Skander Charni, Raeesh Muhammad, Abdulkarem I. Amhamed, Brahim Aissa, Halima Bensmail

International Conference on Thermal Engineering (ICTEA) (2025)

It is becoming urgent to design MOFs with a high capacity of CO2 adsorption due to climate change. Machine learning opens up new possibilities for building MOFs with high CO2 adsorption capacity and understanding the CO2 adsorption mechanism. In this study, we collected 608 data points from several reports, characterized with several key variables such as BET surface area (SBET), total pore volume (VT), pressure (P), temperature (T), Langmuir surface area (SL). Eight engineered features were also used: electronegativity (E), atomic mass (AM), density (D), first ionization (FI), specific heat (SH), atomic radius (AR) and melting point (MP) to predict CO2 uptake. A panoply of machine learning models was built to predict the CO2 adsorption capacity of MOFs. The results indicated that Gradient Boosting Machine (GBM) had the best prediction performance (R2 training = 0.983, R2 testing = 0.945, RMSE training = 1.017, RMSE testing = 1.646). Feature importance study demonstrated that beside pressure and temperature, CO2 adsorption parameters (75%), textures (20%), and MOF electronegativity or some chemical features (5%) are all important in the CO2 adsorption process and that higher value of those parameters tend to negatively affect the adsorption intake.

https://journals.library.torontomu.ca/index.php/ictea/article/download/2635/2164