Predicting stable crystalline compounds using chemical similarity

Hai-Chen Wang,Silvana Botti,Miguel A. L. Marques

DOI: https://doi.org/10.1038/s41524-020-00481-6

IF: 12.256

2021-01-26

npj Computational Materials

Abstract:Abstract We propose an efficient high-throughput scheme for the discovery of stable crystalline phases. Our approach is based on the transmutation of known compounds, through the substitution of atoms in the crystal structure with chemically similar ones. The concept of similarity is defined quantitatively using a measure of chemical replaceability, extracted by data-mining experimental databases. In this way we build 189,981 possible crystal phases, including 18,479 that are on the convex hull of stability. The resulting success rate of 9.72% is at least one order of magnitude better than the usual success rate of systematic high-throughput calculations for a specific family of materials, and comparable with speed-up factors of machine learning filtering procedures. As a characterization of the set of 18,479 stable compounds, we calculate their electronic band gaps, magnetic moments, and hardness. Our approach, that can be used as a filter on top of any high-throughput scheme, enables us to efficiently extract stable compounds from tremendously large initial sets, without any initial assumption on their crystal structures or chemical compositions.

materials science, multidisciplinary,chemistry, physical

What problem does this paper attempt to address?

The problem this paper attempts to address is the efficient discovery of stable crystalline compounds. Specifically, the authors propose a high-throughput scheme based on chemical similarity, generating new potentially stable crystal phases by substituting chemically similar atoms in the crystal structures of known compounds. This method aims to effectively screen stable compounds from a vast initial set of compounds without making any prior assumptions about their crystal structures or chemical compositions. The core objectives of the paper include: 1. **Improving screening efficiency**: Significantly increasing the success rate of identifying stable compounds from a large pool of candidate compounds through the concept of chemical similarity. 2. **Reducing computational cost**: Compared to traditional high-throughput methods, this approach can substantially reduce the consumption of computational resources. 3. **Expanding the range of materials**: Extending beyond main group elements and transition metals to include lanthanides and actinides, thereby broadening the search scope for potential new materials. Through these objectives, the paper hopes to achieve breakthroughs in the discovery of new materials, particularly in fields with specific performance requirements such as electronic, magnetic, and mechanical properties.

Predicting stable crystalline compounds using chemical similarity

Reliable and Practical Computational Prediction of Molecular Crystal Polymorphs

Structurally Constrained Evolutionary Algorithm for the Discovery and Design of Metastable Phases

Predicting crystal form stability under real-world conditions

Large-scale machine-learning-assisted exploration of the whole materials space

A Robust Crystal Structure Prediction Method to Support Small Molecule Drug Development with Large Scale Validation and Prospective Studies

A Hybrid Machine Learning Approach for Structure Stability Prediction in Molecular Co-crystal Screenings

Crystal structure prediction with machine learning-based element substitution

Predicting Synthesizability using Machine Learning on Databases of Existing Inorganic Materials

The computational prediction of pharmaceutical crystal structures and polymorphism

Towards machine learning for microscopic mechanisms: a formula search for crystal structure stability based on atomic properties

Powder X-Ray Diffraction Assisted Evolutionary Algorithm for Crystal Structure Prediction

A complete description of thermodynamic stabilities of molecular crystals

Efficient first-principles prediction of solid stability: Towards chemical accuracy

A method for the automatic generation of a minimal basis set of structural templates for material phase-space exploration

Crystal Structure Prediction of Binary Alloys via Deep Potential

A machine learning potential-based generative algorithm for on-lattice crystal structure prediction

Predicting the phase stability of multi-component high entropy compounds

Enhancing Crystal Structure Prediction by Decomposition and Evolution Schemes Based on Graph Theory

Phase Prediction via Crystal Structure Similarity in the Periodic Number Representation

Unleashing the power of novel conditional generative approaches for new materials discovery