GPT-assisted learning of structure-property relationships by graph neural networks: Application to rare-earth doped phosphors

TL;DR

This paper combines GPT-4 and graph neural networks to efficiently mine data and predict properties of rare-earth doped phosphors, significantly advancing materials discovery with minimal supervision.

Contribution

It introduces a novel workflow integrating GPT-4 data mining with CGCNN modeling for phosphor property prediction, reducing the need for domain expertise.

Findings

Achieved a test R^2 of 0.77 for emission wavelength prediction.

Screened over 40,000 inorganic materials for phosphor properties.

Demonstrated transfer learning to adapt bandgap models for emission wavelength prediction.

Abstract

Applications of machine learning techniques in materials science are often based on two key ingredients, a set of empirical descriptors and a database of a particular material property of interest. The advent of graph neural networks, such as the Crystal Graph Convolutional Neural Network (CGCNN), demonstrates the possibility of directly mapping the relationship between material structures and properties without employing empirical descriptors. Another exciting recent advancement is in large language models such as OpenAI's GPT-4, which demonstrates competency at reading comprehension tasks and holds great promise for accelerating the acquisition of databases on material properties. Here, we utilize the combination of GPT-4 and CGCNN to develop rare-earth doped phosphors for solid-state lighting. GPT-4 is applied to data-mine chemical formulas and emission wavelengths of 264 Eu(II)-doped phosphors from 274 papers. A CGCNN model is trained on the acquired dataset, achieving a test $R^2$ of 0.77. The model is then used to screen over 40,000 inorganic materials to make predictions on the emission wavelengths. We also demonstrate the possibility of leveraging transfer learning to fine-tune a bandgap-predicting CGCNN model towards the prediction of phosphor emission wavelengths. The workflow requires minimal human supervision, little domain knowledge about phosphors, and is generalizable to other material properties.