Key Findings
The integration of machine learning and quantum mechanics simulations has revealed thousands of previously overlooked chemical reactions that facilitate the conversion of carbon dioxide (CO2) into fuels and industrial chemicals. By expanding the reaction network from 152 to 9,389 individual reactions on a copper catalyst, researchers achieved a predicted CO2 conversion rate approximately 40 times higher than the original model. This expanded model correctly identified methanol and carbon monoxide as the major products, aligning with experimental data, whereas the smaller model incorrectly predicted formic acid.
Technical Details
The research team utilized quantum mechanics simulations to establish an initial database of 152 reactions. They then trained machine learning models to estimate energy barriers for additional reactions involving 105 chemical species on the catalyst surface. This automated computational approach allowed the team to identify transformations occurring in a single reaction step, vastly increasing the network’s scale. Experimental validation was conducted through collaborations with the Hindustan Petroleum Corporation Limited’s Green Research and Development Centre and the Agency for Science, Technology, and Research in Singapore.
Background & Context
CO2 recycling via hydrogenation typically relies on catalysts to react CO2 with hydrogen to form methanol. However, simulating every possible step using quantum mechanics is computationally prohibitive, leading researchers to traditionally focus on a limited set of assumed critical reactions. This simplification often omits vital pathways, leading to inaccurate predictions regarding product yields and conversion rates.
Significance & Outlook
The study uncovered a surprising mechanism where hydrogen is transferred to intermediate compounds as molecular H2 rather than as separate atoms, particularly when transferring to intermediates containing oxygen. This challenges conventional chemical assumptions and suggests that catalysts with stronger H2 interactions could enhance methanol production. Beyond CO2 hydrogenation, this computational framework—combining ML, quantum simulations, and reaction rate models—is applicable to other industrial processes, including nitrogen reduction and water splitting.
Source: https://www.sciencedaily.com/releases/2026/10/261013015758.htm
Get our weekly technology intelligence — free
Receive an infographic that lets you judge at a glance whether each field’s analysis report is worth reading.
Subscribe Free — Weekly Tech Intelligence
By subscribing, you’ll receive Troy-Technical’s weekly technology intelligence newsletter.
- Your email and selected fields are used only to deliver the newsletter.
- We never share your information with third parties.
- You can unsubscribe anytime via the link in each email.
See our Privacy Policy for details.
Takes about a minute · Unsubscribe anytime
