14.09.2014 Views

exotic nuclei structure and reaction noyaux exotiques ... - IPN - IN2P3

exotic nuclei structure and reaction noyaux exotiques ... - IPN - IN2P3

exotic nuclei structure and reaction noyaux exotiques ... - IPN - IN2P3

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

MURE: MCNP Utilities for Reactor Evolution<br />

<strong>IPN</strong>O Participation: J. Brizi, S. David, O. Méplan, J. N. Wilson<br />

Collaboration : LPSC Grenoble<br />

L’équipe MURE de l’<strong>IPN</strong> travaille sur la simulation de réacteurs nucléaires innovants et sur les scénarios<br />

associés. Les réacteurs rapides sont sans doute amenés à jouer un rôle dans le futur, du fait de leur capacité<br />

d’une part à atteindre la régénération et réduire ainsi d’un facteur 100 la consommation de minerai d’uranium,<br />

et d’autre part à recycler les actinides mineurs et réduire ainsi la charge thermique et radiotoxique<br />

des déchets produits. Dans ce cadre, nous effectuons des études sur les réacteurs rapides refroidis au sodium,<br />

en envisageant par exemple l’utilisation d’un combustible au thorium dans divers scénarios innovants.<br />

Introduction<br />

Currently there is a renewed interest in the expansion<br />

of the role of nuclear power for energy generation<br />

due to its lack of CO 2 emissions (which<br />

contribute to global warming), its economic viability,<br />

<strong>and</strong> its potential for energy security <strong>and</strong> independence.<br />

The future of nuclear energy may require breeding<br />

(reduction of uranium consumption) <strong>and</strong> optimized<br />

waste management. Innovative technologies have<br />

to be explored in order to reduce considerably the<br />

ore consumption <strong>and</strong> the associated waste production.<br />

Sodium-cooled fast reactors seem to be the<br />

most achievable technology in the coming decades,<br />

<strong>and</strong> can play an important role to launch the<br />

generation 4 technologies. However, the st<strong>and</strong>ard<br />

sodium-cooled reactors face to the problem of the<br />

positive void reactivity <strong>and</strong> a major Minor Actinides<br />

(MA) production if transmutation is not considered.<br />

In this context, we perform neutronic studies on<br />

innovative (or evolutive) sodium-cooled reactors.<br />

These studies are based on MURE (MCNP Utility<br />

for Reactor Evolution), a C++ object-oriented evolution<br />

code that couples the Monte-Carlo transport<br />

code MCNP with a fuel depletion code under given<br />

conditions. Different configurations of a fast sodium<br />

cooled reactor (SFR, with a total power of<br />

1.45 GWe) are investigated, using U/Pu or Th/U<br />

cycle, with or without minor actinides transmutation.<br />

A self-breeding Th/U configuration has been<br />

found, using thorium blankets, which allows reducing<br />

significantly sodium void reactivity.<br />

MURE<br />

Numerical studies of a reactor's neutronics can be<br />

based on deterministic codes or Monte Carlo<br />

codes. The advantage of the former ones is that<br />

they are very fast <strong>and</strong> accurate but only for well<br />

known systems such as conventional UOx PWR<br />

(Pressuriezed Water Reactor) or U/PU FNR (Fast<br />

Neutron Reactor). The study of innovative reactors<br />

required the used of accurate numerical codes,<br />

independent of models, using Monte Carlo techniques<br />

which depend only on the nuclear data li-<br />

braries (cross-sections, ...) <strong>and</strong> allow full 3D description<br />

of the exact geometry. Monte-Carlo codes<br />

require more CPU time for calculations (but are not<br />

prohibitively long) <strong>and</strong> are the only reference for<br />

innovative reactors or fuel cycles. We have developed<br />

a number of tools to couple a Monte Carlo<br />

transport code for neutronics, MCNP with fuel<br />

burn-up, called MURE . This work has taken place<br />

in collaboration with LPSC Grenoble, Subatech<br />

Nantes <strong>and</strong> NRI Czech republic. MURE is written<br />

in an oriented object C++, allowing greater portability<br />

<strong>and</strong> conviviality.<br />

The main part of MURE concerns the burn-up calculation.<br />

It consists of a set of MCNP calculations<br />

to obtains <strong>reaction</strong> rates for given reactor material<br />

composition, followed by a numerical (Runge-<br />

Kutta) integration of the Bateman equations governing<br />

the reactor fuel evolution (fissile burn-up,<br />

appearance of fission products <strong>and</strong> minor actinides)<br />

using MCNP results <strong>and</strong> providing the new<br />

composition for the next MCNP run.<br />

Special attention has been paid to the definition<br />

of evolution conditions: power profile over the<br />

evolution, reactivity control by either neutron poisons<br />

or control rods, refueling capability, ... After<br />

the evolution a graphical interface based on ROOT<br />

allows a global view of results as well as offers the<br />

possibility to make post treatment (radiotoxicity,<br />

heat, ...).<br />

A new set of tools has been recently added in order<br />

to take into account the fuel evolution during<br />

the cooling-down period after irradiation, reprocessing<br />

<strong>and</strong> re-fabrication. This allows to quantify<br />

more precisely the waste production <strong>and</strong> the inventory<br />

of fissile materials. Moreover, the decay of<br />

241 Pu (period 14,4 years) outside the reactor is now<br />

taken into account, <strong>and</strong> the impact of the reprocessing<br />

on the neutronics <strong>and</strong> on the quantity of<br />

241 Am in waste can be precisely quantified.<br />

Waste comparison<br />

The fast reactor studied is the sodium fast reactor<br />

117

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!