Automated Multiscale Approach To Predict Self-Diffusion from a Potential Energy Field.

Area of Science:

• Materials Science
• Chemical Engineering
• Computational Chemistry

Background:

• Estimating gas diffusion in nanoporous materials is crucial for large-scale screening.
• Brute-force molecular dynamics simulations are computationally expensive, especially for systems with low diffusion coefficients.
• Transition state theory offers a more efficient alternative but requires automatic detection of diffusion pathways and transition states.

Purpose of the Study:

• To develop an efficient algorithm for analyzing gas diffusion in nanoporous materials.
• To enable large-scale screening of materials for gas diffusion properties.
• To facilitate transition state theory-based analysis of diffusion processes.

Main Methods:

• An algorithm that analyzes energy grids of moving particles to detect diffusion paths and their directions.
• Identification of energy levels at which diffusion paths form.
• Partitioning of grid coordinates into energy basins and transition states for diffusive systems.

Main Results:

• The algorithm accurately identifies non-diffusive systems.
• It enables a transition state theory-based analysis of diffusion in diffusive systems.
• Tested on methane (CH4) diffusion in zeolites, the method proved accurate, fast, and rigorous.

Conclusions:

• The developed algorithm provides an efficient and accurate method for analyzing gas diffusion in nanoporous materials.
• It overcomes the computational limitations of molecular dynamics for large-scale screening.
• The method is versatile, with no geometric limitations on diffusion tunnels or transition states.