Generative Language Model for Catalyst Discovery

TL;DR

This paper presents CatGPT, a transformer-based language model that generates inorganic catalyst structures, demonstrating its potential as a tool for accelerating catalyst discovery through generative modeling and fine-tuning.

Contribution

Introduction of CatGPT, a pretrained transformer model for inorganic catalyst generation, and its fine-tuning for specific catalyst design tasks.

Findings

High validity and accuracy in generated catalyst structures

Effective fine-tuning for specific catalytic reactions

Potential to accelerate catalyst discovery processes

Abstract

Discovery of novel and promising materials is a critical challenge in the field of chemistry and material science, traditionally approached through methodologies ranging from trial-and-error to machine learning-driven inverse design. Recent studies suggest that transformer-based language models can be utilized as material generative models to expand chemical space and explore materials with desired properties. In this work, we introduce the Catalyst Generative Pretrained Transformer (CatGPT), trained to generate string representations of inorganic catalyst structures from a vast chemical space. CatGPT not only demonstrates high performance in generating valid and accurate catalyst structures but also serves as a foundation model for generating desired types of catalysts by fine-tuning with sparse and specified datasets. As an example, we fine-tuned the pretrained CatGPT using a binary alloy catalyst dataset designed for screening two-electron oxygen reduction reaction (2e-ORR) catalyst and generate catalyst structures specialized for 2e-ORR. Our work demonstrates the potential of language models as generative tools for catalyst discovery.