Topological representation of layered hybrid lead halides for machine-learning using universal clusters

TL;DR

This paper introduces a machine learning approach that uses topological representations via atom-specific persistent homology to predict band gaps in layered hybrid lead halides, aiding the discovery of materials with desired electronic properties.

Contribution

It presents a novel application of invariant topological features for machine learning predictions of electronic properties in hybrid halide materials.

Findings

Topological features improve band gap prediction accuracy.

Method enables efficient screening of new hybrid halide materials.

Approach can be extended to other material properties.

Abstract

Layered hybrid halide compounds offer promising functional properties, particularly tunable band gaps, conductivity, light harvesting thus making them prospective for applications in photovoltaics and optoelectronics. This study exemplifies an approach of predicting band gaps using machine learning models enhanced by invariant topological representations of these materials using the atom-specific persistent homology method in order to facilitate the discovery and design of new hybrid halide materials with tailored electronic properties.