International Journal of Greenhouse Gas Control 128 (2023) 103958

Contents lists available at ScienceDirect

International Journal of Greenhouse Gas Control

journal homepage: www.elsevier.com/locate/ijggc

Cost and life cycle analysis for deep CO2 emissions reduction of
steelmaking: Blast furnace-basic oxygen furnace and electric arc
furnace technologies
Guiyan Zang a, Pingping Sun a, *, Amgad Elgowainy a, Pallavi Bobba a, Colin McMillan b,
Ookie Ma c, Kara Podkaminer c, Neha Rustagi d, Marc Melaina e, Mariya Koleva e
a
Systems Assessment Center, Energy Systems Division, Argonne National Laboratory, 9700 South Cass Avenue, Lemont, IL 60439, United States
b
National Renewable Energy Laboratory, 901 D Street SW Suite 930, Washington, DC 20024, United States
c
U.S. Department of Energy, Strategic Analysis, Office of Energy Efficiency and Renewable Energy, 1000 Independence Ave SW, Washington, DC 20585, United States
d
U.S. Department of Energy, Hydrogen and Fuel Cell Technologies Office, 1000 Independence Avenue, SW Washington, DC 20585, United States
e
U.S. Department of Energy, Hydrogen and Fuel Cell Technologies Office, 15013 Denver West Parkway, Golden, Colorado 80401, United States

A R T I C L E I N F O A B S T R A C T

Keywords: Iron and steel manufacturing is the largest contributor to CO2 emissions among heavy industries worldwide. This
Steelmaking is mostly due to the use of coal in blast furnace-basic oxygen furnace (BF-BOF) process for virgin (primary) steel
Decarbonization production. The electricity generation mix used in the electric arc furnace (EAF) process to recycle scrap steel
Techno-economic analysis
also contributes to the CO2 emission associated with secondary steel production. To decarbonize iron and steel
Life cycle analysis, CO2 emissions
Energy switching
sector, we investigated decarbonization options for BF-BOF and EAF processes, including energy efficiency,
carbon capture and storage, and the use of clean energy sources, in various BF-BOF and EAF process configu­
rations. For each decarbonization approach, we evaluated the CO2 reduction potential via life cycle analysis
(LCA) and estimated the associated cost through techno-economic analysis (TEA). A typical U.S. BF-BOF for
virgin steel production has a cradle-to-gate (CTG) CO2 emissions of 1,990 kg/MT steel with a levelized cost of
steel (LCOS) of $439/MT steel, while a typical U.S. EAF process for secondary steel production in the United
States has a CTG CO2 emissions of 270 kg/MT steel with a LCOS of $365/MT steel. Combining renewable energy
sources and carbon capture, BF-BOF CTG CO2 emissions can be reduced to 16 kg/MT steel, and EAF configu­
rations can achieve similar deep reductions to reach 25 kg/MT steel. The corresponding LCOS with these
decarbonization levels is estimated to increase to $542/MT steel and $348/MT steel, respectively. The estimated
CO2 avoidance costs vary from -$90/MT CO2 to $646/MT CO2, depending on the various decarbonization
technologies and energy prices.

1. Introduction production. In a coke oven, coal is heated to produce coke, which has a
higher carbon content and lower impurities, and is an important raw
Iron and steel manufacturing is the largest contributor to CO2 material for pig iron production (Babich and Senk, 2019; Mayer et al.,
emissions and second largest energy consumer among heavy industries 2019). This large coal consumption, as well as high demand for other
worldwide (IEA, 2020a). The iron and steel industry accounts for 83% of energy sources, results in high greenhouse gas (GHG) emissions in iron
the coal demand in the U.S. manufacturing sector (EIA 2018 and 2020), and steel manufacturing (Ryan et al., 2020). In 2019, the direct GHG
where coal is used as a primary fuel as well as a feedstock for coke emissions from U.S. iron and steel manufacturing were 72 million metric

Abbreviations: CTG, Cradle to gate; MMT, Million metric tons; AHSS, Advanced high strength steel; BF, Blast furnace; BOF, Basic oxygen furnace; EAF, Electric arc
furnace; MT, Metric ton; DOE, U.S. Department of Energy; SOA, State of the art; CC, Carbon capture; IEA, International Energy Agency; LCA, Life cycle analysis; TEA,
Technology-economic analysis; GREET, Greenhouse gases, Regulated Emissions and Energy use in Transportation; COG, Coke oven gas; BFG, Blast furnace gas;
MECS, Manufacturing Energy Consumption Survey; MEA, Mono-ethanol-amine; NG, Natural gas; RNG, Renewable natural gas; DRI, Direct reduced iron; HBI, Hot-
briquetted iron; TIC, Total installed cost; O&M, Operations and maintenance; LHV, Lower heating value; NETL, National Energy Technology Laboratory.
* Corresponding author.
E-mail address: [email protected] (P. Sun).

https://doi.org/10.1016/j.ijggc.2023.103958
Received 17 March 2022; Received in revised form 3 August 2023; Accepted 20 August 2023
Available online 31 August 2023
1750-5836/Published by Elsevier Ltd.
G. Zang et al. International Journal of Greenhouse Gas Control 128 (2023) 103958

tons (MMT) CO2 equivalent, or 6% of the total U.S. GHG emissions from renewable energy sources (both electricity and fuels) rather than fossil
manufacturing sectors (GHGRP, 2019). To achieve the goal of net-zero energy sources (Echterhof, 2021). The direct CO2 emissions from EAF
CO2 emissions by 2050 (IEA, 2020b), the iron and steel manufacturing mills are more than 90% lower than those of BF-BOF mills, and CO2
sector needs a deep reduction of CO2 emissions (Arens et al., 2017). That capture cost increases greatly when CO2 emissions are low owing to
reduction can be achieved through increased energy efficiency, carbon economic of scales (Herron et al., 2014). Thus, CC technology is not used
capture and storage (CC), and the use of cleaner energy (Milford et al., to reduce CO2 emissions in EAF mills.
2013). However, these approaches may increase crude steel production The CO2 emissions reduction in BF-BOF and EAF configurations has
cost given the investment cost of energy-efficient technologies, the been previously studied in terms of increased energy efficiency, with
increased energy consumption of CC, and the higher price of clean en­ and without CC, and with and without the use of renewable energy
ergy. Thus, this work discusses the CO2 emissions reduction potential of sources, respectively. However, none of the decarbonization methods
different approaches with cost tradeoffs to determine the optimal can achieve deep decarbonization individually; a combination of several
decarbonization solution. methods is needed to reach the net-zero emissions target.
In the U.S. steel industry, the two major manufacturing technologies The previous steel decarbonization studies were based on various
are the blast furnace–basic oxygen furnace (BF-BOF) and the electric arc plant parameters and analysis boundaries, thus did not employ a uni­
furnace (EAF) (Jamison et al., 2015). In 2019, a total of 87.8 MMT of form framework to evaluate and compare different decarbonization
steel was produced in the U.S.: 30% from BF-BOF and 70% from EAF options. Consequently, information from these studies could not be
(World steel Association, 2020). BF-BOF consumes mostly iron ore, simply compared to investigate decarbonization options for steel
though scrap (recycled steel) can constitute up to 30% of the raw ma­ manufacturing. A systematic, comprehensive, and quantitative analysis
terial; EAF uses primarily scrap without iron ore (Cavaliere, 2019). U.S. for steel decarbonization options with consistent system boundary and
BF-BOF facilities are over 30 years old (Hasanbeigi and Springer, 2019), baseline is thus needed. This study provides a comprehensive life cycle
on average, with an energy consumption of around 23 GJ/metric ton analysis (LCA) and techno-economic analysis (TEA) of six U.S. BF-BOF
(MT) steel, of which 85% is coal (Jamison et al., 2015). Because U.S. EAF and four U.S. EAF decarbonization methods to achieve deep CO2
facilities use scrap as the primary feedstock, their energy consumption is reduction.
much lower: 6.1 GJ/MT steel (including energy consumption for fin­ The six BF-BOF configurations analyzed include two types of system
ishing processes at 3.8 GJ/MT) of which 59% is electricity (Hasanbeigi designs: current-practice U.S. blast furnace technologies (BF-BOF) and
and Springer, 2019). The different technologies and energy consump­ state-of-the-art (SOA) blast furnace technologies (BF-BOF-SOA). For
tion profiles of the BF-BOF and EAF processes require different CO2 each of these technologies, we analyzed three system designs: the two
reduction options. base cases (BF-BOF and BF-BOF-SOA), cases with CC (BF-BOF-CC and
BF-BOF plants, sometimes called integrated mills, consist of multiple BF-BOF-SOA-CC), and cases in which CC is combined with a change in
processes such as coke making, iron ore agglomeration (Cui et al., 2021), all energy sources from fossil to renewable (BF-BOF-all and BF-BOF-
blast furnace, basic oxygen furnace, refining, and casting processes. SOA-all). The four EAF configurations also include a current technol­
Energy efficiency improvements, such as multifunctional energy sys­ ogy case and state-of-the-art technology case, and for each we analyzed
tems (Jin et al., 2009), solid waste utilization (Griffin and Hammond, the base cases (EAF and EAF-SOA) and the decarbonization cases with
2019), and heat-energy recovery (Chen et al., 2018), are potential ways all energy sources changed from fossil to renewable (EAF-all and EAF-
to reduce energy consumption and CO2 emissions. The U.S. Department SOA-all). The technology readiness level of these configurations and
of Energy’s (DOE’s) Bandwidth Study estimated the energy consumption their available year is listed in Table S1 of the supporting information
of a typical BF-BOF process in the U.S. as the baseline (Jamison et al., (SI). The LCA was conducted using GREET® (Greenhouse gases, Regu­
2015). This study concluded that the energy efficiency improvements lated Emissions and Energy use in Technologies) 2020. GREET is a life-
can reduce energy consumption by 10% achieved through cycle model developed by Argonne National Laboratory to evaluate
state-of-the-art (SOA) BF-BOF technology, such as coke dry quenching, energy and emissions impacts of fuels and products (Wang et al., 2020).
enhanced combustion control, waste gas heat recovery, steam recovery, TEA was conducted using discounted cash flow analysis—the same
ladle management, and bottom stirring (Jamison et al., 2015). Carbon methodology used in DOE’s Ironmaking Process Alternatives Screening
capture and storage is another CO2 emissions reduction approach in Study Volume I: Summary Report (Greene, 2000).
steel production (Rigamonti and Brivio, 2022). Biermann et al. studied In this study, we derived the mass and energy conversion of all
an integrated steel mill to estimate the CO2 capture cost. Their results evaluated BF-BOF and EAF configurations information from various
showed that the lowest capture cost, $33/MT CO2, can be achieved by literature sources, such as DOE reports (Jamison et al., 2015), industrial
partial CO2 capture from blast furnace gas, with 36% carbon avoidance. reports (The Athena Sustainable Materials Institute, 2002), and others.
To achieve 76% carbon avoidance, the capture cost increases to $51/MT The mass and energy conversion data were incorporated in the GREET
CO2 (Biermann et al., 2019). The International Energy Agency (IEA) model to evaluate the CTG CO2 emissions, covering all stages from iron
compared three CO2 capture cases that showed the CO2 avoidance cost ore recovery to steel production. The mass and energy flow data were
to be $56–$81 per metric ton when the CO2 avoidance ratio is 47%–60% also used as input for equipment scaling, and capital and operating costs
(IEA, 2013). Another widely studied approach for industrial steel CO2 evaluation, which are used in the discounted case flow analysis to
emissions reduction is switching from fossil energy to renewable sources calculate the LCOS.
of energy (Kumar et al., 2017). Mandova et al. showed that by using
biomass, a BF-BOF mill can achieve a maximum CO2 emissions reduc­ 2. Methodology
tion rate of 42% (Mandova et al., 2018).
The second type of steel mill, EAF mills, are known as “mini mills” as Increased energy efficiency, carbon capture and storage, and use of
they have only one primary conversion process. The DOE bandwidth renewable energy are three methods that can be used to reduce CO2
study estimated the energy consumption of EAF technology at 2.3 GJ/ emissions from iron and steel manufacturing processes. This method­
MT for crude steel production—90% less than the current typical BF- ology section reviews the BF-BOF and EAF steel production technologies
BOF process (Jamison et al., 2015). Birat et al. reviewed the global en­ using these three methods and includes a discussion of LCA boundaries
ergy consumption of EAF and showed that the best EAF route in practice and assumptions as well as detailed information for TEA analysis. The
had an energy intensity of 2.2 GJ/MT crude steel, of which 1.6 GJ/MT is basic assumptions and conditions for the analysis are listed below.
related to electricity consumption and 0.6 GJ/MT is related to fossil fuel
consumption for preheating (Birat, 2010). The typical CO2 emissions a) The energy and mass conversion data represents U.S. steel industry
reduction approach in EAF mills is to increase energy efficiency and use average value (i.e., does not represent a specific plant).

2
G. Zang et al. International Journal of Greenhouse Gas Control 128 (2023) 103958

b) The steel plants are assumed to operate under steady state (i.e., the agglomeration), lump coke (from the coke oven), and limestone are
transient state energy and mass conversion during startup and shut added to the top of the blast furnace. In the lower section of the blast
down is not considered); furnace, coke is gasified to produce CO through reactions R1 and R2
c) The life cycle analysis focuses on the feedstock and fuel consump­ using hot blast air (24 w% O2) as the gasification agent (IEA, 2013). The
tions, and does not account for the embodied emission during plant CO reacts with iron oxides as the reducing agent to form hot metal (pig
construction, which is likely negligible when allocated to per MT of iron) through reactions R3 to R5.
steel, given the large production throughput over the long plant life /
C + O2 →CO2 ΔH 0 = − 393 kJ mol (R1)
time;
d) The techno-economic analysis uses the average U.S. historical fuel /
C + CO2 →2CO ΔH 0 = +172 kJ mol (R2)
cost, thus the impact of regional and time-dependent price variations
are not covered in this study. /
3Fe2 O3 + CO→2Fe3 O4 + CO2 ΔH 0 = − 47 kJ mol (R3)
/
Fe3 O4 + CO→3FeO + CO2 ΔH 0 = +19 kJ mol (R4)
2.1. Steel production technology
/
FeO + CO→Fe + CO2 ΔH 0 = − 11 kJ mol (R5)
2.1.1. BF-BOF technology and CO2 reduction methods
Fig. 1 shows the mass flow rate and CTG CO2 emissions of a current- The hot metal produced in the blast furnace is routed to the BOF to be
practice baseline U.S. BF-BOF steel mill. The energy consumption data is purified and converted to liquid steel. Up to 30% scrap can be also fed to
from a previous work by Jamison et al. (2015), for a typical BF-BOF BOF (Suopajärvi et al., 2018). Finally, the liquid steel is refined and
process in the United States. The material flow rates of iron ore, scrap, cooled in the refining and casting stages to produce crude steel (Kapoor
limestone, and lime are based on previous studies for a baseline steel et al., 2021). In the BF-BOF configuration, coke oven gas (COG) and
production plant in the United States (The Athena Sustainable Materials blast furnace gas (BFG) are formed in the coke oven and blast furnace,
Institute, 2002). The CO2 emissions are evaluated based on energy respectively, and are used as fuel in the power generation unit to supply
consumption, material consumption, and the carbon balance using heat and power to the entire system (Peacey and Davenport, 2016).
GREET model. The BF-BOF process includes two materials preparation Fig. 1 shows the material flow rates of iron ore, scrap, limestone, and
processes: ore agglomeration (pelletizing and sintering) and coke mak­ lime from a typical steel production plant in the United States, as eval­
ing (He and Wang, 2017). In the ore agglomeration process, iron ore is uated by the Athena Sustainable Materials Institute (The Athena Sus­
crushed and ground to remove impurities and pelletized to form uni­ tainable Materials Institute, 2002). Table 1 shows the energy
formly sized round iron ore pellets. In the sintering process, the ore consumption of the six BF-BOF configurations covered in this study. The
pellets are mixed with iron fines, coke breeze (fine coke), and limestone BF-BOF case represents the typical current technology (current-practice
to form hardened lumps of sinter as feedstock for the blast furnace. The baseline) in U.S. iron and steel manufacturing, which uses 22.7 GJ en­
coke breeze supplies energy for the sintering processes. In the coke oven, ergy (after by-product displacement) to produce 1 metric ton of crude
coking coal is heated to high temperatures in an airless environment to steel. The energy consumption of the BF-BOF (also shown in Fig. 1) is
drive off volatile chemicals, increase carbon content, and produce lump calculated from a U.S. onsite energy consumption database, based on the
coke as the energy supply for the blast furnace (Xu et al., 2020). The tar U.S. Energy Information Agency’s (EIA’s) Manufacturing Energy Con­
and benzol produced from the coke oven are sold to the market as sumption Survey (MECS) data, which includes offsite electricity as well
by-products (IEA, 2013). Although not all coke is produced on-site in U. as steam generation and transmission losses (Jamison et al., 2015).
S. BF-BOF plants, the process is included in this analysis to make the Table 2 shows direct CO2 emissions from the baseline BF-BOF with
analysis boundaries consistent with the energy data source (American additional NG consumption for carbon capture. In the baseline case, the
Coke and Coal Chemicals Institute, 2020). flue gas from coke oven, BF-hot stoves and steam boiler has a CO2
After the materials preparation, ironmaking reactions take place in concentration above 10 Vol%. These flows are used as high
the blast furnace (Suopajärvi et al., 2018). Pellets and sinter (from ore

Fig. 1. Mass flowrate (red arrows) and CTG CO2 emissions (clouds) of the current practice baseline of the BF-BOF process in the U.S.

3
G. Zang et al. International Journal of Greenhouse Gas Control 128 (2023) 103958

Table 1
Energy consumption of six BF-BOF configurations for crude steel production.
Groups BF-BOF BF-BOF-SOA
Energy consumption (GJ/MT steel) BF-BOF (current practice baseline) BF-BOF-CC BF-BOF-all BF-BOF-SOA BF-BOF- SOA-CC BF-BOF- SOA-all

Residual oil 0.152 0.152 0.000 1.316 1.316 0.000

Gasoline 0.003 0.003 0.003 0.003 0.003 0.003
Diesel 0.038 0.038 0.038 0.038 0.038 0.038
NG 2.530 7.150 0.000 0.378 4.727 0.000
RNG 0.000 0.000 7.267 0.000 0.000 6.008
Coking coal 18.15 18.15 16.33 17.01 17.01 15.31
Industrial coal 2.039 2.039 0.000 2.574 2.574 0.000
Biochar 0.000 0.000 3.194 0.000 0.000 3.656
Grid electricity 0.875 1.416 0.000 0.446 0.955 0.000
Clean electricity 0.000 0.000 1.405 0.000 0.000 0.927
By-product − 1.130 − 1.130 − 0.852 − 1.350 − 1.350 − 1.089
Total 22.66 27.82 27.39 20.42 25.27 24.85
Relative to baseline (%) 100% 123% 121% 90% 112% 110%

Table 2
Direct carbon emissions and CO2 captured from BF-BOF baseline case.
Coke oven Sintering BF hot stoves BF other BOF Steam boiler NG for CO₂ capture Total

Direct CO₂ emissions (kg/MT steel) 157 166 404 32 32 968 236 1995
CO₂ concentration (v%) 15% 5% 27% 6% 6% 25% 3% –
CO₂ captured (kg/MT steel) 141 0 364 0 0 871 0 1376
CO₂ capture ratio (%) 7% 0% 18% 0% 0% 44% 0% 69%

concentration CO2 sources for carbon capture (Herron et al., 2014). The 2015). As shown in Table 1, the total energy consumption of the
carbon capture method is mono-ethanol-amine (MEA) with 90% carbon BF-BOF-SOA case is 20.4 GJ/MT crude steel (by accounting for credit of
capture ratio from these sources. Thus, the total CO2 captured from the by-product displacement). BF-BOF-SOA case shows 10% reduction of
BF-BOF baseline case is 1376 kg/MT steel, which, when compared with energy consumption from current BF-BOF (22.7 GJ/MT). More detailed
the total carbon emissions (1995 kg/MT) in the base case, implies a energy consumption information for the six BF-BOF configurations is
carbon capture ratio of 69%. shown in Table S2 of the Supporting Information (SI).
According to case B of IEA’s Iron and Steel CCS Study (Techno-Eco­
nomics Integrated Steel Mill) (IEA, 2013), the heat required for the MEA 2.1.2. EAF technology and CO2 reduction methods
carbon capture process is supplied by natural gas (NG) combustion. The “Mini-mill” steelmaking with EAF as the major reaction unit ac­
NG consumption is 3.0 GJ for each metric ton of CO2 captured (IEA, counts for approximately 26% of world crude steel production and 70%
2013). The total electricity consumption for carbon capture and of U.S. steel production (Hasanbeigi and Springer, 2019). Unlike BF-BOF
compression (to 153 bar) is 0.4 GJ for each metric ton of CO2 captured. that uses iron ore as the primary feedstock, EAF produces steel from
The electricity consumption for carbon capture and compression is scrap (recycled steel), direct reduced iron (DRI), pig iron, and additives,
estimated at 43 MJ/MT CO2 and 354 MJ/MT CO2, respectively (IEA, without iron ore input. In the United States, scrap is used as feedstock in
2013; Zang et al., 2021). almost all U.S. EAF plants (USGS, 2020a). Only four metallic iron plants
Coal, NG, residual oil, and electricity are the major energy inputs for in the U.S. produce a limited amount of reduced iron (DRI) (one DRI
the current BF-BOF plant. Switching energy sources from fossil energy to plant in Louisiana, and three hot-briquetted iron (HBI) plants in Indiana,
renewable energy or low carbon energy, such as bioenergy, can reduce Ohio, and Texas).
steel production carbon intensity and reduce (fossil) CO2 emissions (Luh For a baseline case of U.S. EAF steel production, the energy con­
et al., 2020). For example, biochar has the potential to replace all the sumption, materials flowrate, and CO2 emissions are derived from the
industrial coal for combustion use in the BF-BOF technology and can study of Jamison et al., The Athena Sustainable Materials Institute, and
replace 10% of the coke used in the blast furnace (Mandova et al., 2018; the GREET model. In the EAF process, scrap is melted with a certain
Mousa et al., 2016). Given that residual oil and NG are used primarily as amount of added carbon (e.g., graphite) in order to lower the metallic
fuel for combustion without participating in the major reactions, both of iron melting point, and therefore reduces electricity consumption
them can potentially be replaced by renewable natural gas (RNG) (Cavaliere, 2019). The process of iron-carbon melting is very complex
(Cavaliere, 2019). In addition, grid electricity can be switched to clean and includes heat transfer from the melted liquid metal to the scrap and
(low or zero carbon emissions) electricity from lower-carbon energy from the surface to internal layers of scrap pieces (Gajic et al., 2016).
sources such as nuclear, biomass, and wind/solar energy to reduce CO2
emissions (Arens et al., 2021). The BF-BOF-all case in Table 1 shows the
resultant energy consumption with all the energy switching options. The Table 3
total energy consumption of BF-BOF-all case is 0.4 GJ/MT crude steel Energy consumption of four EAF configurations for steelmaking.
lower than that of the BF-BOF-CC case, because 10% of the coke used in Groups EAF EAF-SOA
the blast furnace is replaced by biochar. Energy consumption EAF (current- EAF- EAF- EAF-
Currently, the U.S. BF-BOF facilities are, on average, over 30 years (GJ/MT steel) practice baseline) all SOA SOA-all

old. Thus, a group of state-of-the-art BF-BOF technologies (BF-BOF-SOA) NG 0.546 0 0.189 0

are listed in Table 1 to reflect recent BF-BOF technology improvements RNG 0 0.546 0 0.189
Grid electricity 1.779 0 1.744 0
(Jamison et al., 2015). The BF-BOF-SOA case uses the most efficient
Clean electricity 0 1.779 0 1.744
technologies or equipment for ore agglomeration, coke making, blast Total 2.325 2.325 1.933 1.933
furnace, and basic oxygen furnace. All processes of BF-BOF-SOA case Relative to baseline 100% 100% 83% 83%
have lower energy consumptions than the BF-BOF case (Jamison et al., (%)

4
G. Zang et al. International Journal of Greenhouse Gas Control 128 (2023) 103958

The energy demands of EAF (shown in Table 3 as current-practice where the liquid fuel includes low-sulfur diesel, gasoline, and residual
baseline) are 1.8 GJ/MT of electricity and 0.5 GJ/MT of NG in the oil, with extraction and transportation CO2 emissions of 12.3, 16.1, and
current typical U.S. EAF case (Jamison et al., 2015). The EAF-all case in 9.3 kg CO2/GJ liquid fuel, respectively. The NG fuel carries 5.9 kg CO2/
Table 3 shows the energy consumption when NG is replaced by RNG and GJ emissions from upstream extraction and transportation processes. It
grid electricity by clean electricity. can be replaced by renewable natural gas, of which the CO2 emission
The two EAF-SOA options in Table 3 reflect recent technology im­ varies with different sources and production technologies. For the pre­
provements in EAF (Jamison et al., 2015). The EAF-SOA case uses the sent study, we show results for RNG from combined waste, which has
most efficient EAF technology available today. As shown in Table 3, the − 57.2 kg CO2/GJ emissions from the production process, based on
total energy consumption of the EAF-SOA case is 1.9 GJ/MT crude steel, GREET 2020. Industrial coal carries 1.5 kg CO2/GJ emission for up­
which is 17% lower than the energy consumption of the current EAF stream extraction and transportation processes; it can be replaced by
technology. The EAF-SOA-all case shows the RNG and clean electricity biochar to reduce CO2 emission. One GJ of biochar can be produced
application potential in the EAF-SOA case. from forest residue with a yield of 25%, using 0.14 GJ heat and 0.04 GJ
power (Crombie and Mašek, 2015, Cong et al., 2018). The upstream CO2
emissions of the biochar are − 86.3 kg CO2/GJ after accounting for the
2.2. LCA analysis and fuel switching CO2 emissions biogenic carbon (Wang et al., 2020).
The direct CO2 emissions from material pretreatment, ironmaking,
Using the same process configurations described above, we evalu­ and steelmaking include fuel combustion emissions and process emis­
ated CO2 emissions from steel production using the GREET model sions (detailed information shown in Table 5). The fuel combustion
(2020). The CO2 emissions analysis from steel production can be con­ emissions factors are from GREET 2020, and the process emissions of
ducted in three scopes: scope I—direct emissions; scope II—CO2 emis­ sintering (27.6 kg CO2/GJ) and blast furnace (21.9 kg CO2/GJ) are from
sions for the electricity supply; and scope III— all the upstream the simulation by the previous study (The Athena Sustainable Materials
emissions of process inputs, such as fuel/material extraction, trans­ Institute, 2002). The LHV, density, and carbon ratio of fuels are shown in
portation, and emissions displacement of by-products (Birat, 2010). Table 6.
Fig. 2 shows the cradle-to-gate (CTG) LCA analysis boundaries used in
this study, which include all the above scopes. We have considered four
major analysis steps of materials/fuel extraction and transportation and 2.3. TEA analysis and fuel switching cost
electricity generation and transmission, materials pretreatment, iron­
making, and steelmaking steps. The LCA analysis is based on the func­ TEA analysis evaluates the LCOS using a discounted cash flow
tional unit of kg CO2 per MT of crude steel produced (kg/MT steel) (Cruz analysis and process-level information from Ironmaking Process Alter­
et al., 2021). natives Screening Study Volume I: Summary Report (Greene, 2000). The
Fig. 2 illustrates the main configurations of five cases for steel pro­ discounted cash flow analysis is broadly used for levelized cost evalua­
duction from BF-BOF and EAF processes. The CO2 emissions associated tion of steel production. The LCOS is the steel price that makes the net
with energy switching are accounted for in the materials/fuel extraction, present value of the steel plant zero when the plant life is assumed to be
transportation, electricity generation and transmission steps (detailed 25 years with a discount rate of 10%. The construction period of a new
information is shown in Table 4). All the data is from GREET 2020, steel plant is two years, with 75% invested during the first 12 months

Fig. 2. LCA analysis boundary for BF-BOF and EAF cases.

5
G. Zang et al. International Journal of Greenhouse Gas Control 128 (2023) 103958

Table 4
CO2 emissions for extraction and transportation of various materials and fuels and for generation of electricity from GREET 2020.
Liquid fuel Gas fuel Solid fuel

Extraction Diesel Gasoline Residual oil NG RNG Coal Char

transportation
GREET 2020 pathway Petroleum LS Petroleum gasoline Petroleum residual NG as stationary RNG combined waste Coal to power Pyrolysis
diesel blendstock oil fuels to NG plant IDL
CO2 emissions (kg/ 12.3 16.1 9.3 5.9 − 57.2 1.5 − 86.3
GJ)

Electricity Source Materials

Extraction U.S. grid Bio-electric Nuclear- Hydro-electric Wind Limestone Lime

transportation electric
GREET 2020 pathway Electric U.S. Electric biomass fired Electric Electric Electric wind Ag-inputs Ag-inputs lime in
mix power nuclear hydroelectric power CaCO3 U.S.
CO2 emissions (kg/GJ) 118 6.5 2.0 0.0 0.0 8.4 1085

Table 5
CO2 emission factors of various fuels.
Liquid fuel Gaseous fuel Solid fuel

Fuel combustion Diesel Gasoline Residual oil NG BFG COG Coal Char
CO2 emissions (kg/GJ) 73.9 68.8 80.6 56.3 278.6 44.1 94.8 85.3
Process emissions Sintering Blast furnace
CO2 emissions (kg/MT steel) 27.6 21.9

Table 6 Table 7
Fuel properties used for life cycle and technical-economic analysis. Equipment installed cost for BF-BOF and EAF technologies.
Material LHV (GJ/MT- Density (kg/ Carbon content (wt Reference Units Reference Scaling
parameter fuel) L) %) scale (MT/ installed cost exponent
year) (US$ million)
Coal fuel 22.65 – 58.57%
Coking coal 28.61 – 74.70% BF-BOF
Residual oil 39.47 0.99 86.80% Coke oven 2277,702 Coal 400 0.80
Gasoline 43.45 0.74 86.30% Sintering 4445,559 Sinter 220 0.80
Diesel 42.61 0.85 87.10% Blast furnace, hot 3894,263 Hot 622 0.80
Coke 31.34 0.00 86.67% stoves, and hot metal metal
NG 47.14 0.00 72.40% desulphurization
Basic oxygen furnace 4323,327 Crude 459 0.80
and steel refining steel
and 25% spent in the second 12 months. The start-up time is 12 months, Continuous slab caster 4000,000 Crude 195 0.80
steel
and revenues are assumed to be 75% those of a normal operating year
Lime production 591,361 Crude 16 0.80
(Greene, 2000). steel
Capital expenditures are the sum of the total installed cost (TIC) and Air separation unit 4323,327 Crude 130 0.80
contingency (5% of the TIC) (IEA, 2013). TIC is evaluated using Eq. (1) steel
(Manzolini et al., 2020), where Cref,i is the reference equipment cost. The Power plant 4323,327 Crude 362 0.80
steel
BF-BOF equipment cost listed in Table 7 is from (IEA, 2013) and the EAF Steam generation plant 4323,327 Crude 139 0.80
equipment cost listed in Table 7 is from (Greene, 2000). Sref,i is the steel
reference equipment size, Si is the real size of the equipment used in this Raw material handling 4323,327 Crude 247 0.80
study, and f is the scaling exponent. Detailed information for the TIC is steel
Pre-operating expenses 4323,327 Crude 21 0.80
shown in Table 7.
steel
∑
n
( / )f Land preparation, site 4323,327 Crude 144 0.80
TIC = Cref ,i × Si Sref ,i (1) development, and steel
i=0 waste disposal
Buildings and site 4323,327 Crude 196 0.80
The annual operations and maintenance (O&M) cost includes the infrastructure steel
fixed O&M cost, variable O&M cost, and “other O&M” cost (e.g. slag Project engineering 4323,327 Crude 201 0.80
processing, on-site haulage, disposal, and landfill). The detailed calcu­ steel
CO₂ capture and 4323,327 Crude 590 0.80
lation processes for fixed and other O&M cost are shown in Table 8, and
compression steel
the material price used to calculate the variable O&M cost is shown in EAF
Table 9. In Table 8, the miscellaneous cost includes services related to Electric arc furnace and 4920,000 Crude 591 0.80
logistics, engineering, analysis, infrastructure, and information. The refining steel
LCOS of each system was evaluated to demonstrate the impacts of the Land preparation, site 4920,000 Crude 119 0.80
development, and steel
selected technology and carbon capture option on the steel price. For waste disposal
TEA analysis, the base cases use market prices in 2019 of the incumbent
energy sources: electricity price of $0.07/kWh (Zang et al., 2021), NG
price of $3.7/GJ (EIA, 2019a), coking coal price of $161/MT (EIA, of $0.03-$0.15/kWh (Wiser and Bolinger, 2019), an RNG price of
2019b), and industrial coal price of $68/MT (EIA, 2019b). To show the $6.6-$19.0/GJ (American Gas Foundation, 2019), and biochar price of
impacts of renewable energy prices on the LCOS, a clean electricity price $403-$747/MT (Bushell, 2018) have been used.

6
G. Zang et al. International Journal of Greenhouse Gas Control 128 (2023) 103958

Table 8 3. Results and discussion

Fixed and other O&M costs for BF-BOF and EAF technologies.
Maintenance (% Personnel Miscellaneous 3.1. Cradle-to-gate energy consumption and CO2 emission reduction
installed cost) ($/MT steel) ($/MT steel) potential
BF-BOF
Coke oven 5.0% 8.52 5.10 With a detailed analysis of BF-BOF subprocess energy consumption,
Sinter production 5.0% 8.81 1.67 the overall process energy flow of a typical U.S. BF-BOF process can be
Blast furnace/hot stoves 4.0% 8.26 3.20 summarized as shown in Fig. 3. The width of each flow line represents
and hot metal
desulphurization plant
the heat content, i.e., the lower heating value (LHV) of each flow based
Basic oxygen steelmaking 5.0% 10.68 3.90 on the mass flowrate. Fig. 3 shows the major energy consumption of all
plant and refining equipment of a typical U.S. BF-BOF plant. For the typical U.S. BF-BOF
Continuous slab caster 8.0% 9.31 1.89 plant, coal is the primary energy input: 18.2 GJ of coking coal and 2.0
Lime production 8.0% 0.74 0.60
GJ of industrial coal for the production of 1 metric ton of crude steel. The
Air separation unit 2.5% 0.91
Power plant 2.5% 1.09 other energy inputs of the BF-BOF process are 0.9 GJ/MT of electricity,
Steam generation plant 2.5% 0.39 0.2 GJ/MT of liquid fuel (a mixture of residual oil, gasoline, and diesel),
CO₂ capture and 2.5% 0.57 and 2.5 GJ/MT of NG. It is worth noting that 0.8 GJ/MT of tar and 0.3
compression GJ/MT of benzol are produced from the BF-BOF system as by-products,
Other personnel cost ($/MT steel)
Central engineering 8.02
and both have CO2 emission displacement credit. After the by-product
Management and admin 12.94 displacement, the total energy consumption of the BF-BOF process is
staff 22.7 GJ/MT, as shown in Table 1.
EAF ($/MT steel) The overall process carbon flow of the current-practice baseline U.S.
EAF personnel cost 6.32
BF-BOF process can be summarized as shown in Fig. 4(a). The width of
EAF other O&M cost 34.97
Refining O&M cost 6.35 the flows shown in Fig. 4(a) represents the carbon content of each flow
based on the mass flowrate shown in Fig. 1. The numbers shown in Fig. 4
indicate kg carbon per metric ton of crude steel produced. For the
current-practice baseline U.S. BF-BOF plant, coal is the primary carbon
Table 9
Materials prices used to calculate the variable O&M cost of BF-BOF and EAF
source: 474 kg of carbon from coking coal and 53 kg of carbon from
technologies. industrial coal for the production of 1 metric ton of crude steel. The
other carbon input to the BF-BOF is 39 kg C/MT steel from natural gas, 4
Material Price Unit Reference
kg C/MT steel from liquid fuel (a mixture of residual oil, gasoline, and
Coking coal 160.77 $/MT (EIA, 2019b) diesel), and 22 kg C/MT steel from other material input, such as lime­
Industrial coal 67.65 $/MT (EIA, 2019b)
stone. It is worth noting that 18 kg/MT of carbon in tar and 7 kg/MT of
Residual oil 0.97 $/gal (EIA, 2020)
Electricity 0.07 $/kWh (Zang et al., 2021) carbon in benzol have CO2 emission displacement credit, and tar and
Natural gas 3.71 $/GJ (EIA, 2019a) benzol are produced from the BF-BOF system as by-products. After the
Gasoline 2.67 $/gal (EIA, 2020) by-product displacement, the total carbon input to the BF-BOF process is
Diesel 3.04 $/gal (EIA, 2020) 567 kg C/MT steel.
Iron ores 66.14 $/MT (USGS, 2020a)
Purchased scrap 249.22 $/MT (USGS, 2020b)
For the entire BF-BOF system, heat and steam are supplied by com­
Dolomite 27.67 $/MT (IEA, 2013) busting coke oven gas (COG) and blast furnace gas (BFG). The carbon
Burnt dolomite 109.48 $/MT (IEA, 2013) content of the COG is 12 kg C/GJ COG based on carbon balance. It is
Crude tar 0.97 $/gal (EIA, 2020) assumed that the carbon input to the coke oven is coking coal, while the
Benzol 0.97 $/gal (EIA, 2020)
carbon outputs of the coke oven are coke, tar, benzol, and COG. The
Coke 107.09 $/MT (EIA, 2019b)
Graphite used in EAF 86.34 $/MT (Greene, 2000) carbon content of the BFG is 76 kg C/GJ BFG based on carbon balance.
EAF electrodes 1530.77 $/MT (Greene, 2000) The carbon input sources to the blast furnace are natural gas, coal, coke,
Lime charged 114.47 $/MT (Greene, 2000) tar, and limestone, and the carbon output of the blast furnace is BFG.
O₂ gas to EAF 0.06 $/Nm3 (Greene, 2000) Fig. 4(a) also shows the direct carbon emissions from each process. The
RNG-min 6.60 $/GJ (American Gas Foundation, 2019)
RNG-max 19.00 $/GJ (American Gas Foundation, 2019)
power generation process has the largest carbon emissions, 264 kg C/
Clean electricity-min 0.03 $/kWh (Wiser and Bolinger, 2019) MT, which discharges 967 kg of direct CO2 emissions for 1 metric ton of
Clean electricity-max 0.15 $/kWh (Wiser and Bolinger, 2019) crude steel produce.
Biochar-min 403.00 $/MT (Bushell, 2018) Fig. 4(b) summarizes the mass and carbon flow of the current-
Biochar-max 747.00 $/MT (Bushell, 2018)
practice baseline U.S. EAF plant. Natural gas and graphite/electrode
(shown as “other sources”) are the primary carbon source for steel
Using these prices of fossil energy and renewable energy, the LCOS of production from EAF. The carbon content of the natural gas is 8.4 kg C/
current steel and future low-carbon steel can be estimated. The CO2 MT steel, while the carbon content in the graphite/electrode is 7.0 kg C/
emission amount is quantified from the CTG LCA analysis by combining MT steel. The EAF process discharges 56.5 kg of direct CO2 emissions
CO2 emissions from different processes. Then the CO2 avoidance cost calculated from the carbon balance.
(CCO2 ,A ) is calculated by using the change in LCOS (ΔCsteel ) divided by Figs. 5(a) and 5(b) show the CTG CO2 emissions from different BF-
the change in CO2 emissions (ΔECO2 ), as shown in Eq. (2). By comparing BOF technologies. For the current-practice baseline U.S. BF-BOF plant,
the CO2 avoidance cost from increased of energy efficiency, CC, and the CTG CO2 emissions are 1990 kg/MT steel. The power generation
energy switching, this study can quantify the impact of different CO2 process has the largest CO2 emissions: 979 kg/MT steel (49% of the
emissions reductions. total) resulting from combustion of COG, BFG, and fuel for power gen­
/ eration and steam production, and the upstream fuel production emis­
CCO2 ,A ($ / MTCO2 ) = − ΔCsteel ($ / MTsteel) ΔECO2 (MT CO2 / MTsteel) sions. The 49% also accounts for the burdens for the refining and casting
(2) processes. The 49% power generation CO2 emissions share is similar to
that found in the study by Birat, which showed that 47% of total CO2
emissions are from power generation (Birat, 2010). The blast furnace
has the second-largest CO2 emissions: 512 kg/MT steel (26% of the

7
G. Zang et al. International Journal of Greenhouse Gas Control 128 (2023) 103958

Fig. 3. Energy Sankey diagram of a typical U.S. steel mill using BF-BOF technology. All values are in the unit of GJ per metric ton of crude steel production. The
width of flow line indicates the quantity of energy based on lower heating value, the red flow represents blast furnace gas, the yellow flow shows the coke oven gas
and the blue, green, brown, and gray flows represent energy flows of electricity, liquid fuel, natural gas, and other energy, respectively.

Fig. 4. The CTG carbon Sankey diagram of (a) BF-BOF current-practice baseline (b) EAF current-practice baseline in the U.S. All the values indicate kg carbon per
metric ton of crude steel production. The width of the flow line indicates the quantity of carbon. The red flow represents blast furnace gas, the yellow flow shows the
coke oven gas, and the blue, green, brown, and gray flows represent carbon flows of other materials, liquid fuel, natural gas, and solid fuel, respectively.

8
G. Zang et al. International Journal of Greenhouse Gas Control 128 (2023) 103958

Fig. 5. LCA results of BF-BOF and EAF. (a) CTG CO2 emissions of current BF-BOF group, (b) CTG CO2 emissions of the BF-BOF-SOA group, (c) CTG CO2 emissions of
current EAF group, and (d) CTG CO2 emissions of the EAF-SOA group. The “power generation” is the internal power generation in the BF-BOF processes.

total), because 1.2 GJ BFG/MT steel and 0.2 GJ COG /MT steel are respectively. The BF-BOF-all-no CC case combines all the energy
combusted to supply heat for the blast furnace. switching options without using carbon capture and storage. The CO2
After identifying the material and energy inputs and quantifying CO2 emissions from the BF-BOF-all-no CC case is 1396 kg/MT steel—30%
emissions, the potential CO2 reduction can be estimated by switching lower than the current-practice baseline of BF-BOF.
the fossil carbon source to a renewable carbon source. Fig. 5(a) shows The BF-BOF-CC case uses NG and grid electricity to supply the energy
the CTG CO2 emissions when the energy source is changed from grid for carbon capture, while the BF-BOF-CC-R case uses RNG and clean
electricity to clean electricity (wind power) (indicated by grid → clean electricity as the energy supply for carbon capture. For both cases, 1376
electricity) and from oil and NG to RNG (oil, NG → RNG). It also shows kg of CO2 can be captured from BF-BOF plants, with energy consumption
the emissions when using biochar to replace 10% of the coke con­ of 3.0 GJ NG/RNG per MT CO2 and 0.4 GJ grid/clean electricity per MT
sumption (maximum amount allowed without modifying BF process) CO2. When the energy inputs for carbon capture are NG and grid elec­
and 100% of the industrial coal consumption in the blast furnace (coke tricity, the CTG CO2 emissions for BF-BOF-CC are reduced by 1025 kg/
→ biochar). The change in electricity source has the potential to reduce MT steel (51% of the total). In contrast, when the energy inputs are from
CO2 emissions by 104 kg/MT steel, and the use of RNG and biochar has renewable sources of RNG and clean electricity, CTG CO2 emissions for
the potential to reduce CO2 emissions by 174 and 318 kg/MT steel, BF-BOF-CC are reduced by 1326 kg/MT steel (67% of total). The case of

9
G. Zang et al. International Journal of Greenhouse Gas Control 128 (2023) 103958

BF-BOF-all, which combines carbon capture and all energy switching Table 10
options, has the lowest CO2 emissions in the BF-BOF group; it has an O&M cost components of the BF-BOF case.
emission of 16 kg/MT steel, a 99% reduction from the current-practice Item cost Cost breakdown Percentage of Annual OPEX
base case shown in Fig. 5(a). (US$ million/y) O&M (%) (US$ million/y)
The CTG CO2 emissions for the BF-BOF-SOA group (with the most Fixed O&M cost 337.8
efficient blast furnace and basic oxygen furnace) are shown in Fig. 5(b), Maintenance 110.0 7.8%
with the energy consumption in Table 1. The CTG CO2 emissions of the Direct labor 159.3 11.3%
BF-BOF-SOA case are 1842 kg/MT steel, which is 7% lower than that of Indirect labor 68.4 4.8%
Variable O&M cost 1050.8
baseline BF-BOF case as a result of its lower energy consumption. The Fuel and reductant 517.4 36.6%
BF-BOF-SOA-CC case captures 1295 kg CO2 for each MT crude steel Iron ore 247.4 17.5%
produced. The amount of CO2 captured in the BF-BOF-SOA case is 6% Purchased scrap 215.4 15.3%
lower than that of the BF-BOF case because the lower fuel consumption Fluxes 34.3 2.4%
Consumables & 36.3 2.6%
of BF-BOF-SOA results in the lower amount CO2 emission. With carbon
other utilities
capture, the CTG CO2 emissions of the BF-BOF-SOA-CC are 877 kg/MT Other O&M cost 23.9
steel, which is 52% lower than BF-BOF-SOA case and 9% lower than the Miscellaneous 43.1 3.1%
BF-BOF-CC case. With all energy switching and carbon capture options, works expense
the CTG CO2 emissions of BF-BOF-SOA-all are 84 kg/MT steel (96% Other OPEX 11.4 0.8%
Slag processing 6.2 0.4%
lower than the current-practice baseline of BF-BOF), as shown in Fig. 5 On-site haulage 0.2 0.0%
(b). Disposal and landfill 3.9 0.3%
These results indicate that increasing energy efficiency has a limited By-product credit − 40.9 − 2.9%
CO2 emissions reduction potential of 7.4%, while CCS has a more sig­ Total O&M 1412 100.0% 1412
nificant CO2 emissions reduction potential of 21.7%. The decarbon­
ization potential of fuel switching ranges from 5.2% to 15.1%.
ore and scrap cost, and labor cost representing 33%, 33%, and 24% of
About 70% of the steel plants in the United States use EAF technol­
the LCOS, respectively. The LCOS of $439/MT steel is close to the
ogy to process scrap and a small amount of DRI feedstock. The CTG CO2
amounts found by previous research that showed a cost of $413/MT
emissions for EAF and EAF-SOA are shown in Fig. 5(c)-(d), with energy
steel in 2020 using the basic oxygen furnace route (Steelonthenet.com,
consumption shown in Table 1. Unlike the BF-BOF process, which
2020a). The BF-BOF-CC case in Fig. 6(a) shows the LCOS when carbon
consists of six major subunits or reaction processes, the EAF pathway has
capture technology is integrated into current-practice baseline BF-BOF
only one main reaction unit: the electric arc furnace. Thus, the CTG CO2
plant. The major capital expenditure for carbon capture is $560
emissions from the EAF cases come only from the consumption of NG,
million per plant for the CO2 capture unit and compression equipment
electricity generation, and process emissions due to the graphite use.
with a CO2 capture ratio of 65%. In Fig. 6(a), the LCOS of the BF-BOF-CC
Graphite is used as the electrode as well as a carbon source to reduce the
case is $506/MT, of which 6% is capital expenditure, 32% fuel cost, 29%
electricity consumption in the electric arc furnace. The CTG CO2 emis­
iron ore and scrap cost, and 22% labor cost, with the remaining 11%
sions of EAF are 270 kg/MT steel—86% lower than the CTG CO2
being electricity, other O&M, and by-product credit. As a result of the
emissions in the BF-BOF case. The NG → RNG and grid → clean elec­
addition of the carbon capture unit, the fuel and electricity cost of the
tricity columns in Fig. 5(c) show the CTG CO2 emissions when NG is
BF-BOF-CC case is $29/MT higher than the BF-BOF case, while the labor
replaced by RNG and grid electricity by clean electricity (wind power).
cost of the BF-BOF-CC case is $7/MT higher than the BF-BOF case. The
The use of RNG and clean electricity has the potential to reduce CO2
LCOS of the BF-BOF-all is $542/MT when the minimum renewable fuel
emissions by 34 kg/MT steel and 210 kg/MT steel, respectively. When
price is used for the TEA analysis, as shown in Fig. 6(a).
the combination of RNG and clean electricity is used, the CTG CO2
The LCOS for the BF-BOF-SOA and BF-BOF-SOA-CC cases is $534/
emissions of EAF can be reduced to 25 kg/MT steel that is 91% lower
MT and $596/MT, respectively, or 22% and 36% higher than the current
than the CO2 emissions from current-practice baseline of EAF. Fig. 5(d)
BF-BOF technology. The capital expenditure in the BF-BOF-SOA case
shows the CTG CO2 emissions for state-of-the-art EAF (EAF-SOA) tech­
and BF-BOF-SOA-CC case is $3.7 billion and $4.6 billion per plant,
nology (with energy consumption data shown in Table 1). The CTG CO2
respectively, as shown in Tables 11 and 12. The high capital expenditure
emission of EAF-SOA is 244 kg/MT steel, which is 10% lower than the
for the construction of a greenfield steel plant adds $112/MT and $138/
current-practice baseline EAF steel production plant in the U.S. After all
MT to the steel production cost for the cases with and without CCS,
energy switching options (to RNG and clean electricity), the CTG CO2
respectively, and results in higher LCOS compared to the current BF-BOF
emissions of EAF-SOA-all are 26 kg/MT steel. If RNG and clean elec­
technology.
tricity are used in combination, the CTG CO2 emissions of EAF can be
Fig. 6(a) also shows the impacts of electricity, natural gas, and bio­
reduced by 89%.
char cost on the LCOS. For the BF-BOF baseline, the costs of electricity,
natural gas, coking coal, and industrial coal are $0.07/kWh, $3.7/GJ,
3.2. LCOS of steel and cost of CO2 avoidance $161/MT, and $68/MT, respectively. When clean electricity is used to
replace grid electricity, the clean electricity cost changes to somewhere
The TEA analysis in this study uses 2019 U.S. dollars based on ma­ between $0.03/kWh and $0.15/kWh (Wiser and Bolinger, 2019). Fig. 6
terials market prices in 2019, shown in Table 9. Given that the typical U. (a) shows a sensitivity analysis with three LCOS using $0.03/kWh,
S. BF-BOF steel mill has been running for more than 30 years, the $0.11/kWh, and $0.15/kWh as clean electricity costs. When the elec­
calculation of the LCOS from current BF-BOF baseline case does not tricity cost increases from $0.03/kW to $0.15/kW, it constitutes 2% to
include capital expenditures. For the BF-BOF case, the fixed O&M cost is 8% of the LCOS, and the LCOS changes from $430/MT to $458/MT.
$338 million per year and includes the cost of maintenance, direct labor, Given that the RNG cost is $6.6 to $19.0/GJ, or 78% to 147% higher
and indirect labor. The variable O&M cost is $1051 million per year and than the market cost of NG ($3.7/GJ), the LCOS of steel made using RNG
includes the cost of fuel and reductant, iron ore, purchased scrap, fluxes, is $446-$477/MT, 2%− 9% higher than the LCOS of steel made using the
consumables, and other utilities. The “other O&M” cost is $24 million current BF-BOF technology. Meanwhile, the high cost of biochar
per year and includes all other expenses and by-products credit ($403-$747/MT) results in an LCOS increase to $478-$531/MT, which
(Table 10). The LCOS (making the net present value zero) of BF-BOF is 9%− 21% higher than the current BF-BOF technology. Fig. 6(a) sum­
case is $439/MT crude steel, as shown in Fig. 6(a). For the BF-BOF marizes the range of crude steel market prices in the U.S. from 2019 to
cost, the three leading cost sources are fuel and reductant cost, iron

10
G. Zang et al. International Journal of Greenhouse Gas Control 128 (2023) 103958

Fig. 6. TEA results of BF-BOF (a) LCOS and (b) CO2 avoidance cost relative to the base line case of BF-BOF. BF-BOF-all-no CC, BF-BOF-all, BF-BOF-all-SOA-no CC and
BF-BOF-SOA-all use RNG, clean electricity, coal and biochar as energy sources to produce crude steel. BF-BOF, BF-BOF-CC, BF-BOF SOA and BF-BOF-SOA-CC use NG,
grid electricity, and coal as energy sources to produce crude steel. The other bars show the impacts of renewable energy’s price modified from the current-practice
baseline of BF-BOF.

2020, with a minimum crude steel market price of $268/MT and a transportation. In general, for a specific steel plant, the decarbonization
maximum crude steel market price of $529/MT (USGS, 2020b). options need to be evaluated by considering accessibility to low carbon
Under the current baseline case assumptions, these results indicate energy sources, availability to CO2 pipeline for transportation, prox­
that all the fuel switching options lead to higher steel cost, due to the imity to CO2 storage site, etc. Fig. 6(b) compares the CO2 avoidance cost
current higher cost of low carbon and renewable energy sources (e.g., of different technology and energy switching options, accounting for all
RNG, biochar) relative to fossil energy sources. The energy efficiency CO2 avoidance from cradle to gate boundaries. The LCOS ($439/MT
increase results in a moderate steelmaking LCOS, based on the current steel) and CTG CO2 emissions (1990 kg/MT steel) of the current-practice
technology level. CCS shows the lowest LCOS due to concentrated CO2 baseline BF-BOF technology are used as the reference. For each tech­
emission from BF-BOF that enables low-cost CO2 capture. It is worth nology and energy switching option, the change of LCOS ΔCsteel and the
mentioning that the CCS option did not consider CO2 transportation and change of CO2 emissionsΔECO2 are calculated relative to the baseline
storage cost since the present study represents a generic case for steel case. The CO2 avoidance cost is evaluated using ΔCsteel divided by ΔECO2 ,
production with no specific CO2 storage site or distance for shown in Eq. (2). For example, the CO2 reduction cost of BF-BOF-CC is

11
G. Zang et al. International Journal of Greenhouse Gas Control 128 (2023) 103958

Table 11 reduction costs of using clean electricity, RNG, and biochar are in the
Capital expenditure for the greenfield BF-BOF-SOA case. range of $− 89/MT to $189/MT, $42/MT to $219/MT, and $123/MT to
BF-BOF-SOA Plant section Cost breakdown (US$ $289/MT, respectively. A cost of $− 89/MT means that when the clean
Million) electricity price is $0.03/kWh, the application of clean electricity in the
Coke oven 489 BF-BOF case can reduce both CO2 emissions and LCOS.
Sintering 252 These results show that CO2 avoidance cost for CCS is much lower
Blast furnace, hot stoves, and hot metal 588 than increasing energy efficiency. The CO2 avoidance cost of fuel
desulphurization switching is largely dependent on renewable energy prices, which could
Basic oxygen furnace and steel refining 451
Continuous slab caster 185
be reduced potentially with technology energy efficiency improvement.
Lime production 13 Fig. 7(a) shows the LCOS of various EAF cases. The current-practice
Air separation unit 124 EAF case has an LCOS of $365/MT crude steel, which is 17% lower than
Power plant 266 the current BF-BOF technology. In Fig. 7(a), scrap, electricity, mainte­
Raw material handling 122
nance, and other O&M costs account for 70%, 9%, and 18% of the LCOS,
Spare parts and first fill 110
Pre-operating expenses 20 respectively. The calculated LCOS of EAF at $365/MT is close to esti­
Land preparation, site development and waste 137 mates by previous studies that calculated the EAF steelmaking route cost
disposal as $385/MT (Steelonthenet.com, 2020b). Because the state-of-the-art
Buildings and site infrastructure 186 technology of EAF-SOA includes a capital investment of $17/MT steel,
Project engineering 191
Utility 417
the LCOS of EAF-SOA is 4% higher than current EAF technology. Fig. 7
Total installed cost 3551 (a) shows the impact of the energy switching on LCOSs of steel produced
Contingency (5% of total installed cost) 178 from EAF technology. When RNG and clean electricity are used for steel
Total investment cost (US$ Million) 3728 production, the LCOS changes from $367/MT to $373/MT and from
$346/MT to $405/MT (in Fig. 6), respectively. The LCOS of EAF-all is
$17/MT lower than that of the current EAF technology, assuming that
Table 12 the clean electricity price in the EAF-all case is $0.03/kWh, which is
Capital expenditure for the BF-BOF-SOA-CC case. $0.04/kWh less than that used in the current EAF plant. All the LCOSs of
BF-BOF -SOA-CC Cost breakdown (US$ the EAF technologies are in the range of crude steel market price vari­
Million) ation, indicating that increased energy efficiency and use of clean energy
Coke oven 489
sources can achieve deep CO2 emissions reduction with attractive crude
Sintering 252 steel production cost.
Blast furnace, hot stoves, and hot metal 588 Fig. 7(b) compares the cost of CO2 avoidance energy consumption
desulphurization and energy switching options. The LCOS ($365/MT steel) and CTG CO2
Basic oxygen furnace and steel refining 451
emissions (270 kg/MT steel) from the current EAF technology are used
Continuous slab caster 185
Lime production 13 as the reference to calculate the change (in LCOS ΔCsteel and in CO2
Air separation unit 124 emissions ΔECO2 ) based on different technology and energy switching
Power plant 344 options. The CO2 avoidance cost of EAF-SOA is $447/MT, because the
Steam generation plant 132
amount of CO2 avoided is only 34 kg/MT steel. For the current EAF
Raw material handling 235
Pre-operating expenses 20
technology, the CO2 avoidance cost when using RNG and clean elec­
Land preparation, site development, and waste 137 tricity adds $46/MT to $242/MT and $− 90/MT to $192/MT, respec­
disposal tively, which is similar to the CO2 avoidance cost for the current BF-BOF
Buildings and site infrastructure 186 technology shown in Fig. 6(b).
Project engineering 191
The steel made in BOF technology with a low levels of “tramp” ele­
CO₂ capture and compression 561
Utility 417 ments is destined for flat products, while the steel produced from the
Total installed cost 4324 EAF is served for billet and bloom1 products (Zhu et al., 2019). The U.S.
Contingency (5% of total installed cost) 272 DOE report of Bandwidth Study on Energy Use and Potential Energy Saving
Total investment cost (US$ Million) 4596 Opportunities in U.S. Advanced High Strength Steels Manufacturing esti­
mated the total advanced high-strength steel (AHSS) production in the
$66/MT CO2, i.e., the cost of incorporating carbon capture into the U.S. to be 1.2 million metric tons, with 80% of it produced using BOF
current BF-BOF plant. This result is in the range of the IEA report (IEA, configurations (DOE Advanced Manufacturing Office, 2017). In order to
2013), while lower than the CO2 capture cost of $80-$110/MT from the meet the material property requirements of different applications, BOF
study by Herron et al., 2014. This is because the present study assumes technology can not be completely replaced by EAF technology given the
that waste heat from steel production configurations reduces the NG former technology yields products with higher quality than the latter.
consumption in the boiler. In contrast, the study by Herron et al., 2014 The volatility of scrap cost, iron ore cost, and energy prices contribute to
designed a standalone boiler to supply heat for the carbon capture the differences in LCOSs between BOF and EAF. The sensitivity analysis
process without using waste heat. of the TEA results are shown in Figs. 8 and 9.
The LCOS of BF-BOF-SOA is $96/MT steel higher than the current
BF-BOF cost, but the CO2 emissions reduction is only 148 kg/MT steel. 4. Conclusions
The BF-BOF-SOA has the highest CO2 avoidance cost, $646/MT CO2,
due to the high capital expense required. The BF-BOF-SOA-CC case Iron and steel manufacturing is the largest CO2 emission source and
combines carbon capture with the most efficient BF-BOF-SOA technol­ the second-largest energy consumer among heavy industries worldwide.
ogy. Although the LCOS of BF-BOF-SOA-CC is $157/MT steel higher To decarbonize steel manufacturing, a detailed assessment is required to
than the current BF-BOF technology, it has a larger CO2 emissions
reduction potential of 1113 kg/MT steel, leading to a CO2 avoidance cost
of $141/MT CO2. Fig. 6(b) also shows the CO2 avoidance cost for energy 1
Flat products are finished rolled steel products like steel strip and plate. A
switching options. The cost of renewable energy sources is the key billet is a semi-finished steel product with a square cross section up to 155 mm x
parameter that impacts the CO2 avoidance cost. In Fig. 6(b), the CO2 155 mm. A bloom is a semi-finished product with a square cross section larger
than 155 mm x 155 mm.

12
G. Zang et al. International Journal of Greenhouse Gas Control 128 (2023) 103958

Fig. 7. TEA results of EAF (a) LCOS and (b) CO2 avoidance cost. EAF-all and EAF-SOA-all use RNG and clean electricity as energy sources to produce crude steel. EAF
and EAF-SOA use NG and grid electricity to produce crude steel. While all the other bars show the impacts of renewable energy’s price modified from the current-
practice baseline of EAF.

Fig. 8. Sensitivity analysis results of BF-BOF configuration.

Fig. 9. Sensitivity analysis results of EAF configuration.

13
G. Zang et al. International Journal of Greenhouse Gas Control 128 (2023) 103958

understand a) the current energy consumption and greenhouse gas Conceptualization, Validation, Supervision, Writing – review & editing.
emission profiles associated with steelmaking, b) the decarbonization Marc Melaina: Conceptualization, Validation, Supervision, Writing –
options and the potential CO2 reduction potential; and c) the economic review & editing. Mariya Koleva: Conceptualization, Validation, Su­
impacts of various decarbonization pathways. pervision, Writing – review & editing.
This study provided a systematic analysis with a consistent system
boundary and harmonized assumptions to evaluate various decarbon­ Declaration of Competing Interest
ization options. Six BOF-BOF and four EAF configurations for steel
decarbonization were analyzed, including plant energy efficiency The authors declare that they have no known competing financial
improvement, energy source switching, and CCS, covering all steps from interests or personal relationships that could have appeared to influence
iron ore recovery to final steel production. The CTG CO2 emissions the work reported in this paper.
analysis indicates that the CO2 emissions of BF-BOF and EAF configu­
rations can be reduced by more than 90% compared with the baseline Data availability
cases by combining carbon capture and energy switching from fossil
fuels to renewable energy sources. Data will be made available on request.
The LCOS (levelized cost of steel) was estimated via techno-
economic analysis using a discounted cash flow analysis model. The
LCOS of the U.S. BF-BOF baseline case is $439/MT steel, and that of the Acknowledgments
U.S. EAF baseline case is $365/MT steel. The application of the carbon
capture increases the LCOS of BF-BOF to $506/MT and the combination This research was supported by the Strategic Analysis Office and the
of carbon capture and renewable energy sources increases the LCOS of Hydrogen and Fuel Cell Technologies Office of the U.S. Department of
BF-BOF to $542/MT. The LCOS of BF-BOF-SOA case increases to $534/ Energy’s Office of Energy Efficiency and Renewable Energy under
MT as a result of the high capital investment of the greenfield BF-BOF- Contract No. DE-AC02-06CH11357.
SOA facility. This work was authored in part by the National Renewable Energy
The CO2 avoidance costs vary from -$90/MT CO2 to $646/MT CO2 Laboratory, operated by Alliance for Sustainable Energy, LLC, for the U.
depending on various technologies and energy prices. The CO2 avoid­ S. Department of Energy (DOE) under Contract No. DE-AC36-
ance cost associated with RNG use is $42/MT CO2 to $242/MT CO2, and 08GO28308. Funding provided by the U.S. Department of Energy Office
that of the application of clean electricity is $− 90/MT CO2 to $192/MT of Energy Efficiency and Renewable Energy Office of Strategic Pro­
CO2, impacted by the price of renewable energy sources. The CO2 grams. The views expressed in the article do not necessarily represent
avoidance cost of carbon capture is $66/MT, and that of BF-BOF-SOA the views of the DOE or the U.S. Government. The U.S. Government
and EAF-SOA is $646/MT CO2 and $447/MT CO2, respectively, retains and the publisher, by accepting the article for publication, ac­
depending on the capital investment. knowledges that the U.S. Government retains a nonexclusive, paid-up,
The present study investigates the decarbonization options that can irrevocable, worldwide license to publish or reproduce the published
be applied to the current BF-BOF and EAF processes, which the domi­ form of this work, or allow others to do so, for U.S. Government
nant iron and steel manufacture processes in the United States. Our purposes.
study benchmarks the U.S. steel sector emission baseline, lays out po­
tential decarbonization options for these existing facilities and quan­
Supplementary materials
tifies the decarbonization amount and cost. We are aware of other low
carbon or emerging technologies for steel production, such as DRI-EAF
Supplementary material associated with this article can be found, in
using natural gas or hydrogen to reduce the CO2 emissions in virgin
the online version, at doi:10.1016/j.ijggc.2023.103958.
steel making. We evaluated these emerging DRI technologies and dis­
cussed the potential of further decarbonization in a separate paper
References
(Zang et al., 2023). These two studies together provide insights to steel
industry technology developers and stakeholders/investors to manu­ American Coke and Coal Chemicals Institute, 2020. U.S. coke plants.
facture low carbon steel, and inform policy makers and the public. Our American Gas Foundation, 2019. Renewable sources of natural gas: supply and emissions
research will shed light on ion/steel manufacture decarbonization di­ reduction assessment. Washington, DC.
Arens, M., Åhman, M., Vogl, V., 2021. Which countries are prepared to green their coal-
rections by identifying decarbonization opportunities with quantifica­ based steel industry with electricity? Reviewing climate and energy policy as well as
tion of emission reduction potential; and provide quantitative the implementation of renewable electricity. Renew. Sustain. Energy Rev. 143,
decarbonization cost information that help reduce investment risks and 110938 https://doi.org/10.1016/j.rser.2021.110938.
Arens, M., Worrell, E., Eichhammer, W., Hasanbeigi, A., Zhang, Q., 2017. Pathways to a
accelerate low carbon manufacture technology deployment. low-carbon iron and steel industry in the medium-term – the case of Germany.
J. Clean. Prod. 163, 84–98. https://doi.org/10.1016/j.jclepro.2015.12.097.
Supporting Information Babich, A., Senk, D., 2019. 13 - Coke in the Iron and Steel Industry, in: Suárez-Ruiz, I.,
Diez, M.A., Rubiera, F.B.T.N.T. in C.C. (Eds.). Woodhead Publishing, pp. 367–404.
10.1016/B978-0-08-102201-6.00013-3.
Additional details on process-level energy consumption of BF-BOF Biermann, M., Ali, H., Sundqvist, M., Larsson, M., Normann, F., Johnsson, F., 2019.
cases and technology readiness level are shown in the Supporting Excess heat-driven carbon capture at an integrated steel mill–considerations for
capture cost optimization. Int. J. Greenh. Gas Control 91, 102833. https://doi.org/
Information. 10.1016/j.ijggc.2019.102833.
Birat, J., 2010. Global Technology Roadmap for CCS in Industry-Steel Sectoral Report.
CRediT authorship contribution statement France. https://www.globalccsinstitute.com/archive/hub/publications/15671/glo
bal-technology-roadmap-ccs-industry-steel-sectoral-report.pdf.
Bushell, A., 2018. Master’s Project. Duke University.
Guiyan Zang: Methodology, Formal analysis, Writing – original Cavaliere, P., 2019. Clean Ironmaking and Steelmaking Processes: Efficient Technologies
draft, Software. Pingping Sun: Conceptualization, Validation, Writing – for Greenhouse Emissions Abatement. Springer, Lecce, Italy.
Chen, Q., Gu, Y., Tang, Z., Wei, W., Sun, Y., 2018. Assessment of low-carbon iron and
original draft. Amgad Elgowainy: Conceptualization, Validation,
steel production with CO2 recycling and utilization technologies: a case study in
Writing – original draft, Resources, Supervision. Pallavi Bobba: Data China. Appl. Energy 220, 192–207. https://doi.org/10.1016/j.
curation. Colin McMillan: Resources, Validation, Writing – review & apenergy.2018.03.043.
editing. Ookie Ma: Conceptualization, Validation, Supervision, Writing Cong, H., Mašek, O., Zhao, L., Yao, Z., Meng, H., Hu, E., Ma, T., 2018. Slow pyrolysis
performance and energy balance of corn stover in continuous pyrolysis-based poly-
– review & editing. Kara Podkaminer: Conceptualization, Validation, generation systems. Energy Fuels 32, 3743–3750. https://doi.org/10.1021/acs.
Supervision, Writing – review & editing. Neha Rustagi: energyfuels.7b03175.

14
G. Zang et al. International Journal of Greenhouse Gas Control 128 (2023) 103958

Crombie, K., Mašek, O., 2015. Pyrolysis biochar systems, balance between bioenergy and the USA. Comput. Chem. Eng. 133, 106602 https://doi.org/10.1016/j.
carbon sequestration. GCB Bioenergy 7, 349–361. https://doi.org/10.1111/ compchemeng.2019.106602.
gcbb.12137. Mandova, H., Leduc, S., Wang, C., Wetterlund, E., Patrizio, P., Gale, W., Kraxner, F.,
Cruz, T.T., da, Perrella, Balestieri, J.A., de Toledo Silva, J.M., Vilanova, M.R.N., 2018. Possibilities for CO2 emission reduction using biomass in European integrated
Oliveira, O.J., Ávila, I., 2021. Life cycle assessment of carbon capture and storage/ steel plants. Biomass Bioenergy 115, 231–243. https://doi.org/10.1016/j.
utilization: from current state to future research directions and opportunities. Int. J. biombioe.2018.04.021.
Greenh. Gas Control 108, 103309. https://doi.org/10.1016/j.ijggc.2021.103309. Manzolini, G., Giuffrida, A., Cobden, P.D., van Dijk, H.A.J., Ruggeri, F., Consonni, F.,
Cui, L., Liu, M., Yuan, X., Wang, Q., Ma, Q., Wang, P., Hong, J., Liu, H., 2021. 2020. Techno-economic assessment of SEWGS technology when applied to
Environmental and economic impact assessment of three sintering flue gas treatment integrated steel-plant for CO2 emission mitigation. Int. J. Greenh. Gas Control 94,
technologies in the iron and steel industry. J. Clean. Prod. 311, 127703 https://doi. 102935. https://doi.org/10.1016/j.ijggc.2019.102935.
org/10.1016/j.jclepro.2021.127703. Mayer, J., Bachner, G., Steininger, K.W., 2019. Macroeconomic implications of switching
DOE Advanced Manufacturing Office, 2017. Bandwidth Study on Energy Use and to process-emission-free iron and steel production in Europe. J. Clean. Prod. 210,
Potential Energy Saving Opportunities in U.S. Advanced High Strength Steels 1517–1533. https://doi.org/10.1016/j.jclepro.2018.11.118.
Manufacturing. DOE Advanced Manufacturing Office, Washington, DC. Milford, R.L., Pauliuk, S., Allwood, J.M., Müller, D.B., 2013. The roles of energy and
Echterhof, T., 2021. Review on the use of alternative carbon sources in EAF steelmaking. material efficiency in meeting steel industry CO2 targets. Environ. Sci. Technol. 47,
Met 11 (2), 222. https://doi.org/10.3390/met11020222. 3455–3462. https://doi.org/10.1021/es3031424.
EIA, 2020. Annual energy outlook 2020 with projections to 2050. Washington, DC. https Mousa, E., Wang, C., Riesbeck, J., Larsson, M., 2016. Biomass applications in iron and
://www.connaissancedesenergies.org/sites/default/files/pdf-actualites/AEO2020% steel industry: an overview of challenges and opportunities. Renew. Sustain. Energy
20Full%20Report.pdf. Rev. 65, 1247–1266. https://doi.org/10.1016/j.rser.2016.07.061.
EIA, 2019a. Natural gas prices [WWW Document]. http://www.eia.gov/dnav/ng/ng_pri_ Peacey, J.G., Davenport, W.G., 2016. The Iron Blast furnace: Theory and Practice.
sum_dcu_nus_m.htm. Elsevier.
EIA, 2019b. Coal explained-coal prices and outlook [WWW Document]. 2019. http Rigamonti, L., Brivio, E., 2022. Life cycle assessment of methanol production by a carbon
s://www.eia.gov/energyexplained/coal/prices-and-outlook.php. capture and utilization technology applied to steel mill gases. Int. J. Greenh. Gas
Energy Information Administration (EIA), 2018. Manufacturing energy consumption Control 115, 103616. https://doi.org/10.1016/j.ijggc.2022.103616.
survey (MECS)–steel industry analysis brief [WWW Document]. https://www.eia.go Ryan, N.A., Miller, S.A., Skerlos, S.J., Cooper, D.R., 2020. Reducing CO2 emissions from
v/consumption/manufacturing/briefs/steel/. U.S. steel consumption by 70% by 2050. Environ. Sci. Technol. 54, 14598–14608.
Gajic, D., Savic-Gajic, I., Savic, I., Georgieva, O., Di Gennaro, S., 2016. Modelling of https://doi.org/10.1021/acs.est.0c04321.
electrical energy consumption in an electric arc furnace using artificial neural Steelonthenet.com, 2020a. Basic oxygen furnace route steelmaking costs 2020 [WWW
networks. Energy 108, 132–139. https://doi.org/10.1016/j.energy.2015.07.068. Document]. 2020. URL https://www.steelonthenet.com/cost-bof-2020.html.
Greene, L., 2000. Ironmaking process alternatives screening study volume I: summary Steelonthenet.com, 2020b. Electric arc furnace steelmaking costs 2020 [WWW
report. Oak Ridge, TN. Document]. URL https://www.steelonthenet.com/cost-eaf-2020.html.
Greenhouse Gas Reporting Program (GHGRP), 2019. Greenhouse gas reporting program Suopajärvi, H., Umeki, K., Mousa, E., Hedayati, A., Romar, H., Kemppainen, A.,
industrial profile: chemicals Sector. Washington, DC. https://ghgdata.epa.gov/gh Wang, C., Phounglamcheik, A., Tuomikoski, S., Norberg, N., Andefors, A.,
gp/main.do#. Öhman, M., Lassi, U., Fabritius, T., 2018. Use of biomass in integrated steelmaking –
Griffin, P.W., Hammond, G.P., 2019. Analysis of the potential for energy demand and Status quo, future needs and comparison to other low-CO2 steel production
carbon emissions reduction in the iron and steel sector. Energy Procedia 158, technologies. Appl. Energy 213, 384–407. https://doi.org/10.1016/j.
3915–3922. https://doi.org/10.1016/j.egypro.2019.01.852. apenergy.2018.01.060.
Hasanbeigi, A., Springer, C., 2019. How clean is the U.S. Steel Industry? An international The Athena Sustainable Materials Institute, 2002. Cradle-to-gate life cycle inventory:
benchmarking of energy and CO2 intensities. San Francisco, CA. Canadian and US steel production by mill type. Ottawa, Canada.
He, K., Wang, L., 2017. A review of energy use and energy-efficient technologies for the USGS, 2020a. Iron ore data sheet-mineral commodity summaries 2020 [WWW
iron and steel industry. Renew. Sustain. Energy Rev. 70, 1022–1039. https://doi. Document]. 2020. URL https://pubs.usgs.gov/periodicals/mcs2020/mcs2020-iron-o
org/10.1016/j.rser.2016.12.007. re.pdf.
Herron, S., Zoelle, A., Summers, W.M., 2014. Cost of Capturing CO2 from Industrial USGS, 2020b. Iron and Steel Scrap Statistics and Information [WWW Document]. USGS.
Sources. NETL. ttps://www.usgs.gov/centers/nmic/iron-and-steel-scrap-statistics-and-information?
IEA, 2013. Iron and steel CCS study (Techno-economics Integrated Steel Mill). Stoke qt-science_support_page_related_con=0#qt-science_support_page_related_con.
Orchard, Cheltenham. http://documents.ieaghg.org/index.php/s/P3rYI5vSh Wang, M., Elgowainy, A., Lee, U., Bafana, A., Benavides, P.T., Bumham, A., Cai, H., Dai,
80SPM7. Q., 2020. Summary of expansions and updates in GREET® 2020. Lemont, IL, US.
International Energy Agency (IEA), 2020a. Iron and steel technology roadmap-towards Wiser, R.H., Bolinger, M., 2019. 2018 Wind technologies market report.
more sustainable steelmaking. Paris. http://www.iea.org/reports/iron-and-steel-te World steel Association, 2020. Steel Statistical Yearbook 2020 Concise Version-A Cross-
chnology-roadmap. section of Steel Industry Statistics 2010-2019. World steel Association, Belgium.
International Energy Agency (IEA), 2020b. World energy outlook 2020. Paris. htt Xu, Q., Zou, Z., Chen, Y., Wang, K., Du, Z., Feng, J., Ding, C., Bai, Z., Zang, Y., Xiong, Y.,
p://www.iea.org/reports/world-energy-outlook-2020. 2020. Performance of a novel-type of heat flue in a coke oven based on high-
Jamison, K., Kramer, C., Brueske, S., Fisher, A., 2015. Bandwidth Study on energy use temperature and low-oxygen diffusion combustion technology. Fuel 267, 117160.
and potential energy saving opportunities in U.S. iron and steel manufacturing. https://doi.org/10.1016/j.fuel.2020.117160.
United States. Zang, G., Sun, P., Yoo, E., Elgowainy, A., Bafana, A., Lee, U., Wang, M., Supekar, S.,
Jin, H., Sun, S., Han, W., Gao, L., 2009. Proposal of a novel multifunctional energy 2021. Synthetic methanol/fischer–tropsch fuel production capacity, cost, and carbon
system for cogeneration of coke, hydrogen, and power. J. Eng. Gas Turbines Power intensity utilizing CO2 from industrial and power plants in the United States.
131. https://doi.org/10.1115/1.3078791. Environ. Sci. Technol. https://doi.org/10.1021/acs.est.0c08674.
Kapoor, I., Davis, C., Li, Z., 2021. Effects of residual elements during the casting process Zang, G., Sun, P., Elgowainy, A., Bobba, P., McMillan, C., Ma, O., Koleva, M., 2023. Cost
of steel production: a critical review. Ironmaking Steelmaking 48 (6), 712–727. and life cycle analysis for deep CO2 emissions reduction for steel making: direct
https://doi.org/10.1080/03019233.2021.1898869. reduced iron technologies. Steel Res. Int., 2200297 https://doi.org/10.1002/
Kumar, U., Maroufi, S., Rajarao, R., Mayyas, M., Mansuri, I., Joshi, R.K., Sahajwalla, V., srin.202200297.
2017. Cleaner production of iron by using waste macadamia biomass as a carbon Zhu, Y., Syndergaard, K., Cooper, D.R., 2019. Mapping the annual flow of steel in the
resource. J. Clean. Prod. 158, 218–224. https://doi.org/10.1016/j. United States. Environ. Sci. Technol. 53, 11260–11268. https://doi.org/10.1021/
jclepro.2017.04.115. acs.est.9b01016.
Luh, S., Budinis, S., Giarola, S., Schmidt, T.J., Hawkes, A., 2020. Long-term development
of the industrial sector–case study about electrification, fuel switching, and CCS in

15