×

Recommended by

Indexed by

Optical materials discovery and design with federated databases and machine learning

Victor Trinquet1, Matthew L. Evans1,2*, Cameron Hargreaves1, Pierre-Paul De Breuck1, Gian-Marco Rignanese1*

1 UCLouvain, Institut de la Matiere Condensée et des Nanosciences (IMCN), Chemin des Étoiles 8, Louvain-la-Neuve 1348, Belgium

2 Matgenix SRL, 185 Rue Armand Bury, 6534 Gozée, Belgium

* Corresponding authors emails: matthew.evans@uclouvain.be, gian-marco.rignanese@uclouvain.be
DOI10.24435/materialscloud:5p-vq [version v1]

Publication date: Aug 05, 2024

How to cite this record

Victor Trinquet, Matthew L. Evans, Cameron Hargreaves, Pierre-Paul De Breuck, Gian-Marco Rignanese, Optical materials discovery and design with federated databases and machine learning, Materials Cloud Archive 2024.114 (2024), https://doi.org/10.24435/materialscloud:5p-vq

Description

Combinatorial and guided screening of materials space with density-functional theory and related approaches has provided a wealth of hypothetical inorganic materials, which are increasingly tabulated in open databases. The OPTIMADE API is a standardised format for representing crystal structures, their measured and computed properties, and the methods for querying and filtering them from remote resources. Currently, the OPTIMADE federation spans over 20 data providers, rendering over 30 million structures accessible in this way, many of which are novel and have only recently been suggested by machine learning-based approaches. In this work, we outline our approach to non-exhaustively screen this dynamic trove of structures for the next-generation of optical materials. By applying MODNet, a neural network-based model for property prediction that has been shown to perform especially well for small materials datasets, within a combined active learning and high-throughput computation framework, we isolate particular structures and chemistries that should be most fruitful for further theoretical calculations and for experimental study as high-refractive-index materials. By making explicit use of automated calculations, federated dataset curation and machine learning, and by releasing these publicly, the workflows presented here can be periodically re-assessed as new databases implement OPTIMADE, and new hypothetical materials are suggested.

Materials Cloud sections using this data

No Explore or Discover sections associated with this archive record.

Files

File name Size Description
campaign.tar.gz
MD5md5:08ca692e704a4e977a86d0a4031f795c
387.5 MiB An archive containing the results from the active learning campaign.
optimade.yaml
MD5md5:e18da5c0fa23b6ab203fc90499c4e2d0
Go to the OPTIMADE API
1.1 KiB YAML config file for use with `optimade-maker` for creating OPTIMADE APIs from this static data.
README
MD5md5:5122483a549c8593f333d0cbb7ecc614
622 Bytes Data description

License

Files and data are licensed under the terms of the following license: Creative Commons Attribution 4.0 International.
Metadata, except for email addresses, are licensed under the Creative Commons Attribution Share-Alike 4.0 International license.

External references

Software (Software implementation)
Journal reference
V. Trinquet, M. L. Evans, et al., Faraday Discussions (2024) doi:10.1039/D4FD00092G
Preprint

Keywords

electronic structure OPTIMADE optical materials refractive index MODNet

Version history:

2024.114 (version v1) [This version] Aug 05, 2024 DOI10.24435/materialscloud:5p-vq