Determining the grain orientations of battery materials from electron diffraction patterns using convolutional neural networks

TL;DR

This paper introduces a convolutional neural network approach to rapidly and accurately determine grain orientations in battery materials from electron diffraction patterns, surpassing traditional pattern matching methods.

Contribution

It is the first to apply deep learning to electron diffraction data analysis, significantly improving speed and accuracy in grain orientation characterization.

Findings

CNN models outperform conventional pattern matching in accuracy

Deep learning reduces analysis time for diffraction patterns

Models trained with dynamical effects data enhance reliability

Abstract

Polycrystalline materials have numerous applications due to their unique properties, which are often determined by the grain boundaries. Hence, quantitative characterization of grain as well as interface orientation is essential to optimize these materials, particularly energy materials. Using scanning transmission electron microscopy, matter can be analysed in an extremely fine grid of scan points via electron diffraction patterns at each scan point. By matching the diffraction patterns to a simulated database, the crystal orientation of the material as well as the orientation of the grain boundaries at each scan point can be determined. This pattern matching approach is highly time intensive. Artificial intelligence promises to be a very powerful tool for pattern recognition. In this work, we train convolutional neural networks (CNNs) on dynamically simulated diffraction patterns of LiNiO2, an important cathode-active material for Lithium-ion batteries, to predict the orientation of grains in terms of three Euler angles for the complete fundamental orientation region. Results demonstrate that these networks outperform the conventional pattern matching algorithm with increased accuracy and efficiency. The increased accuracy of the CNN models can be attributed to the fact that these models are trained by data incorporating dynamical effects. This work is the first attempt to apply deep learning for analysis of electron diffraction data and enlightens the great potential of ML to accelerate the analysis of electron microscopy data, toward high-throughput characterization technique.