Potential of energy efficiency measures in the world steel industry

University of Groningen
CIO, Center for Isotope Research
IVEM, Center for Energy and Environmental Studies
Master Programme Energy and Environmental Sciences
Potential of energy efficiency measures
in the world steel industry
Tjebbe Galama
EES 2012-160 M

Master report of Tjebbe Galama
Supervised by: Tengfang Xu (Lawrence Berkeley National Laboratory)
Prof.dr. H.C. Moll (IVEM)
Dr. C. Visser (IVEM)
University of Groningen
CIO, Center for Isotope Research
IVEM, Center for Energy and Environmental Studies
Nijenborgh 4
9747 AG Groningen
The Netherlands
http://www.rug.nl/fmns-research/cio
http://www.rug.nl/fmns-research/ivem

CONTENTS
1 Summary ................................................................................................................................. 3
2 Samenvatting ........................................................................................................................... 5
3 General Acronyms ....................................................................................................................7
4 Introduction ............................................................................................................................ 9
5 Research scope ....................................................................................................................... 11
5.1 Background ..................................................................................................................... 11
5.2 Research Aim .................................................................................................................. 11
5.3 Research questions.......................................................................................................... 12
5.4 System definition and borders ........................................................................................ 13
5.5 Methodology and data sources ....................................................................................... 16
6 Steel production ..................................................................................................................... 21
7 Steel energy efficiency measures ........................................................................................... 25
8 World steel production and different regions ....................................................................... 29
8.1 Steel production ............................................................................................................. 29
8.2 Economics ...................................................................................................................... 32
9 Results ................................................................................................................................... 35
9.1 Potential of energy efficiency measures in the US steel industry in 2006 .................... 35
9.1.1 Energy savings ............................................................................................................ 38
9.1.2 Carbon mitigation ................................................................................................... 40
9.2 Further results on cost curves for the US steel industry ................................................. 41
9.2.1 Industry structure change ....................................................................................... 42
9.2.2 Cost curve for carbon mitigation ............................................................................ 42
9.2.3 Modified cost of conserved energy ......................................................................... 44
9.2.4 Including or excluding other non-energy benefits ................................................. 47
9.2.5 Changes in discount rate ......................................................................................... 47
9.2.6 Summary of further conclusions in the US steel industry. ..................................... 48
9.3 Results of energy efficiency measures in other world regions ....................................... 49
9.3.1 Production of steel in different world regions ........................................................ 49
9.3.2 Economics of different world regions ...................................................................... 51
9.3.3 Further results ........................................................................................................ 55
9.4 World steel energy savings potential .............................................................................. 57

9.4.1 Energy savings potential .......................................................................................... 57
9.4.2 Comparison to other analysis ................................................................................. 59
9.5 Key assumptions and gaps in knowledge ....................................................................... 60
10 Conclusions ........................................................................................................................ 63
11 Discussion .............................................................................................................................. 65
12 References ............................................................................................................................. 67
Appendix A. Overall Measures ...................................................................................................... 71
Appendix B. Iron Ore Preparation ................................................................................................ 72
Appendix C. Iron Making - Blast Furnace .................................................................................... 76
Appendix D. Iron Making – Basic Oxygen Furnace ..................................................................... 79
Appendix E. Secondary Steelmaking - Electric Arc Furnace (EAF) ............................................. 80
Appendix F. Casting ...................................................................................................................... 86
Appendix G. Hot Rolling ............................................................................................................... 87
Appendix H. Cold Rolling and Finishing ....................................................................................... 91
Appendix I. References ................................................................................................................. 92
Appendix J. Calculation example ................................................................................................ 101
Appendix K. Results of energy efficiency calculation for the US 2006 ...................................... 102

1 SUMMARY
The world steel industry plays a major role in energy use and Greenhouse Gas (GHG) emissions
now and in the future. Implementing energy efficiency measures is among one of the most costeffective
investments that the industry could make in improving efficiency and reducing GHG
emissions. The goal of this thesis was to analyse the potential of energy efficiency measures in
the world steel industry. Through characterizing energy-efficiency technology costs and
improvement potentials, energy savings and carbon mitigation have been developed and
presented for energy efficiency measures for the world steel industry in 2006.
In order to properly analyse the total world steel industry, specific regions have been identified;
the United States (US), Western Europe (WEU), Former Sovjet Union (FSU), India (IND),
China (CHI), Japan (JAP) and Central- and South America (CSA). These regions together are
responsible for over 80% of the world steel production.
First, an in depth study of 72 identified energy efficiency measures in the US steel industry in
2006 was conducted. Data for energy efficiency measure was collected from case studies,
scientific reports, energy audits, etc. Measures were evaluated on cost-effectiveness by
calculating Cost of Conserved Energy (CCE) and total energy savings.
Once each measure was characterized individually, its applicability to the US iron and steel
industry as a whole was assessed. The total of energy savings calculated for each measure
represents the total potential for the US. Next, the potential of only cost-effective measures was
calculated by adding only the energy savings potential of measures which were identified as
cost-effective by CCE calculation. From the energy savings, the carbon mitigation was
calculated. In summary, implementing all energy efficiency measures in the US steel industry
showed major reduction potentials for both the energy use (385 PJ) and carbon emissions (7.2
MtC). Even if only cost-effective measures are taken into account, the energy and carbon
reductions add up to 300 PJ and 6.2 MtC, respectively (equals 26% and 22% of total US steel
industry energy use and carbon emissions).
The assessment of energy efficiency measures in the US lead to a number of assumptions on key
factors influencing the energy efficiency measures for the steel industry in different world
regions. From the assessment in the US, it was clear that the energy use, production, industry
structure and average weighted energy price were the most important factors influencing the
potential of (cost-effective) energy efficiency measures. With these key factors, the analysis was
expanded to the other identified world regions. For these world regions estimations were made
of the effects of energy efficiency measures in in their steel industry, using the identified key
factors and an adaptation of costs data by the PPP-index. The results of these estimations are
provided in Table 1.
3

Table 1 Summary of results of energy savings and carbon reductions in steel industry for different world
regions in 2006
Region
Total final energy savings
(PJ/year)
% of total final energy
consumption
Total cost-effective energy
savings (PJ/year)
% of total final energy
consumption
Total carbon reductions
(MtC/year)
4
United
States
Western
European
Union
Former
Soviet
Union
India China Japan
Central
and
South
America
385 756 620 215 2170 548 200
33% 27% 19% 14% 23% 28% 16%
300 568 421 166 1513 402 148
26% 20% 13% 11% 16% 21% 12%
8 14 12 5 38 11 5
% of total carbon emissions 27% 21% 16% 13% 18% 24% 15%
Total cost-effective carbon
reductions (MtC/year)
6 11 9 4 28 8 4
% of total carbon emissions 22% 17% 12% 10% 13% 18% 12%
The results of the in depth analysis of the US, the estimations for the selected world regions and
an assumption for the remaining other world regions, provided an estimation of the potential
for the world steel industry. Apparently, the application of energy efficiency measures all over
the world resulted in a total potential energy savings of about 5.7 EJ in 2006 (5.3-6.0 EJ when
uncertainties are included). The cost-effective energy savings for the world steel industry are
about 4.1 EJ in 2006 (3.5-5.4 EJ, uncertainties included). The total potential for energy
efficiency measures is about 23% of the world steel final energy consumption. The cost-effective
energy savings add up to about 16% of the final energy consumption.
For carbon mitigation the application of energy efficiency measures all over the world resulted
in a total potential carbon reduction of about 107 MtC in 2006 (101-113 MtC, uncertainties
included). The cost-effective carbon reductions for the world steel industry are about 82 MtC in
2006 (76-101 MtC, uncertainties included). The total potential for energy efficiency measures is
about 20% of the world steel carbon emission mitigation. The cost-effective carbon reductions
add up to about 15% of the total carbon emissions.
In short, the application of energy efficiency measures in the world steel industry offers a major
potential for final energy savings and carbon emission reductions.

2 SAMENVATTING
Voor staalindustriën over de hele wereld staat het reduceren van broeikasgassen hoog in het
vaandel. Een van de meest kostenefficiente manieren om de uitstoot van broeikasgassen te
verminderen, is het gebruik van energie-efficiencymaatregelen. Het doel van deze scriptie is om
een analyse te doen van de energie-efficiencymaatregelen in de wereld staalindustrie. Door de
energie efficiency te karakteriseren op zowel kosten als energiebesparingen, konden de totale
energiebesparing en reductie in koolstofuitstoot berekend worden voor de wereld staalindustrie
in het jaar 2006.
Om een degelijke benadering van de totale wereld staalindustrie te kunnen geven, zijn een
aantal specefieke werelddelen gekozen die in totaal meer dan 80% van de staalproductie
vertegenwoordigen. De geselecteerde werelddelen zijn: Verenigde Staten, West Europa,
Voormalige Sovjet Unie, India, China, Japan en Centraal- en Zuid-Amerika.
Voor dit onderzoek is eerst een analyse gemaakt op 72 geïndentificeerde energieefficiencymaatregelen
voor de staalindustrie in de Verenigde Staten (VS). Data voor deze
maatregelen werd verzameld uit wetenschappelijke tijdschriften, case studies, verslagen van
energie controles, etc. De maatregelen werden beoordeeld op hun kosteneffectiviteit door
middel van Kosten per Bespaarde Energie (KBE) en hun potentiële energiebesparing.
Nadat de data voor alle maatregelen individueel was vastgesteld, is gekeken naar het effect van
deze maatregelen, als zij toegepast worden in de staalindustrie van de VS. De totale
energiebesparing van alle maatregelen voor de Amerikaanse staalindustrie werd op die manier
berekend. Daarnaast werd met behulp van de KBE bepaald welke maatregelen kostenefficienct
zijn en werd ook de totale besparing van deze maatregelen bepaald. De resultaten voor de VS
samengevat: door toepassing van alle gevonden maatregelen is er 385 PJ aan energiebesparing
mogelijk en zal 7.2 Mt aan koolstofemissies worden gereduceerd. Als alleen wordt gekeken naar
maatregelen die kostenefficient zijn, is er 300 PJ aan energiebesparing mogelijk en zal er 6.2 Mt
aan koolstofemissies worden gereduceerd (dit is respectievelijk 26% en 22% van het totale
energieverbruik en koolstofemissie in de staalindustrie in de VS).
Daarnaast zijn er bij de analyse voor de staalindustrie in de VS een aantal essentiële factoren
voor energie-efficiencymaatregelen bepaald. Deze essentiële facotren werden gebruikt om de
effecten van de maatregelen in andere werelddelen te bepalen. De essentiële factoren voor de
analyse van energie-efficiencymaatregelen voor de verschillende werelddelen zijn: het
energieverbruik, productie, structuur van de industrie en een gemiddelde gewogen energieprijs.
Samen met een aanpassing voor de kosten van elke methode, met de zogenaamde PPP-index, is
met deze essentiële factoren een schatting gemaakt van de effecten van energieefficiencymaatregelen
in de ander werelddelen. De resultaten hiervan zijn getoond in tabel 1.
5

Tabel 1. Resultaten voor energie-efficiencymaatregelen in de staalindustrie van verschillende werelddelen
in 2006.
Werelddeel
Totale uiteindelijke
energiebesparing (PJ/jaar)
% van het totale uiteindelijke
energieverbruik
Totale kosteneffectieve
energiebesparing (PJ/jaar)
% van het totale uiteindelijke
energieverbruik
Totale koolstofuitstoot
vermindering (Mt/jaar)
6
Verenigde
Staten
West
Europa
Voormalige
Sovjet Unie
India China Japan
Centraal-
en Zuid-
Amerika
385 756 620 215 2170 548 200
33% 27% 19% 14% 23% 28% 16%
300 568 421 166 1513 402 148
26% 20% 13% 11% 16% 21% 12%
8 14 12 5 38 11 5
% van totale koolstofuitstoot 27% 21% 16% 13% 18% 24% 15%
Total kosteneffectieve
koolstofuitstoot
vermindering (Mt/jaar)
6 11 9 4 28 8 4
% van totale koolstofuitstoot 22% 17% 12% 10% 13% 18% 12%
Met de bevindingen voor de staalindustrie in de VS, de schattingen voor de andere werelddelen
en aanname voor de overgebleven werelddelen, kon een totaal potentieel voor energieefficiencymaatregelen
voor de wereld staalindustrie bepaald worden. De totale energiebesparing
van alle gevonden maatregelen over de hele wereld geven een besparing van ongeveer 5,7 EJ in
2006 (met een onzekerheid van 5,3 tot 6,0 EJ). Als alleen kosteneffectieve maatregelen geteld
worden is de totale besparing 4,1 EJ (met een onzekerheid van 3,5 tot 5,4 EJ). De totale
uiteindelijke energiebesparing voor de wereld staalindustrie komt neer op 23% besparing
wanneer alle maatregelen beoordeeld worden en 16% besparing wanneer alleen kosteneffectieve
maatregelen beoordeeld worden.
De totale reductie van koolstofemissie, met alle gevonden maatregelen in de wereld
staalindustrie in 2006, is 107 Mt koolstof (met een onzekerheid van 101 tot 113 Mt). Als alleen
kosteneffectieve maatregelen gerekend worden is de totale koolstofreductie 82 Mt (met een
onzekerhied 76 tot 101 Mt). De totale koolstofemissiereductie voor de wereldstaal industrie
komt neer op 20% besparing wanneer alle maatregelen beoordeeld worden en 15% besparing
wanneer alleen kosteneffectieve maatregelen beoordeeld worden.
Kort samengevat zijn er zeer grote besparingen, voor zowel energieverbruik als koolstofemissies,
mogelijk door de toepassing van energie-efficiencymaatregelen.

3 GENERAL ACRONYMS
American Iron and Steel Institute (AISI)
Basic Oxygen Furnace (BOF)
Blast furnace (BF)
Bureau of Economic Analysis (BEA).
Carbon dioxide (CO2)
Cost of carbon reduction (CCR)
Cost of conserved energy (CCE)
Crude Steel (CS)
Department of Energy (DOE)
Electric arc furnace (EAF)
Energy Information Administration (EIA)
Energy-climate (EC) models
Environmental Protection Agency (EPA)
Greenhouse gas (GHG)
Gross Domestic Product (GDP)
Integrated assessment models (IAM)
Intergovernmental Panel on Climate Change (IPCC)
International Energy Agency (IEA)
International Iron and Steel Institute (IISI)
Lawrence Berkeley National Laboratory (LBNL)
Manufacturing Energy Consumption Survey (MECS)
Market Exchange Rate (MER)
Modified cost of carbon reduction (MCCR)
Modified cost of conserved energy (MCCE)
Operation and maintenance (O&M)
Purchasing Power Parity (PPP)
Units
Annual carbon savings (tC/yr)
Annual change in O&M costs ($/yr)
Annual energy savings (GJ/yr)
Capital cost ($)
Capital recovery factor (yr -1 )
7

Cost of carbon reduction (CCR, $/MtC)
Cost of conserved energy (CCE, $/GJ)
Energy (in petajoules -PJ, or in gigajoules - GJ)
Energy savings per production (GJ/Tonne)
Lifetime of the mitigation option (years)
Million metric ton, or million tonne (Mt)
Million tonne of carbon (MtC)
Modified cost of carbon reduction (MCCR, $/MtC)
Modified cost of conserved energy (MCCE, $/GJ)
Tonne of hot metal (thm)
8

4 INTRODUCTION
This master thesis is the final assignment for a master degree in energy and environmental
science. A major part of this research has been performed at the Lawrence Berkeley National
Laboratory (LBNL) in Berkeley, California, as a challenging scholar research. Under the
supervision of Tengfang Xu, a database of energy efficiency measures in the US steel industry
was constructed. The application of this database has provided us with useful results on energy
efficiency and carbon mitigation in the US steel industry which in turn can be used to enhance
integrated assessment models on climate change for example. Worrell et al. (2002) and Xu et al.
(2010) have already included extensive work on energy efficiency measures for the US steel
industry in their reports. These results have been updated during my research at LBNL. The
results of this research have been reported in ‘Bottom-up Representations of Energy Efficiency
Technologies as Mitigation Measures for the US Iron and Steel Sector from 1994 to 2010’ by Xu,
Galama and Sathaye (2012).
The master thesis itself will be aimed at estimating the effect of energy efficiency measures in
major steel industries around the world. In order to account for the world steel industry this
thesis will have to extend the database available for the US, to other world regions. This is the
main second challenge of this research: how to include the rest of the world with the data
available for the US. The thesis will be supervised at the University of Groningen by prof. dr.
H.C. Moll.
Steel has become one of most commonly produced materials used in a wide variety of
application worldwide, for example in buildings, automobiles, machines or cans (sealed
containers). The world production of the iron and steel sector was about 1.5 billion metric tons
of crude steel in 2011 (Worldsteel, 2012). Steel is a commodity product produced and
transported worldwide.
The production of materials require a vast amount of energy. One third of the world’s energy
consumption and 36% of the carbon dioxide (CO2) emissions are attributed to manufacturing
industries. Steel production requires a number of energy intensive processes. (Xu, 2010) The
iron and steel industry accounts for about 19% of final energy use (Figure 1; 21.4 EJ) and about
one quarter of the direct CO2 emissions (1,057 Mt of CO2) from all industry sectors in the world
in 2004 (IEA, 2007). The impact of steel production on energy consumption and greenhouse gas
(GHG) emissions is therefore large. At the same time, steel will be a critical material for the
quality of life in 2050 and beyond, and will therefore continue to be a future contributor to
energy consumption and GHG emissions (AISI, 2009).
9

Figure 1: World industrial final energy 2004. (IEA, 2007)
The energy use and emissions of a steel plant are significantly affected by several factors such as
technology, plant size, and quality of raw materials (Xu, 2010). For technology, the adoption of
energy efficiency measures in steel production is key in order to reduce energy consumption and
GHG emissions. Analysing and managing the costs of these technological measures have thereby
become very important for steel producers and policy makers to make accurately decide on
investments in technology.
The use of energy efficiency measures in the manufacturing industries could provide great
savings in energy and energy related carbon emissions (Xu, 2010). The aim of this project is
therefore, to assess the expectations and effects of application of energy efficiency measures on
energy use and carbon emissions in the world steel industry.
Also previous efforts on energy efficiency measure in the US steel industry showed a significant
savings potential at low or even negative cost per unit of energy saved. This potential deserves
more attention than it has received so far (Xu, 2010). The US Department of Energy (DOE)
believes the primary steel production efficiency can be improved on the order of 20 to 30%
based on existing technology in the US. Steam supply systems and motor systems are common
in the steel industry, and have improvements to offer efficiency potentials on the order of 15 to
30% (IEA, 2006). If taking other world regions into account, which currently have lower energy
efficiencies compared to the US, there could be even larger efficiency gains in these regions.
This research will further look into the effects of applying energy efficiency measures in the
world steel industry on energy use and GHG emissions. The goal is to present updated results on
the potential in the US steel industry and include other world regions to provide an estimation
of energy and GHG mitigation in the world steel industry.
10

5 RESEARCH SCOPE
5.1 Background
The main arguments for conducting this research in energy efficiency measures, is that the
world steel industry plays a major role in energy use and GHG emissions now and in the future,
as mentioned in the introduction. In many cases, implementing energy efficiency measures is
among one of the most cost-effective investments that the industry could make in improving
efficiency and productivity, and reducing CO2 emissions (Xu, 2010).
In addition, Kim and Worrell (2002) suggest that an international comparison can contribute to
a better understanding of opportunities for emissions reduction. Iron and steel have a complex
industrial structure, but only a limited variety of processes are applied worldwide and they use
similar energy resources and raw materials, making comparisons worthwhile.
Also, plenty of information is available on the application and effects of implementing energy
efficiency technologies. Many of the energy efficiency technologies have become cost-effective,
and it is important and necessary to continue to incorporate new information on technology
characteristics. Also their evolution and response to energy and carbon price should be
monitored in order to enhance empirical descriptions of the technologies (Xu, 2010). Moreover,
the results of assessing energy efficiency measures could provide detailed information for
different world regions for integrated assessment models. Related policy makers around the
world can use these integrated assessment models, to make more accurate decisions on
stimulating the implementation of technologies (Xu, 2010).
5.2 Research Aim
This research will therefore aim at fulfilling the need for a better understanding of the effects of
energy efficiency measures in the steel industry, in order to make more effective decisions on
energy saving and GHG reduction that come with those energy efficiency measures. Plenty of
data are available on the measures themselves and the amount of energy that can be saved with
a single measure in a specific case. The effects of the combined measures on the industry are
not readily available or based on outdated data. Worrell et al. (2002) and Xu et al. (2010) have
already conducted major research into the potential of energy efficiency measures for the US
steel industry in 1994 and 2002. The aim for this current research at LBNL, will be to update
these data to 2006 and 2010, in order to provide relevant and recent technology data for
integrated assessment models. These integrated assessment models can be used by policy
makers to make accurate decisions, for example on stimulating the implementation of energy
efficiency measures.
This updated research will only encompass the potential of energy efficiency measures in the
steel industry in the US. In order to include the world steel industry, these results should be
extended to include other world regions. For the world steel production this will not be a
realistic goal, if not every region will be researched in detail. Due to time constraints and data
availability it is not possible to provide a detailed insight in the potential of individual energy
11

efficiency measures in each region. A more realistic research goal would therefore be to find a
more general conclusion on the use of energy efficiency measures in the world steel industry. A
more general conclusion can be the total estimated energy savings and carbon reduction of
energy efficiency measures in the world steel energy. This will be the aim of this research, where
the results acquired at LBNL will provide as intermediate results in this thesis.
Summarised, the goal of this thesis is to analyse the potential of energy efficiency measures in
the world steel industry.
5.3 Research questions
The main question will be:
12
What is the energy savings and carbon mitigation potential of energy efficiency measures in
the World steel industry?
The potential is the total amount of energy savings and carbon mitigation that can be achieved
by implementing energy efficiency measures to the extend which is expected to be feasible. In
order to determine the potential, first a major assessment of energy efficiency measures in the
steel industry will be conducted. This will be the key starting point for the research, which will
be performed at LBNL. In order to describe the potential of energy efficiency measures, the
effect of each single implemented measure should be investigated. Therefore, the first subquestion
will be:
Sub-question 1: What energy efficiency measures are available for the mitigation of energy use
and carbon emissions?
Energy efficiency measures for the steel industry in the US have been well described in several
publications (Worrell, 2001; Worrell, 2007; Xu, 2010; Choudhury, 1998; DOE, 2011). These
descriptions will be used for a basis of describing the technologies for energy efficiency
measures. Next, a major effort will be put in updating these measures in the US steel industry,
since the mentioned publications are often based on data of 1990-2000. The measures will be
updated and any possible new measures will be included in the assessment acquired from
similar sources.
This analysis of energy efficiency measures will only encompass the US steel industry, since the
data on energy efficiency measures will be largely based on research in this industry. Therefore,
the second sub-question should include the globalization of these measures. Therefore, more
detailed information on the differences of these regions will have to be acquired:
Sub-question 2: What are the typical energy use and economics in different world regions
concerning the production of steel?
This question will provide insight in the different steel producing regions to assess the effects of
implementing energy efficiency measures. The important factors to assess the differences in
steel production or the baseline production data are: total energy consumption, major industry

processes (Table 5) and steel production (described in chapter ‘World steel production and
different regions’). These data are often provided in statistical reports of the IEA (International
Energy Agency), reports on comparison of industries and statistical bureaus of the different
regions.
The most important factors for assessing the energy efficiency measures in the different regions
will be economic data like the Purchasing Power Parity (PPP) index, expected penetration of
measures and average weighted energy prices. This data can be found in statistical and
economics bureaus of the different regions as well as EIA (Energy Information Administration),
although the penetration of energy efficiency measures will have to be estimated as no specific
data were found.
From the stated differences in steel production for each region, a significant estimation of the
potential of the energy efficiency measures can be made. After the conversion of energy
efficiency measures for different selected regions in the world, an assessment of energy
efficiency measures in the world steel industry can be made.
Sub question 3: What will be the energy savings from the application of energy efficiency
measures in the world steel industry?
The total world steel market will be assessed by defining regions that together produce over 80%
of the world steel production (Table 2). This should provide an overview of the total potential
world energy savings in the steel industry of energy efficiency measures.
5.4 System definition and borders
As already mentioned in the research questions, the database for energy efficiency measures for
the US steel industry will be thoroughly described. When completed, the set of energy efficiency
measures analysed for the US, will be used to assess the energy savings and carbon reduction
potential for other world regions. This is a combination of assessing different aggregate levels of
the steel industry as well as looking at different regions. At a technology level, the effects of
implementing an energy efficiency measure will be described, which is on a low aggregate level
(the lowest for this thesis). The results of all measures combined will be at a country or region
scale, a higher aggregate level. This method is referred to as a bottom-up approach (IEA, 2008).
As the results for one region have been acquired, the effects in other world regions will be
determined based on the results for the US. Eventually these results combined should provide
the desired insight in the world steel industry, the highest aggregate level for this thesis.
First, the definition of steel industry should be further explained. In the steel industry are all
production processes are involved in making rolled steel from either scrap, DRI or iron ore. Any
processes prior to the steel production, like mining and crushing of raw materials are not taken
into account. Also the processes after rolling of steel, like forming or application of steel in cars,
cans or construction material, are excluded as well. An overview of the included processes is
given in chapter ‘Steel production’.
13

The levels which can be identified between the lowest and the highest aggregate level will not all
be incorporated as can be seen in Error! Reference source not found.. The effects of a
certain energy efficiency measures will not be analysed on a process scale or on a plant scale, but
will immediately be translated from the data found on process technologies to a country level. In
Error! Reference source not found. the green arrow represents sub-question 1, the
potential of energy efficiency measures (in process technologies) on the steel industry of the
United States. The second sub-question is presented by the blue arrows, where the results for
the United states will be used to estimate the potential in other world regions.
Figure 2 Representation of the aggregation levels and different regions assessed in this thesis in order to
conclude on the potential for the world steel industry.
The selected regions in Figure 2 represent the main steel producing regions which are
considered to have similar characteristics within each region. These selected regions encompass
a total production of about 82% of the world production of crude steel in 2006, as seen in Table
2. The other remaining regions are about 18% of the total world steel production.
14

Table 2: Production, final energy use and calculated energy intensity of the major iron and steel producing
countries, 2006
Region
Crude Steel
Production
(Metric Mt) 1
% of total
China 419.1 34%
Western European Union 173.2 14%
Former Soviet Union 119.9 10%
Japan 116.2 9%
United States 98.6 8%
Central and South America 45.3 4%
India 49.5 4%
Other 225.3 18%
Total 1,247.1 100%
Notes: All in metric tons. ‘Other’ is calculated by the difference of total and the sum of all regions.
Numbers are rounded off.
1 Source: Worldsteel, 2011
The application of the energy efficiency measures in the process technologies or equipment is
analysed, though the effects on the steel production processes will not be described. Also the
steel production facilities and the steel companies will not be described separately. Only the
total industry structure of the country and its characteristics and the energy efficiency
technologies which can be implemented in process technology. The generalisation of the steel
industry in the US is considered legit, since steel production processes in different companies
usually have similar process characteristics. The steel production in different world regions may
have different industry structures, but the processes used for steel production are similar (Xu,
2010).
For this research the use of energy efficiency measures in the steel industry will be used to
determine energy savings potential for the steel industry. Another large share of observed
differences in energy intensities and CO2 emissions on a plant and country level are explained by
variations in the quality of the resources that are used and the cost of energy (IEA, 2007). While
the quality of resources may have an equally important influence, this is not included in this
research.
Energy efficiency measures for the steel industry only encompass the main production
processes, present in typical steel production facilities. This is the preparation and processing of
raw materials, the actual steel production processes and the finishing processes like hot and cold
rolling (explained in further section). The reference year of all data found for this thesis will be
2006, so production, energy and economic data for the energy efficiency measures and regional
data will be from 2006. This is because for that year more data were available.
15

The results for the US will be used to determine the potential for other regions, therefore only a
number of mentioned key factors will be used to include in the characteristics of these regions.
No detailed reports and data could be found on both the steel industry statistics, like production
data and process-specific energy use, as for case studies or reports on energy efficiency
measures for each of all regions. For all regions, apart from the US, the data used for
determining the potential are limited, and only major factors like industry structure and total
energy use are used to determine the potential of energy efficiency measures in each region.
Finally, the potential of energy efficiency measures is only expressed in energy savings and
energy related carbon savings. Other effects may be important, but will not be taken into
account for this research, such as environmental factors like waste reduction or economic
factors like total cost reductions. While the cost-effectiveness of measures will be an important
factor, the total economic effects of the energy efficiency measures will not be reported, like total
reduced operational costs.
5.5 Methodology and data sources
The system definition provides insight in the subjects and results that can be expected in this
report. This chapter will elaborate on the methods used to acquire sufficient answers to the
research questions. The tools, methods and data sources which will be used to find these
answers will be described.
The assessment of different energy efficiency measures in the world will be similar to the
existing methods used by Worrell (2002) and Xu (2010). The use of Cost of Conserved Energy
(CCE) will be used to rank different energy efficiency measures on cost-effectiveness and energy
savings potential. With the weighted average energy price for every region, the cost-effective
measures for every region will be identified. The maximum potential energy savings from energy
efficiency measures can be calculated for the world, as well as the cost-effective potential energy
savings. From these energy savings, the potential carbon mitigation can be calculated by using
conversion rates of different energy types and their carbon emissions.
For every identified efficiency measure, detailed information should be included in order to be
able to assess their effects. The assessment will be based on the energy savings in GJ per tonne
of steel, the (annualized) investment costs and operating cost-change. With these data the CCE
can be calculated (Meier, 1984). If the cost of conserved energy is lower than the average
weighted energy price, the measure is considered cost effective. The calculation of the CCE is
explained in equations 1 and 2.
16

Where:
∙
∆
d
q = (2)
−n
( 1−
( 1+
d)
)
CCE = Cost of conserved energy for an energy efficiency measure in $/GJ
I = Capital cost ($)
q = Capital recovery factor (yr -1 )
∆E= Annual energy savings (GJ/yr)
M = Annual change in monetizable non-energy costs from O&M changes ($/yr)
B = Annual additional non-energy cost benefits ($/yr)
d = Discount rate
n = Lifetime of the mitigation option (years)
The energy savings and costs data will be multiplied with a penetration rate of each measure.
The penetration rate is a determination of the share of the industry which is expected to be able
to use the energy efficiency technology. Although previously indicated, the processes in the steel
industry are much alike, differences in steel production facilities could inhibit the use of certain
energy efficiency measures in some cases. For example, some measures might only be viable for
plants with smaller scale production. For the US, for every measure the penetration rate was
determined either by expert interviews, literature sources or estimations. However, penetration
rates in other world regions will be very difficult to determine.
From the CCE a ranking can be made of the most cost-effective measures, to produce a cost
curve (Xu, 2010). The cost curve is a graphic representation of the total and cost-effective
potential energy savings of energy efficiency measures. An example is shown in Figure 3.
Basically, in a cost curve it can be observed what is the maximum amount of energy savings
(GJ/tonne) at certain costs ($/GJ). All the measures which have a CCE lower than the average
weighted fuel price are considered cost effective. So in Figure 3 about 3.8 GJ/tonne of steel can
be saved cost-effectively. This explains the exclusion of energy cost savings in the calculation of
CCE. Since the value of CCE is compared to the energy price to determine cost-effective
measures.
In Figure 3 the effect of including or excluding other benefits can be observed. In this case
productivity benefits. If other benefits are included, the cost curve is much lower, which means
that the cost for the same amount of energy savings decrease. From both the cost curve and the
data tables of the cost curve, the energy savings can be determined.
(1)
17

Cost of Conserved Energy ($/GJ)
Discount Rate = 30%
Figure 3 Example of a cost curve with and without including non-energy productivity benefits in the US
iron and steel industry (Worrell et al. 2002)
Then, from the calculated energy savings of fuel and electricity an estimation of the carbon
mitigation can be calculated, to include the carbon mitigation results from energy efficiency
measures (Worldsteel, 2007; Worrell (2001, 2002, 2007); Xu 2010, EPA 2010). To estimate the
carbon emissions reduction, the carbon emissions of each process and electricity have been
determined (Xu, 2012). The carbon emissions for each process have been determined in kg of
carbon emissions per GJ of energy used. Carbon emission reductions from energy savings in one
of the processes can be determined in that way.
In this report only the final energy will be used to present the estimated savings for the steel
industry. It can be discussed to include primary energy for a more accurate determination,
because energy requirements for electricity generation and distribution would then be included.
However, steel production facilities produce electricity from by-products, like BF or coke oven
gas this makes it difficult to estimate the conversion of final energy to primary energy for every
region. Next, the cost-effectiveness of implementing energy efficiency measures in steel
production will be researched. Steel production facilities will not take into account the primary
energy savings. Only the final energy savings are important for cost benefits. A discount rate of
30% is used to determine the annualized investment costs.
In general, overall data availability limits the accuracies of estimating the potential degree of
implementation. For example, the Energy Information Administration reports the uptake of
some energy efficiency measures in the Manufacturing Energy Consumption Survey (MECS),
such as crosscutting technologies like process controls, building controls, waste heat recovery or
adjustable speed drives (EIA, 2009). For other measures specific to the iron and steel industry,
additional literature sources, sector specific statistics, or expert estimates were used. Other key
literature sources used for iron and steel industry specific measures were the “Round-Up’s”
published by the journal Iron & Steelmaker on electric arc furnaces, blast furnaces and
continuous casters (I&SM, 1997a; I&SM, 1997b). A number of additional measures were found
18
21
18
15
12
9
6
3
0
-3
-6
Annual C ost-E ffective P rim ary E nergy S avings
E xcluding N on-E nergy
B enefits: 1.9 G J/tonne
difference: 1.9 J/tonne,
approxim ately 168 P J/year
Including N on-E nergy
B enefits: 3.8 G J/tonne
1994 W eighted A verage P rim ary Fuel P rice ($2.14/G J)
C ost C urve W ithout P roductivity B enefits
C ost C urve Including P roductivity B enefits
0 1 2 3 4 5 6
E nergy S aving s (G J/tonne)

y assessing DOE energy audits in US steel plants. For key technologies, reference lists of
manufacturers such as Voest-Alpine Industrieanlagebau (VAI) (VAI, 1997) were also used.
Once each measure was characterized individually, its applicability to the US iron and steel
industry as a whole was assessed. In principle, in order to estimate the potential for future
uptake of each energy efficiency and GHG-emission reduction measure, each measure was
characterized by the degree to which implementation of the measure can be applied in the US
iron and steel industry. The total of energy savings calculated for each measure represents the
total potential for the US. Next, the potential of cost-effective measures is calculated, by adding
only the energy savings potential for cost-effective energy efficiency measures.
The key factors for the energy savings potential are the total energy consumption, industry
structure and steel production statistics (further explained in chapter ‘World steel production
and different regions’). These data are often provided in statistical reports of the IEA
(International Energy Agency), reports on comparison of industries and statistical bureaus of
the selected regions. The most important factors for assessing the economic side of the energy
efficiency measures in the different regions will be the PPP price index to compare investment
and operating costs, expected penetration of measures and average weighted energy price for
each region. These data will be needed to identify the cost-effective energy efficiency measures
for each region. The latter mentioned factors will be more difficult to determine.
Next, the Market Exhchange Rates (MER) is used to compare the PPP-index of different regions
in the same currency (World Bank, 2006). Examples of PPP indexes in different countries are
the Big Mac index and OECD comparative price levels. On the other hand, technology and
equipment for energy efficiency measures may not be available in each region, since it is not a
common good like gold or wheat (Levi, 2005). So, the PPP index might not be a precise cost
difference for implementing energy efficiency measures in one region, compared to another.
Moreover, it is likely that the energy efficiency technologies origin from western economies, so
the PPP index would not be true in that case. Therefore, in this report the PPP index will only be
used partly. The investment costs for a certain technology will be estimated at 70% of the cost
identical to the US, while 30% of the cost dependent upon the PPP index.
Penetration rates of technologies per region will be a more difficult factor. This can hardly be
scientifically checked. Reports like the Manufacturing Energy Consumption Survey (MECS),
which is issued in the US every four years, provide an approximation of technology penetration
in certain industries. The availability of similar reports for the steel industry for other regions is
marginal.
When the data for energy efficiency measures is converted from the US data to other regions,
similar calculations will be made concerning the CCE of each measure. For each region the
estimated potential for energy efficiency measures will be determined. The cumulative of these
potentials will eventually present a world potential of energy efficiency measures in the steel
industry. Then for the final result, the world potential can be compared to studies of theoretical
minimum energy consumption of steel production. The theoretical minimum energy use for the
steel industry in the world, compared to the total energy use in the steel industry now and the
potential energy use after implementing energy efficiency measures.
19

Because a number of assumptions have to be made to calculate the potential of energy efficiency
measures, these assumptions will be tested on their sensitivity. A number of scenarios will be
included in order to clarify the influence of these assumptions on the outcome. First, the PPP
index is introduced to be included in 30% of the investment and operating costs. Influences of
the PPP index will be assessed by changing it to 0% or 60%, to determine the influence of this
factor. Next, the effect of energy efficiency measures in other world regions may be different. For
example, an energy efficiency measure in the more energy intensive India, may have a larger
effect on the energy consumption in that region than if this measures was applied in the US
(Schumacher, 1998). For some of the measures we will try to find if we can link the energy
savings potential to the energy consumption.
20

6 STEEL PRODUCTION
The world steel industry is one of the largest and energy intensive industries in the world. The
total production of crude steel was 1.4 billion metric tons in 2010 (Table 2). Crude steel, the
basic end product of steel production processes, is produced in many varieties from hard to soft
steel. The hardness of steel is dependent upon the chemical composition of steel. Crude steel will
then be shaped in different forms for different purposes. Also, surface treatments like painting
or selling the of solid steel for forging industries is possible. The main end products in which
steel is used are in the construction, automotive, packaging and appliances industries.
Steel is produced roughly by two types of processing. First, the integrated steel mills produce pig iron
from raw materials (iron ore and coke) using a blast furnace, followed by producing steel from pig iron
using a basic oxygen furnace (BOF). Second, the secondary steel mills produce steel from scrap, pig iron
or direct reduced iron (DRI) using an electric arc furnace (EAF).
Figure 4 provides an scheme for the most typical steel production routes.
Figure 4 Iron and Steel Production Routes
21

For the primary route, pig iron is produced in a blast furnace, using coke in combination with
injected coal or oil, to reduce sintered or pelletized iron ore to pig iron. Limestone is added as a
fluxing agent. Coke is produced in coke ovens from coal. Reduction of the iron ore is the most
energy-consuming process in the production of primary steel. Modern blast furnaces are
operated at various scales, ranging from mini blast furnaces (capacity of 75 ktonnes/year) to the
largest with a capacity of 4 Mtonnes/year. Usually a production facility has more than one blast
furnace, producing up to 10 Mtonnes of crude steel per facility. Besides iron, the blast furnace
also produces some by-products like blast furnace gas (used for heating purposes), electricity (if
top gas pressure recovery turbines are installed) and slags (used as building materials; Xu,
2010).
Primary steel is produced from pig iron by two processes: open hearth furnace (OHF) or basic
oxygen furnace (BOF). The OHF is still used in different configurations, mainly in Former Sovjet
Union and India. OHF is considered an out-dated technology because of productivity and energy
and environmental issues (Gosh, 2008). While OHF uses more energy, this process can use
more scrap (recycled steel) than the BOF process. Nevertheless, BOF process is rapidly replacing
OHF worldwide, because of its greater productivity and lower capital costs. In addition, the BOF
process needs no net input of energy and can even be a net energy exporter in the form of BOFgas
and steam. The process operates through the injection of oxygen, oxidizing the carbon in the
hot metal. This is the main process to transform the hard and brittle pig iron into softer and
ductile steel. Several configurations exist depending on the way the oxygen is injected. The steel
quality can be improved further by various ladle refining processes used in the steel mill (Xu,
2010).
Secondary steel is produced in an electric arc furnace (EAF) using scrap. Scrap steel is melted
and refined, using a strong electric current. Several process variations exist, using either AC or
DC currents, and fuels can be injected to reduce electricity use. Just like the blast furnace, the
EAF produces steel, slag and EAF gas, although by-products are produced in lower amounts.
Direct reduced iron (DRI), or sponge iron, is produced by reduction of the ores below the
melting point in small scale plants (< 1 Mtonnes/year) and it has different (but close) properties
than pig iron. DRI production is growing and nearly 4% of the iron in the world is produced by
direct reduction, of which over 90% uses natural gas as a fuel (Midrex, 1996). DRI serves as a
high quality alternative for scrap in secondary steel making.
Casting and shaping are the next steps in steel production. Casting can be a batch (ingots) or a
continuous process (slabs, blooms, billets). Ingot casting is the classical process and is rapidly
being replaced by continuous casting machines (CCM). In 1998, 83% of global crude steel
production was cast continuously (IISI, 1999). In 2010 this was already 95% (IISI, 2011).
Continuous casting is a significantly more energy-efficient process for casting steel than the
older ingot casting process. The casted material can be sold as ingots or slabs to steel
manufacturing industries. Most of the steel is rolled by the steel industry to sheets, plates, tubes,
profiles or wire. Generally, the steel is first treated in a hot rolling mill. The steel is heated and
passed through heavy roller sections reducing the thickness of the steel. The hot rolling process
produces profiles, sheets, or wire. After the hot rolling process, sheets can be reduced in
22

thickness by cold rolling. Finishing is the final production step, and this may include different
processes such as annealing, pickling, and surface treatment.
These processes have a different demand for energy and thereby different carbon emissions. The
average energy use of these processes has been determined for the US steel industry. The most
energy demanding process is the blast furnace, while processes like steel finishing are relatively
low in energy demand. The total final energy demand for producing primary steel is about 20.1
GJ/tonne of steel. For secondary steel production the final energy demand is about 5.5
GJ/tonne. These and other important findings for the US steel industry are summarised in
Table 3.
Table 3 Final energy, associated carbon emissions, and production in 2006 for the US Steel industry
US Steel production Primary Steel
Secondary
Steel
Total Steel
Production (Mt/year) 42.5 56.1 98.6
Final Energy Consumption (PJ/year) 852 311 1,162
Energy Related Carbon Emissions (MtC/year) 18.9 9.5 28.4
Final energy Intensity (GJ/tonne) 20.1 5.5 11.8
Table 3 shows some characteristics of the primary and secondary steel. First, the energy
intensity of primary steel is considerably more compared to secondary steelmaking. The
difference is due to the secondary steel making does not require the energy consuming reduction
process. The secondary steelmaking requires electricity for melting scrap steel.
23

7 STEEL ENERGY EFFICIENCY MEASURES
This section will explain which energy efficiency measures there are and will include the
identified energy efficiency for this research. Cost curves for 51 measures for improving energy
efficiency in the iron and steel sector were evaluated for 1994 by Worrell et al. (2002). Then,
updated cost curves of measures were developed for the year 2002 for the same set of measures
by Xu (2010). For 2006 additional measures were found, totalling at 75 measures for that years.
In addition, Appendix A to H includes descriptions of the mitigation measures. The following
describes the steps to assess the energy efficiency measures and formulating cost curves.
First, for every measure the energy savings potential per production unit was identified. Next,
the investment costs of the measures are identified and the expected lifetime of the measure.
The lifetime and the investment costs of the measure were used to determine the annualized
investment costs per production unit. To calculate the CCE, these are the most important data.
In addition to data on energy savings and costs, some of the measures had identifiable and
quantifiable additional benefits, such as reduced labour and maintenance or increased yields.
Table 4 enlists a selection of the mitigation technologies and their corresponding benefits for the
iron and steel industry.
25

Table 4 Examples of energy efficiency measures for the Iron and Steel Industry that have other nonenergy
benefits as well as energy benefits (Worrell et al. 2002).
Improved process control
Bottom Stirring / Stirring gas
injection
26
Secondary Steel making
Foamy slag Reduced tap to tap times
Oxy-fuel burners
DC-Arc furnace
Scrap preheating - Tunnel
furnace (CONSTEEL)
FUCHS Shaft furnace
Twin Shell w/ scrap preheating Reduced tap-to-tap time
Average increase in productivity of 9-12% and reduced electrode
consumption of 25%
Net cost savings of $0.9-2.3/tonne from increased yield of 0.5%
Reduced tap-to-tap time of 6% and improved product quality from O2
injection
Reduced tap-to-tap time, reduced electrode use, increased refractory
life and improved stability
Increased productivity by 33%, reduced electrode consumption by
40% and reduced dust emissions
Reduced electrode consumption, reduced flue gas dust emissions by
25%, increased yield of 0.25-2.0% and 20% increased productivity
Integrated Steel making
Coke dry quenching Reduced dust emissions and improved working climate
Pulverized coal injection to 130
kg/tonne hot metal (thm)
Pulverized coal injection to 225
kg/thm
Injection of natural gas to 140
kg/thm
Reduced coke-related emissions
Reduced coke-related emissions
Reduced coke-related emissions
Adopt continuous casting Reduced material losses from about 8% to 2%
Hot charging
Improved material quality, reduced material losses, improved
productivity by up to 6% and potential reduction of slab stocking
Both
Thin slab casting Improved productivity and reduced material losses
Energy efficiency measures have been identified in a wide variety of applications. For example
foamy slag, which results in a natural isolation layer on top of molten steel in an EAF. Another
example is the use of an alternative energy source, for example oxy-fuel burners in EAF furnaces
to reduce use of electricity, or the use of powder coal instead of coke for the blast furnace. In
both cases the primary fuel use is increased, which reduces the need for generating electricity or
producing coke. Other measures included the upgrading of process control or auxiliary
equipment like steam or pressurised air systems. Planning is an issue for ‘hot charging’, which is
again a different type of measure. During the casting process, liquid steel is solidified by cooling,
after which it will have to be reheated before rolling. Hot charging is a way of reducing the
cooling of steel after casting, and charging in reheat furnaces ‘as hot as possible’, reducing the
energy consumption of a heating furnace.

A single energy efficiency measures can reduce the energy use for steelmaking up to 3.5 GJ per
tonne of steel. That particular measure is the application of thin slab casting where casting and
hot rolling are combined in one process. The application of this measure will have a large impact
on the energy use for hot rolling of steel. The investment costs for such a measure are about
$150 per tonne. The CCE calculated for this measure, when the capital recovery factor is
included, is about 12 $/GJ. With weighted average energy prices around 7 $/GJ such a measure
is not cost effective. If other non-energy benefits will be included in the calculation of this
measure, the CCE will be 3 $/GJ. So, this measure is only cost-effective if the other non-energy
benefits are included.
Another important factor for the combined effects of energy efficiency measures is the
production rate. Preliminary results of the research at LBNL indicated that the production in a
certain period has a large effect on the energy savings potential of the measures (Xu, 2012). The
relation is logic, since the energy savings for every measure is defined in GJ/tonne of steel. A
reduction in production will result in a reduction in potential energy savings. Dependent upon
the region in which the production change occurs, the potential for energy efficiency measures
may change to great extent.
The cost-effectiveness of an energy efficiency measure is highly dependent upon the energy
price. If energy prices increase, the cost-effectiveness of energy efficiency measures will increase
accordingly. One example is the increased natural gas price in 2006 in the US. The results found
from the research at LBNL indicated that the increase of energy efficiency potential in 2006 was
partially devoted to the increased natural gas price in that year (Xu, 2012).
27

8 WORLD STEEL PRODUCTION AND DIFFERENT REGIONS
In order to analyse the different world regions, the characteristics of each region are
determined. First, the differences in steel production for each region are determined for the
estimation of energy efficiency potential. Next, the economic differences will be elaborated on to
include the determination of cost-effective measures.
8.1 Steel production
While the steel industry around the world use similar production processes, the structure of the
industry is different for every region. Table 5 summarises the different main steel production
processes. First, the integrated or primary steel industry, which uses blast furnaces and basic
oxygen furnaces, can be as low as 42% in the US in 2004 or as high as 90% in China. As
indicated in the chapter ‘Steel Production’ the production of primary steel requires a lot more
energy compared to secondary steel. It can be expected the average energy intensity of China
steel industry is thus larger than the intensity in the US. The energy intensive production of steel
in China is expected to increase the energy savings potential with energy efficiency measures.
Table 5: Use of steel producing processes in the different regions of the world, 2006
Region
Integrated
Production
(BOF)
Secondary
Production
(EAF)
Integrated
Production
(OHF)
Total
%
China 90% 10% 0% 100%
Western European Union 57% 43% 0% 100%
Former Soviet Union 59% 17% 24% 100%
Japan 74% 26% 0% 100%
United States 42% 58% 0% 100%
Central and South America 60% 38% 0% 99%
India 42% 56% 2% 100%
Other - - - -
Note: Numbers have been rounded off. Central and South America is the only region where numbers
do not add up to 100%, for unknown reason.
Source: Worldsteel, 2011
What is seen in Table 5 is the use of OHF in India, but especially in the Former Sovjet Union.
The use of OHF steelmaking is an out-dated technology for producing steel from pig iron (IEA,
2007). The energy intensity of these regions is therefore expected to be higher compared to
other regions. In order to determine the energy intensity of every region, the total steel
production and energy use are required, described later in this chapter.
29

The production of steel in world has seen some obvious trends in the past decade, as can be seen
in Figure 5. The steel production in China has grown largely from about 150 million tons in 2001
to over 600 million in 2010 (Worldsteel, 2011). Also, steel production in India has grown (250%
growth from 2001 to 2010), while regions like US and Europe remain relatively constant. The
effects of the financial crisis can be seen in a drop in steel production in several regions in 2009.
Except for China and India, every other region encountered a drop in production in 2009,
compared to 2008 (Worldsteel, 2011). Since for this research a single reference year is chosen
(2006), the effect of steel production trends will not be emphasized further in this report.
Figure 5 World steel production trends in the years 2001-2010. (Worldsteel, 2011)
The strong growth in production in China and India, should benefit energy efficient production.
With the expansion of production capacity, one could assume the state of the art technology will
be acquired. However, the increase in production in China and India have likely reduced the
global average energy efficiency per tonne of steel produced. As can be seen in Table 6, in China
and especially India, steel is produced with a high average energy intensity (Worldsteel, 2011;
IEA, 2007; Hasanbeigi, 2011).
30

Table 6: Production, final energy use and calculated energy intensity of the major iron and steel producing
countries, 2006
Region
Crude Steel
Production
(Metric Mt) 1
Final Energy
use (EJ) 2
Energy
intensity
(GJ/ton CS)
China 419.1 9.5³ 22.7
Western European Union 173.2 2.8 16.2
Former Soviet Union 119.9 3.3 27.4
Japan 116.2 1.9 16.8
United States 98.6 1.5³ 14.8
Central and South America 45.3 1.2 27.0
India 49.5 1.6 31.8
Other 225.3 3.1 13.8
Total 1,247.1 24.9 20.0
Notes: All in metric tons. ‘Other’ is calculated from the difference of total and the sum of all regions.
1 Source: Data for 2006 (Worldsteel, 2011)
2 Source: Data for 2004 (IEA, 2007). Includes coke making and blast furnaces, adjusted to 2006 by
production difference.
3 Source: Data for 2006 (Hasanbeigi, 2011)
Discrepancy between sources may exist since the energy intensity of ‘Other’ is relatively low.
Another explanation could be that the smaller countries use the lower investment and lower
energy requiring EAF furnaces. For EAF furnaces average energy intensities lower than 14
GJ/metric ton are usual, which explains a lower energy intensity for ‘Other’. Another
discrepancy is, the result found for the US steel industry is in contrast with the result calculated
by Xu et al. (2012) (see Table 3). This is due to different calculation methods.
In different regions the characteristics can differ largely. For example the energy intensity of the
steel production in India requires almost 32 GJ per tonne of produced steel, whereas the US
only requires 15 GJ per tonne of steel (Table 6). The use of energy efficiency measures in these
regions could therefore have a different effect on the energy use. For example, the
implementation of an energy monitoring and management system is expected to reduce the
energy consumption by a certain percentage. The measure will have an increased effect in a
more energy intensive region. Other measures which replace a certain technology, like the use of
oxy fuel burners for EAF furnaces, are not dependent upon the energy intensity of steel
production. For these measures the effects of the energy efficiency measures are not expected to
change in different world regions.
31

The difference in energy intensity of different regions is substantial. Some examples are the use
of out-dated technologies such as open hearth furnaces, still used in Russia and Ukraine (IEA,
2007). The reason for the low energy efficiency in China is mainly due to a high share of smallscale
blast furnaces, a large supply of coal, a high share of inefficient coking plants and low
quality ore (IEA, 2007).
8.2 Economics
The economics is the second important factor we should elaborate on for the assessing the
energy efficiency potential in the world steel industry. As mentioned in the methodology and
data resources, mostly a PPP index and the average energy prices will be needed for an
estimation of the cost-effective energy efficiency measures. The difference in prices and thereby
the investment and operational cost will be determined with the PPP index. Many types of PPP
indexes exist, and in this report the OECD statistics PPP index will be used (OECD, 2012).
It can be expected that the PPP index will not be an accurate determination of the costs in the
different regions for several reasons. A PPP index is based on numerous products, which also
encompass food and other products not related to steel industry. Moreover, certain energy
efficiency technologies might not be available in every region, where import of technology from
another region will be required. If import of technologies is the case for many regions and
technologies, the use of the PPP index will not be a reasonable estimate of the investment costs.
Therefore, a decision was made on the use of PPP index to a certain extent. In these cases we
used the PPP index for 30%. In addition, the influence of the PPP-index was tested at 0% and
60% for each region.
The PPP index for the different regions is shown in Table 7, where the index for the US is the
reference set at 1. Regions with a lower PPP index, require lower investments for energy
efficiency measures, and are therefore expected to have a larger amount of cost-effective
measures.
Table 7 PPP index for different world regions, compared to the US
Region PPP index
China 0.435
Western European Union 1.042
Former Soviet Union 0.465
Japan 1.07
United States 1
Central and South America 1 0.672
India 2 0.435
1 Central and South America is the average of Chile and Mexico.
2 India PPP index was not retrieved, China PPP index is used here.
Source: (OECD, 2012)
32

Energy price is another important factor for the determination of cost-effective measures.
Higher energy prices in certain regions are expected to yield a larger amount of cost-effective
energy savings from energy efficiency measures. The average weighted energy price for the US
was calculated at $6.76/GJ (Table 8) for the iron and steel industry in 2006.
Table 8 Average weighted energy price determined by share and price of energy types for the US Steel
industry in 2006
Energy type Share (%)
Final Energy
Price ($/GJ) 1
Electricity 14% $ 12,55
Residual fuel oil 2% $ 5,66
Distillate fuel oil 0% $ 13,34
Gas 29% $ 8,08
Coal+other 37% $ 3,36
Coke 18% $ 7,19
Average weighted energy price 100% $ 6,76
1 Source: EIA, 2009
The average weighted fuel price is based on the found share of each energy type in the US steel
industry, based on baseline calculation. The energy prices have been acquired from the
Manufacturing Energy Consumption Survey (MECS) for the US steel industry. Energy prices for
other world regions will differ because of their industry structure and the local prices of different
fuel types.
One specific example of different policies concerning the price of energy is India. India uses
APM (Administered Price Mechanism) prices. These prices are determined by the government,
for public sector companies (Public Enterprise Survey, 2011). India’s steel industry also includes
a number of state owned steel production facilities like SAIL (Steel Authority of India Limited).
Other steel companies, like Tata Steel, are private owned, and have to pay market prices for
energy.
33

9 RESULTS
Next, the results of this thesis research will be presented. First, the results for the US steel
industry will be presented, generally based on the research performed at LBNL. Next, the
expansion of the results to other parts of the world are included.
9.1 Potential of energy efficiency measures in the US steel industry in 2006
The report of Xu et al. (2012) focused on energy efficiency in the US steel industry in the years
1994, 2002, 2006 and 2010, largely because the data for the US steel industry was available for
these years were more complete than other years. The identified measures for that research are
all described in 0 to Appendix I.
This report will focus on the world steel industry in 2006, since most recent data is available in
all different regions for this year. All the cost data (US dollars) are obtained and presented as the
nominal currency values for the respective year (i.e. 2006). In Appendix J an example of the
calculations made for energy efficiency measures is given. Next in Appendix K the most
important results of calculations for energy efficiency measures in the US steel industry for
2006 have been given.
Data on energy efficiency measures and the US iron and steel industry were used to identify the
potential of energy efficiency measures in the US steel industry. Accordingly, for every measure
the Cost of conserved energy (CCE) is calculated. Next to the energy savings of each measure,
the CCE value is the most important information. In order to illustrate the potential of all the
energy efficiency measures identified for the US steel industry, cost curves have been developed.
For each energy efficiency measure the final energy savings in the US steel industry have been
calculated. The measures are subsequently ranked on CCE value from low to high. The CCE of
each measure is then plotted to the cumulative energy savings. Based on the data finding of this
research in Berkeley, Figure 6 shows the cost curves of 72 energy efficiency measures for the US
iron and steel for 2006 (Xu, 2012). As already mentioned in the methodology, each horizontal
curve line is cumulative energy savings of each measure, while the height (Y-axis) of each
vertical line section represents the cost per unit of energy saved corresponding to the same
efficiency measure.
35

Figure 6 cost curve for specific final energy savings of energy efficiency measures in the US steel industry
in 2006
From the cost curves a number of conclusions can be drawn. First, from the shape of the cost
curve, an overall impression on the energy efficiency measures can be attained. As already
mentioned, the measures for which the calculated cost of conserved energy (CCE) is lower than
the average weighted energy price are cost effective. The most cost-effective measures (lowest
CCE values) are displayed on the left, these can even have a negative CCE.
A negative CCE is possible if the other non-energy benefits of a certain measure are larger than
the annualized investment costs of the measure. In other words, without the energy cost savings
taken into account, the measure is cost effective. Though, the potential amount of energy savings
from measures which have a negative CCE is minor. As can be seen in Figure 6, the cost curve
intersects with the X-axis at about 0.2-0.3 GJ/tonne.
According to IPCC (2001), no-regrets opportunities for GHG emissions reduction are the
options whose benefits such as reduced energy costs and reduced emissions of local or regional
pollutants equal or exceed their costs to society, excluding the benefits of avoided climate
change. In this report, a no-regrets option is defined as a GHG reduction option (i.e., via energy
efficiency measure) that is cost effective over the lifetime of the technology compared with a
given energy price, without considering benefits of avoided climate change. Although existence
of no-regret options is not acknowledged by some economists, a number of cost-effective
measures were identified in the US iron and steel sector. There are many factors, including
market barriers and knowledge gap, which contribute to slower adoption of such measures in
the markets (Xu, 2010).
36

The curve section right of the intersection with the X-axis is interesting, here is the rest of the
cost-effective measures. As mentioned in the methodology, the cost-effective measures require a
CCE lower than the average weighted energy price (6.76 $/GJ for the US steel industry in 2006,
Table 8). In Figure 6 an intersection of the cost curve with the average weighted energy price can
be seen at around 3 GJ/tonne (X-axis). So the ranked energy efficiency measures up to a
cumulative energy saving of 3 GJ/tonne in total, can be considered cost effective.
On the far right of the cost curve, energy efficiency measures can be seen which are not cost
effective. These measures have a CCE value, greater than the average weighted fuel price.
Installing the energy saving measures will induce more costs compared to the existing use of
energy. Therefore it is not expected that these measures will be implemented. Though, the cost
curve ends at about 3.9 GJ/tonne, with a CCE for the lowest ranked efficiency measure at 93
$/GJ, which indicates the total potential of the identified energy efficiency measures.
Cost curves can have large negative and positive values, due to the substitution of energy
sources. For example, the use of FUCHS Shaft preheater furnace for EAF steelmaking furnaces
will substitute the use of electricity with natural gas. The total energy savings of this measure is
only a minor amount (0.01 GJ/tonne), though the other benefits of this technology are
significant, over 1 $/tonne. The value of the CCE is thereby very low, -88 $/GJ. From a CCE
perspective, this is an very interesting measure, but the actual energy savings are marginal.
Cost curves provide an overview of identified energy efficiency measures, their total and costeffective
potential. Though the intersections in the graph are rather hard, the result is not
expected to be true for every production facility. As mentioned in the system definition, the
production facilities will not be taken into account in this report. A measure can be cost effective
for one facility, and not cost effective for the other. The intersections are therefore indications of
cost-effective measures, which provide a reasonable estimation of the potential of energy
efficiency measures in the industry. A measure considered not cost effective in this report,
should therefore not be abolished for implementation in any steel production facility. Since the
total US steel industry is assessed, the conditions in some steel producing facilities might benefit
certain efficiency measures. In reality, these energy efficiency measures may be cost efficient in
particular circumstances. The opposite of cost-effective efficiency may also be the case.
To get more insight into the middle section, where most cost-effective measure can be found,
another cost curve is shown in Figure 7. This is the same cost curve as in Figure 6, though the
scale on the Y-axis has changed. Separately identifying energy efficiency measures is easier in
this figure. Clearly, some energy efficiency measures have a larger effect in final energy savings
than others.
37

Figure 7 Cost curves of final energy savings for the US steel industry for 2006 including other non-energy
benefits and including inflation (narrowed scale on Y-axis)
9.1.1 Energy savings
The result of the CCE and cost curves calculation have been summarised in Table 9. The
potential of all the energy efficiency measures combined is provided in that table. Included are
the total production and energy use of the US steel industry. The total production is used to
determine the total savings potential of the energy efficiency measures in both the integrated
and secondary steelmaking. The current total energy consumption is included to calculate the
energy savings percentages. These are the main conclusion from the research on energy
efficiency measures for the US steel industry (Xu et al. 2012).
38

Table 9. Technical potential for energy savings in the US iron and steel making in 2006.
Energy
savings
(GJ/tonne)
Total
production
(MTonnes)
Total
Energy
savings
(PJ)
Current
energy
consumption
(PJ)
Energy
savings
potential
(%)
Integrated steelmaking 5.4 42.5 231 852 27%
Secondary steelmaking 2.7 56.1 154 311 49%
Total annual final energy
savings
Note: Numbers in the table are rounded off
3.9 98.6 385 1162 33%
Table 9 provides specific numbers on the energy efficiency improvement potential for energy us
in the steel industry. It must be noted that the numbers are dependent upon calculations with a
number of assumptions as mentioned in the system definition and methodology. These numbers
provide an indication of the energy efficiency potential of energy efficiency measure, if all
measures can be implemented as expected to our findings.
The reductions of energy are severe, as can be observed from Table 9. The total reduction of
energy in the US steel industry for 2006 was 33%, with a total energy saving of 385 PJ. What is
interesting is the secondary steelmaking savings potential which is large in terms of % but
smaller in terms of actual energy savings. The secondary steel making, which is younger
compared to the integrated steelmaking, has shown an increased attention in the past few
decades, concerning energy reduction and productivity increase. At the same time the
optimization of integrated steel production has shown a decrease in efficiency improvements.
A note for Table 9 is that the total final energy consumption of 1,162 PJ in 2006, differs from the
1.5 EJ found in Table 6. The discrepancy can be partly devoted to the method of assessing the
energy use. The calculation of Xu et al. (2012) is based on the energy use per process.
Discrepancies in energy consumption exist in more cases, the MECS result for final energy use
in the US steel industry in 2006 was 1,318 PJ, while the result for the AISI was 1,098 PJ (MECS,
2010, AISI, 2010). These discrepancies are severe, but the reason for these difference is not
easily traced. Sometimes the steel industry is defined differently. The discrepancies in energy
use will the Energy savings potential (%).
If only the cost-effective energy efficiency measures are taken into account, results will be
different. As already seen in Table 10, the cost-effective measures represent roughly 75% of the
total potential of the energy efficiency measures. Measures with a CCE value lower than the
average weighted energy price of 6.76 $/GJ are considered cost effective. The potential of the
cost-effective measures is summarised in Table 10.
39

Table 10. Cost-effective potential for energy savings in the US iron and steel making in 2006.
40
Energy
savings
(GJ/tonne)
Total
production
(MTonnes)
Total
Energy
savings
(PJ)
Current
energy
consumption
(PJ)
Energy
savings
potential
(%)
Integrated steelmaking 3.7 42.5 158 852 19%
Secondary steelmaking 2.5 56.1 142 311 46%
Total cost-effective final
energy savings
Note: Numbers in the table are rounded off
9.1.2 Carbon mitigation
3.0 98.6 300 1162 26%
With the reduction in energy consumption calculated for every measure, the reduction in carbon
emissions can be estimated. The carbon emissions have been calculated related to every process.
For example, an energy efficiency measure which reduces the energy consumption of the hot
rolling for secondary steel, will reduce the carbon emission by a certain amount, related to the
characteristics of the process (as already mentioned in the methodology). The hot rolling
process requires natural gas, and therefore every GJ saved will result in carbon savings, related
to the GJ of natural gas reduced. This way, for every energy efficiency measure the reduction of
carbon emissions could be calculated for both the total potential as well as the cost-effective
measures. Table 11 provides the results of the calculation of carbon savings from the
implementation of all identified energy efficiency measures in the US steel industry in 2006.
Table 11 Technical potential for carbon reductions in the US iron and steel making in 2006.
Total carbon
reduction
Total carbon
emissions
Carbon
reduction
potential (%)
Integrated steelmaking (MtC/year) 4.0 18.9 21%
Secondary steelmaking (MtC/year) 3.8 9.5 40%
Total technical potential for carbon
reductions (MtC/year)
Note: Numbers in the table are rounded off
7.7 28.4 27%
The carbon reduction potential (%) comes near the energy savings potential, but they are not the
same. A 30% reduction in energy use does not necessarily mean a 30% reduction in carbon
emissions. Apparently, a large number of identified energy efficiency measures are of the kind
with lower carbon emissions reductions. For example, natural gas has lower carbon emissions
per unit of energy (0.014 MtC/PJ) compared to electricity (0.044 MtC/PJ), so energy savings in
natural gas reduce carbon to a lower extent (EIA, 1995). As observed in Table 11, all the energy
efficiency measures combined have a total potential of 7.7 MtC reduction for the total US steel
industry in 2006.

Since the cost-effective energy efficiency measures for the US steel industry for 2006 have also
been identified, the carbon reductions of these measures can be calculated. The carbon
reduction potential of all cost-effective energy efficiency measures is shown in Table 12.
Table 12 Cost-effective potential for carbon reductions in the US iron and steel making in 2006.
Cost-effective
carbon
reduction
Total carbon
emissions
Carbon
reduction
potential (%)
Integrated steelmaking (MtC/year) 2.9 18.9 15%
Secondary steelmaking (MtC/year) 3.3 9.5 35%
Total cost-effective carbon
reductions (MtC/year)
Note: Numbers in the table are rounded.
6.2 28.4 22%
Again the cost-effective carbon reductions percentage is similar to the percentage for the costeffective
energy savings. The total cost-effective carbon savings for the US steel industry in 2006
is about 6.2 MtC, which is 22% of the total carbon emissions in the US steel industry.
Summarised, implementing energy efficiency measures in the US steel industry showed major
reduction potentials for both the energy use an carbon emissions. Even if only cost effective
measures are taken into account, the energy and carbon reductions add up to 26% and 22%,
respectively.
9.2 Further results on cost curves for the US steel industry
The research by Xu et al. (2012) had made some additional conclusions on the use of energy
efficiency measures in the US steel industry. In this paragraph the additional conclusions will be
elaborated on. The additional conclusions provide an insight in the sensitivity and importance of
key factors which led to a better understanding of analysing energy efficiency measures in the
steel industry. Also some suggestions for an advanced calculation of the CCE were provided, in
the MCCE method. The results and their influence on the results is shown in the next paragraph.
In Xu, et al. (2012), additional results for the US steel industry were included for 1994, 2002
and 2010. It was found that the technical potential savings in the integrated steelmaking
decrease in time, which is expected as the share of integrated steelmaking is shrinking. On the
other hand, energy savings in the secondary steelmaking are increasing. In 2006 and 2010 a
large increase was observed for the secondary steel industry, due to additional measures found
mostly for this type of steelmaking. The total final energy consumption of steelmaking was about
1564 PJ for 1994, 1224 PJ for 2002, 1162 PJ for 2006 and 997 PJ for 2010 (Xu, 2012). The total
potential energy savings fluctuate due to production differences in each year, the percentage of
final energy savings potential was increased. The technical potential of energy savings was
approximately 20% in 1994, 24% in 2002 and increased to 33% in 2006 and 2010. The first
conclusions from these results were in fact that both the production difference, industry change
41

and amount of identified energy efficiency measures had a major influence on the results.
Therefore these are the main factors used for the assessment of energy efficiency measures in
the steel industry in other world regions.
9.2.1 Industry structure change
In order to check the influence of industry change, a data set of the year 2002 was used while
different industry structures were introduced. The same set of measures was used to identify
what the influence of the industry structure was on the identified potential for energy efficiency
measure. The industry change from 39% secondary steel production and 61% integrated steel
production, to 61% secondary steel production and 39% integrated steel production (as was the
case from 1994 to 2010) could change the energy savings potential of energy efficiency measures
from about 3.7 GJ/tonne of steel to 3.0 GJ/tonne steel if the same data set (set of measures) was
used, which is a decrease of 19% (Xu, 2012). It was concluded, that with decreasing share of
integrated steel compared to secondary steel over the years, and lower level of energy intensity
in secondary steel production, the total potential energy efficiency measures is (with the same
set of energy efficiency measures) exhibited a decreasing trend (Xu, 2012).
It should be noted, that the use of secondary steel production (with EAF) is still preferred in
terms of energy efficiency. But, the potential of the identified energy efficiency measures in this
report, is lower in secondary steelmaking because of its lower energy intensity compared to
integrated steelmaking.
Next, the difference in baseline data was investigated. The baseline data is for example the
amount of sinter and coke production per ton of steel, or the need for hot rolling of steel.
Differences in baseline data is likely due to a combination of the varying factors, such as annual
production that varied over the years, availability and utilization of technologies in operation,
energy usage in productions, and/or material efficiency (lower material efficiency for hot rolling,
means an increased crude steel production per tonne of end product; Xu, 2012). The baseline
data, where the use of for example coke or sinter were changed showed different results. In the
calculation where the baseline changes were highest (comparison between 1994 and 2006), the
difference in results was only about 0.2 GJ/tonne, about a 6% difference (Xu, 2012). It was
concluded that the change in baseline data was minor, because the amount of sinter needed for
steel production can never change drastically. In other words, you can make about the same
amount of iron from one tonne of sinter. Since the amount of factors influencing the baseline
data are large and the effects of changes in baseline data do not have large influence on potential
of energy efficiency measures, they will be excluded in the further assessment of energy
efficiency measures in the world steel industry.
9.2.2 Cost curve for carbon mitigation
Next to using cost curves for the assessment of energy reduction in the steel industry, cost
curves for carbon mitigation had been introduced as well. The cost curve shows the carbon
mitigation against the Cost of Carbon Reduced (CCR) in a cost curve. In Figure 8 an example of
such a cost curve is shown, here the 2006 cost curve is shown for the US steel industry. From
this cost curve it is obvious that in many cases the reduction of carbon emissions is quite
42

expensive per tonne of carbon. The scale on the Y-axis in Figure 8 is adjusted, but for some
measure up to CCR can go up to $10,000 per tonne of carbon. This is for one extreme case for
coal moisture control for coke making. For that case only minor carbon reductions (0.1
kgC/tonne of steel) can be realised while the investment costs are high ($6.8/tonne of steel).
On the other hand, no general price of carbon mitigation is known. So, the determination of
cost-effective measures based on a CCR cost curve is therefore impossible. The other way
around, when the cost-effective carbon mitigation is known, based on the assessment of costeffective
final energy savings, a price for carbon mitigation can be determined. We found the
cost-effective carbon reductions for the US steel industry in 2006 was about 6.2 MtC. Associated
with cost-effective carbon mitigation is a CCR of about 200 $/tonne of carbon reduced (Figure
8). This was found to be an indication for a carbon price tag, concerning the implementation of
energy efficiency measures (Xu, 2012). Apparently, the application of energy efficiency measures
in the US steel industry is cost effective up to a cost of carbon reduction of 200 $/tonne of
carbon reduced.
Figure 8 Cost curve of cost of carbon reduced for the US steel industry in 2006. (Xu, 2012)
For further research into the rest of the world steel industry, the use of CCR is not expected to
provide new insights for other world regions (apart from the carbon price tag, which might be
different in each region). However, at this moment there is not an obvious conclusion in why the
apparent carbon price tag is around 200 $/tonne of carbon. Investigating carbon price tags
based on cost-effective energy efficiency measures is therefore not expected to be valuable result
for other world regions.
43

9.2.3 Modified cost of conserved energy
Next to the CCE calculation, a suggestion for an advanced method of calculation was introduced
for the energy efficiency measures assessment in the US steel industry. The calculation for the
so-called modified cost of conserved energy (MCCE) differs from CCE calculation as the energy
cost savings are included in the calculation. A new method is proposed to include different
energy types for each of the energy efficiency measures. The main reason for including energy
cost savings in the calculation is the following.
The energy efficiency measures are influenced by the wide variety of energy or fuel types used in
the production processes of the steel industry. For example, energy reduction in the blast
furnace could reduce the need for coal, while an efficiency measure in an EAF could reduce only
electricity. The savings from reduced use of coal are in general less cost effective compared to
electricity savings, mainly due to much lower price for coal (e.g., coal price in 2006 was 3.36
$/GJ, far lower than electricity price of 12.55 $/GJ)(EIA, 2009). Using the concept of CCE to
identify cost-effective measures applicable to the steel sector, we would compare CCE value with
the weighted average energy price for the steel industry even if the fuel or energy type applicable
to an individual measure is homogeneous (e.g., for 2010 a weighted average energy price was
calculated at 6.76 $/GJ while prices per energy type are substantially lower or higher). The
drawback of using traditional CCE and weighted average energy price to characterize costeffectiveness
of an individual measure is that if the measure operates only with electricity, its
cost-effectiveness would be biased (unfavourably) when using weighted fuel price to compare,
because the weighted price is inherently lower than electricity price. In other words, the
measure that was in fact cost-effective (if the right energy price was used) could be otherwise
identified as not cost effective. On the other hand, for a measure that uses only coal to operate,
its degree of cost-effectiveness could be inflated when its CCE is compared with the weighted
average energy price that is higher than coal price itself.
While conventional CCE is a very good index when energy supply (thus price) is homogeneous,
it may lack readiness for evaluating cost-effectiveness of individual measures in different years.
Using the conventional CCE, one has to compare the CCE value with weighted average of energy
price in a particular year. Because fuel prices and energy supplies changed from year to year, the
weighted average of energy price would be a moving target across years, making it more difficult
to readily evaluate cost-effectiveness of measures.
In the following, we have developed and defined an MCCE concept, which incorporates the
annual energy cost savings in the calculation of conventional CCE to address the impacts of
energy and fuel-type variations on actual cost-effectiveness of a measure. Equation 3 shows the
definition of MCCE:
44

Where
∙
∆
MCCE = Modified cost of conserved energy for an energy-efficiency measure (or
mitigation option), in $/GJ
I = Capital cost of mitigation option ($)
q = Capital recovery factor (yr -1 ), see Equation (2)
S = Annual energy cost savings from energy efficiency measure ($/yr)
M = Annual change in monetizable non-energy costs from O&M changes ($/yr)
B = Annual additional non-energy cost benefits ($/yr)
∆E = Annual energy savings (GJ/yr)
Equation 3 is similar to Equation 2 (methodology paragraph), with the difference being that an
additional factor – annual energy cost savings (S) is incorporated to account for energy cost
benefits. The annual energy cost savings are calculated by combining cost savings from using
actual fuels and electricity use for each production process with respective prices in a year.
Because S is expected to be a positive value for each measure, the value of MCCE is expected to
be reduced compared to the conventional CCE value for the same measure. In some cases, the
MCEE can become a negative value. In fact, if the MCCE value of a specific measure is no
greater than zero (“0”), then the measure is considered to be cost effective; in contrast, if the
MCCE value of a specific measure is greater than zero, then the measure is not considered to be
cost effective based upon the available data input. Using MCCE values in comparison with zero
will allow straightforward evaluation of cost-effectiveness of each specific efficiency measure
regardless of whether or not fuel types and prices are homogeneous, because its calculation
already accounts for variations in fuel types, amounts, and energy prices associated with each
individual measure. In Figure 9, an example is given of multiple MCCE cost curves for the US
steel industry in 2010 with different prices per energy type, and their influence on the shape of
the cost curve. The calculation of different cost curves was made to show the influence of energy
prices of single energy types on the MCCE calculation. What is seen in the figure, is the influence
of natural gas seems to have the largest impact on the shape of the costs curve. This is mainly
because a large number of identified measure are included on processes using natural gas as a
main fuel input.
(3)
45

Figure 9 The sensitivity of MCEE cost curves due to energy-price changes in the US steel industry in 2010
(Xu, 2012)
The MCCE method does produce different results, though the amount of cost-effective energy
savings is similar, as can be seen in the intersection of each curve in Figure 9 with the X-axis. In
Xu et al. (2012) the largest difference for the total cost-effective energy savings for the US steel
industry were in 2006. The results for the CCE method and MCCE method were 300 PJ and 322
PJ, respectively. So, a difference of about 7% was observed for using either of these methods.
Moreover, since this new method was introduced in the report of Xu et al. (2012) and has not
been peer reviewed, this method will not be used for the assessment of energy efficiency
measures in other world regions. Also, the type of fuel used per process for other world regions
is not known, while for the US, data could be found about fuel types in the steel industry
processes. The introduction of the MCCE in the US showed a more accurate assessment of
energy efficiency measure, and a more easy comparison of data from different years.
Summarised, the use of MCCE provides more accurate determination of cost-effective measures,
using energy prices of different fuel types in calculation. Apart from the lack of reviews on this
method, the results of MCCE calculation compared to CCE calculation influence results of costeffective
savings up to 7%. The method will therefore not be used in further calculations for the
world steel industry.
46

9.2.4 Including or excluding other non-energy benefits
Another result for the research of Xu et al. (2012) was to look at the influence of other nonenergy
benefits, like productivity gains or reduced labour. In 2006, 25 measures had identified
other non-energy benefits. Overall for 2006, including other non-energy benefits in the cost
curves for the iron and steel industry can significantly decrease the total cost of conserved
energy for the measure with the highest CCE value, from 193 $/GJ to 86 $/GJ to achieve the
same total energy savings of 3.9 GJ/tonne steel in 2006 (Figure 10).
Figure 10 Cost curves for inclusion and exclusion of other non-energy benefits in the US iron
and steel industry (2006)
In summary, inclusion of other non-energy benefits of implementing efficiency measures can
significantly reduce the CCE values and change ranking orders of the CCE value of individual
measures, while achieving the same level of conserved energy. In any further calculations for
steel industry in other world regions, non-energy benefits will always be included in calculation.
9.2.5 Changes in discount rate
The discount rate used for the calculation of CCE (and MCCE) values was determined in the
methodology, a discount rate of 30% would be used for all calculations. However, one could use
a different discount rate. For example, a discount rate of 10% or 20% will reduce the CCE value
of each measure, thus measures will be considered more cost effective.
Logically, a higher discount rate would result in a higher CCE, making the efficiency measure
less cost effective. Therefore, a higher discount would correspond to a smaller set of costeffective
measures than a lower discount rate does. Table 13 presents the total final energy
47

savings from applying all efficiency measures identified for the US, and those from applying
cost-effective efficiency measures for different discount rates..
Table 13 Final energy reductions from cost-effective energy efficiency measures for multiple discount
rates in the US steel industry in 2006.
Final energy reduction by cost-effective measures for different discount measures
Total Final energy savings all identified measures (PJ/year) 385
Total Final energy savings cost-effective measures (discount rate 10%) (PJ/year) 371
Total Final energy savings cost-effective measures (discount rate 20%) (PJ/year) 310
Total Final energy savings cost-effective measures (discount rate 30%) (PJ/year) 300
As can be seen in Table 13 the influence of discount rates is significant on the amount of energy
savings from cost-effective measures. Over 95% of the total potential energy savings become
cost-effective when a discount rate of 10% is used. Choosing a discount rate is therefore very
important because it significantly affects the outcomes from quantifying cost-effectiveness of
energy efficiency measures. The total cost-effective energy savings increase with about 21%.
In summary, while variations in discount rates do not affect the magnitudes of total potential
savings from efficiency measures in any given year, costs of conserved energy increase greatly
with the increase in discount rates. A higher discount rate results in an overall increase in the
total cost of conserved energy. Still, for any further calculations for steel industries in other
world regions, a discount rate of 30% will be assumed.
9.2.6 Summary of other conclusions in the US steel industry.
From the assessment of energy efficiency measures for the US a number of conclusion have been
drawn on the determination of the total potential of energy efficiency measures and the
sensitivity of assessing energy efficiency measures. First of all, the inclusion of other non-energy
benefits has a large influence on the cost-effective energy savings. For 25 measures other nonenergy
benefits were included, while for a lot of measures other benefits had not been identified.
If other non-energy benefits were identified, the influence on the results of cost-effective energy
efficiency measures can be severe, as was demonstrated in Figure 10. Also the use of CCE
method for calculating the potential of energy efficiency measure can be discussed, where the
MCCE provides more accurate results on the cost-effectiveness, if sufficient data is available.
The cost-effectiveness of an individual measure is highly dependent on the selected discount
rate used in the calculation. Quantitatively, the effects of these findings include about a 6%
uncertainty in the identification of total potential of efficiency measures, and 8% for costeffective
measures, or up to 95% cost-effective measures with lower discount rates. Presumably
more uncertainties can be included, like assumption on penetrations rates of technology, but
these could not be easily quantified.
Also the assessment of energy efficiency measures in the US lead to the assumptions on the
major influencing factors for the steel industry in different world regions and on the potential of
48

these energy efficiency measures in those regions. From the assessment in the US it was clear
that the production, industry structure and average weighted energy price were the most
important factors influencing the potential of (cost-effective) energy efficiency measures.
9.3 Results of energy efficiency measures in other world regions
Now we have explained the assessment of energy efficiency measures in the US steel industry,
the application in other world regions will be assessed. This paragraph will demonstrate the
potential of energy efficiency measures in other world regions, based on a number of key factors
influencing the potential of (cost-effective) energy efficiency measures.
As mentioned earlier, the selected world regions other than the US are Western Europe (WEU),
Former Sovjet Union (FSU), India (IND), China (CHI), Japan (JAP) and Central- and South
America (CSA). For each of these regions an assessment is made of the set of energy efficiency
measures which generated the previous results for the US steel industry. In order to provide an
estimation of using the same set of measure for the other world regions a number of adaptation
to the input data have to be made.
9.3.1 Production of steel in different world regions
First the required production data for assessing energy efficiency measure in different world
regions had to be collected. Basic production data were already introduced in chapter ‘World
steel production in different world regions’. The production data needed, consist of the total
production of steel, the industry structure and the total energy use of the steel production. Data
are provided for each of the world regions considered for this research. The results of the
production data have been summarised in Table 14.
Table 14 Steel industry structure and steel production in different world regions in 2006
Region
Steel Industry structure
United
States
Western
European
Union
Former
Soviet
Union
India China Japan
49
Central
and
South
America
Integrated Production (BOF) 42% 57% 59% 42% 90% 74% 60%
Secondary Production (EAF) 58% 43% 17% 56% 10% 26% 38%
Integrated Production (OHF) 0% 0% 24% 2% 0% 0% 0%
Total % 100% 100% 100% 100% 100% 100% 99%
Steel production
Total production (Mt/year) 99 173 120 50 419 116 45
Source: Worldsteel, 2011

Quantities of production in the secondary and integrated industries for every region have been
calculated based on the total production and the industry structure. For example, the Chinese
steel industry had about 90% of their steel produced in the integrated steel industry in 2006,
which is 377 million tonnes of their total steel production of 419 million tonnes. The large share
of integrated steel in China is probably related to their large amount of coal resources. In India
the use of secondary steel production is large since the growth of DRI based steelmaking has had
major growth in the last decade.
As seen in Table 14 the share of steel production in China is the largest of the selected regions.
The US steel industry is one of the smaller regions in terms of steel production, since four other
regions produce more steel. Based on production difference the other world regions will have an
increased potential for energy efficiency measures, merely based on the larger production.
Dissimilar from the production, the energy consumption for each world region cannot be
divided easily in integrated or secondary steelmaking. So for the energy use, an assumption will
have to be made. The energy use for integrated and secondary steel will therefore be estimated,
based on the baseline of energy in the US steel industry. In the US steel industry, the integrated
steel production is 3.6 times as energy intensive compared to the secondary steelmaking. A
similar deviation of energy use is to estimate the total energy use of each industry type in other
world regions. In other words, based on the baseline of US, an estimate was made for the energy
use per industry structure. In Table 6, in chapter ‘world steel production and different world
regions’, the total final energy consumption per region has already been mentioned. In Table 15
the total energy use is included, and an estimated energy use for each industry type in every
regions.
Table 15 Final energy use in steel industries per world region in 2006.
Region
Estimated integrated final
energy use
Estimated secondary final
energy use
50
United
States 1
Western
European
Union 3
Former
Soviet
Union 3
India 3 China 2 Japan 3
Central
and
South
America 3
852 2,325 3,108 1,168 9,234 1,778 1,069
311 484 176 410 283 172 189
Total energy use 1,162 2,809 3,284 1,578 9,518 1,950 1,258
Note: numbers are rounded off
Sources: 1 Xu, 2012, 2 IEA, 2007, 3 Hasanbeigi, 2006
The calculation of Xu, et al. (2012) is based on the energy use per process. In order to determine
the energy use per process for the other world regions, the data of IEA, Hasanbeigi and
Worldsteel is used as a starting point. With the industry structure the production of each type of
steelmaking is analysed. The energy distribution of both the integrated and secondary
steelmaking in the US is then used for the determination of energy use of each steel industry
type in the other world regions. The change in production already has a major effect on the
energy consumption of the other world regions.

As seen in Table 15, China has the largest energy use compared to the other world regions. China
alone is responsible for about 38% of the total final energy use in the world steel industry. What
is also noticeable is the energy efficiency in Japan. While three other regions produce less
secondary steel compared to Japan, Japan is the region with the lowest secondary steelmaking
energy use.
Related to the energy use, carbon emissions could be calculated for steel industries in the
selected regions. The result of the calculated carbon emissions for these industries in 2006 is
shown in Table 16.
Table 16 Energy related carbon emissions in different steel industries per world region in 2006.
Region
Integrated carbon emissions
(MtC/year)
Secondary carbon emissions
(MtC/year)
Total carbon emissions
(MtC/year)
Note: numbers are rounded off
United
States
Western
European
Union
Former
Soviet
Union
India China Japan
Central
and
South
America
24 52 69 26 205 39 24
12 15 5 13 9 5 6
36 66 74 38 213 45 29
As seen in Table 16 again China is largest producer of carbon emissions, which was expected
compared with the energy use in Table 15. What is interesting is the carbon emission in the US
compared to CSA. In terms of energy use, CSA has a larger energy use than the US. However, in
terms of carbon emissions, the US has a larger amount of emissions. This is related to the large
amount of electricity used for EAF production. As mentioned in paragraph 8.1.2, the carbon
emissions per GJ of electricity are large, therefore a lower energy use in the US still results in
higher carbon emissions.
9.3.2 Economics of different world regions
Now to estimate (cost-effective) energy savings, some economic data are needed for each region
as well. The weighted average energy price for steel production in each of the world regions is
the first to be determined. From our analysis of the energy deviation per industry type,
estimated in Table 15, the electricity use of each of the world regions could be estimated. The
distribution of fuel consumption in every world region was expected to be similar to the US,
since no further indication could be found. The energy prices for each energy type for every
region were found. The average weighted energy price for steel production in different world
regions is shown in Table 17.
51

Table 17 Average weighted energy price for steel industry, per world region in 2006
Region
52
Average
Weighted
energy price
($/GJ)
China 6,13
Western European Union 8,12
Former Soviet Union 5,13
Japan 7,97
United States 6,76
Central and South America 6,50
India 4,94
Sources: DECC (2012), BP (2012), Sidorenko (2010), Henderson (2011), IMF (2012), ANEEL (2007), IEA
(2010).
Apparently prices of energy in different regions can differ quite a bit. One influence in the
weighted energy price is the usage of different energy types. Some regions have larger shares of
electricity use compared to others, since there is a larger share of secondary steelmaking. Some
cases for the determination of fuel prices for different fuel types have been difficult. In India the
use of APM (Administered Price Mechanism) for the public sector is used. Part of the steel
industry in India is in the public sector, while the other part of the steel industry has to use the
market price (Public Enterprise Survey, 2011). In this case the average price of different sources
was used as the price for a certain energy type. Another example is the natural gas price for
Russia, for which multiple sources were found, with different prices. Here again, the average of
both sources were used to determine a weighted average energy price. (Sidorenko, 2010, DECC,
2012)
The weighted average energy price is one of the things needed to determine cost-effective
measures for the steel industry in different world regions. Next, the financial data on the energy
efficiency measures might differ from the situation in the US. So both the investment and
operation costs determined for each energy efficiency measure for the US, will have to be
changed to the other regions situation. In order to include a reasonable estimation here, the PPP
index of each region is used to determine the investment and operational costs for other world
regions. The PPP indexes are shown in Table 7 (Paragraph 7.2).
As mentioned in the methodology, the PPP index will only be used for 30% of the investment
and operation costs. For every world region, now the financial data is updated with PPP index to
calculate the CCE value. Subsequently, the total and cost-effective energy savings for these world
regions can be calculated, as shown in Table 18.

Table 18 Total potential and cost-effective energy savings in steel industry in different world regions in
2006
Region
Total energy savings potential
(PJ/year)
United
States
Western
European
Union
Former
Soviet
Union
India China Japan
Central
and
South
America
Integrated final energy savings 236 544 567 135 2087 468 162
Secondary final energy savings 158 212 63 89 126 87 55
Total final energy savings 385 756 620 215 2170 548 200
% of total final energy
consumption
Cost-effective energy savings
(PJ/year)
Integrated cost-effective energy
savings
Secondary cost-effective energy
savings
Total cost-effective energy
savings
% of total final energy
consumption
Note: Numbers are rounded.
33% 27% 19% 14% 23% 28% 16%
158 374 363 87 1400 323 103
142 193 57 79 113 79 46
300 568 421 166 1513 402 148
26% 20% 13% 11% 16% 21% 12%
Some interesting results came from the calculation for different regions. As expected, China has
the largest potential for energy savings, since it also has the highest production rates. India has
the lowest energy savings potential, according to our calculations. This is not completely as one
would expect, since India is one of the most energy intensive steel producers. Also the total
potential of the identified energy efficiency measures is only 14% for India. Since the identified
energy savings for energy efficiency measures were determined in GJ/tonne, it is logic that the
potential of energy efficiency will decrease in countries which have lower production rates.
Also the result for the US show the highest % of total final energy consumption in both the total
potential as well as the cost-effective energy efficiency measures. This could be the result of
discrepancy between sources. For the rest of the regions, the energy consumption data is based
on the IEA (2007) data, while the US data is based on the results by Xu et al. (2012). According
to IEA data the US steel industry has a total energy consumption of 1.5 EJ instead of the 1,162 PJ
now used in the calculations. If IEA data would be used, the % of total final energy consumption
for the US would be 26% and 21% for the total energy savings potential and the cost-effective
savings, respectively.
53

In order to visualize the energy savings, a bar diagram for both the total potential and costeffective
energy savings from energy efficiency measures is provided in Figure 11. In the bar
diagram the results of China having the largest potential becomes even more clear. The other
regions with relatively high potentials are the Western European union and Former Sovjet
Union because of their relatively high production.
Figure 11 World steel energy savings for different world regions in 2006
From the energy savings potential we calculated the carbon mitigation of the energy efficiency
measures in different parts of the world. In Table 19 the potential of each region for the
mitigation of carbon emissions is shown. Not surprising is the fact that in China the potential is
the largest, which would be expected, since the result is based on the energy savings, for which
the conclusion was similar.
54

Table 19 Total potential and cost-effective carbon reductions in steel industry in different world regions
in 2006
Region
United
States
Carbon emission reductions (MtC/year)
Western
European
Union
Former
Soviet
Union
India China Japan
Central
and
South
America
integrated carbon reductions 4 9 10 2 36 8 3
secondary carbon reductions 4 5 2 3 3 3 2
total carbon reductions 8 14 12 5 38 11 5
% of total carbon emissions 27% 21% 16% 13% 18% 24% 15%
Cost-effective carbon emission reductions (MtC/year)
integrated cost-effective carbon
reductions
secondary cost-effective carbon
reductions
total cost-effective carbon
reductions
3 7 7 2 25 6 2
3 4 2 2 3 2 2
6 11 9 4 28 8 4
% of total carbon emissions 22% 17% 12% 10% 13% 18% 12%
Note: Numbers are rounded off
The implementation of the identified energy efficiency measures has a significant effect on the
energy consumption and carbon emissions in the different world regions. In the range of 11% to
26% of the final energy consumption can be saved in the different world regions, while 10% to
22% of the carbon emissions can be reduced. Another obvious conclusion is that the integrated
steel production has a far larger potential in energy and carbon savings compared to the
secondary steelmaking. For the US the difference of secondary and integrated energy savings
was relatively small, but for China and the Formet Sovjet Union the integrated steelmaking
energy savings potential is close to or over 90% of the total potential energy savings.
9.3.3 Further results
The results have been generated with a set of energy efficiency measures initially set up for the
US steel industry. It was expected that the results for other world regions would therefore only
be an estimate of energy efficiency potential, even though the energy efficiency measures have
been altered with a number of assumptions, for example the PPP index.
In case of the PPP index it was assumed that index would only be applicable for 30% of the
investment and operational costs. This assumption was based on the fact that technologies for
energy reduction are likely not to be produced and traded locally. So for the implementation of
55

energy efficiency measures in China, expertise from the US or EU might be required. In that
case the PPP index would not suffice for calculating the costs of each measure.
Then, additional results were produced with a PPP index of 0%, which assumes the prices for
implementing and operating energy efficiency measures in each world region is the same. Next,
the use of a PPP index of 60% was assessed. So the influence of PPP-index on the prices of
efficiency measures is increased. The results of these scenarios are shown in Table 20.
Table 20 The influence of different PPP-index on the amount of cost-effective energy savings from energy
efficiency measures for different world regions in 2006.
Region
Cost-effective final energy
savings (0% PPP)
Cost-effective final energy
savings (30% PPP)
Cost-effective final energy
savings (60% PPP)
Note: numbers are rounded
56
United
States
Western
European
Union
Former
Soviet
Union
India China Japan
Central
and South
America
- 581 431 176 1466 409 163
- 581 431 176 1513 409 166
- 581 463 184 1597 408 169
Apparently using the PPP index to a larger degree will have an effect on the cost-effective energy
efficiency measures. Especially world regions for which the PPP-index deviates by great extend
from the US, the effects of using 0% or 60% of the PPP-index increase the cost-effective savings
by about 5-10%. Conversely, for other world regions with PPP-index close to 1, the effects of this
index are minor, if at all affected.
Next, the assessment of the US steel industry provided some additional findings on the energy
efficiency measures. First, the baseline data in the US could change the potential by about 6%.
Second, the influence of MCCE on the results for cost-effective measures was found to be of
influence up to 8%. Finally, the use of a decreased discount rate could increase the amount of
cost-effective energy savings up to 95% of the total potential energy savings. Taking these
quantifiable uncertainties into account a range could be made for the total potential and costeffective
energy savings for each region, as seen in Table 21.

Table 21 Range of energy savings in the different steel industries in 2006, taking defined uncertainties
into account.
Region
Final energy
savings potential
in PJ (range)
Cost-effective
final energy
savings in PJ
(range)
United
States
385
(362-407)
300
(277-365)
Note: Numbers are rounded off.
Western
Europea
n Union
756
(713-802)
581
(538-719)
Former
Soviet
Union
620
(584-657)
431
(399-589)
India China Japan
215
(202-228)
176
(163-204)
2170
(2047-2300)
1513
(1401-2061)
548
(517-581)
409
(378-521)
Central
and
South
America
200
(188-212)
166
(154-190)
Taking these uncertainties into account, the results for each region are provided with a rather
large range. For China the difference between the lower and higher range for cost effective
energy savings is 660 PJ. The range of cost-effective energy savings in China thereby ranges
from 15% to 22% of the total energy use in the steel industry in China.
9.4 World steel energy savings potential
With the energy and carbon savings for the different world regions, an end result can be made
for the application of energy efficiency measures in the world steel industry. Results for the
different regions will be accumulated for a global estimation.
9.4.1 Energy savings potential
In order to provide an indication of the world steel energy savings of energy efficiency measures
it will first have to be assumed what the energy savings potential in other parts of the world will
be. These are the world regions not selected for this research. For the other regions it is assumed
that the average savings in selected world regions will also be used in other world regions.
According to the energy consumption in Table 6, the share of other world regions is calculated.
The share of other world regions, multiplied by the total energy savings of the selected world
regions provided the result for this final region. The cost-effective energy savings for other world
regions have been calculated similarly.
57

From the total of all the energy saving potential Figure 12 could be generated.
Figure 12 World steel energy savings potentials from energy efficiency measures in 2006
Apparently, the application of energy efficiency measures all over the world resulted in a total
potential energy savings of about 5.7 EJ in 2006 (5.3-6.0 EJ when uncertainties are included).
The cost-effective energy savings for the world steel industry are about 4.1 EJ in 2006 (3.5-5.4
EJ, uncertainties included). The total potential for energy efficiency measures is about 23% of
the world steel final energy consumption. The cost-effective energy savings add up to about 16%
of the final energy consumption.
To put these numbers in perspective, the energy reductions possible by of all identified
measures equal to the output of about 226 coal power plants of 1 GW. The energy savings for
only cost-effective measures equals the output of about 163 coal power plants of 1 GW. So the
potential energy savings for the world are definitely substantial.
Similarly the carbon mitigation can be calculated for energy efficiency measures in the world
steel industry. In Figure 13 the total carbon reductions for the world steel industry are
demonstrated.
58
Energy savings (PJ/year)
6000
5000
4000
3000
2000
1000
0
Total final energy
savings
Total cost effective
energy savings
Other
Central and South America
Japan
China
India
Former Soviet Union
Western European Union
United States

Carbon reductions (MtC/year)
120
100
80
60
40
20
0
total carbon reductions total cost effective
carbon reductions
Figure 13 World steel carbon reduction potential from energy efficiency measures in 2006
For carbon mitigation the application of energy efficiency measures all over the world resulted
in a total potential carbon reduction of about 107 MtC in 2006 (101-113 MtC, uncertainties
included). The cost-effective carbon reduction for the world steel industry are about 82 MtC in
2006 (76-101 MtC, uncertainties included). The total potential for energy efficiency measures is
about 20% of the world steel carbon emission mitigation. The cost-effective carbon reductions
add up to about 15% of the total carbon emissions.
9.4.2 Comparison to other analysis
Other
Central and South America
Japan
China
The potential of energy savings in the existing iron and steel industry has been analysed by the
International Energy Agency. The results of this study have been summarised in Figure 14. The
methodology for this analysis could not be retrieved. Apparently the analysis encompasses the
CO2 reduction potential of best available technologies (BAT). A few key technologies have been
identified, as well as blast furnace and steel finishing improvements. According to the analysis,
the potential of these technologies for the world steel industry is about 340 Mt of CO2 (about 93
MtC; IPCC, 2007).
India
Former Soviet Union
Western European Union
United States
59

Figure 14 CO2 reduction potential of best available technology by IEA analysis. (OECD, 2008)
The total potential of energy efficiency technology for the world steel industry is thus
comparative to the results produced in our analysis. A total potential of 107 MtC in our analysis,
93 MtC in the analysis of IEA. On the other hand, the results of the IEA use reference year 2004,
while 2006 is reference in our calculations. Also, the analysis of the IEA might differ in
methodology, compared to our calculations. Another note is the IEA has selected a five specific
number of technologies, and two assemblies of measures for blast furnace and steel finishing. In
our analysis we use a wide variety of measures and included calculation of merely cost-effective
measures. Apart from these differences, both analyses provide results in a similar range.
9.5 Key assumptions and gaps in knowledge
As already mentioned the energy use and emissions of a steel plant are significantly affected by
several factors such as technology, plant size, and quality of raw materials (Xu, 2010). This
research only encompassed energy efficiency measures in the steel industry. Other factors, like
plant size, are not included, but may as well be effective. Moreover, the data on energy efficiency
measures only encompasses the measures which could be identified in literature sources and for
which sufficient data could be found. In practice, additional energy efficiency measures may be
available.
60

An identified deficit in data is in the penetration rate of energy efficiency measures. Data on
penetration rates of certain technologies is not readily available, therefore only approximations
of this factor can be made.
We assumed the energy efficiency measures will have about the same energy and carbon
mitigation effects in the selected world regions as in the US. Only total production, energy use
and industry structure is changed. No influence of raw material quality is included. Also the
baseline data for other regions is assumed to be similar to the US, so the amount of sinter or
coke produced per tonne of steel is expected to be about the same in any region. In reality, these
results might be different in every region and have a significant effect on the total energy
savings, as indicated in paragraph ‘further results on cost curves for the US steel industry’.
Another difficulty is the assumption that effects of the energy efficiency measures will be similar
in every world region. Some of the energy efficiency measures will have a similar effect in every
world region, while others will have a different effect in other world regions. However, to
determine the effects of certain measures in other regions, each measure should be assessed for
the different region. Taking into account, the limited amount of time for this research, it will not
be viable to do an assessment of each measure in every world region.
Further factors for other regions which have not been taken into account for this research are
the technological advancements of each region. Japan might for example already have
implemented most of the energy efficiency measures or the Former Soviet Union which still
operates a number of out-dated open hearth furnaces. Also the quality of raw materials is not
taken into account, which is difficult to determine, but taken into account for this research.
The identification of key factors was based partly in the research done in the US. This research
was in itself not peer reviewed or reviewed by other scientist. Though, we used the conclusion of
influence of industry change as a key factor for assessing the world steel industry. Although the
assumption seems legit, it is a large leap from one conclusion, and right away using it as a base
for further research.
61

10 CONCLUSIONS
Through characterizing energy-efficiency technology costs and improvement potentials, energy
savings and carbon mitigation for energy efficiency measures were developed and presented for
the world steel industry in 2006. The major sources of influence on the application of energy
efficiency measures are: the baseline year, discount rate, energy price, production, industry
structure (e.g., integrated versus secondary steel making and the number of operating plants),
efficiency measures, share of iron and steel production to which the individual measures can be
applied, and inclusion of other non-energy benefits.
First, studies for the US steel industry concluded in two cases: By application of all identified
measures or application of only cost-effective measures. If only the cost-effective energy
efficiency measures are taken into account, usually the cost-effective measures represent
roughly 75% of the total potential of the energy efficiency measures. Measures with a CCE value
lower than the average weighted energy price of $6.76/GJ for the US, are considered cost
effective.
The total energy savings potential of energy efficiency measures in the US in 2006 was 385 PJ
per year. For cost-effective measures the energy savings would total at 300 PJ. From the energy
savings the carbon reductions could be calculated. The total potential carbon reductions from all
the identified measures in the US in 2006 was 7.7 MtC. The cost-effective carbon reductions for
the US steel industry in 2006 is about 6.2 MtC.
During the analysis of energy efficiency measures in the US steel industry also some additional
results were discovered. First, the concept of MCCE was developed in this report, which help to
accurately characterize the cost-effectiveness of individual measures and facilitate the
identification and comparison of efficiency measures. Second, the inclusion of other nonenergy
benefits has a large influence on the cost-effective energy savings. For 25 measures other
non-energy benefits were included, while for a lot of measures additional benefits had not been
identified. Third, the cost-effectiveness of an individual measure is highly dependent on the
selected discount rate used in the calculation.
Also the assessment of energy efficiency measures in the US led to the assumptions on key
factors influencing the energy efficiency measures for the steel industry in different world
regions. From the assessment in the US, it was clear that the production, industry structure and
average weighted energy price were the most important factors influencing the potential of
(cost-effective) energy efficiency measures. With these key factors, the analysis was expanded to
assessing the world steel industry. For the world steel industry the effects of energy efficiency
measures in different world regions were estimated using the key factors and adaptation of data
with the PPP-index. The results of these estimations are provided in Table 22.
63

Table 22 Summary of results of energy savings and carbon reductions in steel industry for different world
regions in 2006
Region
Total final energy
savings
% of total final energy
consumption
Total cost-effective
energy savings
% of total final energy
consumption
total carbon
reductions
% of total carbon
emissions
total cost-effective
carbon reductions
% of total carbon
emissions
64
United
States
Western
European
Union
Former
Soviet
Union
India China Japan
Central
and
South
America
385 756 620 215 2170 548 200
33% 27% 19% 14% 23% 28% 16%
300 568 421 166 1513 402 148
26% 20% 13% 11% 16% 21% 12%
8 14 12 5 38 11 5
27% 21% 16% 13% 18% 24% 15%
6 11 9 4 28 8 4
22% 17% 12% 10% 13% 18% 12%
The use of the PPP-index was only applied to a certain extent, in all calculations 30%. World
regions for which the PPP-index deviates by great extent from the US, the effects of using 0% or
60% of the PPP-index increase the cost-effective savings by about 5-10%. However, for other
world regions with a PPP-index close to 1, the effects of this index are minor, if at all affected.
From the combined effects for the different world regions, and an assumption for other world
regions which had not been selected for analysis, the results for the world steel industry could be
obtained. Apparently, the application of energy efficiency measures all over the world resulted in
a total potential energy savings of about 5.7 EJ in 2006 (5.3-6.0 EJ when uncertainties are
included). The cost-effective energy savings for the world steel industry are about 4.1 EJ in 2006
(3.5-5.4 EJ, uncertainties included). The total potential for energy efficiency measures is about
23% of the world steel final energy consumption. The cost-effective energy savings add up to
about 16% of the final energy consumption.
For carbon mitigation the application of energy efficiency measures all over the world resulted
in a total potential carbon reduction of about 107 MtC in 2006 (101-113 MtC, uncertainties
included). The cost-effective carbon reduction for the world steel industry are about 82 MtC in
2006 (76-101 MtC, uncertainties included). The total potential for energy efficiency measures is
about 20% of the world steel carbon emission mitigation. The cost-effective carbon reductions
add up to about 15% of the total carbon emissions.

11 DISCUSSION
A short discussion is included on the value of this research, based on the work and initially set
goals of this research. For example, it is likely the results of this research will not be an accurate
representation of the actual energy saving potential in the world steel industry. This would
indicate the report is not useful basis for decisions making for policy makers. This section
should provide insight in possible flaws in the results of our calculations and suggestions for
further research.
First, because of the short time span and the availability of data, this research can only provide
an estimation of the energy savings in the world steel production. To gain a more accurate
assessment, more detailed information for the application of energy efficiency in each of the
world regions should be included in any further research. Whereas now, only for the US a
thorough investigation to energy efficiency measures was made. Based on the comparison with
IEA data in paragraph 8.4.2, apparently these estimations are similar to findings in other
researches on the same topic.
As mentioned in the background, one of the goals of this research was to provide new data for
integrated assessment models (IAM), for example on climate change. These models are used by
policy makers to make more accurate decisions on future policies regarding climate change. If
one would assume the IAM are now lacking in data on energy efficient technology, this research
would provide reasoned estimations on the potential of energy efficiency measures for the steel
industry. Even though these are estimations, if one would assess global climate change, data
from this report can be used.
If one would for example analyse or make decsions for a specific world region, this report only
provides estimations. As mentioned in our gaps in knowledge, the assessment of energy
efficiency measures in the selected world regions is done by changing a few key factors, but
using the same set of measures as for the US. Because of the lack of available data on some
regions, these estimation might already provide a useful estimation. But for regions like the
India or China more detailed reports on energy use and carbon emissions can be found in
different sources (Schumacher, 1998; Hasanbeigi, 2011)
All of these discussion points need to be taken into account for further research in the subject.
What was mainly found is, the lack of data required to enhance the results for the world steel
industry, was the major problem for a proper assessment. By including more regions and
finding out which factors, next to the factors we have described are most important, in the
future more accurate analysis should be possible. And as Xu, 2010 already mentioned, the data
on energy efficiency measures should be continuously updated.
65

12 REFERENCES
AISI - American Iron and Steel Institute, 2010. 2010 Annual Statistical Report, Washington, DC:
AISI.
ANEEL - Agência Nacional de energia eléctrica, Brasil, 2007, Tarifas Médias por Classe de
Consumo, http://www.aneel.gov.br/aplicacoes/tarifamedia/Default.cfm Accessed Sep 2012.
BP, 2012, Coal Prices,
http://www.bp.com/sectiongenericarticle800.do?categoryId=9037186&contentId=7068650,
Accessed Sep 2012.
Choudhury, R., A.K. Bhaktavatsalam, Energy inefficiency of Indian steel industry—Scope for
energy conservation, Centre for Energy Studies, Indian Institute of Technology, May 1998.
Cooney, S., 2007, Steel: Prices and policy issues, CRS Report of Congress, Oct 2007, Cornell
University ILR School
DECC, 2012, Department of Energy and Climate Change, Energy Price statistics,
http://www.decc.gov.uk/en/content/cms/statistics/energy_stats/prices/prices.aspx Accessed
Sep 2012
DOE - Department of Energy, Advanced Energy Manufacturing office, technology deployment,
Steel,
http://www1.eere.energy.gov/manufacturing/tech_deployment/partners/by_industry_list.cfm
?industry=Steel, Accessed Mar 2012
DOE - Department of Energy,2011, Industrial Technologies Program, Energy-Intensive
Processes Portfolio: adressing Key Energy Challanges Across US Industry, DOE/EE-0389
Energy Information Administration (EIA), 1995, Emissions of Greenhouse Gases in the United
States 1987-1994, ftp://www.eia.doe.gov/environment/057394.pdf
Energy Information Administration (EIA), 2009, 2006 Manufacturing Energy Consumption
Survey on manufacturing and industrial energy uses and costs. The United States Department of
Energy, Washington, DC. http://www.eia.doe.gov/emeu/mecs/contents.html, Accessed July
2012.
EPA, Environmental protection agency (2010), Available and emerging technologies for reducing
greenhouse gas emissions from the iron and steel industry, Policies and programs division.
Gosh, A., Chatterjee, A., 2008, Ironmaking and Steelmaking, theory and practice, Prentice-Hall
of India, New-Delhi, ISBN: 978-81-203-3289-8
Hasanbeigi, A., Price, L., Aden, N., A comparison of iron and steel production energy use and
energy intensity in China and the US, Lawrence Berkeley National Laboratory, 2011
IEA - International Energy Agency, 2010, Natural gas in India, working paper.
IEA - International Energy Agency, 2007, Tracking Industrial Energy Efficiency and CO2
emissions, OECD/IEA, Paris, ISBN 978-92-64-03016-9
IEA - International Energy Agency, 2006, Fact Sheet: Energy Technology Perspectives:
Scenarios & Strategies to 2050, http://www.iea.org/papers/2006/industry.pdf Accessed Feb
2012
67

IMF, International Monetary Fund, 2012, Natural gas monthly price,
http://www.indexmundi.com/commodities/?commodity=natural-gas&months=120, Accessed
Sep 2012
IPCC – Intergovernmental Panel on Climate Change, 2007, Contribution of Working Group III
to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, 2007,
Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA
Iron and Steelmaker (I&SM), 1997a. Iron & Steelmaker’s 1997 Blast Furnace Roundup, Iron and
Steelmaker 24(8): 24-2
Iron and Steelmaker (I&SM), 1997b. Electric Arc Furnace Roundup – United States, Iron and
Steelmaker 24(5): 20-39.
Kim, J., Worrell, E., 2002, International comparison of CO2 emission trends in the iron and steel
industry, Energy Policy 30, 827–838
Levi, M.D., 2005, International Finance, 4 th edition, Routledge, New York, USA, ISBN 0-415-
30899-2
Meier, A.K., 1984, The cost of conserved energy as an investment statistic, Sixth Annual
Industrial Energy Technology Conference, Volume 2, Houston, TX, USA, April 1984.
OECD, 2012, Organization for Economic Co-operation and Development, OECD.StatExtract,
PPPs and Exchange rates, http://stats.oecd.org/Index.aspx?datasetcode=SNA_TABLE4,
accessed Sep 2012.
Henderson, 2011, Natural gas prices in Russia – towards export netback?, Oxford institute for
energy studies.
Schumacher, K., J. Sathaye, India’s Iron and Steel Industry: Productivity, Energy Efficiency and
Carbon Emissions, Lawrence Berkeley National Laboratory, October 1998.
Sidorenko, A., 2010, Electricity in Russia, Chapter 16, p. 345-368, Apec Publications.
Voest Alpine Industrieanlagenbau (VAI), 1997. FUCHS Shaft Furnaces, The Power, The
Performance, The Profit, Linz, Austria: Voest Alpine Industrieanlagenbau Gmbh.
World Bank, 2006, Official exchange rate,
http://data.worldbank.org/indicator/PA.NUS.FCRF?page=1, Accessed Oct 2012
Wordsteel Association, 2007, CO2 emission data collection, user guide, version 5.1, Brussels
Worldsteel Association, 2010, Steel Statistical Yearbook 2010, Worldsteel Committee on
Economic Studies, Brussels
Worldsteel Association, 2012, Crude steel production 2011,
http://www.worldsteel.org/dms/internetDocumentList/steel-stats/2011/Crude-steelproduction-2011/document/2011%20steel%20updated%20Feb2012.pdf,
Accessed Mar 2012.
Worrell, E., Neelis, M., 2007, World best practice energy intensity values for selected industrial
sectors, Lawrence Berkeley National Laboratory, LBNL-62806
Worrell, E., Price, L., Martin, N., 2001, Energy efficiency and carbon dioxide emissions
reduction opportunities in the US iron and steel sector, Energy 26, p. 513-536
68

Xu, T., J. Sathaye, C. Galitsky, Development of Bottom-up Representation of Industrial Energy
Efficiency Technologies in Integrated Assessment Models for the Iron and Steel Sector,
Lawrence Berkeley National Laboratory, August 2010.
Xu, T., T. Galama, J. Sathaye. 2012. Bottom-up Representation of Energy Efficiency
Technologies as Climate Mitigation Measures for the US Iron and Steel Sector from 1994 to
2010, US EPA Report, Lawrence Berkeley national Laboratory: Berkeley, CA.
69

Appendix A. Overall Measures 1
The investment cost and operational benefits for energy efficiency measures, have been based
on the findings of Worrell, 1999, where the costs are expressed in 1994$. The costs have been
adjusted for the years 2006 in the database with the GDP price index (BEA, 2011). The appendix
however only shows the 1994$ costs. In case it is explicitly mentioned the measure is only for
2006 and 2010, the costs are in 2006$. If mentioned the measure is only for 2010 the costs are
in 2010$.
Preventative maintenance involves training personnel to be attentive to energy consumption
and efficiency. Successful programs have been launched in many industries (Caffal, 1995;
Nelson, 1994). Examples of good housekeeping in steel making include timely closing of furnace
doors to reduce heat leakage and reduction of material wastes in the shaping steps. We estimate
energy savings of 2% of total energy use, or fuel savings of 0.45 GJ/t of product and electricity
savings of 0.04 GJe/t of product, based on savings experienced at an integrated steel plant in The
Netherlands (Worrell et al., 1993). We assume minimal investment costs for good housekeeping
options ($0.01/t), although training and in-house information are needed, resulting in increased
annual operating costs. Based on good housekeeping projects at Rover (a large car manufacturing
plant in the UK), we estimate annual operating costs of about $11,000 per plant, or approximately
$0.02/t crude steel (Caffal, 1995). We apply this measure to all integrated and secondary
steelmaking in the US in 1994.
Energy monitoring and management systems. This measure includes site energy
management systems for optimal energy recovery and distribution between various processes and
plants. A wide variety of such energy management systems exist (Worrell et al., 1997; Caffal,
1995). Based on experience at the Hoogovens steel mill (The Netherlands) and British Steel (Port
Talbot, UK), we estimate energy savings of 0.5%, or fuel savings of 0.12 GJ/t of product and
electricity savings of 0.01 GJe/t of product, for US integrated sites (Farla et al., 1998; ETSU, 1992).
We estimate the costs of such a system to be approximately $0.15/t crude steel based on the costs
for the system installed at Hoogovens ($0.8M) (Farla et al., 1998). This measure is applied to
100% of US steel production facilities.
Variable speed drive: flue gas control, pumps, fans. From the assessment of the DOE
the amount of energy saved by the application of multiple VFD’s on fans and pumps in three
integrated steel plants, the total energy savings from this measure are about 0.01 GJ/tonne of
steel. Based on two estimated investment costs from the energy audits, the investment costs are
estimated at 0.7$/tonne of steel. Since VFD’s cannot be implemented cost efficient in all steel
plants, the application in 50% of the integrated steel plants is expected.
Steam system optimization. From the assessment of the DOE energy audits the
implementation of a number of measures to reduce the energy consumption for the steam
system could provide a significant energy saving. At a steel finishing plant in the US, a
suggestion was made to improve the following; increase condensate recovery, repair steam
1 Excerpted from Ernst Worrell, Nathan Martin, Lynn Price (1999) Energy Efficiency and Carbon Dioxide Emissions
Reduction Opportunities in the US Iron and Steel Sector. LBNL Report #41724
71

traps, reduce vented steam, reduce steam leaks and reduce steam demand for space heating. The
combined energy savings from the improvements of the steam system were 0.28 GJ/tonne of
steel. In other cases the combined energy savings from improving the steam system would lead
to savings of 0.08 and 0.15 GJ/tonne for respectively an integrated steel plant and a coking
plant. The latter is estimated from cost savings divided by a fuel price of $8 per GJ. Where
repairing steam traps would provide the largest saving and repairing steam leaks would provide
the least saving. The assumed energy saving based on the DOE assessment is 0.18 GJ/tonne of
steel. The investement costs needed to improve the steam system are estimated at $2/tonne of
steel. The optimization is expected to be implemented in 100% of the integrated steel mills,
while for secondary steel mills the implementation is estimated at 50%.
Air system optimization. Air systems are used throughout the steel industry. According to
the assessment of DOE energy audits, the use of some optimizations to this system will lead to
an energy saving of 0.01 GJ/ton from analysis of a steel finishing and hot rolling plant. The
measures include reduction of leaks, reduction of plant pressure, replacement of open blown
nozzles and shutdown of compressors when possible. The investment cost are estimated at
Increase boiler efficiency. From the assessment of the DOE energy audits the optimization
of boilers could potentially increase the boiler efficiency. By optimizing the oxygen content in
the flue gas, reducing the heat loss and preheat the feed water with waste heat, fuel savings of
about 0.02 GJ/tonne can be achieved, estimated from assessments at two different steel plants.
The investment costs are estimated at about $0.3/tonne of steel from the DOE assessment.
Appendix B. Iron Ore Preparation 2
Iron ore is prepared in sinter plants where iron ore fines, coke breeze, water treatment plant
sludges, dusts, and limestone (flux) are sintered into an agglomerated material (US DOE, OIT,
1996). In 1994, 12.1 Mt of sinter were produced in the US (AISI, 1996). Fuel consumption for this
process 26 PJ and electricity consumption was 2 PJ resulting in a primary energy intensity of 2.6
GJ/t sinter.
Sinter plant heat recovery. Heat recovery at the sinter plant is a means for improving the
efficiency of sinter making. The recovered heat can be used to preheat the combustion air for the
burners and to generate high pressure steam which can be run through electricity turbines.
Various systems exist for new sinter plants (e.g. Lurgi EOS process) and existing plants can be
retrofit (Stelco, 1993; Farla et al., 1998). In 1994, only 15% of the blast furnace feed consisted of
sinter; the remainder of the feed was composed of pellets, pelletized at the mining site (AISI,
1996). We apply this measure to all existing sinter plants and estimate the fuel savings (steam and
coke) associated with production of this 12.2 Mt of sinter to be 0.55 GJ/t sinter, based on a
retrofitted system at Hoogovens in The Netherlands, with increased electricity use of 1.5
kWh/tonne sinter (Rengersen et al., 1995). NOx, SOx and particulate emissions are also reduced
2 Two energy efficiency measures that we do not include are the use of higher quality iron ores in iron ore preparation
and reduction of the basicity of the sinter (Aichinger, 1993). These measures are not considered due to lack of data on
current implementation and future potential in the US
72

with this system. The measure has capital costs of approximately $3/t sinter (Farla et al. 1998).
We do not estimate costs for new sinter plants since it is unlikely that such plants will be built in
the US, due to the large investment required. New iron making technologies (discussed below)
aim at the use or lump ore or ore fines, instead of using agglomerated ores.
For sinter plants the use of grade recovery and cascade utilization (GRUC) should provide an
energy saving (Dong, 2010). In this setting the high temperature heat recovered is used for
steam and electricity generation, while lower temperatures are used for drying and preheating
operations. According to an application in a 3.8 Mtonne per year sinter plant, the application of
the GRUC could increase electricity recovery with 15 MW (about 0.12 GJ/tonne sinter) and heat
for direct heat recovery was about 50GJ/hour (about 0.12 GJ/tonne sinter). Compared to the
previously found results on sinter plant heat recovery this option is not more beneficial in terms
of final and primary energy. Therefore older date is used still used in the database.
Reduction of air leakage. Reduction of air leakages will reduce power losses for the fans by
approximately 3-4 kWh/t sinter (Dawson, 1993), and could have a positive effect on the heat
recovery equipment. These savings may need small investments for repair of the existing
equipment. We estimate these costs at $0.1/t sinter capacity. The reduction of air leakage in the
sinter plant should increase the productivity as well. At the Nagoya works at Nippon Steel the
productivity was increased by 5%, after a reduction of 83% of the air leakage (Sakaue, 2009). It
is estimated this will decrease the operational cost per ton by $0.1/tonne sinter.
Increasing bed depth. Increasing bed depth in the sinter plant results in lower fuel
consumption, improved product quality, and a slight increase in productivity. The savings amount
to 0.3 kg coke/t sinter per 10 mm bed thickness increase, and an electricity savings of 0.06 kWh/t
sinter (Dawson, 1993). We assume a bed thickness of 550 mm in 1994, which can be increased to
650 mm. This will result in a fuel savings of 0.09 GJ/t sinter and an electricity savings of 0.002
GJ/t sinter. No investment costs are assumed for this measure.
Improved process control. Improved process controls in various systems have resulted in
energy savings, and many different control systems have been developed. Based on general
experience with industrial control and management systems, the savings may be estimated at 2-
5% of energy use (Worrell et al., 1997). We conservatively use a figure of 2% savings or a primary
energy savings 0.05 GJ/t sinter. Capital costs are assumed to be $0.15/t sinter (See also the
measure on Energy management and monitoring systems). Improved process control does not
only benefit the energy efficiency of the sinter plant, but can also increase the productivity of the
sinter plant. According to Siemens-VAI the productivity of sinter plants can be improved by 12%
by using improved process control (Siemens-VAI, 2011a). The operational benefits are estimated
at $0.23/tonne sinter. The investment costs for the improved process control are estimated at
$0.25/tonne sinter.
Use of waste fuels in the sinter plant can reduce the energy demand in sinter making. The
energy demand in sinter making is met by mixing iron ore with breeze from coke making and gas
in burners. Sinter making is also used to "scavenge" byproducts such as millscale and ironcontaining
dusts and sludges. It is possible to use waste oils (especially from cold rolling mills)
which are currently landfilled (US DOE, OIT, 1996), however the use will be limited by emission
73

limits due to incomplete combustion. A well-monitored combustion process could reduce the use
of gas in the burners (Cores et al., 1996). It is difficult to estimate the savings for this measure,
since it depends on the composition and quantity of lubricants and the installed gas clean-up
system at the sinter plant. However, based on a survey of European mills, the average sludge
production from cold rolling mills is 1 kg/t rolled material. The variation can be large, though,
ranging from 0.01 to 10 kg/t steel. The oil content is less than 10% and the sludge contains around
45-55% iron. While this does not represent much energy, it is beneficial to process this sludge in
the sinter plant to recover the iron losses. About 50% of the sludge is recycled in the sinter plant
in Europe. Along with the oil recovery sludges, there are also oil, creases, and emulsions produced
at a rate of 1.3 kg/t rolled steel (Roederer and Gourtsoyannis, 1996). Assuming that the high
heating value of these oils is the same as that of heavy fuel oil, total oil production is estimated to
be around 1.2 kg oil/t rolled steel (assuming 7.5% in oil recovery sludges and 90% in oils, creases,
and emulsions). We assume a calorific value of 34 MJ/kg, or an energy savings of 41 MJ/t rolled
steel, or 0.18 GJ/t sinter. (Cores et al., 1996). This is measure is applied to integrated plants with
sinter plants on site (allowing for waste recovery), or 74% of the rolling sludges and oils (1.68 PJ).
Bethlehem steel has developed a waste recovery and waste injection system, at a cost of about $25
M to recycle 200 ktons of various materials (Schriefer, 1997). We estimate the tonnage of waste
fuels recycled to be 4,800 tons at an estimated production of 4 Mt rolled steel. With an estimated
sinter production of 3 Mt, this results in a cost of $0.20/t sinter. Also the use of oil in the sinter
plant is minimized in order for the sinter plant to produce consistently according to the
‘Guidelines on Best Available Techniques (BAT) for Sinter Plants in the Iron Industry’ (Finlay,
2004).
Replacement of coke by sunflower seed husks has already been investigated, it was concluded
the substitution of 10% coke by husks is feasible. The use of sunflower seed husks would mean
part of the energy use in the sinter plant is from renewables. Moreover, the productivity is
expected to increase by 6.4% by using sunflower seed husks (Ooi, 2008). There is however no
indication this would improve the energy balance.
Improve charging method
Limonite (brown iron ore) used as a raw material for sintering is inexpensive, but it decreases
the productivity in the sintering process because it combines strongly with water and has a
coarse particle size. These problems can be overcome by using an improved charging method.
The system adopts a drum chute and a segregation slit wire. The purpose of the drum chute is to
reduce the height difference (dropping difference) in material charging, while the segregation
slit wire controls the particle size distribution. Specifically, because a constant particle size is
maintained, the permeability of the sintering mixture is increased, resulting in improved
sintering efficiency, and the material return ratio due to poor sintering is reduced. This system
was developed by a Japanese steelmaker and has been introduced at all its plants in Japan.
Productivity improvement amounts to 5 percent and energy consumption due to coke use
decreases by 0.07 MMBtu/ton sinter (0.08 GJ/tonne sinter) compared to a conventional
charging system (EPA, 2010). The investment cost of this system are estimated to be $2.5 per
tonne, and the operational benefits of the are estimated at $0.1/tonne.
74

Coke Making
Currently there are 50 active coke batteries in the US with a total production in 1994 of 16.6 Mt
coke (Hogan and Koelble, 1996b). Coke making consumed 74 PJ of fuel and 2 PJ of electricity,
resulting in a primary specific energy consumption of 4.9 GJ/t (US DOE, OIT, 1996).
Coal moisture control uses the waste heat from the coke oven gas to dry the coal used for coke
making. The moisture content of coal varies, but it is generally around 8-9% for good coking coal
(IISI, 1982). Drying reduces the coal moisture content to a constant 3-5% (Stelco, 1993) which in
turn reduces fuel consumption in the coke oven by approximately 0.3 GJ/t. According to Eremin
(2011) a reduction in moisture content of 1.1 to 3.5% is possible with pre drying coal. This is less
than the 3-5% assumed by Stelco (1993), therefore the energy benefits are reduced from 0.3
GJ/tonne to 0.22 GJ/tonne. The coal can be dried using the heat content of the coke oven gas or
other waste heat sources. Coal moisture control costs for a plant in Japan were $21.9/t of steel
(Inuoe, 1995). Based on Japanese coke use data in 1990, we assume approximately 450 kg coke/t
of crude steel, resulting in coal moisture control costs of $49/t coke or $14.7/t crude steel. We
apply this measure to 100% of US coke production in 1994.
Programmed heating instead of conventional constant heating of the coke ovens ensures
optimization of the fuel gas supply to the oven at the various stages of the coking process and
reduces the heat content of the coke before charging (IISI, 1982). Use of programmed heat can
lead to fuel savings of about 10% (IISI, 1982). According to Gongfa (2010) the use of intelligent
process control in cokemaking can reduce the energy requirement by 2-3%. In case of
cokemaking this would be about 0.15-0.22 GJ/tonne coke. Small capital costs regarding the
computer control system for the coke oven are incurred. We estimate these costs to be $75K per
coke battery for a large energy management system (derived from Caffal, 1995), which is
equivalent to approximately $0.23/t coke for the coking capacity of the integrated steel mills
(excluding merchant coke producers). This measure is also applied to 100% of US coke production
in 1994.
Variable speed drive coke oven gas compressors can be installed to reduce compression
energy. Coke oven gas is generated at low pressures and is pressurized for transport in the internal
gas grid. However, the coke oven gas flows vary over time due to the coking reactions. We assume
that the compressors are driven with steam turbines, since we lack information on the coke oven
gas compressors in the US, and that this measure can therefore be applied to all US coke making
facilities. Installing a variable speed drive system on a compressor at a coke plant in The
Netherlands saved 6-8 MJ/t coke, at an investment of $0.3/t coke (Farla et al., 1998).
Coke dry quenching is an alternative to the traditional wet quenching of the coke, and this
process reduces dust emissions, improves the working climate, and recovers the sensible heat of
the coke. Dry coke quenching is typically implemented as an environmental control technology.
Various systems are used in Brazil, Finland, Germany, Japan, and Taiwan (IISI, 1993), but all
essentially recover the heat in a vessel where the coke is quenched with an inert gas (nitrogen).
The heat is used to produce steam (approximately 400-500 kg steam/t), equivalent to 800-1200
MJ/t coke (Stelco, 1993; Dungs and Tschirner, 1994). According to the exergy analysis of Errera
(2000) the energy savings, in steam, are 1,632 MJ/tonne of coke. The steam can be used on site
75

or to generate electricity. For new coke plants the costs are estimated to be $50/t coke, based on
the construction costs of a recently built plant in Germany (Nashan, 1992). However, it is very
unlikely that new coke plants will be constructed in the US, so we use retrofit capital costs in the
calculation. Retrofit capital costs depend strongly on the lay-out of the coke plant and can be very
high, up to $70 to $90/GJ saved (Worrell et al., 1993). We assume $70/t coke. Operating and
maintenance costs are estimated to increase by $0.5/t coke. The actual operational savings from
using CDQ are about $7.6/tonne, used for the 2006 and 2010 measures (Hsun Lin, 2009). We
apply this measure to all US coke making facilities.
Appendix C. Iron Making - Blast Furnace
Iron making is the most energy-intensive step in integrated steel making. In 1994 there were 40
blast furnaces in the US, producing 49.3 Mt of iron (AISI, 1995). Iron making consumed 676 PJ
fuel and 4 PJ electricity, resulting in a primary specific energy consumption of 13.9 GJ/t.
One of the main energy efficiency measures in the iron making stage is the injection of fuels into
the blast furnace, especially the injection of pulverized coal (PCI). Pulverized coal injection
replaces the use of coke, reducing coke production and hence saving energy consumed in coke
making (above) and reducing emissions of coke ovens and associated maintenance costs. Coal
injection has increased in recent years due to environmental legislation combined with the high
average age of US coke plants. Closing of old coke plants is leading to increased coke imports. In
1994 coke was mainly imported from Japan, China, and Australia (Hogan and Koelble, 1996b).
Increased fuel injection requires energy for oxygen injection, coal, and electricity and equipment
to grind the coal. The coal replaces part of the coke that is used to fuel the chemical reactions.
Coke is still used as support material in the blast furnace. The maximum fuel injection depends
on the geometry of the blast furnace and impact on the iron quality (e.g. sulfur). Coal injection is
common practice in many European blast furnaces and is increasing in the US to reduce the
amount of coke required. Maximum theoretical coal injection rates are around 280-300 kg/t hot
metal. In the US the coal injection rate varies. A 1994 survey of seven blast furnaces in the US
gave fuel injection rates between 41 and 226 kg/t hot metal (Lanzer and Lungen, 1996). The
highest injection rates, of 225 kg/t, have been reached at USX Gary (Schuett et al., 1997). Coke
replacement rates vary between 85% and 100% (Schuett et al., 1997). We assume that 1 kg of
coke will be replaced by 1.08 kg of injection fuel, a replacement rate of 92%.
The investments for coal grinding equipment are estimated to be $50-55/t coal injected (Farla et
al., 1998). O&M costs show a net decrease due to reduced coke purchase costs and/or reduced
maintenance costs of existing coke batteries, which is partly offset by the increased costs of oxygen
injection and increased maintenance of the blast furnace and coal grinding equipment. We
estimate the reduced operation costs on the basis of 1994 prices of steam coal and coking coal to
be $15/t (IEA, 1995). This is a low estimate, as cost savings of up to $33/t are possible, resulting in
a net reduction of 4.6% of the costs of hot metal production (Oshnock, 1995a).
Pulverized coal injection to 130 kg/t hot metal. In this measure, the average coal injection
rate is increased from the current average of 2 kg/t hot metal (US DOE, OIT, 1996) to 130 kg/t hot
metal for all blast furnaces. This net increase of 128 kg/t hot metal leads to fuel savings of 0.77
GJ/t hot metal with capital costs of $7/t hot metal (Farla et al., 1998). Operation costs will
76

decrease by $2/t hot metal (IEA, 1995). 3 This measure is applied to 80% of all blast furnaces;
injection of natural gas (see below) is applied to the remaining 20%. Injection of pulverized coal
may lead to reduced capacity utilization of the blast furnace (Hanes, 1999). Hence, the economic
benefits may vary by plant.
Pulverized coal injection to 225 kg/t hot metal. In this measure, the injection rate is
increased to 225 kg/t hot metal (as reached at USX Gary blast furnace 13) for the large volume
blast furnaces only (defined as those with production rates of 2.3-3.6 Mt/year, which is
approximately 30% of total production) (Schuett et al., 1997). This leads to fuel savings of 0.57
GJ/t hot metal, with an extra investment of $5.2/t hot metal and reduced operating costs of $1/t
hot metal. According to a mathematical model of de Castro, 2010, the amount of pulverized coal
can be increased to 280 kg/thm.
Injection of natural gas. 4 This measure is only applied to a portion of medium sized furnaces,
defined as those with production rates of 1.3-2.3 Mt/year, represent 20% of total furnaces.
Currently, coal is seen as the favorable injection fuel because of its low price. Injection of natural
gas is an alternative. The blast furnace productivity is increased by about 25-30% when using
pulverized coal injection (de Castro, 2010). Injection of natural gas together with pulverized coal
should provide even further benefits as productivity increase of about 30-35% is expected.
Maximum injection rates are lower than for coal (Oshnock, 1995b). Replacement rates for natural
gas vary between 0.9 and 1.15 kg natural gas/kg coke (Oshnock, 1995b). Natural gas injection tests
by the Gas Research Institute show a maximum injection rate of 130-150 kg/t hot metal, with
estimated investment costs of $4-5/t hot metal (Anonymous, 1995). Assuming a replacement rate
of 1kg natural gas/kg coke, savings from replacing 140 kg of coke are estimated to be 0.9 GJ/t hot
metal. We assume that operating costs will decrease similar to that seen in the lower PCI injection
measure ($2/t hot metal).
Injection of oil up to 130 kg/thm
The injection of oil in the blast could reduce the use of coke by 1.2 tons for every 1 ton of oil that
is injected in the blast furnace (EPA, 2010). The amount of oil can be increased up to 130
kg/tonne of crude steel. The use of waste oil would be a preferred option for blast furnace
injection. The reduction in energy is based upon the reduction for pulverized coal and gas
injection and is estimated at about 0.85 GJ/tonne hot metal. The reduced operational cost due
to the reduced use of coke ovens is estimated at $2/tonne of hot metal. The investment costs for
oil injection are estimated at $5/tonne of hot metal for the adaptations to the tuyere.
3 Costs are calculated as follows: 128kg coal/t hot metal = 0.128t coal/t hot metal * $55 capital costs = $7/t hot metal.
4 The implementation level of this measure will interact with the level of pulverized coal injection. Following further
research, we may revise both this and the pulverized coal injection measure to reflect an increased emphasis on the
use of natural gas over coal due to CO2 concerns. At this time, we do not have adequate data on actual levels of natural
gas injection. Other fuels can also be injected, but we have not included any due to lack of data. Injection of plastic wastes
has been tested at Stahlwerke Bremen in Germany at rates of 30 kg/t hot metal (Janz and Weiss, 1996). Chlorine content
(due to PVC) may lead to dioxin formation, making efficient flue gas control equipment necessary.
77

Top pressure recovery turbines (wet type) are used to recover the pressure in the furnace. 5
Although the pressure difference is low, the large gas volumes make the recovery economically
feasible. The pressure difference is used to produce 15-40 kWh/t hot metal (Stelco, 1993).
Turbines are installed at blast furnaces worldwide, especially in areas where electricity prices are
relatively high (e.g. Western Europe, Japan). The standard turbine has a wet gas cleanup system.
The top gas pressure in the US is generally too low for economic power recovery (I&SM, 1997a&b).
A few large blast furnaces (representing 20% of production) have sufficiently high pressure.
According to Oda, 2007, the use of TRT can be implemented in a large amount of the integrated
steel plants. In Japan and Korea the application is estimated at 100%, while for the US the
application is estimated at only 5% (Oda, 2007). The electricity generated by the TRT is
expected to generate 48 kWh/tonne of hot metal. Future upgrades of blast furnaces might lead to
increasing top pressures to improve productivity. We assume a power recovery of 30 kWh/t hot
metal in the US for 1994 and 2002, with typical investments of about $20/t hot metal (Inoue,
1995) for 20% of the 1994 US blast furnace capacity.
Recovery of blast furnace gas during charging of the blast furnace is designed to recover the
1.5% of gas that is lost during charging. A recovery system has been developed and installed by
Hoogovens in The Netherlands. The savings are estimated to be 66 MJ/t hot metal at a cost of
$0.3/t hot metal (Farla et al., 1998). We assume that such systems can be installed in 60% of US
blast furnace capacity based on an estimate of the number of bell-type charging mechanisms in
the US
Hot blast stove automation can help to reduce the energy consumption of the stoves, increase
the reliability of the operation, increase stove life-time, and optimize gas mix (Beentjes et al., 1989;
Derycke et al., 1990; Kowalski et al., 1990). The energy savings of such systems are estimated to be
between 5% (Beentjes et al., 1989) and 12 to 17% (Derycke et al., 1990). Based on the high fuel
consumption of hot blast stoves in the US (US DOE, OIT, 1996) we assume savings of 370 MJ/t
hot metal (Derycke et al., 1990). The installation of a hot blast stove automation system at Sidmar,
Gent (Belgium) had a payback of two months (Derycke et al., 1990). We assume an investment
cost of $0.3/t hot metal, to be implemented in all small blast furnaces, or 60% of the total US blast
furnace capacity (equivalent to 30.3 Mt in 1994). We assume that all blast furnaces with capacities
over 4500t hot metal/day have already installed automatic control systems.
Recuperator hot blast stove. Hot blast stoves are used to heat the combustion air of the blast
furnace. The exit temperature of the hot blast stove flue gases is approximately 250°C. The heat
can be recovered to preheat the combustion air of the stoves. Various recovery systems have been
developed and implemented (Stelco, 1993). Fuel savings vary between 80 and 85 MJ/t hot metal
(Farla et al., 1998; Stelco, 1993). We assume savings of 80 MJ/t hot metal. The costs of
recuperation systems are high and depend strongly on the size of the stoves (i.e. the blast furnace).
We estimate the costs to be $18-20/GJ saved (Farla et al., 1998), equivalent to $1.4/t hot metal.
5 Top pressure recovery turbines (dry type) use a dry gas clean up system which raises the turbine inlet temperature,
increasing the power recovery by about 25-30% (Stelco, 1993). However, the system is more expensive, estimated at 28
US$/t hot metal (Inoue, 1995). Due to the high costs, we assume that this system will not be implemented on existing
blast furnaces in the US in the near term.
78

An efficient hot blast stove can run without the need for natural gas. We apply this measure to
100% of 1994 US blast furnaces.
Improved blast furnace control systems have been developed in Japan and Europe that
provide improved control over systems currently used in Canada (Stelco, 1993) and presumably in
the US A successful control system has been installed at Rautaruukki Steel Works in Raahe,
Finland, reducing total fuel use to 440-450 kg/t hot metal (Stelco, 1993), and increasing
productivity and flexibility (Pisila et al., 1995). British Steel has developed an expert system for
blast furnace control (Fitzgerald, 1992). We estimate the savings of improved blast furnace control
strategies at half of the savings reached at Rautaruukki, i.e. 0.4 GJ/t hot metal (Pisila et al., 1995),
with the other half attributed to charge material upgrading. Capital costs are estimated to be
$0.5M per blast furnace. With 40 blast furnaces and a combined capacity of 55.5 Mt this is
equivalent to $0.36/t hot metal (Hogan and Koelble, 1996a). No large changes in operating costs
are expected. We apply this measure to 50% of 1994 US blast furnaces.
Appendix D. Iron Making – Basic Oxygen Furnace
Direct reduced iron (DRI), hot briquetted iron (HBI,) and iron carbide are all alternative iron
making processes (McAloon, 1994). Because of the small production quantities (in the reference
year 1994) we do not discuss energy efficiency measures in the alternative iron making processes
separately. In 1994 only one producer (Georgetown Steel) produced 480 kt DRI (Midrex, 1995),
using a gas-based Midrex process built in 1971. The energy consumption of a state-of-the-art
Midrex-unit is 10 to 11 GJ/t iron and 110 kWh/t (Midrex, 1993). DRI is produced through the
reduction of iron ore pellets below the melting point of the iron. DRI is mainly used as a high
quality iron input in electric arc furnace (EAF) plants. The US steel industry also imports DRI
from countries in Latin America. New DRI plants are being constructed in Alabama (a mothballed
plant built originally in 1975 in Scotland) and in Louisiana (a new Midrex Megamod module) and
other plants have been announced. A new alternative iron production process, the iron carbide
process, has been pioneered by Nucor which has one plant operating in Trinidad and another
plant scheduled to be built in Texas. The growing production by EAF plants in the US, high scrap
prices, and the need for high quality inputs due to the expansion of EAF producers in the flat steel
market will increase the future demand for alternative iron inputs.
Steelmaking - Basic Oxygen Furnace (BOF)
In basic oxygen furnace (BOF) steel making a charge of molten iron and scrap steel along with
some other additives (manganese and fluxes) is heated and refined to produce crude steel. BOF
crude steel production in 1994 was 55.3 Mt with fuel and electricity consumption of 19 PJ and 6
PJ, respectively. Primary energy intensity for this process step in our base year (1994) was 0.7
GJ/t.
BOF gas and sensible heat recovery (suppressed combustion) is the single most energysaving
process improvement in this process step, making the BOF process a net energy producer.
By reducing the amount of air entering over the convertor, the CO is not converted to CO2. The
sensible heat of the off-gas is first recovered in a waste heat boiler, generating high pressure
steam. The gas is cleaned and recovered. The total savings vary between 535 and 916 MJ/t steel,
depending on the way the steam is recovered (Stelco, 1993). Suppressed combustion reduces dust
79

emissions and since the metal content of the dust is high, about 50% of the dust can be recycled in
the sinter plant (Stelco, 1993). According to Li (2010) the se of heat recovery on the BOF can
recover up to 12 kWh/tonne electricity. The costs will depend on the need for extra gas holders.
Suppressed combustion is very common in integrated steel plants in Europe and Japan. In the US
no BOF gas seems to be recovered (US DOE, OIT, 1996; Hanes, 1999), so we apply this measure to
100% of US BOF steel making. We assume an energy recovery rate of 916 MJ/t crude steel (Stelco,
1993), with estimated capital costs of 22$/t crude steel, based on plants in Japan (Inoue, 1995)
and The Netherlands (Worrell et al., 1993).
Variable speed drive on ventilation fans. The BOF process is basically a batch process. The
volumes of flue gases vary widely over time, making variable speed drives an option. Large fans
are used in the BOF plant to control air quality. At Hoogovens the use of variable speed drives has
been shown to save power (Worrell et al., 1993) in the BOF, reducing the power demand by
approximately 20%, or 0.9 kWh/t crude steel (Farla et al., 1998). According to the assessment of
DOE energy audits the use of VFD’s on fans can save less than 0.01 GJ/tonne of steel. This
confirms the estimated energy saving of about 0.9 kWh/tonne steel. With total costs of $1M
(1988) the investment costs are $0.2/t crude steel (Farla et al., 1998). We assume that such
variable speed drives could be used in all US BOF steel making facilities.
Appendix E. Secondary Steelmaking - Electric Arc Furnace (EAF)
Electric arc furnace or secondary steel making involves the production of steel from scrap metal
which is melted and refined using electricity in an electric arc furnace (US DOE, OIT, 1996).
Electric arc furnaces are on average smaller capacity compared to blast furnace/BOF capacity and
use less energy. In 1994 there were 122 secondary steel mills with 226 electric arc furnaces. EAF
steel production in 1994 was 35.9 Mt and energy consumption for the furnaces was 6 PJ fuel and
62 PJ of electricity, reflecting a primary energy intensity of 5.5 GJ/t.
Improved process control (neural networks) can help to reduce electricity consumption
beyond that achieved through classical control systems. For example, neural networks or “fuzzy
logic” systems analyze data and emulate the best controller. For EAFs, the first “fuzzy logic”
control systems have been developed using current, power factor and power use to control the
electrodes in the bath (Staib and Bliss, 1995). The average power savings are estimated to be up to
8% (or 38 kWh/t), with an average increase in productivity of 9-12% and reduced electrode
consumption of 25% (Staib and Bliss, 1995). The actual savings depend on the scrap used and the
furnace operation. Furnace maintenance costs are reduced as well. We assume an average
efficiency improvement of 30 kWh/t (or 0.1 GJ/t). In 1994, advanced control systems were
installed at 16 furnaces in the US (Kimmerling, 1997), with a total capacity of 5.8 Mt (equivalent to
9% of the US EAF capacity in 1994). The capital and commissioning costs are estimated to be
$250,000 per furnace, with annual costs savings at roughly $1/t (Kimmerling, 1997). Since the
average capacity of EAF plants was 260 kt/year in 1994, we estimate the capital costs to be
$0.95/t. The measure is assumed to be applicable for 90% of the US EAF capacity.
In a more recent study of modeling an EAF, the optimal controls of an EAF will reduce the
electric energy use by more than 20%. The model had a mean error of 5.22% when compared to
real data (Ferretti, 2006). The model is a theoretical model, so actual application is not
80

guaranteed. Badische Stahl Engineering found savings can be reached with the Virtual Lance
Burner systems (VLB). VLB systems allow automatic operation including free programming of
the operation and totally free adjustment of oxygen and gas. They have been installed in a total
of 40 EAF facilities; an average saving of 40 kWh/tonne (0.14 GJ/tonne) was measured. For
2006 an 2010 therefore an updated saving of 0.14 GJ/tonne is used. Also the power-on-time of
the furnaces was reduced by 5.2 minutes, where the average was about 48 minutes (Opferman,
2008). No updates on the operational or investment costs have been found. It is expected
installation cost have changed apart from usual price index increase.
Flue gas monitoring and control using variable speed drives can reduce the energy use for
the flue gas fans, reducing the heat losses in the flue gas (Stockmeyer et al., 1990; Walli, 1991;
Worrell et al., 1997). The flue gas flow varies over time, which makes the use of variable speed
drives possible. Flue gas VSDs have been installed in various countries (e.g. Germany, UK). The
electricity savings are estimated to be 15 kWh/t (Stockmeyer et al., 1990), with a payback period of
2 to 3 years (Walli, 1991; Worrell et al., 1997). We estimate the capital investments to be $2/t, and
apply this measure to all furnaces with a size of 100 t or larger, equivalent to 50% of the US EAF
capacity.
Tenova installed its Goodfellow EFSOP® system in a 120ton EAF at CMC steel in Sequin, TX.
The use of this off gas based control system lead to a decrease in gas and electricity use of
respectively 9.23 kWh/ton (0.01 GJ/tonne) and 0.03 GJ/tonne and would increase productivity
by 0.03%. The control system resulted in a $3.02 per ton overall savings on energy and
increased yield (Maiolo, 2006). These updated figures will be used for the 2006 and 2010
database.
Ultra high power transformers. Transformer losses can be as high as 7% of the electrical
inputs (CMP, 1992). The losses will depend mainly on the sizing and age of the transformer. When
replacing the transformer it is possible to convert furnace operation to ultra high power,
increasing productivity, as well as reducing energy losses. Ultra high power furnaces are those
with a transformer capacity of over 700 kVA/t heat size. The savings are estimated at 1 kWh/t per
MW power increase. The weighted 1994 average transformer capacity is estimated to be 480
kVA/t heat size for all non-ultra high power (UHP) furnaces. In 1994 38% of EAF capacity can be
classified as UHP furnaces. Many EAF operators have installed new transformers and electric
systems to increase the power of the furnaces, e.g. Co-Steel (Raritan, NJ), SMI (Sequin, TX),
Bayou Steel (Laplace, LA) (Ninneman, 1997). UHP operation might lead to heat fluxes, and
increased refractory wear, making cooling of the furnace panels necessary. This results in heat
losses partially offsetting the power savings. The increased power can be reached by installing new
transformers or paralleling existing transformers. The replacement of a 93 MVA transformer at
Co-Steel (Raritan, NJ) with one rated at 120-144 MVA in 1997 was included in a project totally
costing $6.2M (Ninneman, 1997). This is equivalent to approximately 8.3$/t steel produced. This
is a high cost estimate as the total project costs included other equipment as well. We assume that
all transformers for medium to large furnaces over 15 years old can be replaced by more efficient
equipment. This is equivalent to approximately 115 furnaces with a capacity of 32.2 Mts (40% of
the total EAF capacity). We assume that the losses can be reduced to 4%, saving approximately 14
kWh/t. Transformers are assumed to have a lifetime of 15 years. The total energy savings are
81

estimated to be 17 kWh/t, (14 kWh due to transformer replacement and 3 kWh for upgrading to
UHP).
Bottom stirring/stirring gas injection is done by injecting an inert gas (e.g. argon) in the
bottom of the EAF, which increases the heat transfer in the melt and the interaction between slag
and metal (leading to an increased liquid metal yield of 0.5%) (Schade, 1991). This increased
stirring in the bath can lead to electricity savings of 11 to 22 kWh/t, with annual net production
cost reduction of $0.5 to 1.0/t accounting for increased labor and argon costs, based on tests at
Lukens Steel Co. in 1990 (Schade, 1991). Increased liquid steel yield increases the net cost savings
to $0.9-2.3/t (Jones, 1993). Furnaces with oxygen injection are sufficiently turbulent, reducing the
need for inert gas stirring (see below). We assume power savings of 20 kWh/t and cost savings of
$1.5/t. No data are available on the current application rate in US EAFs. We assume potential
application in 11% of the 1994 EAF capacity (i.e. small AC furnaces without oxygen injection). The
capital costs for retrofitting existing furnaces are estimated to be $0.6/t (1987) (Riley and Sharma,
1987) for increased refractory costs and installing tuyeres. The annual costs for inert gas purchases
are estimated to be $1.1/t (Riley and Sharma, 1987). The productivity increase (excluding saved
energy costs, including saved electrode costs, labor and alloys) is estimated to be $3.1/t (Riley and
Sharma, 1987). The lifetime of the tuyeres is limited to 100-200 heats (Riley and Sharma, 1987),
or approximately 6 months.
Foamy slag practice helps to reduce the heat losses through radiation from the melt by
covering the arc and melt surface with foamy slag. Foamy slag can be obtained by injecting carbon
(granular coal) and oxygen, or lancing of oxygen only. Foamy slag practice seems to be common
with a large number of operators in the US, so the potential savings are limited. However, not all
operators have implemented the practice well. We will assume that all medium to large furnaces
without oxygen injection can still implement this technology. Approximately 30-40% of the 1994
capacity (Jones, 1998) could still implement foamy slag practice, or improve the application. The
net energy savings (accounting for energy use for oxygen production) are estimated at 5-7
kWh/tonne steel (derived from Adolph et al., 1990). Based on the costs of installing oxygen lances
the investments are estimated at approximately 10$/tonne capacity (Jones, 1997b). Foamy slag
practice may also increase productivity through reduced tap-to-tap times, which is equivalent to
an n estimated cost saving of 1.8$/tonne steel (derived from Adolph et al., 1990).
Oxy-fuel burners/lancing can be installed in EAFs to reduce electricity consumption by
substituting electricity with fuels, increase heat transfer and reduce heat losses (foamy slag, see
above). Typical savings range from 2.5 to 4.4 kWh per Nm3 oxygen injected (IISI, 1982; CMP,
1987; Haissig, 1994; Stockmeyer et al., 1990), with common injection rates of 18 Nm3/t (IISI,
1982). The injection rate can be increased to 26 m3/t with increased fuel injection. Natural gas
injection is 10 scf/kWh, or 0.3 m3/kWh, (CMP, 1992), with typical savings of 20-40 kWh/t (Jones,
1996). Approximately 29% of the 1994 capacity (or 16 Mt in medium to large furnaces) has no oxyfuel
burners installed (I&SM, 1997b). These furnaces have an average power consumption of 502
kWh/t. We assume implementation of oxy-fuel burners in 25% of the existing EAF capacity, with
net energy savings of approximately 40 kWh/t. Modification investment costs depend on the
furnace size. With an average EAF size of 110 tons, the investments are estimated to be
approximately $4.8/t (Jones, 1997a). The improved heat distribution leads to reduced tap-to-tap
times of about 6% (CMP, 1995), leading to estimated annual cost savings of $4.0/t (CMP, 1987).
82

Oxygen injection also reduces the nitrogen content of the steel, leading to improved product
quality (Douglas, 1993). We estimate a lifetime of 10 years for this measure.
The use of gas burners in EAF will have significant electricity savings, but will require the use of
extra natural gas. The total energy savings of the use of natural gas instead of electricity are
minor, as the substitution is almost 1 kWh for 1 kWh. This was found from an extensive study in
70 EAF energy balances (Kirschen, 2009). Only in terms of CO2 emissions will the substitutions
of electric power by natural gas significantly benefit. However determination of the CO2
emissions is complex for different EAF and countries (Kirschen, 2009).
Eccentric bottom tapping (EBT). Eccentric bottom tapping is applied in most modern
furnaces, leading to slag-free tapping, shorter tap-to-tap times (increased productivity), reduced
refractory consumption, reduced electrode consumption (0.1 to 0.3 kg/t) and improved ladle life.
EBT helps to reduce energy losses and to improved emissions control. The energy savings are
estimated to be 15 kWh/t (0.05 GJ/t) (CMP, 1992). Reconstructing an existing EAF furnace at
Ipsco, Regina (Saskatchewan, Canada) cost $2.2 M (Ninneman, 1997). The furnace has an annual
production capacity of 688 kt, estimating the retrofit costs at $3.2/t capacity. It is assumed that all
new furnaces have EBT. We assume that EBT can be installed in all medium to large capacity EAF
built before 1986 (29.5 Mts), as the technology was introduced commercially around 1983 (Teoh,
1989), or equivalent to 52% of the production.
DC arc furnaces use direct current (DC) instead of conventional alternating current (AC). In a
DC furnace one single electrode is used, and the bottom of the vessel serves as the anode, resulting
in improved heat distribution in the furnace. This reduces the power consumption. Another major
advantage of DC furnaces is the reduced tap-to-tap time and electrode consumption (down to 1.2-
1.6 kg/t steel) (Macauley and Smailer, 1997; Mueller, 1997), increased refractory life, and
improved stability (Jones,1997b; Stelco,1993). DC technology is applicable to large furnaces (80 -
130 t heat size), and small furnaces are expected to remain AC systems. Larger DC-furnaces
(using two electrodes) are being investigated. The disadvantage of DC-systems is the up to 10-35%
higher capital costs (Jones, 1997b). Currently, the maximum current is restricted due to the use of
one electrode, but UHP DC systems are under development (Palasios and Arana, 1995). In the US,
Charter Steel, Florida Steel, Gallatin Steel, North Star, and Nucor (Hickman, Berkeley, Norfolk)
are using DC furnaces. The 1994 average power consumption of furnaces over 100 ton heat size is
estimated at 473 kWh/t (430 kWh/ton). The Nucor-plant (Hickman) achieves a consumption of
368 kWh/t, 36 Nm3 oxygen and 0.5-1.8 kg electrode (Mueller, 1997). The net energy savings are
estimated at 90 kWh/t (accounting for oxygen production at 0.4 kWh/Nm3 (Hendriks, 1994)).
Compared to new AC furnaces the savings are limited to 10-20 kWh/tonne (Jones, 1998). Based
on a cost-estimate for a 100 ton furnace the net extra investments compared to an AC furnace are
estimated to be $2.7M, or $3.9/t capacity (1991) (CMP, 1991). High efficiency AC/DC converters
with constant power control can decrease the energy use by 2.3% and produce a financial gain of
about 0.7 million euro per year for a 100MW furnace (Ladoux, 2005). Total cost savings are
estimated at $2 to $6/ton (CMP, 1991). This includes electrode cost savings, that are
approximately $2/ton steel (CMP, 1992). We assume annual cost savings (excluding energy costs)
of $2.5/t. Introducing DC furnaces competes with oxygen lancing, fuel injection, post combustion,
and eccentric bottom tapping,. We assume a market penetration of 15% of capacity in the US, of
which two-thirds is assumed to use as a twin shell to preheat scrap (see below). AC arc furnaces
83

still enjoy greater popularity than DC arc furnaces because of some practical advantages (Jones,
2011).
Scrap preheating is a technology that can reduce the power consumption of EAFs through
using the waste heat of the furnace to preheat the scrap charge. Old (bucket) preheating systems
had various problems, e.g. emissions, high handling costs, and a relatively low heat recovery rate.
Modern systems have reduced these problems, and are highly efficient. The energy savings depend
on the preheat temperature of the scrap. Various systems have been developed and are in use at
various sites in the US and Europe, i.e. Consteel tunnel-type preheater, Fuchs Finger Shaft, and
Fuchs Twin Shaft. Twin shell furnaces (see below) can also be used as scrap preheating systems.
All systems can be applied to new constructions, and also to retrofit existing plants.
Consteel process consists of a conveyor belt with the scrap going through a tunnel, down to the
EAF through a “hot heel”. Various US plants have installed a Consteel process, i.e. Florida Steel
(now AmeriSteel, Charlotte, NC) New Jersey Steel (Sayreville, NJ) and Nucor (Darlington, SC),
and one plant in Japan. The installation at New Jersey Steel is a retrofit of an existing furnace
(Lahita, 1995). Besides energy savings, the Consteel-process results in a productivity increase of
33% (Jones, 1997a), reduced electrode consumption of 40% (Jones, 1997a) and reduced dust
emissions (Herin and Busbee, 1996). Electricity use can be decreased to approximately 370-390
kWh/t (Herin and Busbee, 1996) without supplementary fuel injection in retrofit situation, while
consumption as low as 340-360 kWh/t have been achieved (Jones, 1997c) in new plants. We
estimate the electricity savings to be 60 kWh/t for retrofit. The extra investments are estimated to
be $2M (1989) for a capacity of 400-500,000 ton per year (Bosley and Klesser, 1991), resulting in
specific investments of approximately $4.4 to $5.5/t. The annual costs savings due increased
productivity, reduced electrode costs and increased yield are estimated to be $1.9/t (Bosley and
Klesser, 1991). Though, Consteel process is less effective compared to the shaft furnace because
of the low heat transfer. The conveyor travels parallel to the gas flow not penetrating the scrap
(Toulouevski, 2005). Therefore the shaft furnace or twin shell are preferred over the Consteel
process.
FUCHS shaft furnace consists of a vertical shaft that channels the offgases to preheat the scrap.
The scrap can be fed continuously (4 plants installed worldwide) or through a so-called system of
‘fingers’ (15 plants installed worldwide) (VAI, 1997). The optimal recovery system is the ‘double
shaft’ furnace (3 plants installed worldwide), which can only be applied for new construction. The
Fuchs-systems make almost 100% scrap preheating possible, leading to potential energy savings
of 100-120 kWh/t (Hofer, 1997). The energy savings depend on the scrap used, and the degree of
post-combustion (oxygen levels). In the US Fuchs systems have been installed at North Star
(single shaft (1996), Kingman, AZ), North Star-BHP (double shaft (1996), Delta, OH),
Birmingham Steel (finger shaft (1997), Memphis, TN). Two other Finger shaft processes have been
ordered by Chapparel (TX) and North Star (Youngstown, OH). Carbon monoxide and oxygen
concentrations should be well controlled to reduce the danger of explosions, as happened at North
Star-BHP. The scrap preheating systems lead to reduced electrode consumption, yield
improvement of 0.25-2% (CMP, 1997; VAI, 1997), up to 20% productivity increase (VAI, 1997) and
25% reduced flue gas dust emissions (reducing hazardous waste handling costs) (CMP, 1997). A
special system has been developed for retrofitting existing furnaces called the Fuchs Optimized
Retrofit Shaft, with a relatively short shaft. Retrofit costs are estimated at $6/t (Hofer, 1997) for an
84

existing 100 t furnace. Using post-combustion the energy consumption is estimated at 340-350
kWh/t (Jones, 1997d) and 0.7 GJ fuel injection (Hofer, 1996). According to VAI (1997) only 6-
8Nm³/tonne CS, as additional natural gas injection is required. This equals a requirement of
about 0.23-0.31 GJ/tonne (if 39 MJ/Nm³ is assumed). Assumed is the measure will increase fuel
consumption by 0.4 GJ/tonne. The production costs savings amount up to $4.5/t (excluding
saved electricity costs) (Hofer, 1997).
Scrap preheating competes with oxy-fuel injection and post combustion, as these options are
basically integrated in most scrap preheating systems. All furnaces over 70 t capacity could be
retrofitted cost-effectively (Hofer, 1996), or 74% of the 1994 US capacity (using on average 470
kWh/t in 1994), leading to net power savings of approximately 120 kWh/t and increased fuel
consumption of 0.4 GJ/t.
Twin shell furnace. The Twin shell concept comprises two EAF-vessels with a common arc and
power supply system. The system increases the productivity by reducing the tap-to-tap time to
approximately 45 to 50 minutes (Heinrich, 1995, Ninneman, 1997), and reducing energy costs
through reduced heat losses. Also, the hot flue gases of one shell can be used to preheat the second
shell. A twin shell AC plant is estimated to use 393 kWh/t compared to 412 kWh/t, saving 19
kWh/t (Macauley and Smailer, 1997) compared to current state-of-the-art single vessel plants for
a 100% scrap feed. The twin shell DC plant can save even more, 80 kWh/t compared to the 1994
average large scale AC furnace. The twin-shell concept can only be applied in the construction of a
new plant. New plants in the US using the Twin Shell concept are Gallatin Steel, Nucor, Steel
Dynamics, and Tuscaloosa Steel, and the resulting energy use varies for each of these plants. The
EAF at Gallatin steel has two AC furnaces, and consumes approximately 450 kWh/t (Jones,
1997b). DC furnaces can be used as well, reducing the power consumption further (see above). The
Twin Shell concept competes with the scrap preheating processes discussed above. Twin shells
seem to be an appropriate process for mini mills with capacities over 1 Mt per year. Very little cost
data exists on the Twin Shell (Jones, 1997b). The capital cost lay-out is expected to be a little more
(with estimated payback in the US of 2 years), while the production costs are expected to be 6%
lower than that of a single shell (Jones, 1997b). We will assume extra investments of $4-6/t (over
those of a new single shell furnace, based on the investments at Nucor, Berkeley County, SC), and
production cost reduction of $1.1/t (derived from (CMP, 1987), excluding energy cost savings). We
assume application of the DC twin shell concept to 10% of the 1994 production capacity.
Siemens EAF Quantum claims to be able to produce steel with an electricity energy
consumption of 280 kWh/tonne (Siemens-VAI, 2011b). This is substantially lower than the
average electricity consumption in the US Steel market. This newer type of furnace is expected
to increase fuel consumption by 0.3 GJ/tonne, compared to normal EAF operations. The
electricity savings are estimated at 0.58 GJ/tonne. The operational benefits of such a production
system is estimate at $4/tonne and the investment costs are estimated $2.5/tonne of steel. This
measure is only expected to be included in 2010.
Increased usage of hot metal in the EAF can reduce the energy use by about 35-50
kWh/tonne. Also tap-to-tap times can be reduced by about 5-10 minutes (Duan, 2009). Hot
metal is however only available in EAF built near a blast furnace or basic oxygen furnace and is
therefore estimated to be applicable to only 10% of the EAF furnaces. The operational costs of an
85

EAF with hot metal use are expected to be reduced by 1.5$/tonne due to the reduced tap-to-tap
time. The investment costs are estimated at 9$/tonne. This measure is only included for 2006
and 2010.
post-combustion of CO gas for preheating of scrap in the EAF furnace can be a benefit
for the energy consumption. According to Grant (2000), the benefits are not always viable in all
EAF production facilities. Careful analysis of the system should be done to assess the benefits to
the energy balance. It is therefore expected this technology can be applied in only 35% of all
EAF’s. From a few case studies the average electricity savings were 35-40 kWh/tonne (Grant,
2000). It would also reduce tap-to-tap times with 3-6%, and a power on time reduction of about
10%. We assume an operational benefit of about $0.9/tonne. This measure is only included for
2006 and 2010.
Appendix F. Casting
Once crude steel is produced it is cast into different shapes (billets, blooms, slabs, or ingots).
Molten steel is poured into a tundish and then released into a mold of one or more strands. A
majority of steel in the US is continuously cast which reduces the need for several intermediate
process steps. In 1994 we estimate that casting energy use was 17 PJ fuel and 15 PJ of electricity
resulting in a primary energy intensity of 0.7 GJ/t (US DOE, OIT, 1996).
Efficient ladle preheating. The ladle of the caster (and the BOF vessel) is preheated with gas
burners. Heat losses can occur through lack of lids and through radiation. The losses can be
reduced by installing temperature controls (Caddet, 1989), installing hoods, by using recuperative
burners (Caddet, 1987), use of oxygen burners (Gitman, 1998), or by efficient ladle management
(reducing the need for preheating). Oxygen burners for ladle preheating are used by many steel
companies in the US already (Gitman, 1998), but use can be expanded considerably. No data are
available on the actual energy use for preheating ladles in the US steel industry. Therefore, we
assume typical fuel use of approximately 0.04 GJ/t crude steel (Worrell et al., 1993). Efficient
preheating will reduce energy use by 50% or 0.02 GJ/t crude steel, with an estimated payback
time of 1.1 year (taking into account savings on ladle handling), or $0.06/t product, assuming a
gas price of $2.8/GJ (IEA, 1995).
According to an application of efficient ladle preheating with so called DOC burner technology
on existing preheaters resulted in a 50% reduction in fuel use for ladle preheating. In that case
the energy reduction was 0.028 GJ/tonne. The net cost savings from the use of the DOC burner
technology would result in a $0.29 per tonne cost reduction (Kelly, 2010). This updated
information is only applied for 2006 and 2010.
Thin slab casting A more advanced technology, near net shape casting, reduces the need for
hot rolling because products are cast closer to their final shape. It is a new technology integrating
casting and hot rolling in one process. Pioneered in the US by Nucor at the Crawfordsville and
Hickmann plants, various plants are operating, under construction, or ordered worldwide.
Originally designed for small scale process-lines, the first integrated plants constructed (Acme,
US; Posco, Korea) or announced the construction of thin slab casters (Germany, Netherlands,
Spain) with capacities up to 1.5 Mt/year (Worrell and Moore, 1997). Currently, four suppliers
(Germany (2), Austria and Italy) supply this technology. We base our description on the CSP-
86

process developed by SMS (Germany) as it represents most of the capacity installed worldwide.
Energy savings are estimated to be 4.9 GJ/t crude steel (primary energy). The energy
consumption of a CSP-plant is 94 MJ fuel per ton for the reheating furnace and electricity use of
43 kWh/t (Flemming, 1995). The investments for a large scale plant are estimated to vary between
$110/t and $180/t product (Anon, 1997a; Anon., 1997b, Schorsch, 1996). We assume therefore an
investment cost of $134/t crude steel, with estimated operation cost savings of between $25/t and
$46/t product (derived from Ritt, 1997 and Hogan, 1992, Schorsch, 1996). We therefore assume
an operation cost savings of $31/t crude steel. The potential capacity of thin slab casting is
estimated to be 20% of US integrated production and 64% of secondary steel. 6
Use of dry rolls instead of water cooled rolls in tunnel furnaces is an application only
for thin slab casters where a tunnel oven is used for the even temperature distribution in the
slab before rolling. If ceramic rolls can be used instead of water cooled rolls, the energy savings
on fuel use for the tunnel oven are estimated to be about 0.05 GJ/tonne (Duraloy, 2011). Also
the ovens will require shorter heat up times, which is estimated to save about $0.01/t product.
The investment costs are estimated at $5/t product (Duraloy, 2011). This measure is only
applied for 2006 and 2010.
Proper sealing on ladle furnace preheating will improve efficiency since ladles can have
large gaps through which flue gasses escape. The sealing of these can result in an energy saving.
The total estimated energy savings from ladle preheating from the DOE energy audits are about
0.07 GJ/tonne. The expected investment costs are estimated at $0.2/tonne. It is assumed all
ladle furnaces are in EAF facilities, where in about 50% of the cases proper sealing can be
applied.
Appendix G. Hot Rolling 7
After casting, the shaped products are further rolled to produce sheet, strip, plate, and other
structural products (US DOE, OIT, 1996). In 1994, 79.6 Mt of steel was hot rolled with an
estimated energy requirement of 259 PJ fuel and 56 PJ of electricity, resulting in a primary
energy intensity of 5.4 GJ/t. This energy intensity is relatively high compared to other countries
and additional data is required to improve this estimate (US DOE, OIT, 1996).
6 Estimate for the potential of thin slab casting in integrated mills is estimated to be 60% of integrated hot strip and
sheet production in 1994 or 11 Mt (AISI, 1996). Estimated potential for secondary mills is based on implementation in
slabs in minimills not currently continuously cast. These estimates will need to be refined in the future.
7 An additional measure is efficient power use in the rolling mill, which can reduce the power demand of the hot rolling
mill. Current hot strip mill power use in US is estimated to be 220 kWh/t (0.8 GJ/t) (US DOE, OIT, 1996). A modern hot
strip mill has a power consumption of about 105 kWh/t (0.4 GJ/t) (Worrell et al., 1993). Thus, installation of a modern
hot strip mill could represent a savings of up to 115 kWh/t (0.4 GJ/t). One component in these mills is motors which are
used for the rolling as well as in quench pumps. The quench pumps in a hot rolling mill are estimated to use 2.5 kWh/t
(Anon., 1994), on which savings of 42-76% are feasible through the application of variable speed drives and installing
control equipment. This system required an investment equivalent to 0.24$/t product saving 1.9 kWh/t hot rolled steel
(7 MJe/t). Reduced maintenance costs amount to 0.02$/t product (Anon., 1994). This measure needs further
quantification before it can be included in the analysis.
87

Hot charging is used to charge slabs at an elevated temperature into the reheating furnace of the
hot rolling mill. The slabs can be charged at various temperatures. Higher charging temperatures
will save more energy. The implementation of the technique depends on the lay-out of the plant,
and the distance between the caster and the hot rolling mill. In some plants the caster and
reheating furnace are “next door” making hot charging less costly (e.g. LTV in Cleveland and
Usines Gustav Boel, Belgium). Handling and transport of the slabs (i.e. a so-called ‘hot
connection’) is required if there is more distance between the caster and the rolling mill (Worrell
et al., 1993). Hot charging not only saves energy, but also improves material quality, reduces
material losses, improves productivity (by up to 6%), and may reduce slab stocking (Ritt,1996).
Care should be taken to descale the slab before charging in the reheating furnace (Caddet, 1990a).
The measure competes with thin slab casting (because in thin slab casting the slab is coupled
through a reheating furnace to the rolling stands) and direct rolling. A few plants in the US now
hot charge a portion of the production, e.g. LTV (Cleveland), USS (Fairfield), Bethlehem (Burns
Harbor), and Geneva Steel, although generally only a small percentage of the slab production (10-
15%) is hot charged (Ritt, 1996). We assume that 60% of cold rolled products (36% of the slabs)
can ultimately be “hot charged”, depending on the lay-out of the plants. A plant-by-plant analysis
is required to determine the actual potential. Assuming a charging temperature of 700°C, the
savings may be up to 0.6 GJ/t “hot charged” steel based on experiences at Bethlehem Steel at
Burns Harbor (Ritt, 1996). Additional annual costs savings amount up to $1.15/t “hot charged”.
Investment costs will strongly depend on lay-out and are estimated to be $15/t hot rolled steel
based on experience at LTV (Wakelin, 1997).
Process control in hot strip mill saves energy and increases productivity and quality of the
rolled steel products (Heesen and Burggraaf, 1991; Schriefer, 1996; Vergote, 1996). Although
direct energy savings may be limited, the indirect energy savings may be substantial due to
reduced rejection of product, improved productivity, and reduced down-time. Based on a system
installed at Sidmar (Belgium) the share of rejects was reduced from 1.5% to 0.2% and down-time
was reduced from more than 50% of the time to 6%. The costs of rolling were reduced from $7/t to
$4.7/t (Vergote, 1996). Similar systems have been installed in mills in many countries. We
estimate the energy savings based on the reduced rejection rate and improved productivity to be
9% of fuel use. We assume this to be equivalent to 0.3 GJ/t product. The investment costs for the
Sidmar plant were estimated to be $2M for a hot strip mill with a capacity of 2.8 Mt (Serjeantson,
1987), equivalent to $0.7/t product. This measure will be applicable to all slabs that are not cast in
a thin-slab caster or sold, i.e. 69% of the total steel production. The lifetime of process control
equipment is estimated at 10 years.
Recuperative burners in the reheating furnace can reduce energy consumption. Industry-wide
average savings for the metals industry are estimated to be up to 30% (Worrell et al., 1997).
Energy use in a reheating furnace will depend on production factors (e.g. stock, steel type),
operational factors (e.g. scheduling), and design features. Therefore, in practice energy
consumption can vary widely between 0.6 and 3.0 GJ/t (Flanagan, 1993), with the low figures due
to hot charging (see above). Based on a survey of 151 furnaces (representing 20% of Western world
steel production) in Japan, Australia, UK and Canada, it was found that 18% of the furnaces had
no heat recovery and 75% had separate heat recovery (Flanagan, 1993). As no specific US data
were available, we assume a similar distribution for the US Installing recuperative or regenerative
88

urners may require substantial changes in the furnace construction and may have high
investment costs. New designs have typically low NOx emissions, despite higher flame
temperatures. We assume installing regenerative burners in 20% of the furnaces used in hot
rolling mills, saving approximately 25% on fuel in these (mostly small) furnaces, based on
experiences in the UK (Flanagan, 1993), or roughly estimated at 0.7 GJ/t product. The
investments for a 12t/hour furnace were approximately $2-3/t. We assume $2.5/t product. The
burners are expected to have a lifetime of approximately 10 years.
Insulation of furnaces using ceramic low-thermal mass insulation materials (LTM) can reduce
the heat losses through the walls further than conventional insulation materials. A survey of steel
reheating furnaces in the steel industry in four countries (not including the US) showed that
approximately 30% of the furnaces had ceramic fiber linings (Flanagan, 1993). We assume a
similar figure for the US steel industry. For a continuous furnace, the savings of implementing
ceramic fiber lining are estimated to be 2-5% (Flanagan, 1993). We assume savings of 0.16 GJ/t
product. We assume that 30% of the furnace capacity can be equipped with ceramic lining during
maintenance and reconstruction (assuming an approximate life-time of 30 years) in the period
until 2005. Although we did not find recent cost data, we assume relative large investments of
approximately $10/t product, derived from de Beer et al. (1994). The lifetime is estimated at 10
years.
From the DOE assessments the improvement of insulation on the reheating furnaces would
decrease the energy use by 0.04 GJ/tonne or 0.24 $/tonne. The mentioned amount of 0.16 GJ
per tonne is therefore a high estimate for 2006. The estimated total savings from furnace
insulation is therefore 0.2GJ/tonne. From analysis of the DOE reports, the improvement of
furnace insulation was suggested to 9 of the 12 plants which an analysis of reheating furnaces
was done. These updated figures are used in the 2006 and 2010 data.
Controlling oxygen levels and variable speed drives on combustion air fans on the
reheating furnace helps to control the oxygen level, and hence optimize the combustion in the
furnace, especially as the load of the furnace may vary over time. The savings depend on the load
factor of the furnace and control strategies applied. Two cases from the UK steel industry
demonstrate the variety. Implementing a variable speed drive combustion fan on a walking beam
furnace at Cardiff Rod Mill (UK) reduced the fuel consumption by 48% with a payback period of
16 months (1985 UK conditions) (Caddet, 1994). Another example (without installing variable
speed drives) is a walking beam furnace for reheating billets, saving approximately 2% on fuel use,
with a payback of one year (1990 UK conditions) (Flanagan, 1993). We conservatively assume
savings of 10% (after previous measures have been introduced), equivalent to 0.33 GJ/t product,
at an investment of 0.5$/t product. As no data is available on the current penetration of VSDs in
reheating furnaces, we assume that this measure can be implemented in half of the furnaces, with
a lifetime of approximately 10 years.
According to the DOE assessments the optimization of oxygen levels would reduce the energy
consumption with on average 0.035 GJ/tonne. The use of VSD on combustion air fans could
reduce the fuel and electricity use of the combustion fans. Due to over engineering of equipment
the typical flow range of a fan is about 60-80% (Vesel, 2007). If a VSD is used to keep the fan at
70%, this could save about 60% in electricity use of the fan (Saidur, 2010). This would mean a
89

minor reduction of electric energy of about 0.001 GJ/tonne of crude steel. The application of a
VSD on a boiler house, reduced the fuel consumption with 2.5%, with a payback period of about
3.2 months (Kilicaslan, 2005). It is estimated the technology would reduce energy in reheat
furnaces with about 3% which is about 0.05 GJ/ton. Also the investment for a boiler was about
$9,000, it is therefore estimated the investment for reheating furnaces is about 0.1 $/tonne
(Kilicaslan, 2005). These updated figures have been used for the 2006 and 2010 data.
Energy efficient drives in the hot rolling mill can replace the currently used conventional
AC drives. The efficiency of large AC drives (> 200 kWe) is estimated to be 91-97% (Worrell and
Moore, 1997). High efficiency motors can save approximately 1-2% of the electricity consumption
(de Almeida and Fonsesca, 1997). Assuming an electricity demand of 200 kWh/t rolled steel, the
electricity savings are estimated to be 4 kWh/t, or 0.01 GJ/t product. Replacement costs are
estimated to be $5/kW (the extra costs compared to that of an ordinary drive) (de Almeida and
Fonsesca, 1997), equivalent to $0.05/kWh-saved, or $0.2/t rolled steel. Large motors have
generally a lifetime of 20 years (de Almeida and Fonsesca, 1997). According to Rosenberg (1997)
the average penetration of efficient motors in all industrial applications is between 6 and 8%. We
assume that 50% of the motors will be replaced at the above mentioned costs.
Waste heat recovery from cooling water. Waste heat can be recovered from the cooling
water of the hot strip mill. When ejected, the rolled steel is cooled by spraying water at a
temperature of 80 o C. An absorption heat pump (or heat transformer) has been installed at
Hoogovens (The Netherlands) to generate low pressure steam (1.7-3.5 bar, 130 o C), which is
delivered to the grid on the site. Fuel savings are estimated to be 0.04 GJ/t product, with an
increased electricity consumption of 0.15 kWh/t (Farla et al., 1998). Investment costs are 42
Dfl/GJ-saved equivalent to $0.8/t product (Worrell et al., 1993), with increased O&M costs
estimated at $0.07/t product. The heat transformer could be applied with all quench water in the
hot rolling mills, e.g. 69% of the total production. The life time is estimated to be 15 years.
Ceramic wall in reheat furnace. A ceramic porous wall in a reheating furnace for billets
resulted in energy savings of up to 22% on the reheating of billets (Tucker, 2008). The flue gas
will pass through the porous wall, the heat is absorbed by the materials and for a substantial
part reradiated back into the furnace. The energy savings for this measure are estimated to be
about 0.3 GJ/tonne of hot rolled steel. The investment costs have been estimated to be about
$2/tonne of hot rolled steel. Application of the technology is expected not be applied in any steel
furnaces in 2006. In 2010 maybe 5% of the steel factories may have applied this method. The
potential is expected to be applicable for all the steel plant with reheating furnaces. This option
is only included in 2010.
Reduce reheat furnace door opening losses. From the assessment of DOE audits a
reduction in energy loss by the furnace door opening should be possible according to an audit in
four steel plants in the US. The estimated energy savings from reducing these losses are
0.014GJ/tonne of steel. The estimated investment needed for the reduced losses through door
openings is $0.2/tonne of steel. Assumed to be implemented in all non-thin-slab casting
facilities. This measure has only been added in 2006 and 2010 data.
90

Appendix H. Cold Rolling and Finishing 8
Steel that has been hot rolled may be cold rolled and further finished to make a product thinner
and smoother. In 1994, 31.7 Mt (35%) of product was cold rolled, all in integrated mills. Based on
fuel consumption of 43 PJ and electricity consumption of 15 PJ, the primary energy intensity was
2.8 GJ/t.
Heat recovery on the annealing line can be done through steam generation using the waste heat,
or by installing regenerative or recuperative burners in the annealing furnace (Meunier and
Cambier, 1993). We aggregate the various energy saving opportunities in one measure, as the total
energy consumption in the annealing stage is limited. Energy use for batch annealing is estimated
at 1.0 GJ/t fuel and 25 kWh/t, and for continuous annealing 0.8 GJ/t and 45 kWh/t (IISI, 1982).
Energy use can be reduced by up to 40% (Meunier and Cambier, 1993), by implementing heat
recovery (using regenerative burners), improved insulation, process management equipment, as
well as variable speed drives. We estimate the savings at 0.3 GJ fuel/t and 3 kWh/t. All cold rolled
steel is assumed to be treated in the annealing furnace, i.e. 30.9 Mt (1994). The total potential
energy savings are estimated at 9 PJ. The investment costs are estimated at $2.7/t, based on
practices at Hoogovens (The Netherlands).
Reduced steam use in the pickling line. In the pickling line heat escapes through
evaporation from the hydrochloric acid bath. The bath is normally heated to temperatures of 95°C
(IISI, 1982). The IISI (1982) reports that steam use can be reduced by 5kg/t, with an assumed
steam use of 30 kg/t, by installing a system of lids and floating balls on top of the bath. This is
equivalent to savings of 17%. For the US steel industry we estimate the savings (including boiler
losses) to be 0.19 GJ/t. At a production of 32 Mt cold rolled product, the total fuel savings are
estimated to be 6 PJ. No investment cost data were available for this study. We estimate the costs
on the basis of a conservative estimate by de Beer et al. (1994) at $2.8/t.
Automated monitoring and targeting system. Installing an automated monitoring and
targeting system at a cold strip mill can reduce the power demand of the mill, as well as reducing
effluents. A system installed at British Steel at Brinsworth Strip Mills, reduced the energy demand
of the cold rolling mill by approximately 15-20%, depending on the load factor (Caddet, 1990b).
The savings are estimated to be 60 kWh/t assuming an average electricity consumption of 360
kWh/t (US DOE, OIT, 1996). We assume the implementation of a similar system, at installation
costs of $1.1/t product ($0.63/t crude steel) (Caddet, 1990b), for half of the cold strip mills in the
US steel industry, or 17% of the total steel production.
8 One measure in cold rolling is continuous annealing, which will reduce the heat losses of the batch furnaces but
demands relative high investment costs. We do not assume implementation of this measure as an energy efficiency
measure.
91

Appendix I. References
Adolph, A., G. Paul, K.H. Klein, E. Lepoutre, J.C. Vuilermoz, and M. Devaux, 1990. “A New
Concept for Using Oxy-Fuel Burners and Oxygen Lances to Optimize Electric Arc Furnace
Operation,” La Revue de Metallurgie – CIT 87(1): 47-53.
American Iron and Steel Institute, 1995. 1994 Annual Statistical Report, Washington, DC: AISI.
American Iron and Steel Institute, 1996. 1995 Annual Statistical Report, Washington, DC: AISI.
American Iron and Steel Institute, 2010. 2010 Annual Statistical Report, Washington, DC: AISI.
Anonymous, 1994. “Energy Saving VSD Quench Pumps,” Steel Times, April: 150.
Anonymous, 1995. “Natural Gas Injection Tests Show Benefits,” Iron & Steelmaker 22(10): 12.
Anonymous, 1996. “BS announces 70M pound Investment,” Ironmaking and Steel making,”
23(1).
Anonymous, 1997a. “$137 Million Expansion Project to Increase Capacity to 1.5 Mt,” Iron & Steel
Maker 24(2).
Anonymous, 1997b. “Hoogovens envisage l’installation d’une coulee continue des brames minces,”
La Revue de Metallurgie-CIT 94(3): 583
Bureau of Economic Analysis (BEA). 2011. National income and product accounts -NIPAs, 2011.
Section1 all tables. GDP and Other Major NIPA Aggregates. US Department of Commerce.
http://www.bea.gov/.
Beentjes, P.A., S.P.A.M. Wokke, and J. van Breda, 1989. “Benefits of Automation of Hot Blast
Stove Operation,” Proceedings 1989 AISE Iron and Steel Exposition and Annual Convention,
September, 18-21, 1989, Pittsburgh, PA: AISE.
Bosley, J. and D. Klesser, 1991. The Consteel Scrap Preheating Process, CMP Report 91-9, Center
for Materials Production, Pittsburgh, PA.
CADDET, 1987. A Horizontal Ladle Preheating Station fired with a Self-Recuperative Burner
Improves Steel Production Operations, (Project 2B.F06.1451.87.UK), Sittard, The Netherlands:
CADDET.
CADDET, 1989. “Improved Design for Foundry Ladle Pre-Heaters,” CADDET Energy Efficiency
Register 3.0 (project UK-89-003), Sittard, The Netherlands: CADDET.
CADDET, 1990a. “Energy Saving by Scale Removal from Charging Slab,” CADDET Energy
Efficiency Register 3.0 (project JP-90-022), Sittard, The Netherlands: CADDET.
CADDET, 1990b. “Computer-based Monitoring and Targeting on a Rolling Mill,” Result 139
(Project UK 90.056/2B.FO3), Sittard, The Netherlands: CADDET.
92

Caddet, 1994. “Variable Speed Drive on a Large Continuous Furnace Combustion Air Fan,”
CADDET Energy Efficiency Register 3.0 (project UK-94-452), Sittard, The Netherlands:
CADDET.
Caffal, C., 1995. “Energy Management in Industry,” CADDET Analyses Series 17, Sittard, The
Netherlands: Caddet.
Center for Materials Production, 1987. Technoeconomic Assessment of Electric Steel making
Through the Year 2000, EPRI/CMP, Report 2787-2, October 1987.
Center for Materials Production, 1991. Direct Current Electric Arc Furnaces, Tech Commentary
CMP-063, CMP, Pittsburgh, PA.
Center for Materials Production, 1992. Electric Arc Furnace Efficiency, EPRI/CMP, Report 92-10,
Pittsburgh, PA: CMP.
Center for Materials Production, 1995. Coal & Oxygen Injection in Electric Arc Furnaces, Tech
Bulletin CMP 95-7TB, CMP, Pittsburgh, PA.
Center for Materials Production. 1997. Electric Arc Furnace Scrap Preheating. Tech
Commentary, Pittsburgh, PA: Carnegie Mellon Research Institute.
Cores, A., A. Formoso, M. Sirgado, J. L. Verduras, and L. Calleja. 1996. "Recovery of Potential
Thermal Energy of Rolling Mill Waste Oil Through Sintering," Ironmaking and Steel making
23(6): 486-492.
Dawson, P.R., 1993. “Recent Developments in Iron Ore Sintering, Part 4: The Sintering
Process,” Ironmaking and Steel making 20(2): 150-159.
de Almeida, A. and P. Fonsesca, 1997. “Characterisation of the Electricity Use in European Union
and the Savings Potential 2010,” in: A. de Almeida, P. Bertoldi and W. Leonhard (eds.), Energy
Efficiency Improvements in Electric Motors and Drives, Berlin, Germany: Springer Verlag.
de Beer, J.G., M.T. van Wees, E. Worrell and K. Blok, 1994. ICARUS-3, The Potential of Energy
Efficiency Improvement in The Netherlands from 1990 to 2000 and 2015, Utrecht, The
Netherlands: Department of Science, Technology & Society, Utrecht University.
de Castro, J.A., Nogami, H., et al. (2002), Numerical investigation of simultaneous injection of
pulverized coal and natural gas with oxygen enrichment to the blast furnace, Isij International
42(11), p. 1203-1211
Derycke, J., R. Bekaert, P. Cousein, L. Bonte, and H. Bruneel, 1990. “Automation of Hot Blast
Stove Operation at Sidmar: Control and Optimisation of Energy Consumption,” Ironmaking and
Steel making 17(2): 135-138.
DOE energy audits, Steel – Plant assessments performed for plants in the steel industry.
http://www1.eere.energy.gov/manufacturing/tech_deployment/partners/by_industry_list.cfm
?industry=Steel Visited Mar, 2012.
93

Dong, H., Jie Li, et al., 2010, Grade Recovery and Cascade Utilization of Residual Heat in Sinter
Cogeneration System, Proceedings of the 2010 Asia-Pacific Power and Energy Engineering
Conference (APPEEC 2010), p. 5
Douglas, J., 1993. “New technologies for Electric Steel making” EPRI Journal, October/
November 1993, pp.7-15.
Duan, J. P., Y. L. Zhang, et al. (2009), EAF steelmaking process with increasing hot metal
charging ratio and improving slagging regime, International Journal Of Minerals Metallurgy
And Materials 16(4), p. 375-382
Dungs, H. and U. Tschirner, 1994. “Energy and Material Conversion in Coke Dry Quenching
Plants as Found in Existing Facilities,” Cokemaking International 6(1): 19-29.
Duraloy (2011), Duraloy dry rolls for use in heating zones, Duraloy technologies, Scottsdale PA,
USA, http://www.duraloy.com/csptunnelrolls.html, visited Mar 2012.
EPA, Environmental protection agency (2010), Available and emerging technologies for reducing
greenhouse gas emissions from the iron and steel industry, Policies and programs division.
Eremin, A., V. G. Mishchikhin, et al. (2011) Using coke-battery flue gas to dry coal batch before
coking, Coke and Chemistry 54(3), p. 77-85
Errera, M.R., Milanez, L.F., (2000), Thermodynamic analysis of a coke dry quenching unit,
Energy Conversion and Management 41(2), p. 109
ETSU, 1992. “Reduction of Costs Using an Advanced Energy Management System,” Best Practice
Programme, R&D Profile 33, Harwell, UK: ETSU
Farla, J.C.M., E. Worrell, L. Hein, and K. Blok, 1998. Actual Implementation of Energy
Conservation Measures in the Manufacturing Industry 1980-1994, The Netherlands: Dept. of
Science, Technology & Society, Utrecht University.
Ferretti, I., Zanoni, S., Zavanella, L. (2006), Energy Efficiency in a steel plant using optimizationsimulation,
Università degli Studi di Brescia, Brescia, Italy.
Finlay, P., (2004), Guidelines on Best Available Techniques (BAT) for Sinter Plants in the Iron
Industry,
http://www.pops.int/documents/meetings/bat_bep/2nd_session/egb2_followup/draftguide/5
D2IronSinteringDRAFTc.pdf, visited March, 2012.
Fitzgerald, F., 1992. “Energy Use and Management in British Steel Plc.,” Ironmaking and Steel
making 19(2): 98-106.
Flanagan, J.M., 1993. “Process Heating in the Metals Industry,” CADDET Analyses Series 11,
Sittard, The Netherlands: CADDET.
Flemming, G., 1995. Personal Communication with G. Flemming, SMS, Dusseldorf, Germany.
94

Gitman, G., 1998. American Combustion, Inc., Personal Communication, August 12 th , 1998.
Gongfa, L., Kong Jianyi, et al. (2010), Intelligent Control of Coke Oven Production Process,
Advanced Materials Research 129-131, p. 198-203
Grant, M.G., (2000), Principals and strategy of EAF post combustion, 58 th Electric Furnace
Conference – Orlando (USA) –Nov 12-15th, 2000.
H. Sakaue (2009), Improvement of Production by Decreasing Air Leak at Nagoya No. 3
Sintering Plant, Tetsu To Hagane-Journal Of The Iron And Steel Institute Of Japan 95(7), p. 52-
55
Haissig, M., 1994. “Enhancement of EAF Performance by Injection Technology” Steel Times,
October 1994 pp.391-393.
Hanes, C., 1999. USS/Kobe Steel, Personal communication, June 1999.
Heesen, G.J. and D.H. Burggraaf, 1991. “New Process Control System of Hoogoven’s Hot Strip
Mill - Key to Improved Product Quality,” Ironmaking and Steel making 18(3): 190-195
Hendriks, C.A., 1994. Carbon Dioxide Removal From Coal-Fired Power Plants, Kluwer Academic
Publishers, Dordrecht, The Netherlands.
Herin, H.H. and T. Busbee, 1996. “The Consteel® Process in Operation at Florida Steel” Iron &
Steelmaker 23(2): 43-46.
Hofer, L., 1996. Electric Steel making with FUCHS Shaft Furnace Technology, Linz, Austria:
Voest Alpine Industrieanlagenbau Gmbh, VAI.
Hofer, L., 1997. Personal communication, Voest Alpine Industrieanlagenbau Gmbh, Linz, Austria,
25 September 1997.
Hogan, W.T., 1992. Capital Investment in Steel, A World Plan for the 1990’s, New York, NY:
Lexington Books.
Hogan, W.T., and F.T. Koelble, 1996a. “Fewer Blast Furnaces, but Higher Productivity,” New Steel
12(11): 62-66.
Hogan, W.T., and F.T. Koelble, 1996b. “Steel’s Coke Deficit: 5.6 Million Tons and Growing,” New
Steel 12(12): 50-60.
Hsun Lin, P., Wang., P., Huang, T., (2009), Exergy Analysis of a Coke Dry Quenching System,
China Steel Technical Report, No. 22, pp. 63-67
I&SM, 1997a. “Iron & Steelmaker’s 1997 Blast Furnace Roundup,” Iron and Steelmaker 24(8):
24-2
I&SM, 1997b. "Electric Arc Furnace Roundup – United States," Iron and Steelmaker 24(5): 20-
39.
95

Inoue, K., 1995. “The Steel Industry in Japan: Progress in Continuous Casting.”
International Energy Agency, 1995. Energy Prices and Taxes, First Quarter 1995, Paris: IEA.
International Iron and Steel Institute (IISI). 1999. IISI web site:
http://www.worldsteel.org/steeldatacentre/lgcountry.htm. Brussels: IISI.
International Iron and Steel Institute, 1993. World Cokemaking Capacity, Brussels, Belgium:
IISI.
International Iron and Steel Institute, Committee on Technology, 1982. Energy and the Steel
Industry, Brussels, Belgium: IISI.
Janz, J. and W. Weis, 1996. “Injection of Waste Plastics into the Blast Furnace of Stahlwerke
Bremen,” La Revue de Metallurgie-CIT 93(10): 1219-1226.
Jones, J. A. T. 1996. "New Steel Melting Technologies: Part III, Application of Oxygen Lancing
in the EAF." Iron and Steelmaker 23(6): 41-42.
Jones, J. A. T. 1997a. "New Steel Melting Technologies: Part X, New EAF Melting Processes."
Iron and Steelmaker 24(January): 45-46.
Jones, J. A. T. 1997b. Electric Arc Furnace Evolution: In Search of the Optimal Design. Bechtel
Corporation..
Jones, J. A. T. 1997c. "New Steel Melting Technologies: Part XVI, CONSTEEL Process." Iron
and Steelmaker 24(July): 47-48.
Jones, J. A. T. 1997d. "New Steel Melting Technologies: Part XV, Fuchs Shaft Furnace." Iron
and Steelmaker 24(June): 43-45.
Jones, J.A.T., 1993. Increased EAF Productivity through Improved Operating Efficiency, Nupro
Corporation.
Jones, J.A.T., 1998. Acutus Gladwin, Personal Communication, August 8 th , 1998.
Jones, R. T., Q. G. Reynolds, et al. Some myths about DC arc furnaces, Journal Of The South
African Institute Of Mining And Metallurgy 111(10), p. 665-673
Kelly, J., Dentella, F., Recanati, A., Visus, J., Miclo, E., (2010) Oxygen-Enhanced Ladle
Preheating Systems: Improved Tap-to-Tap Cycle Time and Operating Cost Reductions, AISTech
2010 - The Iron & Steel Technology Conference and Exposition, Pittsburgh, Pa.
Kilicaslan, I., and E. Ozdemir (2005), Energy economy with a variable speed drive in an oxygen
trim controlled boiler house, Transactions of the ASME. Journal of Energy Resources Technology
127(1), p. 59-65.
Kimmerling, K., 1997. Personal communication and reference list, Neural Applications
Corporation, Coralville, IA (26 August 1997).
96

Kowalski, W., K-H. Peters, W. Cronert, P. Kuhn, and D. Sucker, 1990. “Optimierung der Brenner
von Winderhitzern im Hinblick auf einen hohen CO-Ausbrand,” Stahl u. Eisen 110(11): 41-50.
Ladoux, P., Postiglione, G., Foch, H., Nuns, J., (2005), A comparative study of AC/DC Converters
for High-Power DC arc furnace, IEEE Transactions on Industrial Electronics 52(3), p. 747 – 757
Lahita, J.A., 1995. “The Consteel® Process in Operation at New Jersey Steel Corporation”
Proceedings 5 th European Electric Steel Congress, Paris, June 19-23, 1995.
Lanzer, W. and H.B. Lungen, 1996. “Roheisenerzeugung in Nordamerika,” Stahl und Eisen
116(8): 61-69.
Li, H. F., W. J. Bao, et al. (2010) Energy recovery and abatement potential of CO2 emissions for
an integrated iron and steel making enterprise, Science China-Technological Sciences 53(1), p.
129-133
Macauley, D. and R.M. Smailer, 1997. “Engineering Fundamentals for a Least Cost/Flexible Steel
making Solution” Paper presented at 25 th Advanced Technology Symposium on New Melting
Technologies, St. Petersburg Beach, FL, May 11-14, 1997.
Maiolo, J., Butazakhti, M., Li, W.L. (2006), Developments towards an Intellegent Electric Arc
Furnace at CMC Texas using Goodfellow EFSOP(R) Technology, Tenova Goodfellow Inc.,
Mississauga, ON.
McAloon, T.P., 1994. “Alternate Ironmaking Update,” Iron & Steelmaker 21(2): 37-39 + 55.
Meunier, H. and M. Cambier, 1993. “Use of Furnace Modelling to Improve Energy Efficiency in
the Deepdrawing Steel Sheet Industry,” P.A. Pilavachi (ed.), Energy Efficiency in Process
Technology, Amsterdam: Elsevier Applied Science.
Midrex, 1993. The Midrex Direct Reduction Process, Charlotte, NC: Midrex Direct Reduction
Corporation.
Midrex, 1995. 1994 World Direct Reduction Statistics, Charlotte, NC: Midrex Direct Reduction
Corporation.
Mueller, E.G., 1997. “High Production Meltshops: Trends and Innovations” Paper presented at
25 th Advanced Technology Symposium on New Melting Technologies, St. Petersburg Beach, FL,
May 11-14, 1997.
Nashan, G., 1992. “Conventional Maintenance and the Renewal of Cokemaking Technology,” In:
IISI, Committee on Technology, The Life of Coke Ovens and New Coking Processes under
Development, Brussels: IISI.
Nelson, K., 1994. “Finding and Implementing Projects that Reduce Waste,” in: R. Socolow, C.
Andrews, F. Berkhout and V. Thomas, Industrial Ecology and Global Change, Cambridge, UK:
Cambridge University Press.
Ninneman, P., 1997. “New Melt Shops and Rolling Mills” New Steel 13 (9): pp.40-58.
97

Oda, J., Akimoto, K. (2007), Evaluation of Energy Saving and CO2 Emission Reduction
Technologies in Energy Supply and End-use Sectors Using a Global Energy Model, Transaction
on electrical and electronic engineering, IEEJ Trans; 1: p72–83
Ooi, T., et al. (2008), The study of sunflower seed husks as a fuel in the iron ore sintering
process, Minerals Engineering 21, p167–177
Opfermann, A., Riedinger, D. (2008), Energy efficiency of electric arc furnace, International Arab
Iron and Steel Conference 2008, Qatar, Badische Steel Engineering GmbH
Oshnock, T.W., 1995a. “Pulverized Coal Injection for Blast Furnace Operation, Part IV,” Iron &
Steelmaker 22(2): 41-42.
Oshnock, T.W., 1995b. “Pulverized Coal Injection for Blast Furnace Operation, Part VI,” Iron &
Steelmaker 22(4): 49-50.
Palasios, J.M. and J.L. Arana,1995. “Research and Development of the EAF in Europe”
Proceedings 5 th European Electric Steel Congress, Paris, June 19-23, 1995, pp.391-400.
Pisila, E., S. Kallo, T. Ahalo, and K. Heinanen, 1995. “High Productivity Operation of Rautaruukki
Blast Furnaces,” La Revue de Metallurgie-CIT 92(3): 375-380.
Rengersen, J., Oosterhuis, E., de Boer, W.F., Veel, T.J.M. and Otto, J. 1995. “First Industrial
Experience with Partial Waste Gas Recirculation in a Sinter Plant,” Revue de Metallurgie-CIT 3
92 pp. 329-335 (1995).
Riley, M.F. and S.K. Sharma, 1987. “An Evaluation of the Technical and Economic Benefits of
Submerged Inert Gas Stirring in an Electric Arc Furnace” Iron & Steelmaker 14(6): 27-32.
Ritt, A., 1996a. “The Benefits of Hot-Charging Slabs,” New Steel 12(7): 34-37+44.
Ritt, A., 1996b. “Building More Powerful Melt Shops,” New Steel 12(11).
Ritt, A., 1997. “Acme Rolls 0.030 Inch Hot Band,” New Steel 13(5).
Roederer, C. and L. Gourtsoyannis, 1996. Coordinated Study Steel-Environment, Luxembourg:
European Commission, DG-XII.
Rosenberg, M, 1997. “The United States Motor Systems Baseline: Inventory and Trends,” in: A. de
Almeida, P. Bertoldi and W. Leonhard (eds.), Energy Efficiency Improvements in Electric Motors
and Drives, Berlin, Germany: Springer Verlag.
Saidur, R., Ahamed, J.U., Masjuki, H.H., Energy, (2010), Exergy and economic analysis of
industrial boilers, Energy Policy 38, p.2188–2197
Schade, R.J., 1991. Bottom Stirring in an Electric Arc Furnace, Center for Metals Production,
Pittsburgh, PA, February 1991.
Schorsch, L. L., 1996. “Why Minimills Give the US Huge Advantages in Steel,” McKinsey
Quarterly (2):44-55.
98

Schriefer, J., 1996. “Improving Quality by Better Process Control,” New Steel 12(4): 81-83.
Schriefer, J., 1997. “Reaping the Value from Dust and Slag,” New Steel 13(2): 24-33.
Schuett, K.J., and D.G. White, 1997. “Record Production on US Steel Gary Works’ No. 13 Blast
Furnace with 450 Pounds/THM Co-Injection Rates,” Iron and Steelmaker, 24(3): 65-68.
Serjeantson, R., R. Cordero and H. Cooke, 1987. Iron and Steel Works of the World, 9 th edition,
Worcester Park, UK: Metal Bulletin Books Ltd.
Siemens-VAI (2011a), Metals technologies, Solutions for sinterplants, Order No. E10001-M3-
A6-V3-7600.
Siemens-VAI (2011b), Metals technologies,Simetal EAF Quantum, Order No. E10001-M3-A125-
V2-7600, claimed at:
http://www.industry.siemens.com/datapool/industry/industrysolutions/metals/simetal/en/SI
METAL-EAF-Quantum-en.pdf
Staib, W.E. and N.G. Bliss, 1995. “Neural Network Control System for Electric Arc Furnaces”
Metallurgical Plant & Technology International 2: 58-61.
Stelco, 1993. Present and Future Use of Energy in the Canadian Steel Industry, Ottawa, Canada:
CANMET.
Stockmeyer, R., K-H. Heinen, H. Veuhoff, and H. Siegert, 1990. “Einsparung von elektrischer
Energie am Lichtbogenofen durch eine neue Ausqualmregelung” Stahl u. Eisen 110(12): 113-116.
Teoh, L.L., 1989. “Electric Arc Furnace Technology: Recent Developments and Future Trends”
Ironmaking and Steel making 16(5): 303-313.
Toulouevski, Y.N., Zinurov, I.Y., (2010), Innovation in Electric arc furnaces, Heidelberg,
Springer, ISBN 978-3-642-03800-6
Tucker, R. J., J. Ward, et al. (2008), Development and experimental testing of a porous ceramic
wall to improve the thermal performance of steel billet reheating furnaces, Journal Of The
Energy Institute 81(3), p. 135-142
US Department of Energy, Office of Industrial Technologies, 1996. Energy and Environmental
Profile of the US Iron and Steel Industry, Washington DC: US DOE, OIT.
Uematsu, H., 1989. “Control of Operation and Equipment Prevents Coke Oven Damage,”
Ironmaking Conference Proceedings, Warrendale, PA: Iron and Steel Society.
VAI, 1997. FUCHS Shaft Furnaces, The Power, The Performance, The Profit, Linz, Austria: Voest
Alpine Industrieanlagenbau Gmbh,
Vergote, H, 1996. “New Technologies in Process Control for Hot Strip Mills,” Iron & Steelmaker
23(2): 21-25.
99

Vesel, R., J. Gregory, et al. (2007), Drive technology enables combustion optimization, Power
Engineering 111(4), p. 50.
Wakelin, D.H., 1997. Personal communication with David H. Wakelin, Manager of Development
Engineering, LTV Cleveland Works, December 1.
Walli, R.A., 1991. Adjustable Speed Drives for Electric Arc Furnace Air Pollution Control
Systems, Center for Metals Production, Pittsburgh, PA, December 1991.
Worrell, E. and C. Moore, 1997. “Energy Efficiency and Advanced Technologies in the Iron and
Steel Industry,” in: Proceedings 1997 ACEEE Summer Study on Energy Efficiency in Industry,
Washington, DC: ACEEE.
Worrell, E., J.G. de Beer, and K. Blok, 1993. “Energy Conservation in the Iron and Steel Industry,”
in: P.A. Pilavachi (ed.), Energy Efficiency in Process Technology, Amsterdam: Elsevier Applied
Science.
100

Appendix J. Calculation example
In order to provide some insight in the calculation of the total energy savings potential for
energy efficiency measures in the different steel industries. Here an example calculation of one
energy efficiency measures will be shown.
For this example we will take the ‘Improved blast furnace control systems’. For this measure an
energy saving of 0.4 GJ/tonne of hot metal was found. The capital costs were estimated at $0.36
per tonne of hot metal, while there were no changes in operating costs expected. It was expected
the measures was applicable to 50% of the US steel industry. For this example, just like in other
calculation, we will use a discount rate of 30% with a usual measure lifetime of 10 years.
The capital recovery factor is described in the formula below,
d
q =
−
( 1−
( 1+
d)
n
)
From this formula we find a capital recovery factor of 0.32.
Then to calculate the CCE we use the following formula below,
∙∆
The investment cost of the measure are given above, $0.36 per tonne of hot metal, and no other
non-energy benefits are included (so M and B are 0). Delta E is the energy savings, which is 0.4
GJ/ tonne of hot metal in this case. So the CCE can be calculated, which is $0.28/GJ.
This is much lower than the average weighted fuel price, which for the US was $6.76/GJ, so the
measure can be considered cost-effective.
With a total production of integrated steel in the US in 2006 of 37.9 Mt of hot metal, the
measure could save a potential of 15.2 PJ in the US. Provided the measure is expected to be
applicable for only 50%, as mentioned above, the total potential of this measure 7.1 PJ for an
average cost of $0.28 per GJ saved.
101

102
Appendix K. Results of energy efficiency calculation for the US 2006
2006 BASELINE
Secondary Steelmaking
Steelmaking Electric Arc Furnace
Applied
Carbon
Savings
Applied
Final
Energy
Savings
Cost of
Measure
Operation
Cost Change
Measure
Lifetime
Applied
Carbon
Savings Final CCR Final CCE
Primary
Capital
Recovery
CCE Final MCCE Factor
kgC/tonne
steel (GJ/tonne) (US$/tonne) (US$/tonne) (years) (M illion kgC) US$/tC (US$/GJ) (US$/GJ) (US$/GJ)
Real discount
rate
Improved process control (neural network) 3.3 0.07 0.63 (0.66) 10.0 324 (139.45) (6.21) (2.28) (18.76) $0.32
Fluegas Monitoring and Control 0.5 0.01 0.73 (0.49) 15.0 49 (526.42) (19.71) (8.47) (30.90) $0.31
Transformer efficiency - UHP transformers 1.2 0.03 0.71 - 15.0 116 183.13 8.15 3.00 (4.40) $0.31
Bottom Stirring / Stirring gas injection 0.2 0.00 0.05 (0.16) 0.5 20 (215.77) (9.60) (3.53) (22.15) $2.44
Foamy slag 0.1 0.00 1.47 (0.26) 10.0 11 1,925.44 85.68 31.50 73.13 $0.32
Oxy-fuel burners 1.5 0.00 1.09 (0.32) 10.0 151 23.32 16.63 0.41 (124.83) $0.32
Eccentric Bottom Tapping (EBT) on existing furnace 0.1 0.00 0.12 - 20.0 7 518.10 23.05 8.48 10.50 $0.30
DC-Arc furnace 1.7 0.04 0.57 (0.37) 30.0 165 (116.29) (5.17) (1.90) (17.72) $0.30
FUCHS Shaft furnace 2.7 0.01 1.54 (1.03) 30.0 270 (206.53) (88.66) (3.66) (175.40) $0.30
Twin Shell w/ scrap preheating 0.4 0.01 z (0.08) 30.0 36 141.06 6.28 2.31 (6.27) $0.30
Recover heat from waste gas 0.3 0.02 0.91 (0.09) 10.0 31 651.70 8.94 8.94 2.32 $0.32
Post combustion of CO gas 2.6 0.06 1.37 (0.41) 10.0 252 12.55 0.56 0.21 (11.99) $0.32
Increased usage of hot metal 0.4 0.01 0.51 (0.09) 10.0 38 207.27 9.22 3.39 (3.33) $0.32
Secondary Casting
Efficient ladle preheating 0.1 0.00 0.01 - 10.0 6 74.12 0.89 0.89 (7.18) $0.32
Proper sealing on ladle furnace preheating 0.2 0.02 0.06 - 10.0 24 76.61 0.92 0.92 (7.15) $0.32
Near net shape casting/thin slab casting 9.4 0.57 28.54 (6.66) 20.0 924 207.86 3.44 2.68 (5.38) $0.30
Use dry rolls in tunnel ovens for TSC 0.9 0.07 0.74 (0.00) 20.0 84 258.70 3.12 3.12 (4.96) $0.30
Secondary Hot Rolling
Process control in hot strip mill 1.3 0.10 0.27 - 10.0 131 65.92 0.90 0.90 (7.17) $0.32
Recuperative burners 3.1 0.23 1.04 - 10.0 307 108.65 1.49 1.49 (6.59) $0.32
Insulation of furnaces 0.4 0.03 1.96 - 10.0 41 1,521.14 20.86 20.86 12.79 $0.32
Reduce losses from furnace door opening 0.1 0.00 0.08 - 10.0 6 434.61 5.96 5.96 (2.11) $0.32
Controlling oxygen levels and VSDs on combustion air fans 0.1 0.01 0.03 - 15.0 11 86.57 1.27 1.21 (6.95) $0.31
Energy-efficient drives in the rolling mill 0.2 0.00 0.12 - 20.0 20 174.86 7.78 2.86 (4.77) $0.30
Waste heat recovery from cooling water 0.2 0.01 0.33 0.03 15.0 17 774.06 10.29 10.54 2.28 $0.31
General Technologies
Preventative Maintenance 4.0 0.13 0.01 0.01 20.0 393 4.23 0.13 0.10 (7.32) $0.30
Optimizing the steam system 2.4 0.08 0.57 - 20.0 233 72.46 2.15 2.15 (4.41) $0.30
Increase efficiency of boilers 0.2 0.01 0.09 - 20.0 17 152.16 4.52 4.52 (2.04) $0.30
Optimizing the air system 0.2 0.01 0.14 - 20.0 17 243.45 7.24 2.66 (5.31) $0.30
Variable speed drive: flue gas control, pumps, fans 0.1 0.00 0.20 - 5.0 8 966.69 28.74 10.57 16.19 $0.41
Energy monitoring and management system 0.7 0.02 0.10 - 5.0 71 56.65 1.68 1.34 (5.76) $0.41
Integrated Steelmaking
Iron Ore Preparation (Sintering)
Sinter plant heat recovery 1.1 0.04 0.27 - 10.0 109 80.50 2.30 2.34 (2.49) $0.32
Reduction of air leakages 0.0 0.00 0.03 (0.01) 10.0 4 (6.81) (0.30) (0.11) (12.85) $0.32
Increasing bed depth 0.2 0.01 0.00 - 10.0 17 0.12 0.00 0.00 (5.03) $0.32
Improved process control (sinter plant) 0.1 0.00 0.02 (0.02) 10.0 8 (142.33) (4.24) (3.79) (9.63) $0.32
Use of waste fuels in the sinter plant 0.0 0.00 0.00 - 10.0 4 15.73 0.45 0.45 (4.42) $0.32
Improved process control (sinter plant) 0.1 0.01 0.21 (0.01) 10.0 14 410.63 11.79 11.79 6.92 $0.32
Coke Making
Coal moisture control 0.1 0.02 6.78 - 10.0 14 15,095.87 92.94 92.94 78.64 $0.32
Programmed heating - coke plant 0.1 0.02 0.03 - 10.0 12 82.05 0.51 0.51 (13.79) $0.32
Variable speed drive coke oven gas compressors 0.0 0.00 0.04 - 15.0 0 2,747.72 16.92 16.92 2.62 $0.31
Coke dry quenching 1.1 0.17 9.68 (0.81) 18.0 104 2,003.29 12.33 12.33 (1.96) $0.30
Iron Making (Blast Furnace)
Pulverized coal injection to 130 kg/thm 1.0 0.06 0.73 (0.21) 20.0 102 11.19 0.19 0.19 (5.19) $0.30
Pulverized coal injection to 225 kg/thm 0.9 0.06 0.67 (0.08) 20.0 93 124.82 2.08 2.08 (3.30) $0.30
Injection of natural gas to 140 kg/thm 1.2 0.07 0.50 (0.18) 20.0 113 (23.71) (0.39) (0.39) (5.77) $0.30
Injection of oil up to 130 kg/thm 1.1 0.07 0.46 (0.15) 20.0 107 (13.48) (0.22) (0.22) (5.60) $0.30
Top pressure recovery turbines (wet type) 2.4 0.05 7.94 - 15.0 233 1,026.64 45.68 16.80 33.13 $0.31
Recovery of blast furnace gas 0.1 0.01 0.04 - 15.0 12 107.90 1.79 1.79 (3.58) $0.31
Hot blast stove automation 1.4 0.09 0.09 - 5.0 140 25.83 0.43 0.43 (4.95) $0.41
Recuperator hot blast stove 0.5 0.03 0.69 - 10.0 50 439.15 7.30 7.30 1.93 $0.32
Improved blast furnace control systems 2.1 0.13 0.15 - 5.0 209 28.67 0.48 0.48 (4.90) $0.41
Steelmaking
BOF gas + sensible heat recovery 6.2 0.41 12.23 - 10.0 614 634.48 9.57 8.88 1.35 $0.32
Variable speed drive on ventilation fans 0.1 0.00 0.11 - 10.0 6 578.83 25.76 9.47 13.21 $0.32
Integrated Casting
Efficient ladle preheating 0.1 0.01 0.02 - 10.0 8 65.19 0.89 0.89 (5.34) $0.32
Proper sealing on ladle furnace preheating 0.2 0.02 0.04 - 10.0 20 67.38 0.92 0.92 (5.31) $0.32
Thin slab casting 5.6 0.30 14.96 (3.49) 20.0 550 182.81 3.44 2.68 (3.84) $0.30
Use dry rolls in tunnel ovens for TSC 0.7 0.05 0.56 (0.00) 20.0 72 227.53 3.12 3.12 (3.11) $0.30
Integrated Hot Rolling
Hot charging 0.6 0.05 1.48 (0.15) 20.0 62 474.00 6.50 6.50 (1.57) $0.30
Process control in hot strip mill 0.8 0.06 0.16 - 10.0 78 65.92 0.90 0.90 (7.17) $0.32
Recuperative burners 0.7 0.05 0.25 - 10.0 73 108.65 1.49 1.49 (6.59) $0.32
Insulation of furnaces 0.3 0.02 1.48 - 10.0 31 1,521.14 20.86 20.86 12.79 $0.32
Reduce losses from furnace door opening 0.0 0.00 0.05 - 10.0 5 336.91 4.62 4.62 (3.45) $0.32
Controlling oxygen levels and VSDs on combustion air fans 0.1 0.01 0.02 - 15.0 9 86.57 1.27 1.21 (6.95) $0.31
Energy-efficient drives in the rolling mill 0.1 0.00 0.05 - 20.0 8 174.86 7.78 2.86 (4.77) $0.30
Waste heat recovery from cooling water 0.1 0.01 0.22 0.02 15.0 11 774.06 10.29 10.54 2.28 $0.31
Integrated Cold Rolling and Finishing
Heat recovery on the annealing line 1.3 0.09 0.98 - 10.0 131 237.05 3.62 3.42 (3.65) $0.32
Reduced steam use in the pickling line 0.8 0.06 1.08 - 10.0 80 431.82 6.15 6.15 (0.93) $0.32
Automated monitoring and targeting system 1.8 0.04 0.27 - 5.0 177 60.62 2.70 0.99 (9.85) $0.41
General
Preventative Maintenance 3.5 0.17 0.01 0.01 20.0 342 3.69 0.07 0.07 (5.62) $0.30
Optimizing the steam system 2.4 0.12 0.86 - 20.0 239 107.35 2.15 2.15 (3.23) $0.30
Increase efficiency of boilers 0.2 0.01 0.13 - 20.0 17 225.45 4.52 4.52 (0.86) $0.30
Optimizing the air system 0.1 0.00 0.10 - 20.0 9 360.71 7.24 2.66 (5.31) $0.30
Energy monitoring and management system 0.6 0.03 0.08 - 5.0 62 49.34 0.99 0.92 (4.70) $0.41
Variable speed drive: flue gas control, pumps, fans 0.0 0.00 0.15 - 5.0 4 1,432.30 28.74 10.57 16.19 $0.41