Element selection for crystalline inorganic solid discovery guided by unsupervised machine learning of experimentally explored chemistry

Vasylenko, Andrij, Gamon, Jacinthe, Duff, Benjamin, Gusev, Vladimir, Daniels, Luke, Zanella, Marco, Shin, Felix, Sharp, Paul, Morscher, Alexandra, Chen, Ruiyong, Neale, Alex, Hardwick, Laurence, Claridge, John, Blanc, Frederic, Gaultois, Michael, Dyer, Matthew and Rosseinsky, Matthew (2020) Element selection for crystalline inorganic solid discovery guided by unsupervised machine learning of experimentally explored chemistry. [Data Collection]

Data Catalogue DOI: 10.17638/datacat.liverpool.ac.uk/1157

Original publication URL: https://doi.org/10.1038/s41467-021-25343-7

Description

The selection of the elements to combine delimits the possible outcomes of synthetic chemistry because it determines the range of compositions and structures, and thus properties, that can arise. For example, in the solid state, the elemental components of a phase field will determine the likelihood of finding a new crystalline material. Researchers make these choices based on their understanding of chemical structure and bonding. Extensive data are available on those element combinations that produce synthetically isolable materials, but it is difficult to assimilate the scale of this information to guide selection from the diversity of potential new chemistries. Here, we show that unsupervised machine learning captures the complex patterns of similarity between element combinations that afford reported crystalline inorganic materials. This model guides prioritisation of quaternary phase fields containing two anions for synthetic exploration to identify lithium solid electrolytes in a collaborative workflow that leads to the discovery of Li3.3SnS3.3Cl0.7. The interstitial site occupancy combination in this defect stuffed wurtzite enables a low-barrier ion transport pathway in hexagonal close-packing.

Keywords:

Machine Learning, Li ion conductor, Wurtzite

Divisions:

Faculty of Science and Engineering > School of Physical Sciences

Depositing User:

Luke Daniels

Date Deposited:

11 Oct 2021 13:29

Last Modified:

11 Oct 2021 13:29

DOI:

10.17638/datacat.liverpool.ac.uk/1157

URI:

https://datacat.liverpool.ac.uk/id/eprint/1157

Additional details

Creators:	Vasylenko, Andrij Gamon, Jacinthe Duff, Benjamin Gusev, Vladimir Daniels, Luke Zanella, Marco Shin, Felix Sharp, Paul Morscher, Alexandra Chen, Ruiyong Neale, Alex Hardwick, Laurence Claridge, John Blanc, Frederic Gaultois, Michael Dyer, Matthew Rosseinsky, Matthew
Research funder:	Grant Number Name of the Funder EP/N004884 EPSRC EP/P020194 EPSRC FIRG007 Faraday Institution SOLBAT
Date:	10 September 2020
Statement on legal, ethical and access issues :	No access restrictions, data can be made immediately available.
Additional information:	The raw ICSD-2017 data used in this study is available at https://www.github.com/lrcfmd/PhaseFieldRanking. The distribution of the phase fields’ rankings, computed phase field’s energy profile (convex hull) and experimental data generated in this study are available here. The X-ray crystallographic coordinates for the structure reported in this study have been deposited at the Cambridge Crystallographic Data Centre (CCDC), under deposition number 2023329. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif. The code (implementation of VAE and Phase Field Ranking method) developed for this work is available at https://www.github.com/lrcfmd/PhaseFieldRanking. https://doi.org/10.5281/zenodo.5113851
Resource language:	English
Publisher:	University of Liverpool
Contact email address:	l.m.daniels@liverpool.ac.uk
Last Modified:	11 Oct 2021 13:29