Machine learning for materials discovery: two-dimensional topological insulators

Abstract

One of the main goals and challenges of materials discovery is to find the best candidates for each interest property or application. Machine learning rises in this context to efficiently optimize this search, exploring the immense materials space, consisting of simultaneously the atomic, compositional, and structural spaces. Topological insulators, presenting symmetry-protected metallic edge states, are a promising class of materials for different applications. However, further, development is limited by the scarcity of viable candidates. Here we present and discuss machine learning-accelerated strategies for searching the materials space for two-dimensional topological materials. We show the importance of detailed investigations of each machine learning component, leading to different results. Using recently created databases containing thousands of ab initio calculations of 2D materials, we train machine learning models capable of determining the electronic topology of materials, with an accuracy of over 90%. We can then generate and screen thousands of novel materials, efficiently predicting their topological character without the need for a priori structural knowledge. We discover 56 non-trivial materials, of which 17 novel insulating candidates for further investigation, for which we corroborate their topological properties with density functional theory calculations. This strategy is 10$\times$ more efficient than the trial-and-error approach while few orders of magnitude faster and is a proof of concept for guiding improved materials discovery search strategies.

arXiv commentAccepted for publication in Applied Physical Review. For Supplemental Materials please contact one of the authors

References in corpus (38)

• Topological Insulators
• Topological insulators and superconductors
• Quantum Spin Hall Effect in Graphene
• $Z_2$ Topological Order and the Quantum Spin Hall Effect
• Topological Crystalline Insulators
• Quantum Spin Hall Effect and Topological Field Effect Transistor in Two-Dimensional Transition Metal Dichalcogenides
• Two-dimensional materials from high-throughput computational exfoliation of experimentally known compounds
• New Tolerance Factor to Predict the Stability of Perovskite Oxides and Halides
• A General-Purpose Machine Learning Framework for Predicting Properties of Inorganic Materials
• Topological Insulator Materials
• Higher-Order Topological Insulators
• Colloquium: Topological Band Theory
• The Computational 2D Materials Database: High-Throughput Modeling and Discovery of Atomically Thin Crystals
• A Performance and Cost Assessment of Machine Learning Interatomic Potentials
• SISSO: a compressed-sensing method for identifying the best low-dimensional descriptor in an immensity of offered candidates
• Computing topological invariants without inversion symmetry
• On-the-fly Closed-loop Autonomous Materials Discovery via Bayesian Active Learning
• Z2Pack: Numerical Implementation of Hybrid Wannier Centers for Identifying Topological Materials
• Topology-Scaling Identification of Layered Solids and Stable Exfoliated 2D Materials
• Topological Insulators in Random Lattices
• The 2021 Quantum Materials Roadmap
• Transport in two-dimensional topological materials: recent developments in experiment and theory
• Two-Dimensional Arsenene Oxide: A Realistic Large-gap Quantum Spin Hall Insulator
• False-positive and False-negative assignments of Topological Insulators in Density-Functional Theory and Hybrids
• Abundance of ${\mathbb{Z}}_2$ topological order in exfoliable two-dimensional insulators
• Discovering new two-dimensional topological insulators from computational screening
• Toward Realistic Amorphous Topological Insulators
• Atomic positions independent descriptor for machine learning of material properties
• Detection of Topological Materials with Machine Learning
• Computational Search for Magnetic and Non-magnetic 2D Topological Materials using Unified Spin-orbit Spillage Screening
• Artificial Intelligence for High-Throughput Discovery of Topological Insulators: the Example of Alloyed Tetradymites
• Two-dimensional topological materials discovery by symmetry-indicator method
• Disorder effects of vacancies on the electronic transport properties of realistic topological insulators nanoribbons: the case of bismuthene
• Discovery of Higher-Order Topological Insulators using the Spin Hall Conductivity as a Topology Signature
• Structural and electronic properties of realistic two-dimensional amorphous topological insulators
• Conductance of quantum spin Hall edge states from first principles: the critical role of magnetic impurities and inter-edge scattering
• Dual Topological Insulator Device with Disorder Robustness
• Materials property prediction using symmetry-labeled graphs as atomic-position independent descriptors

Cited by in corpus (14)

• Performance Assessment of Universal Machine Learning Interatomic Potentials: Challenges and Directions for Materials' Surfaces
• Connecting Higher-Order Topology with the Orbital Hall Effect in Monolayers of Transition Metal Dichalcogenides
• AI-driven inverse design of materials: Past, present and future
• 2024 roadmap on 2D topological insulators
• Topological thermal transport
• Machine Learning - Driven Materials Discovery: Unlocking Next-Generation Functional Materials - A review
• Twistronics of Janus transition metal dichalcogenide bilayers
• Exploring the structural and optoelectronic properties of natural insulating phlogopite in van der Waals heterostructures
• Topogivity: A Machine-Learned Chemical Rule for Discovering Topological Materials
• Linear Jacobi-Legendre expansion of the charge density for machine learning-accelerated electronic structure calculations
• Phyllosilicates as earth-abundant layered materials for electronics and optoelectronics: Prospects and challenges in their ultrathin limit
• Discovering two-dimensional magnetic topological insulators by machine learning
• Theory of Glide Symmetry Protected Helical Edge States in WTe${}_2$ Monolayer
• Predicting Many Crystal Properties via an Adaptive Transformer-based Framework