Hydrogen use in Steel: Tata Steel, the Netherlands
Following the Dutch Constitutional Climate decision in 2015 [i], the emissions reduction target in the Netherlands was set at 49% by 2030.
For the industry, this meant the elimination of an annual 14 million tonnes of CO2 by 2030.
To reach such an ambitious self-imposed but legally binding emission reduction, the Dutch government implemented two complementary measures to create the framework to enable industrial decarbonization: a carbon tax and a feed-in contractual subsidy mechanism (SDE++). The former will reach 125 €/tCO2 by 2030, providing a “stick” to reduce the market advantage of non-sustainable products. The latter will work as a “carrot”, subsidising the most cost-efficient CO2 reductions in industry and pushing it to act on the low-hanging fruits.
Tata Steel in Ijmuiden is the biggest steel producer in the Netherlands, producing 6.62 million tonnes of steel per year. Today the steel is produced burning coal in a blast furnace, resulting in 12.31 million tonnes of CO2eq per year. These emissions include both the direct emissions from the plant and the emissions resulting from the electricity production from the off-gasses. Almost 10 thousand people are employed at the steel mill [ii]. Thus, finding a solution to decarbonise the plant and avoiding the production to shift to other countries is crucial both to avoid carbon leakage and protect workers.
Direct reduction of iron ore with hydrogen
DRI is a manufacturing process to produce steel using hydrogen to reduce iron instead of coking coal. Hydrogen is used as the main reduction agent in the process of making sponge iron through the direct reduction of iron ore in a shaft furnace. The resulting sponge iron is then used in the production of steel in an electric arc furnace.
This process has the potential to decrease the emissions of steel to close to zero, but only under the condition that the hydrogen going into the process has a low carbon footprint. Hybrit, the first pilot direct reduction plant, was completed in 2020 in Luleå (Sweden) showing that this technology is technically possible and can become economically viable in the long term [iii]. A blast furnace cannot be reconverted into a Hydrogen DRI system, since that conversion requires complete rebuilding of the plant. However, conversion from natural gas to hydrogen is possible.
There are both advantages and disadvantages to this shift to ironmaking with hydrogen. On one hand, the ease of the conversion ensures that existing plants running on fossil gas can be converted to hydrogen when it will be available. On the other hand, projects running on unabated natural gas risk being promoted as sustainable because they are “hydrogen-ready”.
DRI with green hydrogen
Converting the Tata steel plant in Ijmuiden to DRI fuelled with green hydrogen would require a massive deployment of renewable capacity, both to produce hydrogen and the electricity input used in the process. Overall, producing 6.62 million tonnes of steel using exclusively renewable hydrogen and electricity will require 21.2 TWh of renewable electricity [iv]. This is more than the total wind production in the Netherlands in 2020 [v].
One of the most recent and biggest wind farms in the Netherlands is Gemini. It has a nameplate capacity of 600 MW and generates 2.6 TWh of electricity per year [vi].
To produce the green hydrogen needed to decarbonise Dutch steel production, eight wind farms the size of one of the biggest existing wind farms in the Netherlands would be needed.
Currently, the deployment target of the Netherlands for 2040 is to achieve 11 GW of capacity by 2030 [vii]. Assuming an average production factor as the Gemini one, this would entail that the Dutch wind production will reach 47.6 TWh in 2030. An additional 44% of deployment would thus be needed to cover the hydrogen and electricity demand of the steel mill.
DRI with blue hydrogen
Blue hydrogen offers a less electricity-intensive alternative for hydrogen DRI. If one were to convert the same plant but fuelling it with blue hydrogen instead, the energy input required for the plant would be 1.3 Mt of natural gas and 1.4 TWh of renewable electricity [viii]. In this case, the emissions of the plant will be highly dependent on two factors: the methane leakage rate upstream and the CO2 capture rate in the hydrogen production process.
Blue hydrogen would need only 7% of the electricity input of green hydrogen, but to keep emissions low very high CO2 capture rate and very low methane leakage rate are needed.
Keeping the assumption of the methane leakage as above and assuming a capture rate of 90%, the final emissions of this process would be around 0.37 Mt of CO2eq [ix].
Producing low-carbon hydrogen is key
Hydrogen Direct Reduction Iron (DRI) requires hydrogen and electricity as energy inputs. In order for it to be truly low carbon, both of these ingredients must come with no associated emissions. There are currently two main avenues to produce hydrogen with such a low carbon footprint, the so-called blue and green hydrogen. However, both of these technologies bear risks.
Green hydrogen to be truly low carbon must be produced with additional renewable generation.
Electricity must thus come from a renewable source and hydrogen needs to have a carbon footprint as close as possible to 0 gCO2/gH2 [x].
Blue hydrogen must come with close to zero methane leakage and a very high capture rate to be able to comply with low carbon requirements.
What are the pros and cons of blue and green hydrogen?
Does not need carbon capture and storage infrastructure.
Potential lowest climate footprint of any technology. GHG emissions are very close to zero when using exclusively renewable electricity.
No extraction or use of fossil resources.
Very large deployment of renewable generation required
New additional renewables are required to prevent cannibalising renewables intended for grid decarbonisation
21 TWh of renewable electricity required is greater than all current Dutch wind generation
Competition with other uses of renewable electricity such as direct electrification of transport and domestic heating
Use of grid electricity to produce hydrogen would significantly increase overall emissions (read more here)
Contained electricity demand & reduced wait for green H2 at scale
GHG emissions are very low dependent on high capture rate, permanent storage, and low methane leakage
May switch to or blend with green hydrogen as it becomes available
Continue fossil gas extraction and use
Requires CO2 transport and storage infrastructure to be available
Requires very low upstream methane emissions
High CO2 capture rate required
‘Hydrogen-ready’ means the use of fossil gas
The Direct Reduction Iron (DRI) can be done with natural gas instead of hydrogen, this process is known as Midrex. Although it’s not very common in Europe, this process is employed widely in areas where natural gas supply is abundant and cheap, such as Northern Africa [xi]. As DRI is normally run with unabated gas, it results in large greenhouse gas emissions. The total emissions depend on the process itself, as well as the leakage rate of methane in the extraction and transport phases. For the Midrex process, a common gas DRI technology, emissions are around 1.1 – 1.2 kg CO2/kg steel [xii], thus lower than the more common blast furnace process, whose emissions lie between 1.6 and 2.0 kg CO2/kg steel. Blast furnaces cannot be retrofitted to work with the Midrex process, therefore newly built plants are needed if a technological change is envisaged.
What are the pros and cons of using fossil gas for steelmaking?
Allows for a cut in emissions when compared to current Blast Furnace, dependent on low methane leakage
Can be converted to hydrogen
Commercially available technology in use since 1970s
Not a net-zero technology. Steel produced is neither green nor low carbon
Continuation of natural gas extraction and use
Will need to be converted to low-carbon hydrogen to reach low/zero emissions
Risk locking in emitting technology if the hydrogen for the reconversion is not made available
Direct Reduction Iron (DRI) with Natural Gas at Ijmuiden
If the Tata steel plant in Ijmuiden was to be converted to a natural gas DRI plant, to produce its 6.62 million tonnes of steel a year the plant would need to be supplied with 1.3 Mt of natural gas as well as 1.4 TWh of electricity [xiii]. Assuming the published emissions for the Midrex process at 1.1 kg CO2/kg steel [xiv], this would entail 7.282 Mt of CO2e emissions per year.
Calculating the emissions directly, including emissions from natural gas consumption, CO2 equivalent emissions from fugitive methane [xv] and electricity consumed from the current Dutch electricity grid [xvi] results in emissions of 5.494 Mt of CO2e a year. The level of upstream emissions plays a big role in the climate effect of such a switch. Since methane has 28 times the global warming potential than CO2 [xvii], a relatively small increase in the methane leakage rate could change drastically the climate impact of such a project.
How do the different technologies compare in their resource use and emission reductions?
Energy consumption and emissions for Ijmuiden, current technologies versus potential replacement technologies, per tonne of steel.
References and footnotes
[i] Sabin Center for Climate Change Law. 2022. Urgenda Foundation v. State of the Netherlands. Available at: http://climatecasechart.com/climate-change-litigation/non-us-case/urgenda-foundation-v-kingdom-of-the-netherlands/
[ii] Tata Steel. 2020. Tata Steel in Europe Sustainability Report 2019/2020. Available at: https://www.tatasteeleurope.com/sites/default/files/TSE%20Sustainability%20report%202019-20%20(EN).pdf
[iii] The LKAB-SSAB process uses high fine iron ore concentrate containing more than 71 per cent iron resulting in higher efficiency and lower energy requirements than with more traditional iron ores.
[iv] Own calculations using data from Kushnir et al. (2020), available at: https://www.sciencedirect.com/science/article/pii/S0959652619330550
[v] 15.3 TWh, IEA data
[vi] Gemini. 2022. About Gemini Wind Park. Available at: https://www.geminiwindpark.nl/about-gemini-wind-park.html
[vii] Government of the Netherlands. 2022. Offshore Wind Energy. Available at: https://www.government.nl/topics/renewable-energy/offshore-wind-energy
[viii] Own calculations using data from Al-Qahtani et al. (2021), available at: https://www.sciencedirect.com/science/article/pii/S0306261920314136
[ix] Own calculations using data from Al-Qahtani et al. (2021), available at: https://www.sciencedirect.com/science/article/pii/S0306261920314136
[x] Currently the threshold in the Sustainable Finance Taxonomy is set at 3 gCO2/gH2.
[xi] Midrex. 2019. 2018 World Direct Reduction Statistics. Available at: https://www.midrex.com/wp-content/uploads/Midrex_STATSbookprint_2018Final-1.pdf
[xii] Midrex. 2020. Direct from Midrex: First Quarter 2020. Available at: https://www.midrex.com/wp-content/uploads/Midrex-2020-DFM1QTR-Final.pdf
[xiii] Own calculations using data from IIP (2022) for Midrex technology, available at: http://www.iipinetwork.org/wp-content/Ietd/content/direct-reduced-iron.html
[xiv] Midrex. 2020. Direct from Midrex: First Quarter 2020. Available at: https://www.midrex.com/wp-content/uploads/Midrex-2020-DFM1QTR-Final.pdf
[xv] Upstream methane leakage of 0.03% and Dutch transport leakage of 0.01% at GWP100
[xvi] 328gCO2/KWh in 2020, source EEA.
[xvii] The global warming potential is expressed in GWP 100. If GWP 20 is considered this figure increases to 84 times.