Modeling Gas Adsorption and Mechanistic Insights into Flexibility in Isoreticular Metal-Organic Frameworks Using...

Area of Science:

• Materials Science
• Computational Chemistry
• Nanotechnology

Background:

• Metal-organic frameworks (MOFs) possess unique porous structures ideal for gas storage applications.
• Accurate prediction of MOF properties and gas adsorption mechanisms is crucial but computationally demanding.
• Existing simulation methods face limitations with large, periodic MOF systems.

Purpose of the Study:

• To develop accurate and computationally efficient methods for simulating gas adsorption in MOFs.
• To investigate hydrogen (H2) and methane (CH4) adsorption in isoreticular metal-organic frameworks (IRMOFs).
• To explore the suitability of high-dimensional neural network potentials (HDNNPs) for MOF simulations.

Main Methods:

• Construction of HDNNPs for IRMOFs using a fragmentation technique at the density functional theory (DFT) level.
• Implementation of an "adsorption-relaxation" model combining molecular dynamics (MD) and grand canonical Monte Carlo (GCMC) simulations.
• Simultaneous MD and GCMC simulations to study H2 and CH4 adsorption isotherms.

Main Results:

• HDNNPs accurately reproduced DFT-level energies and forces for MOF systems.
• Simulations using HDNNPs showed excellent agreement with experimental gas adsorption values.
• The UFF4MOF classical force field demonstrated limitations for adsorption-relaxation simulations.
• Predicted CH4 uptake in IRMOF-10 under extreme conditions significantly exceeded classical force field predictions.

Conclusions:

• HDNNPs offer a viable and accurate approach for atomistic simulations of MOFs, overcoming computational limitations.
• The "adsorption-relaxation" model with HDNNPs provides detailed mechanistic insights into gas adsorption processes.
• HDNNPs enhance the predictive power for gas storage in MOFs, especially under challenging conditions.