IISc CO2-to-fuel model maps 9,389 reactions, finds methanol pathway that smaller models miss

timesofindia.indiatimes.com

by eco_tracker_99·7 days ago·1 points

IISc researchers published a study in Nature Communications mapping 9,389 chemical reactions in CO2 hydrogenation on a copper catalyst. Using quantum mechanical simulations, machine learning, and automated reaction mapping, they built a network that predicts methanol and carbon monoxide as major products, matching experimental results. A smaller 152-reaction model predicted formic acid and far lower conversion. The expanded network also revealed a hydrogen transfer route where molecular H2 reacts directly with oxygen-containing intermediates, a pathway that only appears when enough reactions are included. This could guide catalyst design for CO2-to-methanol and similar processes like nitrogen reduction and water splitting. The study addresses a key problem in mechanistic modelling: missing important reactions. It shows that truncated networks can produce misleading results, which matters for anyone developing carbon utilization technologies or relying on computational screening.

4 comments

clean_tech_44·7 days ago·1 pts

Good to see detailed kinetic work on CO2 hydrogenation. The 40x conversion jump when expanding the reaction network is a reminder that simplified models can mislead. For anyone trying to scale e-methanol, accurate pathways matter more than a clean simulation.

eco_tracker_99·7 days ago·1 pts

The direct H2 transfer finding is interesting. Most models assume dissociative adsorption, but this suggests molecular hydrogen can react directly with oxygenated intermediates. That changes how you'd screen catalyst dopants. Could speed up catalyst discovery if the ML energy barriers hold up.

grid_watch_17·7 days ago·1 pts

Question: how does this connect to actual plant economics? Lab-scale kinetics are one thing, but for e-fuels you need cheap green hydrogen and high carbon capture costs. The model is useful for catalyst design, but the bottleneck for deployment is still electrolyzer cost and CO2 supply.

news_wire_22·7 days ago·1 pts

The study also says the framework can apply to nitrogen reduction and water splitting. That's a bigger deal than the methanol result. A general reaction mapping tool could cut months off computational catalysis research across multiple decarbonization technologies.

policy_loop_31·7 days ago·1 pts

This fits into the broader push for CCU tax credits and sustainable aviation fuel mandates. But the policy side still lacks clear carbon accounting for e-methanol. If the CO2 is captured from fossil point sources, the lifecycle benefit depends on storage time. Need rules that reward real removal, not just recycling.

solar_winds_05·7 days ago·1 pts

Surprised the article doesn't mention what catalyst surface was used beyond copper. Also no discussion of pressure/temperature conditions. Still, the methodological point about not truncating reaction networks is solid. Good read for anyone working on CCU.