Advancing CH4/H2 separation with covalent organic frameworks by combining molecular simulations and machine learning.
Gokhan Onder Aksu1, Seda Keskin1
1Department of Chemical and Biological Engineering, Koc University Rumelifeneri Yolu, Sariyer 34450 Istanbul Turkey skeskin@ku.edu.tr +90 212 338 1362.
This study introduces a computational method using machine learning to discover new COF materials for efficient methane/hydrogen separation. The approach accelerates the identification of high-performing adsorbents for gas separation applications.
Area of Science:
- Materials Science
- Computational Chemistry
- Chemical Engineering
Background:
- Methane/hydrogen (CH 4 /H 2 ) separation is crucial for natural gas purification and hydrogen energy.
- Existing adsorbents like metal-organic frameworks (MOFs) have limitations in performance and discovery.
- Covalent Organic Frameworks (COFs) show promise but require efficient screening methods.
Purpose of the Study:
- To develop and apply a high-throughput computational screening approach combined with machine learning (ML) for identifying optimal COFs for CH 4 /H 2 separation.
- To evaluate both synthesized and hypothetical COFs (hypoCOFs) for their adsorption-based separation capabilities.
- To accelerate the discovery of novel COF materials with superior performance compared to existing adsorbents.
Main Methods:
- Grand Canonical Monte Carlo (GCMC) simulations were used to study the adsorption of CH 4 /H 2 mixtures in 59,840 synthesized and hypothetical COFs under Pressure-Swing Adsorption (PSA) and Vacuum-Swing Adsorption (VSA) conditions.
- Machine learning models were developed based on simulation results to predict CH 4 /H 2 adsorption properties of remaining hypothetical materials.
- Structural and chemical properties of top-performing COFs were analyzed to guide the search for new materials.
Main Results:
- The study screened 597 synthesized COFs and 7,737 hypothetical COFs, identifying top candidates with high CH 4 selectivity and working capacity.
- Top-performing hypothetical COFs demonstrated CH 4 selectivities and working capacities exceeding those of synthesized COFs and MOFs.
- ML models accurately predicted COF separation performance in seconds, significantly reducing computational cost.
Conclusions:
- The proposed computational approach accurately and efficiently assesses COF materials for CH 4 /H 2 separation.
- This method significantly accelerates experimental efforts in designing and discovering new high-performing COF adsorbents.
- The study unlocks the potential of COFs for advanced gas separation technologies.
More Related Videos
12:05Preparation of Hydrophobic Metal-Organic Frameworks via Plasma Enhanced Chemical Vapor Deposition of Perfluoroalkanes for the Removal of Ammonia
Published on: October 10, 2013
07:45Electrophoretic Crystallization of Ultrathin High-performance Metal-organic Framework Membranes
Published on: August 16, 2018
Related Concept Videos
02:27Predicting Molecular Geometry
02:00Molecular Models
03:24Hybridization of Atomic Orbitals I
02:36Molecular Geometry and Dipole Moments
02:18Conformations of Ethane and Propane
Staggered conformation is a low energy and more stable conformation with the C-H bonds on the front carbon placed at 60°dihedral angles relative to the C-H bonds on the back carbon, leading to a reduced torsional strain. In staggered...