Originalarbeit

BHM (2017) Vol. 162 (1): 7–13 DOI 10.1007/s00501-016-0561-8 © Springer-Verlag Wien 2017

Comparing the CO2 Emissions of Different Steelmaking Routes Barbara Rammer, Robert Millner, and Christian Boehm Primetals Technologies Austria GmbH, Linz, Austria Received October 19, 2016; accepted November 14, 2016; published online January 5, 2017 Abstract: Steps to reduce greenhouse emission gases (CO2) are increasingly becoming a matter of priority for the industry, particularly for the iron and steel industry, which is the largest consumer of the world’s generated energy on a global basis. This paper outlines the principal factors necessary for evaluating CO2 emission rates as the basis for comparing the net CO2 emission figures for different iron and steel production routes from cradle to grave. Calculations show that CO2 emissions considerably depend on the assumptions for the system borders and some external factors, e.g. the specific emission values for grid electricity. Based on the calculations carried out, it was also shown that already today the COREX®/DR/EAF combination can show CO2 emissions that are by far lower compared to any other coal based steelmaking configuration. The topics presented in this paper in relationship to the concerns of the iron and steelmaking industry are rarely dealt with in metallurgical journals. However, they are becoming increasingly important because of environmental, political, and especially economic reasons.

tionsrouten dienen. Die Annahmen für Systemgrenzen und einigen externen Faktoren, wie z.B. Emissionswerten für Netzelektrizität, sind wesentliche Faktoren zur Ermittlung der gesamten CO2 Emissionen einer bestimmten Produktionsroute. Basierend auf den durchgeführten Berechnungen zum Vergleich der verschiedenen Verfahrensrouten (Hochofen, Schmelzreduktion, Direktreduktion, COREX®-Direktreduktions-Verbund, Elektrostahlerzeugung, Konverterstahlerzeugung) konnte gezeigt werden, dass die COREX®/DR/EAF Kombination zu CO2 Emissionswerten weit unter allen anderen kohlebasierten Stahlerzeugungsrouten führen kann. Die in diesem Artikel präsentierten Themen werden in metallurgischen Zeitschriften nur selten behandelt, sind aber – insbesondere in Anbetracht der politischen, ökonomischen und umwelttechnischen Hintergründe – von immer stärker wachsender Bedeutung.

Keywords: CO2 emissions, Blast furnace, COREX®, MIDREXTM, FINEX®

1. Introduction

Vergleich der CO2 Emissionen verschiedener Verfahren zur Stahlproduktion Zusammenfassung: Die Reduktion von Treibhausgasen (CO2) in industriellen Prozessen wird mehr und mehr zum prioritären Thema, insbesondere für die Eisen- und Stahlindustrie, die den weltweit größten Energiekonsumenten darstellt. In dieser Studie werden die prinzipiellen Faktoren vorgestellt, die als Basis für die Ermittlung der CO2Emissionswerte unterschiedlicher Eisen- und StahlprodukC. Boehm () Primetals Technologies Austria GmbH, Turmstrasse 44, 4031 Linz, Austria [email protected]

BHM (2017), 162. Jg., Heft 1

Schlüsselwörter: CO2 Emissionen, Hochofen, COREX®, MIDREXTM, FINEX®

To avoid the expected implications from climate change and global warming, global climate conferences and agreements frequently target the reduction of CO2 as the major greenhouse gas in our atmosphere. The first political decision to take steps to mitigate climate change and to move from a voluntary approach to a regulatory approach with legally binding commitments was marked by the 1997 Kyoto Protocol [1]. It set targets for nearly 40 industrialised nations to cut emissions but allowed them to meet these targets by paying developing countries to cut emissions on their behalf, through an international market in carbon offsets. So, the Protocol’s biggest, most direct business impact was to spawn an international trade in emissions permits [2]. At the Paris Climate Conference (COP21) in December 2015, 195 countries adopted the first-ever universal, legally

© Springer-Verlag Wien

Rammer et al.

7

Originalarbeit

Figure 1: Diminishing returns from improved pig iron production efficiency [5]

binding global climate deal. This agreement goes far beyond prior political consensus and sets out a global action plan to put the world on track to avoid dangerous climate change by limiting global warming to “well below 2°C above pre-industrial levels by reducing greenhouse gas emissions to 40 gigatonnes” [3]. The heavy industry sector, and especially steel production, is a major source of greenhouse gas emissions. Together, the steel and cement industries accounted for 8% of global energy use and 15% of global anthropogenic CO2 emissions in 2012 [4]. It is well-known that iron and steel making processes that use coke ovens, blast furnaces, and basic oxygen steel furnaces generate most CO2. For this traditional route, decades of effort to reach world-leading levels of energyefficiency finally lead to a point where further reduction of energy produces more and more diminishing results (Figure 1) and the theoretical absolute minimum energy needed for pig iron production lies only marginally below a modern state-of-the-art blast furnace (Table 1). But still, the steel industry represents the world’s largest industrial source of carbon dioxide emissions, mainly due to the requirement for coal to convert ore into molten iron [8]. According to the World Steel Association [9], global crude steel production has grown over the last decade from 904 Mt in 2002 to 1670 Mt in 2014, despite the 2008 financial crisis.

One possible consequence from these findings is that climate policies may lead to a significant relocation of energy-intensive industries (such as steel making) away from OECD countries, which might also result in higher global emissions [10]. Others are to follow alternative iron- and steelmaking routes that are less carbon-intensive, such as direct reduction processes, EAF-steelmaking from scrap or future developments like hydrogen steelmaking. Right at the moment, the share of electric arc furnaces and direct reduction in secondary and primary steel making, which generate much less CO2 per tonne of crude steel produced, is already increasing [11]. To compare the efficiency of such modern iron and steelmaking routes with regards to CO2 emissions, several models have been developed which describe different parts of the process. They play a key part in informing decision makers about future trends in the energy system and are an essential tool to evaluate future trends and developments.

2. Basics for CO2 Emission Calculation 2.1 Definition of Emission Scopes according to Greenhouse Gas Protocol The global nature of greenhouse gas emission effects requires a system-wide approach and a solution where all factors adding to the total amount of CO2 emission are considered. To achieve such a global view on the whole steel production route from mining of raw materials to liquid steel, both direct and indirect emissions must be taken into consideration. Direct emissions are those emissions which are directly produced during industrial processes, and indirect emissions are those emissions which are produced during, for example, the generation of electricity at power stations or during the mining, preparation, and transport of raw materials. In this context, the Greenhouse Gas Protocol [12] defines three scopes, describing the different parts of production routes within the steelmaking process (Figure 2).

TABLE 1

Energy demand and CO2 emissions of actually built blast furnaces compared with the theoretical and practical minimum values: Further efficiency gains are possible, but relatively limited [6] Energy needs [GJ/t pig iron] blast furnace in year 1850

100

current average blast furnace

~20

CO2 emission [t CO2/t pig iron] ~1.4

current blast furnaces (best practice)

15.0

1.3

practical minimum energy needs *)

10.4

1.1

absolute minimum energy needs **)

9.8

0.9

*)

calculation including typical energy losses – e.g. reactions and melting of gangue, coal/ coke ashes and additives [7] **) theoretical minimum energy needed to produce liquid iron from iron oxide by reduction with carbon

8

Rammer et al.

© Springer-Verlag Wien

BHM (2017), 162. Jg., Heft 1

Originalarbeit

Figure 2: Definition of Scopes 1, 2, and 3 in iron and steelmaking

Scope 1: All direct GHG emissions. Scope 2: Indirect GHG emissions from consumption of purchased electricity, heat, or steam. Scope 3: Other indirect emissions, such as the extraction and production of purchased materials and fuels, transport-related activities in vehicles not owned or controlled by the reporting entity, electricity-related activities (e.g. T&D losses) not covered in Scope 2, outsourced activities, waste disposal, etc.

2.2 Emission Values for Power Generation The Greenhouse Gas Protocol states that Scopes 1 and 2 should be reported as a minimum, and therefore electricity consumption is a key component of almost all CO2 emission calculations. The nation-wide specific CO2 emission value serves as the basis for the calculation of indirect emissions, which are of particular importance for steel production routes with a high electric power consumption (EAF route) as well as for possible CO2 credits in steel plants where electricity is generated from export gases (LD/BOF route). This value differs from country to country and is

based on the different ratios of hydroelectric, nuclear, and thermal power employed for the generation of electricity (Table 2).

2.3 Steel Production Routes and Definition of Borders Figure 3 shows various liquid-steel-production routes. The different steelmaking routes lead to different products (e.g. hot metal from blast furnace or DRI from a direct reduction plant) that are not comparable at first glance. To allow comparability, all routes have to be levelled by defining a common end product – liquid steel – and accordingly by integrating a BOF for smelting reduction and an EAF for the DR (direct reduction) route into the calculation. The calculations presented in this paper include Scopes 1, 2, and 3 according to the Greenhouse Gas Protocol and therefore also include production and transportation of raw materials. For the Blast Furnace Route, the coking plant and sinter plant are assumed to exist on site. Coal and ore mining are assumed to happen overseas, and an appropriate share of emissions for overseas transport are included in the calculations.

TABLE 2

Specific emission values for grid electricity, excluding heat generation (Basis 2011) [13] Country

kg CO2/kWh

Country

kg CO2/kWh

Norway

0.002

South Korea

0.504

Sweden

0.023

United Kingdom

0.509

France

0.071

Russia

0.513

Brazil

0.093

USA

0.547

Austria

0.177

World

0.624

Canada

0.180

Iran

0.631

Venezuela

0.208

Germany

0.672

Spain

0.343

Indonesia

0.685 0.973

Argentina

0.392

China

Italy

0.411

South Africa

1.069

Japan

0.443

Bulgaria

1.166

OECD Europe

0.452

India

1.333

BHM (2017), 162. Jg., Heft 1

© Springer-Verlag Wien

Rammer et al.

9

Originalarbeit

raw material mining and transport. For COREX® in Germany, this Scope 1 value (2 194 kg CO2) can be found as the sum of the top section in Table 3 (“Scope 1: Direct GHG emissions”). To maintain comparability with the other routes (as mainly coal is charged into the COREX®/FINEX® melter gasifier and in-situ gasified), for the blast furnace, the coking plant, and the sinter plant are kept within the calculation borders and their emissions included in the results.

3. Calculation Results The Steel Production Routes examined in this paper are:

Figure 3: Liquid steel production routes and borders of the CO2 calculations

2.4 CO2 Credits for Slag and Off Gas Utilization Valuable by-products of the steel-making process – such as slag and all kinds of off gases – are usually either used directly in the process or sold to an external consumer. They do carry a CO2 load though, which has to be taken into consideration to present a justified picture of the whole process. In the calculation it is assumed that all produced gases are used within the steel plant to a maximum extent to replace electricity needed for the process and thereby reduce the total CO2 emission. Those gases that cannot be directly used within the steel plant itself are assumed to be utilized in a modern combined-cycle power plant with an efficiency of 45%. The credit is calculated from the amount of electricity produced by the specific country’s emission value for grid electricity. The consideration of these credits is inevitable for the description a realistic process scenario, since no blast furnace, COREX®, or FINEX® plant can economically operate without further utilization of its off gas, and no COREX® or FINEX® plant will be built without appropriate gas utilization. Even though this fact is neglected in many calculations nowadays, it will have to play a role in future carbon footprint studies. Table 3 shows an example of a CO2 calculation for the COREX®–LD/BOF + CCPP (combined cycle power plant) route including credits for gas and slag utilization with Germany as a country-basis.

2.5 Operator-Specific Calculation Model without Consideration of Credits or Emission Factors for Grid Electricity To emphasize the importance of the correct definition of borders, an alternative calculation was carried out, considering only Scope 1 for the different production routes. Therefore, this calculation was conducted with zero CO2 emitted for power production, no CO2 credit for the utilization of any off gas or slag, and exclusion of emissions for

10

Rammer et al.

Blast Furnace with BOF (20% scrap) and CCPP COREX® with BOF (20% scrap) and CCPP FINEX® with BOF (20% scrap) and CCPP (with internal gas-recycling in FINEX®) EAF Furnace (100% scrap) MidrexTM with EAF (20% scrap) FINORED® with EAF (20% scrap) COREX® - DR combination with EAF (20% scrap) and CCPP Figure 4 shows the calculated CO2 emissions for the different routes mentioned above. On the left, the three smelting reduction + BOF routes are listed, in the middle, the EAF variants, and on the right, a combination of COREX® smelting reduction with a direct reduction plant is shown. The different (blue) columns assigned to one production route derive from the consideration of different nationwide specific CO2 emission values for power generation. They range from 0.023 kg CO2/kWh for Sweden to 1.333 kg CO2/kWh for India. A significant dependence on these values can be seen in the results, showing that the right choice of location for a specific plant concept is crucial for the successful minimization of greenhouse gas emissions. The orange columns depict the results from the operator-specific calculation model (as described in Section 2.4 - zero CO2 emitted for power production, no CO2 credit for the use of any off gas or slag, and exclusion of emissions for raw material production and transport). When comparing these orange values of the different process routes, it has to be taken into account that the high brutto CO2 emissions for COREX® or FINEX® derive from the non-utilization of their off gas. This approach – although viable for other steel making routes – is under no circumstances representing a realistic situation for a COREX® or FINEX® plant. As noted in Section 2.4, no COREX® or FINEX® plant will ever be built without further utilization of the gas as economics of COREX® and FINEX® depend on the utilization of the gas. What is clear to see in Figure 4 is the difference between the classic BOF-steelmaking route based on smelting reduction with blast furnace or COREX®/FINEX® on the left and the production routes based on the use of natural gas (FINORED®/MIDREXTM–EAF) on the right. Concerning the CO2 emission, production routes based on the use of natural gas are advantageous compared to

© Springer-Verlag Wien

BHM (2017), 162. Jg., Heft 1

Originalarbeit

TABLE 3

Example of a CO2 calculation for the COREX®-LD/BOF + CCPP route (Germany), to illustrate how credits for gas and slag utilization are integrated Scope 1: Direct GHG emission Process Related Emissions

905 kg CO2 ®

CO2 due to combustion of CO and CH4 in COREX Export Gas

1 180 kg CO2

CO2 due to combustion of CO and CH4 in Converter Gas

108 kg CO2

2 194 kg CO2

Scope 2: GHG emissions from imports of electricity, heat, or steam CO2 due to electrical power generation for COREX®

0.672 kg CO2/kWh

47 kg CO2

CO2 due to electrical power gen. for ASU (Oxygen for COREX®)

0.672 kg CO2/kWh

133 kg CO2

CO2 due to electrical power generation for LD Plant

0.672 kg CO2/kWh

7 kg CO2

CO2 due to electrical power gen. for ASU (Oxygen for the LD Plant)

0.672 kg CO2/kWh

15 kg CO2

201 kg CO2

Scope 3: Other indirect GHG emissions CO2 due to production and transportation of coal (same as iron ore)

29 kg CO2

CO2 due to production and transportation of coke

83 kg CO2

CO2 due to production and transportation of iron ore (lump ore, fine ore)

23 kg CO2

CO2 due to production and transportation of pellets

123 kg CO2

CO2 due to production and transportation of sinter

0 kg CO2

CO2 due to production and transportation of HWKS (steel plant wastes)

0 kg CO2

CO2 due to production and transportation of limestone

1 kg CO2

CO2 due to production and transportation of dolomite

2 kg CO2

CO2 due to production and transportation of quartz

0 kg CO2

CO2 due to production and transportation of LPG

0 kg CO2

CO2 due to production and transportation of DRI/HBI

0 kg CO2

CO2 due to production and transportation of scrap

4 kg CO2

CO2 due to production and transportation of fine coal

0 kg CO2

CO2 due to production and transportation of burned lime

73 kg CO2

338 kg CO2

Total CO2 emissions Gross CO2 emissions without credits

2 733 kg CO2

Credit for granulated slag

-249 kg CO2

Credit for sold LD slag

-2 kg CO2

Credit for converter gas recovery

0 kg CO2

Credit for converter heat recovery

-14 kg CO2

Credit for power generation (CX Gas) with CCPP (Eta=45%; 0.672 kg CO2/kWh)

-932 kg CO2

Credit for power generation (Converter Gas) with CCPP (Eta=45%; 0.6722 kg CO2/kWh)

-59 kg CO2

NET CO2 emissions unter consideration of credits

1 476 kg CO2

production routes based on smelting reduction. This is due to the use of reformed natural gas for reduction work which contains large amounts of hydrogen. As a result, large parts of the reducing gas is oxidized to H2O instead of CO2. The blast furnace, COREX®, and FINEX® processes utilize coke and coal respectively, and thus carbon is the primary energy source and reductant (Figure 4), leading to CO and CO2 as the main gas emission. From the GHG emission point of view, the COREX® and FINEX® processes resemble each other very closely. The difference in the calculation results derives from the fact that, for the FINEX® calculation, additional gas recycling was added. This means that FINEX® off gas is recycled back into the smelting reduction process, creating higher gas use efficiency and the CO2 credit for

BHM (2017), 162. Jg., Heft 1

-1 258 kg CO2

export gas use is accordingly smaller. This gas recycling can be applied both to the COREX® and the FINEX® processes and is advantageous in countries with low specific emission values for grid electricity. The most favourable conditions when it comes to CO2 emission are achieved by the scrap + EAF route, but this is limited due to the availability of high-quality scrap necessary for high-quality steel. Another noticeable difference between the two basic routes lies within their dependency on the nation-wide specific CO2 emission value for the generation of electric power. Whereas “high” specific CO2 emission values (e.g. in China and India with their high percentage of thermal power generation) lead to a lowered total CO2 emission

© Springer-Verlag Wien

Rammer et al.

11

Originalarbeit

Figure 4: Comparison of CO2 emissions of liquid steel production routes

of the smelting reduction + BOF processes, such “high” values are rather disadvantageous for the direct reduction + EAF routes. The reason lies within the use of the process off gas as a replacement for purchased electricity. If those gases are utilized in a combined cycle power plant, much lower emission values are achieved than in the nation-wide electricity grid and the produced electricity is much “greener” than the typical thermal power generated electricity received from the grid. In the calculation, this is represented by the CO2 credit for the generation of electricity. In the FINORED®/MidrexTM–EAF routes, no off gas is available for power generation and the EAF is a very electric power-intensive technology. Therefore, the CO2 footprint worsens with high CO2 emission values, such as in China or India. In the combined COREX®+DR plant production route, COREX® export gas is used as a reduction gas in the DR (direct reduction – e.g. MidrexTM) plant. This route is advantageous in “low” specific CO2-emission-value countries due to the low CO2 emission figures for electric steelmaking as well as due to the utilization of the COREX® export gas for the production of direct reduced iron (DRI). This production route has already been successfully implemented at ArcelorMittal South Africa, Saldanha Steel Works, South Africa, and Jindal South West Steel, India, where, for each ton of coal, two tons of iron (one ton of hot metal and one ton of DRI) are produced. The COREX® hot metal and the DRI are subsequently melted in the EAF to liquid steel. Omitting Scopes 2 and 3 in the calculations (orange bars in Figure 4) gives results rather similar to those of Sweden – the country with the lowest nation-wide CO2 emission value for electric power generation. This shows that the main influence from Scopes 2 and 3 comes from purchased electricity for the process. In this case of very low CO2 emissions caused by electricity production, the blast furnace is the most advantageous route in smelting reduction processes. The direct reduction + EAF routes – naturally – benefit most from low emission values, but they are

12

Rammer et al.

only feasible if sufficient amounts of natural gas are available at a reasonable cost. If the latter is not the case, the combined COREX®+DR plant is an excellent alternative for producing steel at much lower CO2 emissions than the blast furnace or other smelting reduction technologies. In countries with very high CO2 emission values, the smelting reduction technologies become more advantageous as no EAF is required for the production of liquid steel. In countries with very high emission values which usually also have high costs for natural gas, alternative smelting reduction technologies (COREX® and FINEX®) become more beneficial – both with regards to CO2 emissions and economic feasibility.

4. Summary Calculations have been carried out, taking into account Scopes 1, 2, and 3 (according to the Greenhouse Gas Protocol) for different steelmaking routes. Depending on the local site conditions and especially the specific CO2 emission value for the generation of electric power, the COREX® and FINEX® processes offer significant potential for achieving low greenhouse gas emission figures. Although the direct reduction + EAF routes benefit most from low emission values, they are only economically feasible if sufficient amounts of low-cost natural gas are available at a reasonable cost. If the latter is not the case, the combined COREX®+DR plant is an excellent alternative for producing steel at much lower CO2 emissions than the blast furnace or other smelting reduction technologies.

References 1. Rousaki, K.: Market mechanisms for greenhouse gas reduction, IEA Coal Research, 2001 2. Wynn, G.: The likely winners and losers from a Paris climate agreement, Chinadialogue, Dec., 2015 3. United Nations Framework Convention on Climate Change: Adoption of the Paris Agreement, Dec., 2015

© Springer-Verlag Wien

BHM (2017), 162. Jg., Heft 1

Originalarbeit

4. IEA Energy Technology Perspectives 2015 International Energy Agency, Paris, 2015 5. Smil, V.: Steel Industry in Long-Term Perspective: Efficiencies, Opportunities and Limits, ICEF 2nd annual Meeting, Tokyo, 2015 6. Bennett, S.: Perspectives on CCS and its application to the steel sector, 79th Session of the Steel Committee, Paris, 2015 7. Fruehan, R.J.; Fortini, O.; Paxton, H.W.; Brindle, R.: Theoretical Minimum Energies to Produce Steel for Selected Conditions, prepared for the U.S. Department of Energy, Washington, DC, 2000 8. Allwood, J. M.; Cullen, J. M.; Carruth, M. A.; Cooper, D. R.; McBrien, M.; Milford, R. L.; Moyniham, M. C.; Patel, A. C. H.: Sustainable Materials with Both Eyes Open, UIT Cambridge, Cambridge, UK, 2012

BHM (2017), 162. Jg., Heft 1

9. World Steel Association, Steel Statistical Yearbook, World Steel Association, Brussels, Belgium, 2015 10. Babiker, M. H.: Climate change policy, market structure, and carbon leakage, J. Int. Econ., 65 (2005), No. 2, p. 421 11. Olivier, J. G. J.; Janssens-Maenhout, G.; Peters, J. A. H. W.: Trends in Global CO2 Emissions, PBL Netherlands Environmental Assessment Agency, 2015 Report, 2015 12. http://www.ghgprotocol.org/ (May 27th 2016) 13. Brandner, M.; Sood, A.; Wylie, C.; Haughton, A.; Lovell, J.: Electricity-specific emission factors for grid electricity, ecometrica, 2011, http://ecometrica.com/assets/Electricity-specific-emission-factorsfor-grid-electricity.pdf (May 30th 2016)

© Springer-Verlag Wien

Rammer et al.

13