Article Metrics


Online attention

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

DOI: 10.1038/s41467-021-25343-7 DOI Help

Authors: Andrij Vasylenko (University of Liverpool) , Jacinthe Gamon (University of Liverpool) , Benjamin B. Duff (University of Liverpool) , Vladimir V. Gusev (University of Liverpool) , Luke M. Daniels (University of Liverpool) , Marco Zanella (University of Liverpool) , J. Felix Shin (University of Liverpool) , Paul M. Sharp (University of Liverpool) , Alexandra Morscher (University of Liverpool) , Ruiyong Chen (University of Liverpool) , Alex R. Neale (University of Liverpool) , Laurence J. Hardwick (University of Liverpool) , John B. Claridge (University of Liverpool) , Frédéric Blanc (University of Liverpool) , Michael W. Gaultois (University of Cambridge) , Matthew S. Dyer (University of Liverpool) , Matthew J. Rosseinsky (University of Liverpool)
Co-authored by industrial partner: No

Type: Journal Paper
Journal: Nature Communications , VOL 12

State: Published (Approved)
Published: September 2021
Diamond Proposal Number(s): 23666

Open Access Open Access

Abstract: 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.

Journal Keywords: Electronic materials; Energy modelling

Subject Areas: Chemistry, Information and Communication Technology, Materials

Instruments: I11-High Resolution Powder Diffraction

Added On: 27/09/2021 14:17


Discipline Tags:

Inorganic Chemistry Information & Communication Technologies Artificial Intelligence Data processing Materials Science Chemistry

Technical Tags:

Diffraction X-ray Powder Diffraction