Screening of physical solvents for carbon capture through a combined reverse engineeringmachine learning -CAMD approach
Résumé
Reducing CO₂ emissions is crucial to mitigating climate change. At the moment, the use of amines in Carbon Capture has major drawbacks. Although current physical solvents can partially overcome these limitations, they exhibit poor selectivity towards acid gases and require demanding operating conditions. In this work, a novel Computer Aided Molecular Design and Screening procedure is proposed as an alternative to trial-anderror approaches for identifying new solvent candidates. By leveraging predictive physical property models, tailored distribution functions, Health, Safety and Environment criteria and a Genetic Algorithm for the generation of molecular structures, the methodology is first validated on benchmark solvents to then proceed screening an extensive database of molecules. An unconventional alternative to current physical solvents is identified.
Domaines
| Origine | Fichiers produits par l'(les) auteur(s) |
|---|---|
| Licence |