Coarse-grained versus fully atomistic machine learning for zeolitic imidazolate frameworks

TL;DR

This paper compares simplified coarse-grained and detailed atomistic machine-learning models to evaluate how much chemical information can be effectively simplified in modeling zeolitic imidazolate frameworks.

Contribution

It introduces a comparative analysis of coarse-grained versus fully atomistic ML models for ZIFs, assessing the impact of simplification on model accuracy.

Findings

Coarse-grained models capture essential local environment features.

Fully atomistic models provide higher accuracy in local environment prediction.

Simplification may be sufficient for certain properties, reducing computational cost.

Abstract

Zeolitic imidazolate frameworks are widely thought of as being analogous to inorganic AB$_{2}$ phases. We test the validity of this assumption by comparing simplified and fully atomistic machine-learning models for local environments in ZIFs. Our work addresses the central question to what extent chemical information can be "coarse-grained" in hybrid framework materials.