Deep learning with reflection high-energy electron diffraction images to predict cation ratio in Sr$_{2x}$Ti$_{2(1-x)}$O$_{3}$ thin films

TL;DR

This paper demonstrates how deep learning applied to reflection high-energy electron diffraction images can accurately predict cation ratios in Sr$_{2x}$Ti$_{2(1-x)}$O$_{3}$ thin films, enabling real-time, quantitative analysis during synthesis.

Contribution

The study introduces a deep learning approach that transforms qualitative diffraction data into quantitative stoichiometry predictions, revealing new correlations and enabling in situ control of thin film growth.

Findings

Accurate prediction of cation ratio with small dataset of 31 samples.

Identification of a new correlation between diffraction features and stoichiometry.

Demonstration of ML transforming qualitative diagnostics into quantitative measurements.

Abstract

Machine learning (ML) with in situ diagnostics offers a transformative approach to accelerate, understand, and control thin film synthesis by uncovering relationships between synthesis conditions and material properties. In this study, we demonstrate the application of deep learning to predict the stoichiometry of Sr$_{2x}$Ti$_{2(1-x)}$O$_{3}$ thin films using reflection high-energy electron diffraction images acquired during pulsed laser deposition. A gated convolutional neural network trained for regression of the Sr atomic fraction achieved accurate predictions with a small dataset of 31 samples. Explainable AI techniques revealed a previously unknown correlation between diffraction streak features and cation stoichiometry in Sr$_{2x}$Ti$_{2(1-x)}$O$_{3}$ thin films. Our results demonstrate how ML can be used to transform a ubiquitous in situ diagnostic tool, that is usually limited to qualitative assessments, into a quantitative surrogate measurement of continuously valued thin film properties. Such methods are critically needed to enable real-time control, autonomous workflows, and accelerate traditional synthesis approaches.