CaltechAUTHORS
  A Caltech Library Service

Machine learning of optical properties of materials - predicting spectra from images and images from spectra

Stein, Helge S. and Guevara, Dan and Newhouse, Paul F. and Soedarmadji, Edwin and Gregoire, John M. (2019) Machine learning of optical properties of materials - predicting spectra from images and images from spectra. Chemical Science, 10 (1). pp. 47-55. ISSN 2041-6520 . PMCID PMC6334722. http://resolver.caltech.edu/CaltechAUTHORS:20180730-104921622

[img] PDF - Published Version
Creative Commons Attribution Non-commercial.

1659Kb
[img] PDF - Submitted Version
See Usage Policy.

3213Kb
[img] Archive (ZIP) - Supplemental Material
Creative Commons Attribution Non-commercial.

6Kb
[img] Archive (ZIP) - Supplemental Material
Creative Commons Attribution Non-commercial.

7Kb
[img] PDF - Supplemental Material
Creative Commons Attribution Non-commercial.

517Kb

Use this Persistent URL to link to this item: http://resolver.caltech.edu/CaltechAUTHORS:20180730-104921622

Abstract

As the materials science community seeks to capitalize on recent advancements in computer science, the sparsity of well-labelled experimental data and limited throughput by which it can be generated have inhibited deployment of machine learning algorithms to date. Several successful examples in computational chemistry have inspired further adoption of machine learning algorithms, and in the present work we present autoencoding algorithms for measured optical properties of metal oxides, which can serve as an exemplar for the breadth and depth of data required for modern algorithms to learn the underlying structure of experimental materials science data. Our set of 178 994 distinct materials samples spans 78 distinct composition spaces, includes 45 elements, and contains more than 80 000 unique quinary oxide and 67 000 unique quaternary oxide compositions, making it the largest and most diverse experimental materials set utilized in machine learning studies. The extensive dataset enabled training and validation of 3 distinct models for mapping between sample images and absorption spectra, including a conditional variational autoencoder that generates images of hypothetical materials with tailored absorption properties. The absorption patterns auto-generated from sample images capture the salient features of ground truth spectra, and band gap energies extracted from these auto-generated patterns are quite accurate with a mean absolute error of 180 meV, which is the approximate uncertainty from traditional extraction of the band gap energy from measurements of the full transmission and reflection spectra. Optical properties of materials are not only ubiquitous in materials applications but also emblematic of the confluence of underlying physical phenomena yielding the type of complex data relationships that merit and benefit from neural network-type modelling.


Item Type:Article
Related URLs:
URLURL TypeDescription
https://doi.org/10.1039/c8sc03077dDOIArticle
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6334722PubMed CentralArticle
https://doi.org/10.26434/chemrxiv.6726317DOIDiscussion Paper
ORCID:
AuthorORCID
Stein, Helge S.0000-0002-3461-0232
Gregoire, John M.0000-0002-2863-5265
Additional Information:© 2019 The Royal Society of Chemistry. Open Access Article. This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence. All publication charges for this article have been paid for by the Royal Society of Chemistry. The article was received on 11 Jul 2018, accepted on 24 Oct 2018 and first published on 25 Oct 2018. This study is based upon work performed by the Joint Center for Artificial Photosynthesis, a DOE Energy Innovation Hub, supported through the Office of Science of the U.S. Department of Energy (Award No. DE-SC0004993). There are no conflicts to declare.
Group:JCAP
Funders:
Funding AgencyGrant Number
Department of Energy (DOE)DE-SC0004993
Royal Society of ChemistryUNSPECIFIED
PubMed Central ID:PMC6334722
Record Number:CaltechAUTHORS:20180730-104921622
Persistent URL:http://resolver.caltech.edu/CaltechAUTHORS:20180730-104921622
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:88353
Collection:CaltechAUTHORS
Deposited By: George Porter
Deposited On:30 Jul 2018 17:58
Last Modified:12 Feb 2019 19:40

Repository Staff Only: item control page