Actually this is a very nice basic question, i.e. when putting it a bit more general along the lines: ‘Does it generally (a) make sense and is it (b) thermodynamically possible to make steel using metallo-thermic redox reactions? Furthermore, and more specific, would the thermite reaction be an effective basis for making steel?

The concept of making (green) steel using metallo-thermic redox reactions, such as scenarios like the thermite reaction, is interesting but also a bit complex. To evaluate its thermodynamic feasibility and effectiveness, one has to look at the principles of metallo-thermic reduction (and the specific thermodynamics of the thermite reaction, viz, ur question).

Metallo-thermic redox reactions involve the reduction of metal oxides using a more reactive metal as the reducing agent. These reactions can be represented generally as Mx Oy + R - > M + Ry Oz , where Mx Oy is the metal oxide u wanna reduce (such as e.g., iron oxide), and R is a more reactive metal (e.g., aluminium such as for the thermite case).

The thermite reaction is thus only a specific type of metallo-thermic reaction (see many others in the corresponding Ellingham diagrams) where aluminum is used to reduce iron oxide to produce iron and aluminum oxide, i.e. Fe2 O3 + 2 Al -> 2 Fe + Al2 O3 .

This reaction is very exothermic (when u try it urself be VERY careful, do it outside, keep ur distance – it is dangerous when not done with protective care ….), providing a significant amount of heat with Delta H = -852 kJ/mol… that’s a lot of heat !

To assess the complete thermodynamic feasibility and generalize it, u of course need to consider the Gibbs free energy change Delta G of the reaction. For thermite the reaction is of course thermodynamically favorable as Delta G is negative.

However … let me pore some water into the nice drink…. While thermodynamically feasible, the practicality and effectiveness of using s t like a thermite reaction for (green) steel production hinges on many other factors too. The thermite reaction is highly exothermic and generates a lot of heat, which could be beneficial. However, controlling the reaction to produce steel of the desired quality and in large quantities is very very challenging. Aluminum, used as the reducing agent, is very expensive in itself and very energy-intensive to produce.

This could offset all the (cost, scaling, environmental etc) benefits unless a low price and very sustainable source of cheap (scrap… ??) aluminum is utilized. Also, the thermite reaction is difficult to control, as it tends to be highly localized and produces molten iron at very high temperatures, which can complicate the manufacturing process. In other words, we do not want more heat than is really needed. Additionally, the reaction produces aluminum oxide as a byproduct, which must be managed or recycled effectively.

Given the challenges mentioned, alternative methods for (green) steel production might be more viable, such as hydrogen-plasma based smelting reduction, hydrogen-based reduction (e g DRI) and more efficient blast furnace operations etc etc. Hydrogen-based reduction uses hydrogen instead of carbon to reduce iron ore, producing water as a byproduct. This method is promising as it significantly reduces greenhouse gas emissions. Even electrolysis is thinkable as it involves the direct reduction of iron ore using an electric current in molten oxide electrolysis cells (depends then on grid footprint of course).

Furthermore, while the thermite reaction is thermodynamically feasible (and used for welding etc) for reducing iron oxide to produce iron, its practical application for large-scale green steel production is in my humble view really limited by factors such as efficiency, sustainability, cost, process control, and byproduct management etc.

I attach a link with a few open access papers where some of these aspects are discuss or at least touched.

Good luck !

https://drive.google.com/drive/folders/1cW5JonOxvGEtbjZ_wW4LWXojK3rvUg53?usp=sharing

these papers, related this topics, are all open access:

https://pubs.acs.org/doi/full/10.1021/acs.chemrev.2c00799

https://www.nature.com/articles/s41586-019-1702-5

https://www.nature.com/articles/s41586-023-06901-z

https://www.sciencedirect.com/science/article/abs/pii/S135964542100313X

https://www.sciencedirect.com/science/article/abs/pii/S1359645421003517

https://www.sciencedirect.com/science/article/pii/S1359646222000720

https://www.nature.com/articles/s41529-023-00397-8

https://onlinelibrary.wiley.com/doi/full/10.1002/advs.202300111

https://arxiv.org/pdf/2201.13356

In theory, this is a good idea and it is indeed feasible, while there are many problems that need to be solved in the actual implementation process. For example, how to reduce the price of metallic aluminum, in terms of unit value, if using more expensive aluminum to reduce slightly cheaper iron is obviously uneconomical. In addition, intermetallic compounds such as FeAl, FeAl2, etc. are easily formed between Fe and Al. How to control these processes in actual production? These side reactions not only consume reducing agents but also affect the purity of the final product, as well as related reduction reactions of oxides of other metals in raw materials that are not pure enough, et al.