THE GREENING OF STEEL
April 16, 2021 § 5 Comments
Transportation has bad climate change related PR. All sectors combined (including aviation) account for about 13% of global CO2 production, whereas just steel and concrete add up to 15%. Estimates vary, but inescapable is the conclusion that we have not given steel and concrete the attention that we have heaped on transportation to mitigate CO2 production. To exacerbate matters, the world is on an infrastructure expansion spree, including more recently the Biden administration in the US. More infrastructure equates to more concrete and steel. That is more CO2 emissions. Unless we do something about it as we have with electric vehicles and hybrid vehicles.
Mitigating CO2 emissions from concrete and steel is more straightforward than from vehicles because they are what we refer to as point sources. Vehicle tailpipes are distributed, making capture, and disposition of the CO2, prohibitively difficult. Technically doable with pressure swing adsorption methods, but logistically tricky in release to regenerate the adsorbent and subsequent handling of the CO2. A decent analogy is NOx capture with urea, requiring canister replacement, a nuisance to many consumers. This difficulty led to alternative non-intrusive means such as the Lean NOx Trap, with the attendant VW deception.
First a bit of a primer on iron and steel making. Iron ore is largely iron oxide and must be reduced to iron. This is accomplished primarily in blast furnaces, which are shaft furnaces where the reactants are fed at the top and the metal is taken out of the bottom. The iron oxides are reduced by gases produced from coke, which is a derivative of coal. The reaction products include iron and CO2. The iron is then converted to steel by reducing the carbon content and by addition of other alloying elements for properties such as strength and corrosion resistance. Each metric ton (tonne) of steel produces a staggering 1.8 tonnes of CO2.
The Direct Reduction Iron (DRI) process is a means for reducing the carbon footprint. The process temperatures are low, and the iron never in a molten state. The reducing agent is syngas, a mixture of CO and H2. The combination reduces the emissions to 0.6 tonnes CO2 per tonne steel. In a variant, hydrogen alone is the reducing agent, and in a further green variant, the hydrogen is from renewable sources such as electrolysis of water using renewable electricity. However, unlike in the blast furnace process, there is no mechanism for removal of impurities in the ore. Consequently, only high-grade iron ore is tolerated, and this limits DRI to about 7% of the total market because such ore is in relatively short supply and much more costly.
The most promising route to the greening of steel is through CO2 capture at the blast furnace. Unlike flue gases from a power plant, blast furnace flue gas is concentrated, typically 30% CO2. As a result, removal processes are more effective. Today we are on the brink of capture costs below USD 40 per tonne CO2. Carbon credits may be purchased in Europe for about USD 55 per tonne. A recent New York Times story suggests that this will keep rising, with one analyst predicting prices above USD 150. If a major CO2 producer such as steel or cement is forced to buy credits, the price is certain to go up. When the capture cost is below the price for credits, the industry has an incentive to simply collect the gas. However, merely capturing accomplishes little if the gas is not permanently sequestered in what are known as sinks.
One such sink is subsurface storage in oil and gas reservoirs depleted of the original fluid, or in saline aquifers. While feasible, often with costs lowered by using abandoned wells, debate centers on permanence of the storage and the risk of induced seismicity (earthquakes). A variant with an important distinction is injection into reactive minerals such as basalt, with the formation of a non-water-soluble carbonate, which certainly is permanent. However, these wells are more costly because existing abandoned wells are unlikely to be in locations with suitable mineralogy. The exception to that would be abandoned geothermal wells, which could be proximal to igneous rock from the basalt family. However, there are not too many of those, and they are geographically constrained.
Mineralization as a genre is being pursued vigorously, with systems already commercial, although the tonnage being sequestered is still low. Done on the surface in reactors, the resulting carbonate of Na, Ca or Mg can have uses. Monetization even at small profit still renders the capture cost effective. Since, in my opinion, capture costs are heading in the right direction, and already at acceptable numbers, the focus ought to shift to sinks with scalability. Scalability is usefully defined as an aspirational goal of 0.5 gigatonnes CO2 per year by 2040. But goals short of that are fine if several approaches are proven viable.
Endeavors to achieve these goals could be materially assisted by appropriate policy action by the various federal governments. All forms of renewable energy have received subsidies or loan guarantees at some stage in their development. This has resulted in wind and solar being an established part of the electricity portfolio. Similarly, electric vehicles have received subsidy support. The greening of steel and cement ought to receive the same attention. For example, the Biden administration’s infrastructure bill ought to include provisions for preferential purchase of green steel and cement, at premium pricing.
Technology is approaching a tipping point for serious inroads into making steel and concrete green *. Public policy must keep pace.
*For the times they are a changin’ from “The Times They Are a-Changin’” performed and written by Bob Dylan, 1964
Vikram Rao
April 16, 2021
Research Triangle Energy Consortium 
Very interesting thoughts, Vik. Thanks for posting. I’d like to learn more about the ore grade limitations for the direct reduced iron process. Could you point to any relevant resources that you’re aware of?
This information is not easy to find, for some reason. However, Midrex, located in Charlotte, and the designer of the leading version of DRI, has a publication: https://www.midrex.com/wp-content/uploads/Midrex-2019-DFM4QTR-Final.pdf . Table 1 shows the limitations of all constituents, including the gangue.
Terrific, thanks!
Thanks for pointing out this important problem, and offering the best available solution(s) for it, Vikram. I also watched the YouTube video in which you and your son talked about his.
As the solution needs the attention of people with public policy responsibilities, if you haven’t done this already, I suggest conveying these info to such people (in Biden administration), climate advocates like Greta Thunberg, and the advocates in Congress (e.g., AOC). I thought about posting comments about these to their social media accounts. But those are so busy that they may never see them. Hopefully, you have more direct connections to such people, through your network, RTEC or RTI.
I have approached someone who influences some of these folks. Let’s see.