Machine learning spectral indicators of topology

Abstract

Topological materials discovery has emerged as an important frontier in condensed matter physics. While theoretical classification frameworks have been used to identify thousands of candidate topological materials, experimental determination of materials' topology often poses significant technical challenges. X-ray absorption spectroscopy (XAS) is a widely-used materials characterization technique sensitive to atoms' local symmetry and chemical bonding, which are intimately linked to band topology by the theory of topological quantum chemistry (TQC). Moreover, as a local structural probe, XAS is known to have high quantitative agreement between experiment and calculation, suggesting that insights from computational spectra can effectively inform experiments. In this work, we leverage computed X-ray absorption near-edge structure (XANES) spectra of more than 10,000 inorganic materials to train a neural network (NN) classifier that predicts topological class directly from XANES signatures, achieving F${}_1$ scores of 89% and 93% for topological and trivial classes, respectively. Additionally, we obtain consistent classifications using corresponding experimental and computational XANES spectra for a small number of measured compounds. Given the simplicity of the XAS setup and its compatibility with multimodal sample environments, the proposed machine learning-augmented XAS topological indicator has the potential to discover broader categories of topological materials, such as non-cleavable compounds and amorphous materials, and may further inform field-driven phenomena in situ, such as magnetic field-driven topological phase transitions.

References in corpus (18)

• A General-Purpose Machine Learning Framework for Predicting Properties of Inorganic Materials
• The space group classification of topological band insulators
• On-the-fly Closed-loop Autonomous Materials Discovery via Bayesian Active Learning
• High-throughput calculations of magnetic topological materials
• Angle-resolved photoemission spectroscopy and its application to topological materials
• Building Blocks of Topological Quantum Chemistry: Elementary Band Representations
• All Topological Bands of All Nonmagnetic Stoichiometric Materials
• Topological Materials Discovery from Crystal Symmetry
• Unsupervised machine learning and band topology
• Classification of Local Chemical Environments from X-ray Absorption Spectra using Supervised Machine Learning
• Topological correspondence between magnetic space group representations
• Machine Learning Topological Phases with a Solid-state Quantum Simulator
• Symmetry indicators of band topology
• Topological classification and diagnosis in magnetically ordered electronic materials
• Computational Search for Magnetic and Non-magnetic 2D Topological Materials using Unified Spin-orbit Spillage Screening
• High-throughput search for magnetic topological materials using spin-orbit spillage, machine-learning and experiments
• Application of the induction procedure and the Smith Decomposition in the calculation and topological classification of electronic band structures in the 230 space groups
• Determining electronic properties from L-edge X-ray absorption spectra of transition metal compounds with artificial neural networks

Cited by in corpus (7)

• Recent Advances and Applications of Deep Learning Methods in Materials Science
• Topological thermal transport
• Topogivity: A Machine-Learned Chemical Rule for Discovering Topological Materials
• Discovering two-dimensional magnetic topological insulators by machine learning
• Closed-loop Error Correction Learning Accelerates Experimental Discovery of Thermoelectric Materials
• Self-Supervised Generative Models for Crystal Structures
• Structural Descriptors and Information Extraction from X-ray Emission Spectra: Aqueous Sulfuric Acid