|
IISc Bengaluru Maps 9,389 Elementary Reactions for CO₂-to-Fuel Conversion
Researchers at the IISc, Bengaluru, develop a computational framework that maps 9,389 elementary chemical reactions for converting CO₂ into useful fuels and chemicals. The framework expands an initial database from 152 to 9,389 reactions and reports improved prediction of CO₂ conversion and major products consistent with experimental results.
IISc Computational Framework for CO₂-to-Fuel Conversion:
| Dimension | Key Details |
|---|---|
| Catalyst-surface species | The network includes 105 chemical species present on the catalyst surface. |
| Outcome with 152-reaction model | A model limited to 152 reactions predicts formic acid as the main product and underestimates CO₂ conversion. |
| Outcome with expanded model | The expanded model predicts a 40-fold increase in CO₂ conversion and identifies methanol and carbon monoxide as major products, consistent with experimental results. |
| Less conventional pathway identified | The model identifies a pathway where molecular hydrogen (H₂) transfers hydrogen directly to intermediates without first splitting into individual atoms. |
| Technologies used | The framework comprises quantum-mechanical simulations, machine learning (ML), and automated reaction mapping. |
| Quantum-mechanical simulations | Quantum-mechanical simulations estimate reaction energies and barriers based on quantum mechanics. |
| Machine learning role | Machine learning (ML) predicts energy barriers for additional reactions to reduce computational effort. |
| Automated reaction mapping role | Automated reaction mapping identifies potential reactions among chemical species to build a comprehensive reaction network. |
| Applications | The framework supports catalyst design that favors pathways leading to methanol, carbon utilisation for converting captured CO₂ into valuable products, and wider applications to CO₂ reduction with other catalysts, nitrogen reduction, and water splitting. |