Speeding up simulation of diffusion in zeolites by a parallel synchronous kinetic Monte Carlo algorithm

Area of Science:

• Computational chemistry
• Materials science
• Chemical engineering

Background:

• Simulating molecular behavior in confined spaces, like zeolites, is crucial but computationally demanding.
• Traditional kinetic Monte Carlo methods struggle with extended time and length scales.

Purpose of the Study:

• To evaluate a discrete synchronous parallel kinetic Monte Carlo algorithm for simulating aromatic hydrocarbon diffusion in zeolites.
• To determine the algorithm's efficiency based on processor count and domain size.

Main Methods:

• Application of a discrete synchronous parallel kinetic Monte Carlo algorithm.
• Investigation of simulation efficiency concerning processor allocation and domain partitioning.
• Analysis of aromatic hydrocarbon diffusion dynamics within zeolite structures.

Main Results:

• The parallel kinetic Monte Carlo algorithm demonstrates significant efficiency gains.
• Achieving high efficiency is possible with optimized domain size selection.
• The method effectively extends the accessible spatial and temporal scales for simulations.

Conclusions:

• The discrete synchronous parallel kinetic Monte Carlo algorithm is a viable and efficient tool for studying molecular diffusion in zeolites.
• Optimized domain decomposition is key to unlocking the algorithm's potential for large-scale simulations.
• This approach facilitates deeper understanding of molecular transport in microporous materials.