Why could steel industry de-carbonisation be long and difficult?

News Analysis




Why could steel industry de-carbonisation be long and difficult?

This Project Blue Opinion is the first of several short publications which will look at steel industry de-carbonisation. This following paragraphs discuss the challenges faced by the industry for iron making, especially when it comes to replacing blast furnaces with electric arc furnaces. In future Opinion pieces, we will discuss the most likely scenarios for the future, with their technical, economical, and political implications.

Over the past decades, the global steel industry has been driven by China’s industrialisation and urbanisation. China’s steel production rose from 130Mt in 2000 to 1,065Mt in 2020, a staggering 11% CAGR, with China’s market share increasing from 15% to 57%.

China’s economy was primarily focussed on fixed assets investments, with large infrastructure developments.  To support this, China’s steel industry was built upon the blast furnace/basic oxygen route (BF/BOF), which provides large volumes of production and economies of scale.

China’s steel industry has now peaked and with climate change becoming a major global narrative, the steel industry’s de-carbonisation will become the main feature of the sector for the decades to come. The steel industry accounts for about 7% of total carbon emissions globally and 15% in China, levels which must be cut to meet the targets set by various regulatory authorities.

Steel is produced through the BF/BOF route by reducing iron ore with coking coal. After melting, the liquid iron is refined in a basic oxygen furnace (BOF) to finalise the steelmaking process.  The other main steelmaking option is using an Electric Arc Furnace (EAF) which melts scrap (recycled steel) through an electric arc that arises when electrodes contact the metal. The BF/BOF route accounts for about 70% of the world steel production with wide differences between countries. When it comes to emissions, an average BF generates about 2t of CO2 per tonne of steel compared with 0.5t for an only-scrap EAF. 

With such a difference, why not replace BFs with EAFs?

The first reason is economics. Integrated steel mills are capital intensive with an estimated capex of US$1Bn per tonne per year of steel produced. A BF/BOF lifetime can well exceed 50 years with relining taking place every 15-20 years. A furnace relining cost would range US$50-200M depending on the location.

The average age of BFs in service also varies from 40-55 years in Europe, Japan, and North America to less than 15 years in China. A gradual replacement of older BFs is a possibility in developed economies (although several mills keep extending the life of their existing facilities), but the BFs operating in China will continue to produce for several decades. This is also the case in South Korea and in India, with both countries operating more recently commissioned BFs.

The second reason is that size matters.  Large blast furnaces can produce up to 4Mtpy of steel while the average EAF output is 1.5Mtpy. When a country is in its development phase, a steel industry based on BF/BOF allows larger volumes of production and economies of scale to respond to expanding demand, as has been the case in China. The BF/BOF route also allows the production of all types of steel. India is currently expanding its steel industry and more BF/BOFs will be built. Replacing blast furnaces with EAFs would implicitly translate into a downsizing of a local steel industry.

A third reason is steel quality. Scrap (recycled steel) contains many impurities such as copper or tin which cannot be removed in a regular shredding and magnetic separation process and using only scrap in an EAF does not allow it to produce the steel quality required for certain products such as automotive sheets. This is why EAFs produce mainly long steel products, while most flat products are produced through the BF/BOF route.

A fourth reason is scrap availability. Besides the quality issue above mentioned, there is not enough scrap on a global basis.  Most of it comes ‘obsolete scrap’ which is increasingly generated as an economy matures. Developed economies can generate large quantities of scrap metal, but this is not the case for countries such as China. The current scrap-to-steel ratio (e.g. the estimated scrap consumption divided by the crude steel production) is estimated at 38% with large disparities between North America where the ratio is close to 75% and China where it is less than 20%. Scrap availability will only rise gradually over the coming years but the scrap-to-steel ratio is forecast to remain below 50% until 2050.

A fifth reason is DRI availability. Direct Reduced Iron (DRI) is a solid metallised form of iron ore produced by a DRI plant that operates at a temperature below the melting point of the feed ore. A DRI plant does not use coking coal as a reducing agent, but carbon monoxide and hydrogen derived from natural gas or coal. DRI is mixed with scrap in the EAF to enhance the metallics and allows for a higher quality of steel products. However, the direct reduction process does not remove the gangue present in the ore (mostly alumina Al2O3 and silica SIO2) and the melting generates high slag volumes which can cause significant iron losses. To limit this loss, the DRI material must have a high Fe content (at or above 67% Fe) and less than 2.5% of gangue impurities (combined Al2O3 and SIO2).

The problem is that the amount of iron ore suitable for DRI production is limited. The usual DRI feed are DR-grade pellets or lump ore but there is a lack of pelletizing-grade iron ore concentrates suitable to produce DR pellets. Beneficiation of low-grade ore is a possibility but means large volumes of mined material with a low mass recovery. Some Direct Shipped Ores (DSO) with high Fe content of 60-62% mined in the Pilbara region of Australia could be beneficiated but often display a low magnetism which makes the beneficiation process less effective and more costly. In 2022, total DRI production was 123Mt, a volume by far insufficient to meet the potential EAFs demand.

In Europe, announced DRI projects total 40-50Mt by 2030 to achieve the EU de-carbonization targets, aiming at lowering emissions by at least 30% from their 2018–19 levels. Such a target is ambitious given the EU’s 2022 DRI production of 0.6Mt. What will be the reducing agent is also unclear, and assuming that it would be hydrogen, the source of power necessary for the hydrogen production will determine the future level of carbon emissions.

On average, an EAF-DRI mill still generates 1.5t of CO2 per tonne of steel, because the reducing agent for DRI is coal (mostly in India and in China) or natural gas, which are both fossil fuels. A full de-carbonisation would imply using ‘green’ hydrogen as a reducing agent, e.g., produced through a renewable source of energy. Even so, such a plant would not be able to produce all types of steel, in terms of quality, due to the scrap component.   

In summary, while replacing BFs with EAFs can contribute to steel industry de-carbonisation to some extent, economics, scale and product quality requirements, scrap availability and DRI availability means that this cannot happen everywhere. In general, EAFs lack the necessary flexibility to be the only solution to the steel industry de-carbonisation.

Both steelmakers and iron ore producers are currently engaged in finding suitable options, technical and economic, to cut emissions. Optimising and maximising existing operations should be the first step.  This will be discussed in our next Project Blue Opinion paper on the subject.