How much GHG emissions can be cut by new EAFs in steel production? 

As evidence of their climate action, a number of legacy coal-intensive steelmakers have announced or started the construction of electric arc furnaces (EAFs). This is the case of ArcelorMittal with an EAF currently under construction in Gijon (Spain) and another one recently re-announced for Dunkirk (France). In Austria, voestalpine is installing two EAFs (one in Donawitz and in Linz) scheduled to come online in 2027. In Japan, Nippon Steel has been granted about 1.5 billion EUR of subsidies from the government to build a new EAF in Yawata and resume and expand production of already existing EAFs in Hirohata and Shunan.

In very simple terms, an EAF is a huge box that takes iron as an input, melts it using a lot of electricity, and after some chemical adjustments outputs steel in liquid form – a function called steelmaking. EAFs have started to become common in steel production in the second half of the 20th century. Therefore, they are not new: according to the Global Energy Monitor, there are currently over 800 EAFs in operation in the world. Another 100 or so are under construction, and almost 200 new ones have been announced.

Because EAFs mostly run on electricity, they directly emit relatively small quantities of greenhouse gases (GHG). However, unlike integrated coal-based steel plants (BF-BOF) which can produce their own energy to do all the processing steps from raw materials to steel, EAFs just make steel. It means that an EAF alone cannot replace an integrated coal-based BF-BOF steel plant. To cover the same range of functions as a BF-BOF plant, an EAF needs to be complemented with two elements: a source of electricity, and a source of iron.

By the same token, accurately assessing the difference in climate impact between BF-BOF steel and EAF-made steel requires to consider not only the GHG emissions coming from the steel plant, but also from the installations that produce energy and iron, even if they are not located on the same site as the steel plant and do not belong to the company making the steel.

Regarding electricity, almost all existing EAFs simply run on power from the grid. Related GHG emissions are decreasing together with the relative decline of coal and growing share of renewable sources in electricity generation, but steelmakers have so far taken very few significant steps of their own to accelerate this general trend. EAFs newly announced or already under construction are not accompanied either by direct investment in clean electricity generation capacity or power purchase agreements despite the fact that EAFs consume massive amounts of electricity. On the energy side, the decarbonisation potential of EAFs is therefore largely dependent on how the GHG intensity of grid electricity evolves.

As for iron, EAFs are quite flexible, and while they are today mostly used to remelt and recycle steel, they can in fact take any or a combination of the following types, each coming with their specific climate footprint:

• recycled iron and steel (scrap), considered as a zero-GHG emissions iron input;
• pig iron made from iron ores in a coal-fed, CO2-intensive blast furnace (the BF part of BF-BOF plants);
• direct reduced iron also made from iron ores, but in a process different from the blast furnace – the direct reduction process – and able to run on other reducing materials than coal, in particular gas and hydrogen.

For a company seeking to decarbonise, scrap is obviously the number one choice, and most EAF announcements do also mention plans for higher scrap use. However, unless demand for steel falls drastically, steelmakers are likely to struggle to find scrap in sufficient quantities and/or of sufficient quality to meet volumes and product requirements. As a result, scrap inputs would need to be complemented with iron made from iron ores in an ironmaking process (blast furnace or direct reduction).

For legacy coal-intensive steelmakers, often operating more than one blast furnace, a relatively easy solution to this problem is to keep at least one blast furnace and use the resulting pig iron for their new EAF. On the other hand, because pig iron production is by far the most CO2-intensive part of the process chain from raw materials to steel (close to 1.4 tCO2e per tonne of crude steel), continued reliance on pig iron greatly limits the decarbonisation potential of EAFs.

An example is provided by the ResponsibleSteel-certified U.S. Steel Big River plant in Arkansas (USA), equipped with two EAFs. The plant reports a GHG intensity of 1.34 tCO2e per tonne of crude steel with a 57.3% scrap share, the remaining iron being pig iron transported from US Steel blast furnaces located elsewhere in the country. While this emissions level is much lower than for average BF-BOF steel (2.33 tCO2 per tonne of crude steel), it is significantly higher than for average scrap-based EAF steel (0.68 tCO2 per tonne of crude steel).

The gap between the GHG intensity of US Steel Big River-made steel and of the average scrap-based EAF steel also highlights that even though EAFs are used in both cases, the source of iron is a key determinant of actual overall climate impact of steel production.

This holds a lesson for the future EAFs that have been announced or are being built by legacy coal-intensive steelmakers: though such investments are welcome, without information on the iron feedstock of these EAFs, it is impossible to assess how much GHG emissions they will actually cut. Therefore, steelmakers should disclose this information, especially when their EAF projects are supported with hundreds of millions of euros of taxpayers’ money.

Commentary written by: Romain Su