Several governments, industry actors, think tanks and academia have published iron and steel decarbonisation pathways between 2020 and 2025, both at global and South African level. The low-carbon options to reduce the emissions of existing assets and build near-zero emissions new assets are well identified, but many of these options are not economically viable. In particular, high carbon prices and low-cost renewable electricity and hydrogen are essential for narrowing the competitiveness gap with conventional, fossil fuel-based production routes of iron and steel. This chapter showcases external analyses proposing decarbonisation pathways for the steel sector globally and in South Africa. The chapter furthermore presents the identified low-carbon technologies selected in this project to carry out an economic assessment.
Implementing the OECD Framework for Industry’s Net‑Zero Transition in South Africa
Abstract
2.1.1. Global steel decarbonisation pathways
The steel sector is at a crossroads: before 2030, 71% of existing coal-based blast furnaces (1090 Mt) will reach the end of their lifespans and require major reinvestments (Agora Industry, Wuppertal Institute and Lund University, 2021[1]). Meanwhile, the demand for steel is projected to continue growing in the next three decades, as demand increases in emerging and developing economies (IEA, 2020[2]). Relying on coal-based production to meet this demand would result in long-term carbon lock-in, risk stranded assets, threaten jobs and make achieving a 1.5°C-aligned pathway unattainable.
Steel decarbonisation requires a deep transformation of the sector, relying on a mix of technology shifts, energy transitions and material efficiency. Various scenarios for steel decarbonisation by 2050 exist,1 each with varying assumptions and relative importance of mitigation options. However, there is consensus that the 2050 steel industry will be made up of a mix of low-carbon technologies and that the global steel transformation needs to start this decade. An OECD analysis of corporate decarbonisation plans produced by the major steel producers in each region underlines that all regions are considering a wide range of decarbonisation solutions (OECD, 2024[3]).
There is a growing momentum for the steel sector decarbonisation. Over 90% of steelmaking capacity is located in countries committed to net-zero targets (OECD, 2024[3]). Efforts such as the E3G Steel Policy Scorecard provide insights into the speed, pathways and policies being adopted by G7 countries and other key steel producers to decarbonise their steel industries. Amongst steel companies, the vast majority have set ambitious decarbonisation objectives, with two-thirds targeting net-zero emissions and many adopting Scope 1 and Scope 2 targets (OECD, 2024[3]). However, the level of detail in the decarbonisation strategies varies drastically between companies and many would benefit from a more comprehensive approach and further implementation details.
Despite mounting pressure to decarbonise, the steel sector remains off track to achieve net-zero targets by 2050. Over the past decade, global emissions have continued to rise, driven largely by the dominance of the blast furnace-basic oxygen furnace (BF-BOF) route, which is the most carbon-intensive (IEA, 2024[4]). This trend shows no signs of slowing as an estimated 65 Mt of new high-emission blast furnaces are either planned or under construction, expected to come online by the end of 2026 (IEA, 2024[4]).
According to the IEA Net Zero Emissions by 2050 Scenario, increased scrap recycling and mass deployment of innovative technologies are key levers for reducing emissions in the steel sector. A combination of avoided demand, energy efficiency, hydrogen-based production and electrification are expected to make up the bulk of emissions reductions in the steel sector (see Figure 2.1). Until 2030, electrification via EAF and energy efficiency are the main low-hanging fruits, as many profitable projects can already be deployed. These same mitigation measures are projected to play an even greater role by 2050, with hydrogen and CCUS also playing a prominent role. At sector level, most of the emissions reduction are likely to take place after 2030, as many low-carbon options to decarbonise the emissions-intensive ironmaking step of the value chain are not yet commercially viable (IEA, 2023[5]).
Source: (IEA, 2023[5]).
Other modelling exercises differentiate between policy approaches and timeframes for action. For example, the Mission Possible Partnership Carbon Cost scenario illustrates how the sector might decarbonise with coordinated action to support low-emissions steelmaking this decade, while the Technology Moratorium scenario assumes limited progress this decade, before constraining investments in the steel sector to low-emissions technologies from 2030 onwards. The main short-term improvements are related to increased scrap use in EAF and the use of natural gas to produce DRI to replace coal-based ironmaking processes. In the longer run, these scenarios showcase carbon capture and more importantly hydrogen as the main two technologies to achieve deep decarbonisation for the sector (see Figure 2.2). Diverse other steel sector decarbonisation modelling exercises have been conducted with varying levels of emphasis placed on select mitigation measures and varying pace of implementation.
Annual emissions (Scope 1 and Scope 2), in Gt CO2:
Note: The “2050 emissions – 2020 static technology composition” bars in both panels represent what annual emissions would be in 2050 if projected steel demand were met by the same technologies in the same proportions as in 2020. This is not the same as the Baseline scenario, in which some production technology changes occur.
2.1.2. Main technologies and mitigation measures
Carbon capture, use and storage (CCUS) may play an important role in decarbonising steel, especially in countries with a young fleet of BF-BOF or natural gas-based DRI plants. Its deployment in the steel sector will largely depend on capture costs, as well as the availability of CO₂ pipelines and storage sites and the costs to develop these infrastructures. However, carbon capture is often seen as a transition technology. Indeed, there are multiple emission points in steel plants, especially for the BF-BOF route, which would increase the cost required to achieve full decarbonisation. Moreover, most of the planned CCU projects related to the steel industry aim to produce fuels. Therefore, most CCU solutions lead to emissions when these fuels are burnt, even though they substitute current fossil fuel-based fuels that create higher lifecycle emissions. CCUS is still at the demonstration stage today, and few CCUS projects are currently operational in the steel industry globally (Global CCS Institute, 2024[7]). However, CCUS is the technology most commonly cited in corporate decarbonisation plans of large steel producers.2 This is mostly related to the decarbonisation of existing assets, as very few companies plan to implement CCUS for new plants (OECD, 2024[3]). The IEA Net Zero by 2050 Roadmap foresees several CCUS-equipped process technologies being deployed in parallel, including innovative blast furnace retrofits and natural gas-based DRI production with carbon capture. It estimates that CCUS-equipped production methods will account for 37% of iron production by 2050, with 399 Mt of CO2 captured (IEA, 2023[5]).
Direct reduced Iron (DRI) is currently used for about 10% of global steel production (OECD, 2019[8]), and is a promising technology to decarbonise ironmaking via the use of renewable hydrogen. For DRI production to be compatible with a 1.5°C-aligned trajectory without having to capture carbon emissions, renewable hydrogen must be used. Natural gas-based DRI can be a key transition solution, as it is mature technology that can then switch to use renewable hydrogen only with limited retrofits (Yu et al., 2021[9]). This is especially the case when natural gas infrastructure already exists, as building new natural gas-based infrastructure may increase the risk of emissions lock-in. Renewable hydrogen-based DRI, with and without the use of natural gas as a transition fuel, represents two-thirds of the announced low-emissions projects in the iron and steel industry (LeadIT, 2025[10]). Renewable hydrogen-based DRI accounts for nearly half of iron-based steel production by 2050 in the 2023 IEA Net Zero Emissions by 2050 Scenario (IEA, 2023[5]).
Electric arc furnaces (EAFs) are the most common technology for recycling steel. Today, 500 to 550 Mt of steel per year are produced through EAFs, including around 400 Mt of recycled steel scrap (OECD, 2024[11]) Net-zero scenarios for the steel sector suggest that the share of scrap input in steel manufacturing will increase, with the IEA Net Zero Emissions by 2050 Scenario projecting it making up 48% by 2050 (IEA, 2023[5]). While more than 90% of scrap is already recovered (Climate Group, 2025[12]), the scrap quantity used in EAF could further increase and its quality could improve with new recycling technologies and incentives for scrap collection, sorting and processing (OECD, 2024[11]).
Material and energy efficiency measures have significant potential to reduce emissions in the steel sector, although they cannot achieve near-zero or net-zero emission. These solutions are often low-hanging fruits deployable in the short-term, allowing for rapid progress on decreasing steel sector emissions. The basic principle of material efficiency in the steel sector is to use less steel to meet the same demand. This can be achieved by minimising the use of materials through improved design, lengthening the functional life of steel products and facilitating steel recycling. Material efficiency alone reduces demand for cement and steel by 20%, saving around 1 700 Mt CO2 (IEA, 2021[13]). In construction for example, substituting high-strength steel for regular steel can achieve a CO2 reduction of around 30% in steel columns and around 20% in steel beams due to the reduced amount of steel required for the same function (World Steel Association, n.d.[14]).
Significant energy efficiency gains already have and can still be made in the steel sector. Improvements in energy efficiency have led to reductions of about 60% in energy required to produce a tonne of crude steel since 1960 (World Steel Association, n.d.[15]). However, there are still additional efficiency gains to be made, with estimates indicating that a combination of energy efficiency and increased deployment of best available technologies can reduce emissions in the steel sector by 15-25% (de Villafranca Casas et al., 2024[16]).
Several other technologies for steel decarbonisation exist but are currently at lower technology readiness levels. This includes for example iron ore electrolysis, which presents another promising but less mature option. By directly using electricity to reduce iron ore, it avoids energy losses associated with hydrogen production in hydrogen-DRI.
2.1.3. Steel decarbonisation scenarios in South Africa
A strategic combination of mature and disruptive technologies offers the most viable route to achieving net-zero emissions by 2050 for the South African iron and steel industry. Mature solutions such as renewable electrification (accounting for 14% of emissions reductions), energy efficiency improvements (10%), fuel switching (10%) and increased recycling (6%) form a critical foundation. Disruptive technologies, including renewable hydrogen for DRI and CCUS for existing BF-BOF assets, hold the potential to achieve an additional 50% reduction in emissions (see Figure 2.3). In parallel of scenarios developed at country-level, companies can develop their own decarbonisation plans. Box 2.1 provides details of the decarbonisation roadmap of ArcelorMittal South Africa, the largest steel producer in South Africa.
Note: The graph shows the potential of each technology to reach the net-zero target during the period 2025-2050.
Source: Authors based on (National Business Initiative, 2023[17]).
ArcelorMittal South Africa (AMSA), the major steel producer in the country, released in January 2023 a decarbonisation roadmap to decarbonise its steel production. It targets to reduce its carbon intensity by 25% from a 2018 baseline of 2.9 tonnes of CO2 equivalent per tonne of crude steel to 2.16 t CO2e crude steel by 2030, and 85% by 2050 to 0.4 t CO2e/t crude steel. This roadmap anticipates a 25% reduction in costs by 2030.
The roadmap is divided in three different phases:
2023–2027: The roadmap envisages BF-BOF undergoing “no-regret” improvements between 2023 and 2027, including increasing scrap metal use, energy and operational efficiency and contract Power Purchase Agreements.
2027–2030: AMSA aims to start the commissioning of an EAF and close one of the two blast furnaces in Vanderbijlpark.
2030–2050: The company aims to accentuate the transitioning from natural gas to renewable hydrogen for steel production, further integrating carbon capture and use (CCU) and renewable hydrogen technologies, and increasing the use of low-carbon electricity sources.
Furthermore, the company is exploring the production of green DRI at the Saldanha Works plant, by reusing and adapting assets that have been mothballed in 2020 but are still being maintained. As of April 2024, a pre-feasibility report was nearing completion, with decision to be made to progress to feasibility.
Notes: 1) t CO2e/t crude steel, 2018 baseline – 2050; 2) the decarbonisation roadmap is affected by strategic business decisions announced after the publication of the roadmap, including the potential wind-down or monetization of plants.
Renewable electricity is a central component in steel decarbonisation, with the potential to mitigate 14% of the steel sector’s emissions. When considering the entire value chain, electricity accounts for 50% of the energy input in the EAF route and 7% in the BF-BOF route, highlighting its critical importance to steel production (World Steel Association, 2021[21]). Expanding the share of electricity from low-carbon sources can significantly accelerate steel decarbonisation, especially as the share of renewables in South Africa’s power mix is estimated to rise from 10% in 2015 to 37% by 2030 (IRENA, 2020[22]). To achieve this, South Africa targets to rollout at least 3 GW of renewable energy per annum from 2025 onwards, ramping up to 5 GW per annum by 2030 (Government of South Africa, 2025[23]), in addition to the installed renewable power generation capacity of 11 GW in 2024 (IRENA, 2025[24]).
Hydrogen and CCUS are projected to bring the highest emission reduction in South Africa's iron and steel sector, contributing half of the total emissions reduction potential by 2050. Near-zero emissions are achievable when DRI processes are powered by hydrogen derived from renewables electricity. South Africa views this technology as a strategic opportunity, with ambitious plans to scale hydrogen production from 0.02 Mt in 2025 to 7 Mt by 2050 (Department of Trade, Industry and Competition, 2022[25]). CCUS can play a crucial role in decarbonising the existing assets in the BF-BOF route, enabling the capture of CO2 released during production and retrofitting existing steel assets until they reach the end of their life cycle (National Business Initiative, 2023[17]).
2.2.1. Selection of low-carbon technologies
The selection of low-carbon options for the Framework implementation relies on four principles:
1. The cases cover only a subset of technologies. The project does not aim to build comprehensive sectoral roadmaps but rather to analyse how to improve the viability of critical technologies to realise deep emissions reductions through policy and financing solutions. Selecting a few options does not imply that they constitute the only solutions for decarbonisation. A wide range of options will be necessary to decarbonise these sectors in a holistic way, but only a subset will be covered by the implementation of the Framework.
2. The cases should focus on options which directly impact the manufacturing process, have a significant emission reduction potential and be consistent with South Africa’s strategy to reach net-zero emissions. The cases can include transition technologies that meet all these criteria.
3. The technologies should be technically implementable in the short to medium term (i.e., within 10 years). Thus, only technologies with a Technology Readiness Level (TRL) of at least 6-7, and/or with examples of industrial deployment at the international level, are considered. This can also help ensure that South Africa builds on the global trends in industry decarbonisation.
4. The cases should benefit from the policy and financing solutions that the OECD Framework will propose. Therefore, technologies that are already technically and economically viable and widely implemented are less relevant in the context of this project. On the contrary, disruptive technologies that are technically viable but struggle to reach economic viability corresponds well to the scope of the Framework.
TAC members suggested focusing on the steps of the steel value chain that generate most emissions, i.e., ironmaking and steelmaking, rather than downstream processes, such as hot and cold rolling. Furthermore, they proposed providing economic assessments that could apply to both existing and new assets. Given South Africa’s potential to produce renewable hydrogen at a globally competitive cost and the government’s recognition of this technology’s potential to decarbonise several hard-to-abate sectors, TAC members proposed prioritising options that could integrate hydrogen.
Based on these insights, the PSC selected three economic assessments for the Framework implementation:
Case 1 (hereinafter BF-(CCU)-BOF): Revamping of an existing blast furnace with carbon capture and use to produce methanol.
Case 2 (hereinafter H2-DRI-EAF): renewable hydrogen-based Direct Reduced Iron, coupled with an electric arc furnace.
Case 3 (hereinafter S-B EAF): improvement of scrap-based electric arc furnace with (i) scrap preheating, (ii) ultra-high-power transformers (leading to higher productivity and energy savings) and (iii) biomass use to replace coal, considering the use of invasive alien plants.3
In the following sections, the results for BF-(CCU)-EAF and H2-DRI-EAF are presented together, as both aim to decarbonise primary steel production, i.e., steel produced from iron ore. The results for S-B EAF, which aim to reduce emissions from secondary steel production, i.e., steel produced from scrap, are presented separately.
If renewable hydrogen has the potential to bring the direct emissions of ironmaking to near zero in H2-DRI-EAF, carbon capture in integrated steel plants could effectively decarbonise only a part of their operations (see Figure 2.5 and Figure 2.6). Even though they occur out of the boundaries of steel plants, it is also important to consider indirect emissions across the value chain when implementing projects.
Carbon capture rates could reach 95% as per industry claims (Global CCS Institute, 2024[26]), but most global carbon capture projects have reached a maximum rate of 80% (Transition Asia, 2024[27]). Moreover, even though most of the CO2 is generated in the BF, there are many emissions source points in an integrated steel plant (such as the sinter plant, the coke ovens, the BOF or the internal power plant where flue gasses are generally directed for combustion). Therefore, the carbon capture can be seen as a transition technology to reduce the emissions of existing polluting plants. Finally, for BF-(CCU)-BOF, the carbon captured is not stored but converted into methanol, which would eventually be combusted and released into the atmosphere, although the process prevents the production of methanol from fossil fuel sources.
In the case of H2-DRI-EAF, using renewable electricity is critical to limit indirect emissions. Yet, it is worth noting that while onshore wind and solar PV assets do not emit CO2 when they generate electricity, they have a carbon footprint due to the emissions that occurred to manufacture them.4 Even though the overall emissions across the value chain are around 95% lower than for coal-based steelmaking, it is essential to optimise the project design, capacity load of renewable assets and electrolysers, and to facilitate the recycling of these assets at the end of their lifetime to minimise the overall climate impact.
Note: The emissions of the BF-BOF route depicted here amount to 1.8 t CO2/t steel. It does not include the emissions related to the downstream part of the process (e.g. rolling mills) and is aligned on international benchmark rather than on the specific data of South African plants.
Note: indirect emissions DRI include the total emissions of renewable hydrogen production, mainly due to the embedded carbon footprint in the renewable power generation assets necessary to provide electricity to the project.
2.2.2. Methodology for the economic assessment
Calculations, indicators and main assumptions
An Excel tool has been developed by the OECD to carry out the economic assessments. It includes:
A list of technical, financial, economic and environmental parameters for each production route.
A financial module featuring simplified income statement and cash flows calculations.
A “cockpit” to allow a quick modification of selected technical and financial assumptions, pre-defined price scenarios for key parameters and assumptions of key levers and solutions to improve the business case of low-carbon options. These levers include concessional finance parameters, subsidies, economic instruments and market-based instruments. Their impact is presented in Chapter 3.
A result tab to display the main parameters, as well as the results of sensitivity analyses:
The Levelised Cost of Steel (LCOS), i.e., a measure of the average net present cost of steel production over a project’s lifetime, expressed in USD2025/t steel. The analysis of LCOS is presented in this Chapter.
The Net Present Value (NPV) of the projects, relative to benchmarks, expressed in million USD2025. The analysis of NPV is presented in Chapter 3.
For the ease of calculations, Case 1 (BF-(CCU)-BOF) and Case 2 (H2-DRI-EAF) are based on a plant with 2 Mtpa steel production capacity and Case 3 (S-B EAF) on a plant with 0.5 Mtpa steel production capacity, corresponding to the typical size of the corresponding world-class assets. The scope of each Case is described in Figure 2.7, Figure 2.8 and Figure 2.9. For Case 1 (BF-(CCU)-BOF) and Case 2 (H2-DRI-EAF), the tool has been used to compare the results with a benchmark case, the relining of a conventional BF-BOF plant. As Case 3 (S-B EAF) corresponds to improvements of an existing EAF, the calculations are compared to another benchmark case, the same EAF without these improvements.
Note: CCU: Carbon Capture and Usage; BOF: Basic Oxygen Furnace; CC: Continuous Casting; e-methanol: methanol synthesised from captured carbon dioxide together with hydrogen; HRM: Hot Rolling Mill.
Note: DRI: Direct Reduced Iron; EAF: Electric Arc Furnace; CC; Continuous Casting; HRM: Hot Rolling Mill.
Note: DRI: Direct Reduced Iron; EAF: Electric Arc Furnace; CC; Continuous Casting; HRM: Hot Rolling Mill.
As the analysis is forward-looking and considers investment projects being carried out in the short to mid-term, assumptions on key parameters value have been made until 2055. When possible, assumptions are based on historical, current or projected data specific to South Africa. When such information was not available or not considered robust enough, further data have been collected through a review of academic and grey literature and discussed with TAC members. The main assumptions considered to carry out the economic assessment are described in Table 2.1.
Note: CAPEX: capital expenditures; MeOH: methanol; Mtpa: million tonnes per annum. The assumptions have been discussed in Technical Advisory Committee (TAC) meetings.
Limitations of the model
While the model provides an estimate of the LCOS and NPV for the selected cases, this remains a high-level estimate rather than a calculation specific to a given project. Therefore, some parameters and results may vary in real-world projects compared to the base case presented here. The model focusses on the ironmaking and steelmaking steps of the process but does not account for elements that could be decisive for actual projects. For instance:
The downstream part of the processes is not modelled. Moreover, there is no differentiation of steel grades or considerations on the steel product portfolio.
The access to infrastructure or cost to build infrastructure is out of the model scope. For instance, the model does not account for transportation costs if renewable electricity, hydrogen and steel production assets are not co-located, nor for investment in port infrastructure that may be needed for export sales.
While the key assumptions and sensitivity analyses can capture a wide array of circumstances, some factors such as operational issues, difficulty to secure the raw materials (e.g. DR grade pellets, scrap, biomass), alternative options to optimise costs (e.g. opportunity cost to valorise the BF gas in a power plant instead of methanol production), or steel demand in the domestic and export markets are not modelled.
The global market dynamics of the steel sector, notably the potential impact of international trade on steel prices, are not considered in the model.
The model is based on a project finance approach. If the project is implemented through corporate finance, the financing conditions and the impact of some instruments (e.g. tax rebates) may vary.
Report on DRI plant economics in developing country
The Oxford Institute for Energy Studies released a report in 2024 on the “Financing a hydrogen DRI plant in a developing country”. The report underscores that key requirements for financing world-scale projects include a robust commercial structure with long-term contracts that essentially fix the green premium for the product and lock in supply of the specialised iron ore required and support from the governments of both the host and offtaker’s countries. Carbon pricing, along with protective trade measures such as carbon border adjustments, will be critical to ensure economic viability over the long run. Access to low-cost, long-term agency financing will also be key to the project’s success while a suite of mechanisms also exists to leverage public support in the host and offtaker’s countries.
The report also highlights potential impact on the iron and steel industry structure, as industry investors will seek to locate new plants in the lowest-cost locations. Thus, ironmaking plants are expected to be located in regions of lowest-cost renewable electricity (after full firming), and steel plants may either be co-located to utilise the low-cost power or located in end markets.
Tool to assess the Levelised Cost of Steel of various project archetypes
In a series of reports titled “Unlocking the first wave of breakthrough steel investments”, the Energy Transitions Commission presents the key conditions to develop a viable investment case for breakthrough primary steelmaking technologies in the United Kingdom, Southern Europe, France and the United States. The reports are accompanied by a financial tool to model the financial performance of breakthrough iron and steel project archetypes.
Tool to assess the Levelised Cost of Hydrogen and derivatives in various countries, including South Africa
The International PtX Hub, led by GIZ, has developed an Excel tool called the “Business Opportunity Analyzer” to support policymakers and project developers in understanding their country’s specific PtX cost potentials. The “Business Opportunity Analyzer” has a global scope and includes a deep dive analysis for a few countries, including South Africa.
The Excel tool enables users to calculate the trade costs of PtX products (e.g. green ammonia, e-methanol and synthetic fuels). This “pre-feasibility” level information will help policymakers and project developers identify regions with promising business opportunities within the PtX value chain, supporting the development of targeted policies to stimulate investment.
Note: Power-to-X (PtX) is the process of converting renewable electricity (power) into various end products (“X”).
2.2.3. Economic assessment for BF-(CCU)-BOF and H2-DRI-EAF
The calculation of the Levelised Cost of Steel (LCOS) shows that both BF-(CCU)-BOF and H2-DRI-EAF technologies are more expensive than the conventional BF-BOF production route. In the base case scenario, the LCOS of BF-(CCU)-BOF is 26% higher than BF-BOF, and the H2-DRI-EAF route is 42% more expensive than BF-BOF (see Figure 2.10). This hints that both technologies require specific support or market solutions to become economically viable.
The breakdown of the LCOS highlights the main drivers of the technology costs for each production route (see Figure 2.11). BF-(CCU)-BOF has a relatively similar cost structure as the benchmark BF-BOF, however with higher capital expenditures (CAPEX) due to the carbon capture assets, and higher operations costs driven by the additional energy costs of the carbon capture. H2-DRI-EAF has a very different cost structure, where renewable hydrogen represents more than 30% of the LCOS. While renewable hydrogen costs are presented as an external input, they are mainly driven by the CAPEX costs of renewable power generation assets and electrolysers. Thus, in the case of a vertically integrated project,5 CAPEX would represent more than a third of the LCOS of H2-DRI-EAF. This indicates that a cost reduction of the renewable power and electrolyser technologies and DR plant, and/or a reduction of the cost of capital, could play a pivotal role in improving the competitiveness of this production route.
Note: (i) for BF-(CCU)-BOF, the costs presented in the figure include carbon capture but do not include the production costs specific to methanol production; (ii) for H2-DRI-EAF, the costs related to renewable hydrogen are driven by electricity costs, that represent around two-thirds of the renewable hydrogen production costs.
The LCOS is highly dependent on the raw materials, energy and CO2 price. Figure 2.12 depicts the results of a first-level sensitivity analysis, estimating the LCOS of BF-BOF, BF-(CCU)-BOF and H2-DRI-EAF by changing individual parameters with alternative price assumptions described in Table 2.1. As the scrap intake of the three production routes are relatively similar in the base case, an increase of the scrap price does not modify much their comparative LCOS. If renewable hydrogen and renewable electricity prices are held constant during the entire lifetime of the project (which reflects the typical case of a vertically integrated project, or projects with long-term power purchase and hydrogen purchase agreements), the LCOS of H2-DRI-EAF increases significantly, whereas the LCOS of other production routes remains rather constant. Lastly, the alternative scenario with a gradual increase of CO2 price benefits the H2-DRI-EAF route the most. As the CO2 emissions of the BF-BOF benchmark case is below the reported emissions of the two blast furnaces in operation in South Africa, the cost increase could increase further for the BF-BOF and BF-(CCU)-BOF cases. While this indicates that carbon pricing could be instrumental to improving the competitiveness of H2-DRI-EAF, this scenario would lead to a 10% increase in the minimum steel production price in the country, hampering the competitiveness of the domestic steel industry.
Granular sensitivity analyses are presented in Figure 2.13 on four key parameters that influence the relative competitiveness of the three production routes: renewable hydrogen price (Panel A), CO2 price (Panel B), CAPEX price (Panel C) and the scrap share in the H2-DRI-EAF route (Panel D):
Panel A shows the LCOS for different levels of renewable hydrogen price, assuming in each scenario that the renewable hydrogen price remains constant during the entire project lifetime. It shows that the LCOS of H2-DRI-EAF is 62% higher than the LCOS of BF-BOF for a renewable hydrogen price of USD 5/kg, and 27% for a renewable hydrogen price of USD 2/kg. As no project of renewable hydrogen production worldwide has yet confirmed production costs below USD 3/kg, H2-DRI-EAF would require targeted incentives to be cost competitive with BF-BOF.
Panel B highlights that the “sweet spot” for BF-(CCU)-BOF and H2-DRI-EAF to compete with BF-BOF ranges in USD 100-150/t CO2. As the CO2 emissions of the BF-BOF benchmark is below the reported emissions of the two blast furnaces in operation in South Africa, the carbon price enabling the low-carbon technologies to achieve a similar LCOS as BF-BOF could be lower, if the current assets do not reduce their emissions to align with international standards. This price range correspond to the maximum observed carbon price applied for the industry sector.
Panel C underscores the risk faced by the selected low-carbon technologies in case the project CAPEX increases, which may occur due to tight market conditions, limited availability of technology suppliers, or schedule and budget deviations during project implementation. This is particularly true for a vertically integrated H2-DRI-EAF project, which has the higher capital intensity among all the proposed options. For this production route, a 30% CAPEX increase compared to the base case leads to a 12% increase in the LCOS.
Panel D reflects a potential optimisation of the H2-DRI-EAF case, compared to BF-(CCU)-BOF and BF-BOF. Indeed, while the scrap share in BOF is capped at around 25% for technical reasons, EAF can use a flexible intake of DRI and scrap intake, from 100% DRI to 100% scrap. This provides the H2-DRI-EAF route with the possibility to optimise its costs based on the price and availability of DR pellets and scrap, the production costs of DRI, the steel grades to be produced and the capacity factors of various assets. While scrap costs are cheaper than hydrogen-based DRI, it is assumed that a minimum 30% DRI intake would be necessary given the limitations of quality and volume of scrap. While the LCOS of H2-DRI-EAF remains higher than the LCOS BF-BOF in the sensitivity analysis, it reaches cost parity with BF-(CCU)-BOF for scrap shares between 50% and 70%, depending on whether the capacity of the DR plant is reduced to adjust to the higher scrap share in the EAF.
Note: in Panel C and D, the subcase H2-DRI-EAF (plant boundaries only) corresponds to the case where renewable hydrogen and electricity are sourced externally and decreasing over time; the subcase H2-DRI-EAF (incl. RE+H2 assets) corresponds to a vertically integrated project. In panel D, the H2-DRI-EAF (incl. CAPEX impact) considers that the DR plant capacity could be reduced given the higher scrap share in the EAF, which reduces the CAPEX and production costs (in particular hydrogen) for the whole project.
2.2.4. Economic assessment for scrap-based EAF
Many technologies are available or are emerging to reduce the greenhouse gas (GHG) emissions of a scrap-based EAF. The economic assessment in this section focuses on three of these options, namely (i) scrap preheating use, (ii) ultra-high-power transformers and (iii) biomass use to replace coal, considering the use of invasive alien plants (IAP). The main assumptions considered for the economic assessment of these technologies are displayed in Table 2.2.
Note: CAPEX: capital expenditures; biomass costs are based on the delivered cost of Invasive Alien Plants. The estimated costs vary greatly according to the source, as the exact scope of investment and adjustments needed to integrate the technologies vary greatly from one source to another and may also differ in each steel plant.
Sources: Authors, based on (UNDP, 2017[31]; Hasanbeigi, Price and Arens, 2013[32]; Elango et al., 2023[33]; United States Environmental Protection Agency, 2012[34]; European Steel Technology Roadmap, 2021[35]; Salimbeni et al., 2023[36]; Chireshe, 2021[37]; The Japan Iron and Steel Federation, 2023[38]; WWF, 2021[39]).
Scrap preheating and the use of ultra-high-power transformers can improve the LCOS compared to a reference EAF, and the use of biomass only shows a limited LCOS increase (see Figure 2.14). This demonstrates that, as highlighted in global and national pathways, energy efficiency and incremental improvements on existing assets can already be economically viable. The sensitivity analysis in Figure 2.15 shows that the conclusions remain similar when the CAPEX of the project varies according to the range of values available in the literature. In the case of biomass, the supply price may greatly differ according to the quantity of the resource, as well as its energy content and the collection, sorting and transportation costs. Under the conditions of this simplified economic assessment, securing a stable supply of biomass of around USD 50/t is essential to reach economic profitability.
2.2.5. Conclusion of the economic assessment and sensitivity analyses
The economic assessment highlights that several incremental upgrades of EAF technologies can be economically viable, but that BF-(CCU)-BOF and H2-DRI-EAF are respectively 26% and 42% more expensive than the benchmark BF-BOF route, in a base case scenario. This viability gap is mainly due to high capital costs, insufficient access to low-cost renewable electricity, a lack of market incentives for low-emissions steel and low carbon prices.
The sensitivity analysis shows that variations of key parameters such as CAPEX or raw materials and energy prices in realistic ranges in the context of South Africa do not change the conclusions of the base case scenario. This hints that improving several of these parameters through tailored economic and financial support measures will be necessary to make the selected low-carbon technologies attractive for investors.
References
[1] Agora Industry, Wuppertal Institute and Lund University (2021), Global Steel at Crossroads, https://www.agora-energiewende.org/fileadmin/Projekte/2021/2021-06_IND_INT_GlobalSteel/A-EW_236_Global-Steel-at-a-Crossroads_WEB_V2.pdf.
[20] ArcelorMittal South Africa (2025), ArcelorMittal South Africa Navigates Challenging Market: Strategic Alternatives Explored for Long-Term Sustainability, https://www.arcelormittalsa.com/Portals/0/News%20release%20-%20ArcelorMittal%20South%20Africa%20Navigates%20Challenging%20Market%20-%20Strategic%20Alternatives%20Explored%20for%20Long-Term%20Sustainability.pdf (accessed on 6 August 2025).
[19] ArcelorMittal South Africa (2024), ArcelorMittal South Africa: Media Site visit presentation, https://arcelormittalsa.com/Portals/0/ArcelorMittal%20South%20Africa%20Media%20site%20visit%20presentation.pdf (accessed on 18 June 2025).
[18] ArcelorMittal South Africa (2023), Decarbonisation roadmap, https://arcelormittalsa.com/Portals/0/ArcelorMittal%20South%20Africa%20Decarbonisation%20Roadmap%20(January%202023)%20(2).pdf.
[37] Chireshe, T. (2021), Technical and economic pre-feasibility of biomass waste utilization for production of sustainable aviation fuel.
[12] Climate Group (2025), The role of scrap in steel decarbonisation: Key facts and considerations for the construction sector, https://www.theclimategroup.org/sites/default/files/2025-05/The-role-of-scrap-steel-in-decarbonisation-April-2025.pdf.
[16] de Villafranca Casas, M. et al. (2024), Climate targets by major steel companies: An assessment of collective ambition and planned emission reduction measures, https://www.sciencedirect.com/science/article/pii/S2666278723000272.
[25] Department of Trade, Industry and Competition (2022), Proposed South African Green Hydrogen (GH2) Commercialisation Strategy, https://www.thedtic.gov.za/wp-content/uploads/Powerpoint-Summary-Green-Hydrogen-Commercialisation-Strategy.pdf.
[33] Elango, S. et al. (2023), Evaluating Net-zero for the Indian Steel Industry: Marginal Abatement Cost Curves of Carbon Mitigation Technologies, https://www.ceew.in/sites/default/files/How-Can-India-Decarbonise-For-Net-Zero-Sustainable-Steel-Production-Industry.pdf.
[28] Energy Transitions Commission (2023), Unlocking the First Wave of Breakthrough Steel Investments, https://www.energy-transitions.org/publications/breakthrough-steel/.
[35] European Steel Technology Roadmap (2021), Improve the EAF scrap route for a sustainable value chain in the EU Circular Economy scenario, https://www.estep.eu/assets/Publications/Improve-the-EAF-scrap-route-Roadmap-Final-V2-3.pdf.
[7] Global CCS Institute (2024), Collaborating for a Net-Zero Future, https://www.globalccsinstitute.com/wp-content/uploads/2024/11/Global-Status-Report-6-November.pdf.
[26] Global CCS Institute (2024), State of the Art: CCS Technologies 2024, https://www.globalccsinstitute.com/resources/publications-reports-research/technologycompendium2024/.
[23] Government of South Africa (2025), South African Renewable Energy Masterplan, https://sawea.org.za/sites/default/files/content-files/Documents%20%26%20Presentations/2025/SAREM_2025_LAYOUT_DIGITAL.pdf.
[32] Hasanbeigi, A., L. Price and M. Arens (2013), Emerging Energy-efficiency and Carbon Dioxide Emissions-reduction Technologies for the Iron and Steel Industry, https://eta-publications.lbl.gov/sites/default/files/6106e-steel-tech.pdf.
[4] IEA (2024), The Breakthrough Agenda Report, https://iea.blob.core.windows.net/assets/6633c3f9-2fdc-473b-a472-e2e91cba85ba/BreakthroughAgendaReport2024.pdf.
[5] IEA (2023), Net Zero Roadmap: A Global Pathway to Keep the 1.5 °C Goal in Reach, https://www.iea.org/reports/net-zero-roadmap-a-global-pathway-to-keep-the-15-0c-goal-in-reach.
[13] IEA (2021), Net Zero by 2050: A Roadmap for the Global Energy Sector, https://iea.blob.core.windows.net/assets/deebef5d-0c34-4539-9d0c-10b13d840027/NetZeroby2050-ARoadmapfortheGlobalEnergySector_CORR.pdf.
[2] IEA (2020), Iron and Steel Technology Roadmap, http://oecd.org/content/dam/oecd/en/publications/reports/2020/10/iron-and-steel-technology-roadmap_040b14d5/3dcc2a1b-en.pdf.
[30] International PtX Hub (2023), Tool: Business Opportunity Analyzer, https://ptx-hub.org/publication/tool-business-opportunity-analyzer/.
[24] IRENA (2025), Renewable Capacity Statistics, https://www.irena.org/Publications/2025/Mar/Renewable-capacity-statistics-2025.
[22] IRENA (2020), Renewable Energy Prospects: South Africa, https://www.irena.org/-/media/Files/IRENA/Agency/Publication/2020/Jun/IRENA_REmap_South_Africa_report_2020.pdf.
[10] LeadIT (2025), Green Steel Tracker, https://www.industrytransition.org/green-steel-tracker/.
[6] Mission Possible Partnership (2022), Making Net-Zero Steel Possible: An industry-backed, 1.5°C-aligned transition strategy, https://3stepsolutions.s3-accelerate.amazonaws.com/assets/custom/010856/downloads/Making-Net-Zero-Steel-possible.pdf.
[17] National Business Initiative (2023), Decarbonising South Africa’s Heavy Manufacturing Sector, https://www.nbi.org.za/report/decarbonising-south-africas-heavy-manufacturing-sector/.
[3] OECD (2024), “Addressing steel decarbonisation challenges for industry and policy”, OECD Science, Technology and Industry Policy Papers, No. 171, OECD Publishing, Paris, https://doi.org/10.1787/e6cb2f3c-en.
[11] OECD (2024), “Unlocking potential in the global scrap steel market: Opportunities and challenges”, OECD Science, Technology and Industry Policy Papers, No. 170, OECD Publishing, Paris, https://doi.org/10.1787/d7557242-en.
[8] OECD (2019), Low and zero emissions in the steel and cement industries: Barries, technologies and policies, https://portal.oecd.org/eshare/env/pc/Deliverables/CEFIM/Cross-cutting/Decarbonising%20industry/South%20Africa/Literature/GGSD2019_Steel%20and%20Cemement_Final.pdf.
[36] Salimbeni, A. et al. (2023), “Techno-Economic feasibility of integrating biomass slow pyrolysis in an EAF steelmaking site: A case study”, Applied Energy, Vol. 339, https://doi.org/10.1016/j.apenergy.2023.120991.
[40] Schlömer, S. et al. (2014), 2014: Annex III: Technology-specific cost and performance parameters, https://www.ipcc.ch/site/assets/uploads/2018/02/ipcc_wg3_ar5_annex-iii.pdf.
[38] The Japan Iron and Steel Federation (2023), ASEAN Technologies Customized List 2023 version Part-1: EAF (v.4.0), https://www.jisf.or.jp/en/activity/climate/Technologies/documents/TCL_ASEAN_EAF_ver4.0_2023.pdf.
[29] The Oxford Institute for Energy Studies (2024), Ironing out the way for Green Steel: Financing a Green Hydrogen-based Ironmaking Plant in an Emerging Market, https://www.oxfordenergy.org/wpcms/wp-content/uploads/2024/11/CM10-Financing-a-green-hydrogen-based-ironmaking-plant-in-an-emerging-market.pdf.
[27] Transition Asia (2024), Explainer – Carbon Capture in the Steel Sector; BF-BOF abatement, https://transitionasia.org/explainer-carbon-capture-in-the-steel-sector/.
[31] UNDP (2017), Energy efficient technology packages for electric arc furnace, https://sameeeksha.org/books/Electric-Arc-Furnace.pdf.
[34] United States Environmental Protection Agency (2012), Available and Emerging Technologies for Reducing Greenhouse Gas Emissions from the Iron and Steel Industry, https://www.epa.gov/sites/default/files/2015-12/documents/ironsteel.pdf.
[21] World Steel Association (2021), Energy use in the steel industry, https://worldsteel.org/wp-content/uploads/Fact-sheet-Energy-use-in-the-steel-industry.pdf.
[14] World Steel Association (n.d.), Circular Economy, https://worldsteel.org/wider-sustainability/circular-economy/.
[15] World Steel Association (n.d.), Energy use in the steel industry, https://worldsteel.org/wp-content/uploads/Fact-sheet-Energy-use-in-the-steel-industry.pdf.
[39] WWF (2021), Full pre-feasibility of ethanol supply for production of sustainable aviation fuel.
[9] Yu, S. et al. (2021), Decarbonizing the steel sector in Paris-compatible pathways, E3G, https://www.e3g.org/wp-content/uploads/1.5C-Steel-Report_E3G-PNNL-1.pdf.
← 2. In terms of percentage of companies mentioning CCUS in their decarbonisation roadmaps.
← 3. There exist diverse alternative biomass feedstock options in the country. Invasive Alien Plants have been selected for this study as they are a substantial candidate resource and because using them in EAF could contribute to the country’s objective to eradicate invasive plant species. Broader considerations on the sourcing of sustainable biomass are described in Chapter 3.
← 4. The median lifecycle emissions of renewable electricity sources range between 10-50 gCO2e/kWh, compared to 490 gCO2e/kWh for electricity produced from natural gas (combined cycle) and 820 gCO2e/kWh for electricity produced from coal (Schlömer et al., 2014[40]).
← 5. In this report, a vertically integrated H2-DRI-EAF project corresponds to the situation where the project developer investment envelope includes the power renewable assets, the hydrogen production plant and the steel plant.