A Machine‐Learning‐Based Approach for Solving Atomic Structures of Nanomaterials Combining Pair Distribution Functions with Density Functional Theory. Issue 13 (12th February 2023)

• Record Type: Journal Article
• Authors: Kløve, Magnus; Sommer, Sanna; Iversen, Bo B.; Hammer, Bjørk; Dononelli, Wilke
• Is Part Of: Advanced materials. Volume 35:Issue 13(2023)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2023-02-12
• Subjects: crystal structure prediction -- density functional theory -- global optimization -- machine learning -- nanomaterials -- pair distribution function; Materials -- Periodicals; Chemical vapor deposition -- Periodicals; 620.11
• DOI: 10.1002/adma.202208220
• Languages: English
• ISSNs: 0935-9648

Abstract

Abstract: Determination of crystal structures of nanocrystalline or amorphous compounds is a great challenge in solid‐state chemistry and physics. Pair distribution function (PDF) analysis of X‐ray or neutron total scattering data has proven to be a key element in tackling this challenge. However, in most cases, a reliable structural motif is needed as a starting configuration for structure refinements. Here, an algorithm that is able to determine the crystal structure of an unknown compound by means of an on‐the‐fly trained machine learning model, which combines density functional theory calculations with comparison of calculated and measured PDFs for global optimization in an artificial landscape, is presented. Due to the nature of this landscape, even metastable configurations and stacking disorders can be identified. Abstract : Determination of crystal structures of nanocrystalline compounds is a great challenge in solid‐state chemistry. In this work, an algorithm is introduced, which is able to determine the crystal structure of an unknown compound by means of an on‐the‐fly trained machine learning model that combines density functional calculations with comparison of calculated and measured pair distribution functions for global optimization.