
george.hunter@sepa.org.uk
LIST OF PARTICIPANTS
Dr Wolfgang IRREK
Wuppertal Institut für Klima, Umwelt, Energie GmbH
- Research Grp. Energy, Transport & Climate Policy
Döppersberg 19
DE – 42103 Wuppertal
Tel.: +49 202 2492 164
Fax: +49 202 2492 198
wolfgang.irrek@wupperinst.org
Mr Benoît JAQUET
Comité local d'Information et de Suivi du lab. de
Bure (CLIS) – Préfecture de la Meuse
40, rue du Bourg
FR – 55000 Bar le Duc
Tel.: +33 3 29 77 55 40
Fax: +33 3 29 77 64 49
b.jaquet@clis-bure.com
543
Dr Alan J HOOPER
NDA – Radioactive Waste Management Directorate
Curie Avenue – Harwell
UK – Didcot, Oxon OX11 0RH
Tel.: +44 1235 825 401/451(secretary)
Fax: +44 1235 825 289 or 831 239
alan.hooper@nda.gov.uk; lindsay.hardy@nda.gov.uk
Dr Stephan HOTZEL
GRS mbH
Schwertnergasse 1
DE – 50667 Köln
Tel.: +49 221 2068 858
Fax:
stephan.hotzel@grs.de
Mr Michel HUGON
European Commission – DG RTD.J.2
Rue de la Loi 200 – CDMA 1/52
BE – 1049 Bruxelles
Tel.: +32 2 296 57 19
Fax: +32 2 295 49 91
michel.hugon@ec.europa.eu
Dr Victor IGNATIEV
Russian Research Center "Kurchatov Institute"
Kurchatov Square 1
RU – 123182 Moscow
Tel.: +7 499 196 71 30
Fax: +7 499 196 86 79
ignatiev@quest.net.kiae.su
Mr Richard IVENS
FORATOM – Institutional Relations
Rue Belliard 65
BE – 1040 Bruxelles
Tel.: +32 2 505 32 14
Fax: +32 2 502 39 02
richard.ivens@foratom.org
Mr Pekka JARVILEHTO
European Commission – DG ENV.C.5
Rue de la Loi 200 - BU-5 06/152
BE – 1049 Bruxelles
Tel.: +32 2 299 21 15 -
Fax:
pekka.jarvilehto@ec.europa.eu

Ms Aurélie JESTIN
SOGEDEC
Z.I. Digulleville
FR – 50100 Beaumont Hague
Tel.: +33 2 33 01 84 83
Fax: +33 2 33 01 84 99
ajestin@onet.fr
LIST OF PARTICIPANTS
Mr Lawrence H JOHNSON
NAGRA - National Cooperative for the Disposal of
Radioactive Waste
Hardstrasse 73
CH – 5430 Wettingen
Tel.: +41 56 437 12 34
Fax: +41 56 437 12 07
lawrence.johnson@nagra.ch
Dr Siegfried KELLER
BGR (Federal Institute for Geosciences & Natural
Resources)
Stilleweg 2
DE – 30655 Hannover
Tel.: +49 511 643 2397
Fax: +49 511 643 3694
s.keller@bgr.de
Ms Claire KERBOUL
Commission nationale d'Évaluation (CNE) des
Recherches sur la Gestion des Déchets radioactifs
Tour Mirabeau – 39-43, quai André Citroën
FR – 75015 Paris
Tel.: +33 1 40 58 89 05
Fax: +33 1 40 58 89 38
c.jouvance.cne@wanadoo.fr
Dr Thomas KIRCHNER
European Commission – DG TREN.H.2.002
Rue Henry Schnadt 1 – EUFO 04/389
LU – 2530 Luxembourg
Tel.: +352 4301 36732
Fax: +352 4301 30139
thomas.kirchner@ec.europa.eu
Dr Dietrich KOCH
S&B Industrial Minerals GmbH
Schmielenfieldstrasse 78
DE – 45772 Marl
Tel.: +49 2365 804 275
Fax: +49 2365 804 287
s.jakubik@ikominerals.com
544
Mr Hans JIVANDER
Östhammars Kommun – GMF
P.O. Box 66
SE – 742 21 Östhammar
Tel.: +46 17 386 000 - +46 70 692 63 78
Fax: +46 173 175 37
hans.jivander@osthammar.se
Mr Gun Hyo JUNG
FNC Technology Co. – Seoul National University
San 56-1 – RM.#312, BLDG.#135
KR – 151-742 Shilim-Dong, Gwanak-Gu
Tel.: +82 2 872 6411
Fax: +82 2 872 6089
ghjung@fnctech.com
Mr Paul KENNEDY
The Food Standards Agency
Room 415B, Aviation House – 125 Kingsway
UK – London WC2B 6NH
Tel.: +44 207 276 8703
Fax: +44 207 276 8788
paul.kennedy@foodstandards.gsi.gov.uk
Mr Suk Hoon KIM
FNC Technology Co. – Seoul National University
San 56-1 – RM.#312, BLDG.#135
KR – 151-742 Shilim-Dong, Gwanak-Gu
Tel.: +82 2 872 6411
Fax: +82 2 872 6089
kuni0808@fnctech.com
Dr Michael KNAACK
TÜV Nord e.v / ETS
Decommissioning & Waste Management
Grosse Bahnstrasse 31
DE – 22525 Hamburg
Tel.: +49 40 8557 2740
Fax: +49 40 8557 2429
mknaack@tuev-nord.de
Dr Siegfried KÖSTER
Bundesministerium für Wirtschaft u. Arbeit (BMWI)
Wirtschaft u. Technologie
Referat III B 3
DE – 53107 Bonn
Tel.: +49 228 615 2855 - +49 817 1986 3552
Fax: +49 228 615 3174
siegfried.koester@bmwi.bund.de

LIST OF PARTICIPANTS
Mr Vladislav KROŠELJ
Agency for Radioactive Waste Management (ARAO)
Parmova 53 -
SI – 1000 Ljubljana
Tel.: +386 1 236 32 10
Fax: +386 1 236 32 30
vladislav.kroselj@gov.si
Mr Philippe LALIEUX
ONDRAF/NIRAS – Long-Term Management System
Avenue des Arts 14
BE – 1210 Bruxelles
Tel.: +32 2 212 10 82
Fax: +32 2 218 51 65
p.lalieux@nirond.be
Mr Karel LEMMENS
SCK•CEN – Waste & Disposal Department
Boeretang 200
BE – 2400 Mol
Tel.: +32 14 33 31 38
Fax: +32 14 32 35 53
klemmens@sckcen.be
Mrs Xiang Ling LI
EIG EURIDICE ESV
Boeretang 200
BE – 2400 Mol
Tel.: +32 14 33 27 76
Fax: +32 14 32 12 79
xli@sckcen.be
Prof. Dr. Simon LÖW
ETH Zürich – Swiss Federal Institute of Technology
- Engineering Geology
Wolfgang Pauli Strasse 15
CH – 8093 Zürich
Tel.: +41 44 633 32 31 - +41 44 633 27 42 (Secr.)
Fax: +41 44 633 11 08
simon.loew@erdw.ethz.ch
Dr Alfredo LUCE
ENEA – CR Saluggia
Via Crescentino 41
IT – 13040 Saluggia (VC)
Tel.: +39 0161 483 342
Fax: +39 0161 483 381
alfredo.luce@saluggia.enea.it
545
Mr Thibaud LABALETTE
ANDRA
Parc de la Croix-Blanche – 1-7, rue Jean Monnet
FR – 92298 Châtenay-Malabry Cedex
Tel.: +33 1 46 11 83 25 - +33 1 46 11 81 26 (Secr.)
Fax: +33 1 46 11 82 23
thibaud.labalette@andra.fr
Mr Michael LECOMTE
CEA Valrhô - Marcoule – Nuclear Energy Division
B.P. 17171
FR – 30207 Bagnols-sur-Cèze Cedex
Tel.: +33 4 66 79 65 52
Fax:
michael.lecomte@cea.fr
Dr Frank LEMY
Federal Agency for Nuclear Control (FANC)
Rue Ravenstein 36
BE – 1000 Bruxelles
Tel.: +32 2 289 21 29 - +32 493 603 817
Fax: +32 2 289 21 12
frank.lemy@fanc.fgov.be
Dr Virpi LINDFORS
Östhammars Kommun – GMF
P.O. Box 66
SE – 742 21 Östhammar
Tel.: +46 70 98 00 417
Fax: +46 173 175 37
virpi.lindfors@osthammar.se
Dr Gustaf LÖWENHIELM
Swedish Radiation Safety Authority (SSM)
Solna Strandväg 96
SE – 171 16 Stockholm
Tel.: +46 8 799 40 00 - +46 70 209 18 45
Fax: +46 8 799 40 10
gustaf.lowenhielm@ssm.se
Mr Derek MACSON
11, Laurier Street
CA – Gatineau, Québec K0A 0R5
Tel.: +1 819 956 8234
Fax:
derek.macson@pwgsc.gc.ca

Dr Patrick MAJERUS
Ministry of Health
Villa Louvigny – Allée Marconi
LU – 2120 Luxembourg
Tel.: +352 2478 5670
Fax: +352 467 522
patrick.majerus@ms.etat.lu
Ms An MARIEN
SCK•CEN
Boeretang 200
BE – 2400 Mol
Tel.: +32 14 33 32 00
Fax:
amarien@sckcen.be
Dr Maria MARQUES FERNANDES
Paul Scherrer Institut (PSI)
Laboratory for Waste Management
OHLD/111
CH – 5232 Villigen PSI
Tel.: +41 56 310 23 10
Fax: +41 56 310 35 65
maria.marques@psi.ch
Mr Pedro Luis MARTÍN MARTÍN
CIEMAT – DMA - DAG - BIG
Avda. Complutense 22
ES – 28040 Madrid
Tel.: +34 91 346 61 42
Fax: +34 91 346 65 02
pedro.lmartin@ciemat.es
Mr Juan Carlos MAYOR
ENRESA – Division of Science & Technology
C/Emilio Vargas 7
ES – 28046 Madrid
Tel.: +34 91 566 82 17
Fax: +34 91 566 81 65
jmaz@enresa.es
Mr Bruce McKIRDY
Nuclear Decommissioning Authority (NDA)
Radioactive Waste Management Directorate
Curie Avenue – Harwell
UK – Didcot, Oxon OX11 0RH
Tel.: +44 1235 825 432 - +44 7901 516 433
Fax: +44 1235 825 593
bruce.mckirdy@nda.gov.uk
LIST OF PARTICIPANTS
546
Prof. Mario MARIANI
Politecnico di Milano
Via Ponzio 34/3
IT – 20133 Milano
Tel.: +39 02 2399 6385
Fax:
mario.mariani@polimi.it
Dr Jan MARIVOET
SCK•CEN – Waste and Disposal Department
Boeretang 200
BE – 2400 Mol
Tel.: +32 14 33 32 42
Fax: +32 14 32 35 53
jmarivoe@sckcen.be
Ms Sophie MARRION
SOGEDEC
BP 45 ZA Les Tomples
FR – 26701 Pierrelatte
Tel.: +33 4 75 96 24 01
Fax: +33 4 75 96 18 34
smarrion@onet.fr
Mr John MATHIESON
NDA – International Stakeholder Relations
Curie Avenue – Harwell
UK – Didcot, SN8 4HN
Tel.: +44 1235 826 797 - +44 7836 389 668
Fax: +44 1235 825 459
john.mathieson@nda.gov.uk
Dr Charles McCOMBIE
McCombie Consulting
Täfernstrasse 11
CH – 5405 Baden
Tel.: +41 62 871 69 73 - +41 79 239 74 86
Fax:
charles.mccombie@mccombie.ch
Dr Irena MELE
Agency for Radioactive Waste Management
(ARAO)
Parmova 53
SI – 1000 Ljubljana
Tel.: +386 1 236 32 26
Fax: +386 1 236 32 30
irena.mele@gov.si

LIST OF PARTICIPANTS
Dipl.-Ing. Michael MENTE
BGR (Federal Institute for Geosciences & Natural
Resources) – Geozentrum Hannover
Stilleweg 2
DE – 30655 Hannover
Tel.: +49 511 643 2246 - +49 179 971 6434
Fax: +49 511 643 3694
michael.mente@bgr.de
Dr Pauline MICHEL
Forschungszentrum Karlsruhe (FZK) GmbH
Hermann-von-Helmholz-Platz 1
DE – 76021 Karlsruhe
Tel.: +49 7247 824 306
Fax: +49 7247 823 927
pauline.michel@ine.fzk.de
Dr Giorgio MINGRONE
SOGIN SpA – Waste management engineering
Via Torino 6
IT – 00187 Roma
Tel.: +39 06 8304 0511
Fax: +39 06 8304 0911
mingrone@sogin.it
Dr Tiziana MISSANA
CIEMAT – DMA – DAG – BIG
Avda. Complutense 22
ES – 28040 Madrid
Tel.: +34 91 346 61 85 - +34 91 346 61 39
Fax: +34 91 346 65 42
tiziana.missana@ciemat.es
Dr Philippe MONTARNAL
CEA Saclay – DEN/DM2S
Bâtiment 454
FR – 91191 Gif-sur-Yvette Cedex
Tel.: +33 1 69 08 25 75 -
Fax: +33 1 69 08 52 42
philippe.montarnal@cea.fr
Mr Jim MORSE
NDA – Decommissioning & Clean Up Directorate
Herdus House – Westlake Science & Technology
Park
UK – Moor Row, Cumbria CA24 3HU
Tel.: +44 1925 802 265 - +44 7971 919 007
Fax: +44 1925 802 014
jim.morse@nda.gov.uk
547
Dr Gaston MESKENS
SCK•CEN
Boeretang 200
BE – 2400 Mol
Tel.: +32 14 33 21 67 - +32 476 422 635
Fax: +32 14 32 16 24
gaston.meskens@sckcen.be
Ms Jitka MIKŠOVÁ
Radioactive Waste Repository Authority (RAWRA)
Geological Repository Development
Dlážd�ná 6
CZ – 110 00 Praha 1
Tel.: +420 221 421 580 - +420 724 337 869
Fax: +420 221 421 544
miksova@rawra.cz
Mr Jean-Paul MINON
ONDRAF/NIRAS
Avenue des Arts 14
BE – 1210 Bruxelles
Tel.: +32 2 212 10 13
Fax: +32 2 218 51 65
jp.minon@nirond.be
Dr Jörg MÖNIG
GRS mbH – Long-term safety analyses Dept.
Theodor-Heuss-Strasse 4
DE – 38122 Braunschweig
Tel.: +49 531 80 12 202
Fax: +49 531 8012 211
joerg.moenig@grs.de
Dr Helge C. MOOG
GRS mbH
Theodor-Heuss-Strasse 4
DE – 31822 Braunschweig
Tel.: +49 531 8012 224
Fax: +49 531 8012 10224
helge.moog@grs.de
Ms Perrine MUGNIER
Ambassade de France
8 B, Boulevard Joseph II
LU – 1840 Luxembourg
Tel.: +33 4 57 27 12 34
Fax:
perrine.mugnier@diplomatie.gouv.fr

Mr Zoltán NAGY
PURAM – Department of Science & Technology
Puskas Tivadar Street 11
HU – 2040 Budaörs
Tel.: +36 75 519 536
Fax: +36 75 519 589
nagyzoltan@t-online.hu
LIST OF PARTICIPANTS
Dr Gheorghe NEGUT
National Agency for Radioactive Waste (ANDRAD)
Campului 1
RO – 115400 Mioveni-Arges
Tel.: +40 24 829 12 00 - +40 74 223 9633
Fax: +40 24 829 14 00
gheorghe.negut@andrad.ro
Mr Kaj NILSSON
Oskarshamn Municipality
P.O. Box 706
SE – 572 28 Oskarshamn
Tel.: +46 491 887 50 - +46 70 630 56 01
Fax: +46 491 889 88
kaj.nilsson@oskarshamn.se
Mr Sumio NIUNOYA
Obayashi Corporation – c/o NAGRA
Hardstrasse 73
CH – 5430 Wettingen
Tel.: +41 56 437 13 35
Fax: +41 56 437 13 17
obayashi@nagra.ch
Dr Ulrich NOSECK
GRS mbH
Theodor-Heuss-Strasse 4
DE – 38122 Braunschweig
Tel.: +49 531 8012 247
Fax: +49 531 8012 200
ulrich.noseck@grs.de
Mr Péter ORMAI
PURAM
Puskas Tivadar Street 11
HU – 2040 Budaörs
Tel.: +36 23 445 990 - +36 30 9892 414
Fax: +36 23 423 181
peter.ormai@rhk.hu; - ormaipeter@hotmail.com
548
Ms Christina NECHEVA
European Commission – DG TREN.H.2.003
Rue Robert Stumper 10 – EUFO 4/293
LU – 2557 Luxembourg
Tel.: +352 431 38800
Fax: +352 4301 30139
christina.necheva@ec.europa.eu
Ms Rodica NICOLAE
CITON – Technol. & Engineering for Nuclear Projects
Str. Atomistilor 409
RO – 077125 Bucharest-Magurele
Tel.: +40 21 404 60 53
Fax: +40 21 457 44 31
nicolaer@router.citon.ro
Dr Karl-Fredrik NILSSON
EC – JRC Petten – Institute for Energy
Westerduinweg 3
NL – 1755 ZG Petten
Tel.: +31 224 56 5420 - +31 651 389 771
Fax: +31 224 56 5641
karl-fredrik.nilsson@jrc.nl
Mr Sumio NIUNOYA
Obayashi Corp. – Nucl. Waste Technology Dept.
2-15-2 Konan, Minato-Ku
Shinagawa Intercity Tower B
JP – 108-8502 Tokyo
Tel.: +81 3 5769 1309
Fax: +81 3 5769 1975
niunoya.sumio@obayashi.co.jp
Prof. Yuri NOVIKOV
Petersburg Nuclear Physics Institute
Orlova Roscha 1
RU – 188300 St. Petersburg (Gatchina)
Tel.: +7 813 713 6167
Fax: +7 813 713 7196
novikov@pnpi.spb.ru
Dr Gérald OUZOUNIAN
ANDRA – Affaires internationales
Parc de la Croix-Blanche – 1-7, rue Jean Monnet
FR – 92298 Châtenay-Malabry Cedex
Tel.: +33 1 46 11 81 96
Fax: +33 1 46 11 82 68
gerald.ouzounian@andra.fr

Prof. Jaroslav PACOVSKÝ
Czech Technical University (CTU)
Centre of Experimental Geotechnics (CEG)
Thákurova 7
CZ – 166 29 Praha 6
Tel.: +420 224 354 302
Fax: +420 224 354 330
pacovsky@fsv.cvut.cz
Mr Ulli PALMER
Stoller Ingenieurtechnik GmbH
Bärensteinerstrasse 27-29
DE – 01277 Dresden
Tel.: +49 351 212 39 30
Fax:
info@stoller-dresden.de
Mr Eero PATRAKKA
Posiva Oy
Olkiluoto
FI – 27160 Eurajoki
Tel.: +358 2 8372 3700
Fax: +358 2 8372 3709
eero.patrakka@posiva.fi
Dr Michel PERDICAKIS
CNRS – Laboratoire de Chimie physique &
Microbiologie pour l'Environnement (LCPME)
405, rue de Vandoeuvre
FR – 54600 Villers-les-Nancy
Tel.: +33 3 83 68 52 23
Fax: +33 3 83 27 54 44
perdi@lcpme.cnrs-nancy.fr
Mr Angel PEREZ SAINZ
European Commission – DG RTD/J/1
Rue de la Loi 200 – CDMA 5/13
BE – 1049 Bruxelles
Tel.: +32 2 296 15 96
Fax: +32 2 299 52 43
angel.perez-sainz@ec.europa.eu
Mr Stig PETTERSSON
SKB – Repository Technology
Blekholmstorget 30
SE – 101 24 Stockholm
Tel.: +46 8 459 85 28 - +46 70 632 63 58
Fax: +46 8 579 386 11
stig.pettersson@skb.se
LIST OF PARTICIPANTS
549
Mrs Maria Isabel PAIVA
Instituto Tecnologico e Nuclear (ITN)
Dpt of Radiological Protection and Nuclear Safety
Estrada Nacional N 10 – Apartado 21
PT – 2686-953 Sacavém
Tel.: +351 21 994 63 18
Fax: +351 21 955 19 95
ipaiva@itn.pt
Ms Marjatta PALMU
Posiva Oy – Research
Olkiluoto
FI – 27160 Eurajoki
Tel.: +358 2 8372 3855 - +358 50 561 3427
Fax: +358 2 8372 3808
marjatta.palmu@posiva.fi
Ms Delphine PELLEGRINI
IRSN – DSU/SSD
31, av. de la Division Leclerc
FR – 92220 Fontenay-aux-Roses Cedex
Tel.: +33 1 58 35 74 53 -
Fax: +33 1 58 35 77 27
delphine.pellegrini@irsn.fr
Mr Antonio PEREIRA
Oskarshamn Municipality
P.O. Box 706
SE – 572 28 Oskarshamn
Tel.: +46 8 5537 86830
Fax: +46 8 5537 8601
antonio@physto.se
Dr Claudio PESCATORE
OECD – Nuclear Energy Agency (NEA) – AEN/PR
Radiation Protection & Waste Management
2, rue André-Pascal
FR – 75016 Paris Cedex 16
Tel.: +33 1 45 24 10 48
Fax: +33 1 45 24 11 45
claudio.pescatore@oecd.org
Dr Konstantin POTIRIADIS
Greek Atomic Energy Commission (GAEC)
Environmental Radioactivity
Patriarchou Grigoriou & Neapoleos 1
GR - 153 10 Athens (Aghia Paraskevi, Attikis)
Tel.: +30 210 650 6779
Fax: +30 210 650 6754
cpot@eeae.gr

Ms Katerina PTACKOVA
European Commission – DG RTD/J/2
Rue de la Loi 200 – CDMA 1/60
BE – 1049 Bruxelles
Tel.: +32 2 298 69 70
Fax: +32 2 295 49 91
katerina.ptackova@ec.europa.eu
Mr Tanguy RADOMME
SUEZ-TRACTEBEL S.A.
TRACTEBEL Engineering
Avenue Ariane 7
BE – 1200 Bruxelles
Tel.: +32 2 773 83 71
Fax:
tanguy.radomme@tractebel.com
Mr Rodolphe RAFFARD
ANDRA
Parc de la Croix-Blanche – 1-7, rue Jean Monnet
FR – 92298 Châtenay-Malabry Cedex
Tel.: +33 1 46 11 81 20
Fax:
rodolphe.raffard@andra.fr
Dr Rodney S. READ
RSRead Consulting Inc.
117, Sheep River Bay
CA – Okotoks, Alberta T1S IR3
Tel.: +1 403 938 2579 - +1 403 850 5754
Fax: +1 403 938 7680
rsread@rsrci.com
Dr Pascal REILLER
CEA Saclay – DEN/DANS/DPC/SECR
Bâtiment 391
FR – 91191 Gif-sur-Yvette Cedex
Tel.: +33 1 69 08 43 12 - +33 1 69 08 21 00
Fax: +33 1 69 08 54 11
pascal.reiller@cea.fr; reiller@azurite.cea.fr
Ms Lucie RIAD
Regionförbundet Uppsala Län
Kungsgatan 41
SE – 751 48 Uppsala
Tel.: +46 18 182 161
Fax: +46 18 182 115
lucie.riad@regionuppsala.se
LIST OF PARTICIPANTS
550
Mr Octavio QUINTANA TRIAS
European Commission – DG RTD/J
Rue de la Loi 200 – CDMA 5/106
BE – 1049 Bruxelles
Tel.: +32 2 298 93 30
Fax: +32 2 296 42 52
octavi.quintana-trias@ec.europa.eu
Ms Maria RADU
CITON – Radwaste R&D Program
Str. Atomistilor 409
RO – 077125 Bucharest-Magurele
Tel.: +40 21 404 60 27
Fax: +40 21 457 44 31
radum@router.citon.ro
Dr Michel RAYNAL
La Doline
FR – 26400 Beaufort-sur-Gervanne (Drôme)
Tel.: +33 4 75 76 41 07 - +32 473 916 386
Fax:
md.raynal@sfr.fr
Ms Hannah REED
ATKINS Ltd
240 Aztec West – Park Avenue
UK – Almondsbury, Bristol BS32 4SY
Tel.: +44 1454 288 128
Fax: +44 1454 616 480
hannah.reed@atkinsglobal.com
Ms Alexandra REY CUBERO
Univ. Politècnica de Catalunya (UPC)
Diagonal 647
ES – 08034 Barcelona
Tel.: +34 93 401 69 97
Fax:
sandra.rey@upc.edu
Dr John W. ROBERTS
University of Sheffield
Department of Engineering Materials
Portobello Street
UK – Sheffield S1 3JD
Tel.: +44 114 222 6028
Fax: +44 114 222 5943
j.w.roberts@sheffield.ac.uk

Mr Alvaro RODRÍGUEZ BECEIRO
ENRESA – Waste & Fuel Engineering
C/Emilio Vargas 7
ES – 28043 Madrid
Tel.: +34 91 566 82 07
Fax: +34 91 566 81 69
arob@enresa.es
Dr Vincenzo ROMANELLO
Forschungszentrum Karlsruhe (FZK) GmbH
Institut für Kern- und Energietechnik (IKET)
Hermann-von-Helmholz-Platz 1
DE – 76344 Eggenstein-Leopoldshafen
Tel.: +49 7247 823 406
Fax: +49 7247 823 824
vincenzo.romanello@iket.fzk.de
Mr Esko RUOKOLA
Radiation and Nuclear Safety Authority (STUK)
Waste Management Section
Laippatie 4/14
FI – 00881 Helsinki
Tel.: +358 9 7598 8305 - +358 40 533 80 20
Fax: +358 9 7598 8670
esko.ruokola@stuk.fi
Dr Laszlo SAGI
KFKI Atomic Energy Research Institute (AEKI)
Konkoly Thege 29-33
HU – 1121 Budapest
Tel.: +36 30 510 4774
Fax:
sagi@aeki.kfki.hu
Dr.rer.nat. Klaus SALZER
Institut für Gebirgsmechanik (IfG) GmbH
Frederikenstrasse 60
DE – 04279 Leipzig
Tel.: +49 341 336 00215
Fax: +49 341 336 00308
klaus.salzer@ifg-leipzig.de
LIST OF PARTICIPANTS
Dr Anastasia SAVIDOU
National Centre "Demokritos" (NCSR)
Inst. of Nuclear Technology – Radiation Protection
Patriarchou Grigoriou & Neapoleos 1
GR – 153 10 Athens (Aghia Paraskevi, Attikis)
Tel.: +30 210 650 38 77
Fax: +30 210 653 34 31
savidou@ipta.demokritos.gr
551
Prof. Klaus-Jürgen RÖHLIG
Technische Universität Clausthal (TUC)
Institut für Endlagerforschung
Adolph-Römer-Strasse 2a
DE – 38678 Clausthal-Zellerfeld
Tel.: +49 5323 724 920
Fax: +49 5323 722 810
klaus.roehlig@tu-clausthal.de
Dr André RÜBEL
GRS mbH
Theodor-Heuss-Strasse 4
DE – 38122 Braunschweig
Tel.: +49 531 8012 243
Fax: +49 531 8012 211
andre.ruebel@grs.de
Dr Thomas (Tom) RYAN
Radiological Protection Institute of Ireland (RPII)
Regulatory Services Division
3, Clonskeagh Square – Clonskeagh Road
IE – Dublin 14
Tel.: +353 1 269 7766
Fax: +353 1 269 7437
tryan@rpii.ie
Mr Peter SALZER
National Nuclear Fund
Prievozska 30
SK – 821 05 Bratislava
Tel.: +421 33 599 20 82
Fax:
salzer@njf.sk
Dr David SAVAGE
Quintessa Ltd.
32, St. Albans Avenue
UK – Bournemouth, BH8 9EE
Tel.: +44 1202 514 304
Fax: +44 1202 514 304
davidsavage@quintessa.org
Dr Thorsten SCHÄFER
Forschungszentrum Karlsruhe (FZK) GmbH
Institut für nukleare Entsorgung (INE)
Postfach 3640
DE – 76021 Karlsruhe
Tel.: +49 7247 825 494
Fax: +49 7247 823 927
thorsten.schaefer@ine.fzk.de

Prof. Christian SCHRÖDER
Université libre de Bruxelles (ULB) – B ATir
Av. Franklin D. Roosevelt 50 – CP 194/2
BE – 1050 Bruxelles
Tel.: +32 2 650 27 74
Fax: +32 2 650 27 43
christian.schroeder@ulb.ac.be
Ms Jantine SCHRÖDER
SCK•CEN
Boeretang 200
BE – 2400 Mol
Tel.: +32 14 33 21 71
Fax:
jschrode@sckcen.be
Dr Bernhard SCHWYN
NAGRA
Hardstrasse 73
CH – 5430 Wettingen
Tel.: +41 56 437 12 31
Fax: +41 56 437 13 17
bernhard.schwyn@nagra.ch
Ms Ellie SCOURSE
ATKINS Ltd
240 Aztec West – Park Avenue
UK – Almondsbury, Bristol BS32 4SY
Tel.: +44 1454 288 684 - +44 7811 200 984
Fax:
ellie.scourse@atkinsglobal.com
Mr Wolf K SEIDLER
27, rue Auguste Mounié
FR – 92160 Antony
Tel.: +33 1 46 74 52 71 - +33 6 22 51 48 55
Fax: +33 1 46 74 52 71
wolfseidler@aol.com
LIST OF PARTICIPANTS
Mr Manuel Lorenzo SENTIS
Swiss Federal Nuclear Safety Inspectorate (HSK)
CH - 5232 Villigen-HSK
Tel.: +41 56 310 39 93
Fax: +41 56 310 39 95
manuel.sentis@hsk.ch
552
Mr Christoph SCHRÖDER
European Commission – DG TREN.H.2
Rue Robert Stumper 10 – EUFO 04/381
LU – 2557 Luxembourg
Tel.: +352 4301 32118
Fax:
christoph.schroeder@ec.europa.eu
Dr Kristof SCHUSTER
BGR (Federal Institute for Geosciences & Natural
Resources)
Stilleweg 2
DE – 30631 Hannover
Tel.: +49 511 643 38 19
Fax: +49 511 643 28 68
kristof.schuster@bgr.de
Mr Edouard SCOTT-DE-MARTINVILLE
IRSN – Stratégie, Développement & Relations ext.
31, av. de la Division Leclerc
FR – 92262 Fontenay-aux-Roses Cedex
Tel.: +33 1 58 35 82 02
Fax: +33 1 58 35 85 09
edouard.scott-de-martinville@irsn.fr
Ms Cecelia SEIDLER
27, rue Auguste Mounié
FR – 92160 Antony
Tel.: +33 1 46 74 52 71
Fax: +33 1 46 74 52 71
ceceliaseidler@aol.com
Mr Patrik SELLIN
Swedish Nuclear Fuel & Waste Management Co.
(SKB)
Blekholmstorget 30
SE – 101 24 Stockholm
Tel.: +46 8 459 8522 - +46 70 668 0194
Fax: +46 8 579 38 611
patrik.sellin@skb.se
Mr Jean-Bernard SERVAJEAN
EDF Bruxelles
Avenue des Arts 21
BE – 1000 Bruxelles
Tel.: +32 2 289 61 46
Fax: +32 2 289 61 49
jean-bernard.servajean@edf.fr

Mr Mats SJÖBORG
Östhammars Kommun – GMF
Gjutargatan 6
SE – 742 34 Östhammar
Tel.: +46 70 630 44 51
Fax: +46 173 175 37
mats.sjoborg@liberal.se
Dr Alain SNEYERS
SCK•CEN – Waste & Disposal Unit
Boeretang 200
BE – 2400 Mol
Tel.: +32 14 33 31 37
Fax: +32 14 32 35 53
asneyers@sckcen.be
Ms Mila STRAHILOVA
European Court of Auditors
Rue Alcide de Gasperi 12
LU – 1615 Luxembourg
Tel.: +352 4398 47771
Fax: +352 4398 48771
mila.strahilova@eca.europa.eu
LIST OF PARTICIPANTS
Dr Dieter STÜHRENBERG
BGR (Federal Institute for Geosciences & Natural
Resources)
Stilleweg 2
DE – 30655 Hannover
Tel.: +49 511 643 28 77
Fax: +49 511 643 36 94
dieter.stuehrenberg@bgr.de
Ms Rosa SUREDA
Univ. Politècnica de Catalunya (UPC)
Diagonal 647
ES – 08034 Barcelona
Tel.: +34 6 549 31184
Fax:
rosa.maria.sureda@upc.edu
Dr István SZÜCS
MECSEKÉRC Ltd.
Esztergár L. U. 19
HU – 7633 Pécs
Tel.: +36 721 535 389 - +36 30 2672 712
Fax: +36 721 535 388
szucsistvan@mecsekerc.hu
553
Dr Jürgen SKRZYPPEK
Gesellschaft für Nuklear-Service (GNS) mbH
Hollestrasse 7A
DE – 45127 Essen
Tel.: +49 201 109 1480
Fax: +49 201 109 1134
juergen.skrzyppek@gns.de
Dr Mountaka SOULEY
INERIS – École des Mines
Parc de Saurupt
FR – 54042 Nancy
Tel.: +33 3 83 58 42 89 -
Fax: +33 3 83 53 38 49
mountaka.souley@ineris.fr
Dr Dankward STRUWE
Forschungszentrum Karlsruhe (FZK) GmbH
Institut für Reaktorsicherheit
Hermann-von-Helmholz-Platz 1
DE – 76021 Eggenstein-Leopoldshafen
Tel.: +49 7247 822 637 - 041/17724716 KFK D
Fax: +49 7247 823 718
struwe@irs.fzk.de
Mr Lennart SUNNERHOLM
Östhammars Kommun – GMF
Hallmansgatan 35
SE – 742 32 Östhammar
Tel.: +46 70 98 00 417
Fax: +46 173 175 37
virpi.lindfors@osthammar.se
Dr Ji�í SVOBODA
Czech Technical University (CTU)
Faculty of Civil Engineering
Thákurova 7
CZ – 166 29 Praha 6
Tel.: +420 601 236 444 - +420 607 102 650
Fax: +420 224 354 330
jiri.svoboda@seznam.cz
Mr Michael TICHAUER
IRSN – DSU/SSIAD
31, av. de la Division Leclerc
FR – 92262 Fontenay-aux-Roses Cedex
Tel.: +33 1 58 35 78 25
Fax: +33 1 58 35 79 76
michael.tichauer@irsn.fr

Dr Ján TIMUL'ÁK
DECOM Slovakia, spol. s.r.o.
Sibírska 1
SK – 917 01 Trnava
Tel.: +421 33 599 2075
Fax: +421 33 599 1645
timulak@decom.sk
Prof. Pierre TOULHOAT
INERIS
Parc Technologique Alata
FR – 60550 Verneuil-en-Halatte
Tel.: +33 3 44 55 68 45 - +33 6 11 79 60 13
Fax: +33 3 44 55 66 00
pierre.toulhoat@ineris.fr
Mr Julian TRICK
British Geological Survey (BGS)
Kingsley Durham Centre – Nicker Hill, Keyworth
UK – Nottingham NG12 5GG
Tel.: +44 1159 363 538
Fax:
jkt@bgs.ac.uk
Mr Daniele UGOLINI
EC – JRC Ispra
Via Enrico Fermi 1 – Office: 84 00/0218
IT – 21020 Ispra (VA)
Tel.: +39 0332 783 595
Fax: +39 0332 785 077
daniele.ugolini@ec.europa.eu
Dr Elie J J VALCKE
SCK•CEN – Waste and Disposal Expert Group
Boeretang 200
BE – 2400 Mol
Tel.: +32 14 33 31 33
Fax: +32 14 32 35 53
evalcke@sckcen.be
Mr Theofiel VAN RENTERGEM
SPF Economie, P.M.E., Classes moyennes et
Énergie – Division of Nuclear Applications
Boulevard du Roi Albert II, n° 16
BE – 1000 Bruxelles
Tel.: +32 2 277 64 52
Fax: +32 2 277 52 06
theofiel.vanrentergem@economie.fgov.be
LIST OF PARTICIPANTS
554
Dr Joseph (Joe) TOOLE
WorleyParsons
35, St. Ninians Road
UK – Stirling FK8 2HE
Tel.: +44 1786 477 324
Fax:
joe.toole@worleyparsons.com
Dr Öivind TOVERUD
Swedish Radiation Safety Authority (SSM)
Office of Nuclear Waste Safety
Solna Strandväg 96
SE – 171 16 Stockholm
Tel.: +46 8 698 84 53
Fax: +46 8 799 94010
oivind.toverud@ssm.se
Mr Francesco TROIANI
ENEA/RAD/LAB Saluggia – CR Casaccia
Via Crescentino 41
IT – 13040 Saluggia (VC)
Tel.: +39 0161 483 291
Fax: +39 0161 483 555
francesco.troiani@saluggia.enea.it
Dr Jan UHLI�
Nuclear Research Institute Rež (NRI) plc
Fluorine Chemistry Dept.
Husinec-�ež, cp 130
CZ – 250 68 Rež
Tel.: +420 266 173 548
Fax: +420 266 173 531
uhl@ujv.cz
Dr Luc VAN DEN DURPEL
LISTO BVBA
Groenstraat 35
BE – 9250 Waasmunster
Tel.: +32 3 296 70 18 - +32 473 86 56 47
Fax: +32 3 296 71 06
vddurpel@listo.be; vddurpel@pandora.be
Dr Radek VAŠÍ�EK
Czech Technical University (CTU) – CEG
Thákurova 7
CZ – 166 29 Praha 6
Tel.: +420 224 354 918
Fax: +420 224 354 330
radek.vasicek@fsv.cvut.cz

Mr Paolo VENTURONI
ANSALDO Finmeccanica
Avenue des Arts 21
BE – 1000 Bruxelles
Tel.: +32 2 286 80 00
Fax: +32 2 286 80 19
bruxoffice@finmeccanica.org
Mr Jan VERSTRICHT
EIG EURIDICE ESV – c/o SCK-CEN
Boeretang 200
BE – 2400 Mol
Tel.: +32 14 33 27 86
Fax: +32 14 32 12 79
jan.verstricht@sckcen.be
LIST OF PARTICIPANTS
Dr Holger VÖLZKE
Federal Institute for Materials Research & Testing
(BAM)
Unter den Eichen 87
DE – 12205 Berlin
Tel.: +49 30 8104 1334
Fax: +49 30 8104 1337
holger.voelzke@bam.de
Mrs Anna VOSECKOVA
Czech Liaison Office for R&D
Rue du Trône 98
BE – 1050 Bruxelles
Tel.: +32 2 514 66 72
Fax:
voseckova@tc.cz
Mr Simon WEBSTER
European Commission – DG RTD/J/2
Rue de la Loi 200 – CDMA 1/70
BE – 1049 Bruxelles
Tel.: +32 2 299 04 42
Fax: +32 2 295 49 91
simon.webster@ec.europa.eu
Mr Robert WIEGERS
IBR Consult B.V.
De Giesel 14
NL – 6081 PH Haelen
Tel.: +31 475 59 55 64
Fax: +31 475 59 11 17
r.wiegers@ibrconsult.nl
555
Dr Ewoud VERHOEF
COVRA n.v.
Spanjeweg 1
NL – 4380 AE Vlissingen
Tel.: +31 113 616 670
Fax: +31 113 616 650
ewoud.verhoef@covra.nl
Dr Maria Victoria VILLAR GALICIA
CIEMAT
Dept° de Impacto Ambiental de la Energía
Avda. Complutense 22
ES – 28040 Madrid
Tel.: +34 91 346 63 91
Fax: +34 91 346 65 42
mv.villar@ciemat.es
Dr Dusan VOPALKA
Czech Technical University (CTU)
B�ehová 7
CZ – 115 19 Praha 1
Tel.: +420 224 358 206
Fax: +420 222 320 861
vopalka@fjfi.cvut.cz
Dr Jan Richard WEBER
BGR (Federal Institute for Geosciences & Natural
Resources) – Section B2.3: Long-term Safety
Stilleweg 2
DE – 30655 Hannover
Tel.: +49 511 643 2438 -
Fax: +49 511 643 3694
jan.weber@bgr.de
Mr Klaus WIECZOREK
GRS mbH – Final Repository Safety Research Division
Theodor-Heuss-Strasse 4
DE – 38122 Braunschweig
Tel.: +49 531 8012 229
Fax: +49 531 8012 211
klaus.wieczorek@grs.de
Dr Peter WIKBERG
Swedish Nuclear Fuel & Waste Management Co.
(SKB)
Blekholmstorget 30
SE – 101 24 Stockholm
Tel.: +46 8 459 84 63 - +46 70 646 51 07
Fax:
peter.wikberg@skb.se

Dr. Eng. Shuichi YAMAMOTO
Obayashi Corporation – LLW Project
2-15-2 Konan, Minato-Ku – Shinagawa Intercity
Tower B
JP – 108-8502 Tokyo
Tel.: +81 3 5769 1860
Fax: +81 3 5769 1960
yamamoto.shuichi@obayashi.co.jp
Mr Björn ZENNER
S&B Industrial Minerals GmbH
Schmielenfieldstrasse 78
DE – 45772 Marl
Tel.: +49 2365 804 275
Fax: +49 2365 804 287
b.zenner@ikominerals.com
LIST OF PARTICIPANTS
556
Ms Nadja ŽELEZNIK
Agency for Radioactive Waste Management
(ARAO)
Parmova 53
SI – 1000 Ljubljana
Tel.: +386 1 236 32 15 - +386 31 363 477
Fax: +386 1 236 32 30
nadja.zeleznik@gov.si
Dr Piet ZUIDEMA
NAGRA – Disposal of Radioactive Waste
Hardstrasse 73
CH – 5430 Wettingen
Tel.: +41 56 437 12 87
Fax: +41 56 437 13 17
piet.zuidema@nagra.ch

European Commission
EUR 24040 Euradwaste '08 - Seventh European Commission Conference on the Management and
Disposal of Radioactive Waste - Community Policy & Research and Training Activities
Luxembourg: Publications Office of the European Union
2009 — 570 pp. — 17.6 x 25.0 cm
ISBN 978-92-79-13105-9
doi: 10.2777/46864

Publications for sale:
• via EU Bookshop (http://bookshop.europa.eu);
How to obtain EU publications
• from your bookseller by quoting the title, publisher and/or ISBN number;
• by contacting one of our sales agents directly. You can obtain their contact details on the Internet
(http://bookshop.europa.eu) or by sending a fax to +352 2929-42758.
Free publications:
• via EU Bookshop (http://bookshop.europa.eu);
• at the European Commission’s representations or delegations. You can obtain their contact details
on the Internet (http://ec.europa.eu) or by sending a fax to +352 2929-42758.