Article Machine-learning-assisted discovery of polymers with high thermal conductivity using a molecular design algorithm

Stephen Wu ; Yukiko Kondo ; Masa-aki Kakimoto ; Bin Yang ; Hironao Yamada ; Isao Kuwajima SAMURAI ORCID ; Guillaume Lambard SAMURAI ORCID (National Institute for Materials Science) ; Kenta Hongo ; Yibin Xu SAMURAI ORCID ; Junichiro Shiomi ; Christoph Schick ; Junko Morikawa ; Ryo Yoshida

41524_2019_203_MOESM1_ESM.avi
Download all files (12 MB)
Collection

Citation
Stephen Wu, Yukiko Kondo, Masa-aki Kakimoto, Bin Yang, Hironao Yamada, Isao Kuwajima, Guillaume Lambard, Kenta Hongo, Yibin Xu, Junichiro Shiomi, Christoph Schick, Junko Morikawa, Ryo Yoshida. Machine-learning-assisted discovery of polymers with high thermal conductivity using a molecular design algorithm. npj Computational Materials. 2019, 5 (), 66.

Description:

(abstract)

Machine learning technology has a great potential to accelerate the discovery of innovative functional materials. We demonstrate a successful discovery of novel high thermal conductivity polymers that was inspired by machine-learning-assisted polymer chemistry. The achievement was made possible by the interplay of a machine intelligence trained with a polymer database called PoLyInfo. Using a Bayesian molecular design algorithm trained to recognize quantitative structure-property relationships on thermal conductivity and other targeted polymeric properties, we identified thousands of promising designed polymers. Three were selected for monomer synthesis and polymerization based on chemical insights on their potential for further processing in real applications.

Rights:

Keyword: machine learning, polymer, thermal conductivity, molecular design

Date published: 2019-06-21

Publisher: Springer Science and Business Media LLC

Journal:

  • npj Computational Materials (ISSN: 20573960) vol. 5 66

Funding:

  • MEXT | Japan Science and Technology Agency
  • MEXT | Japan Society for the Promotion of Science 15H02672
  • MEXT | Japan Society for the Promotion of Science JP18K18017
  • MEXT | Japan Society for the Promotion of Science 16H06439

Manuscript type: Publisher's version (Version of record)

MDR DOI:

First published URL: https://doi.org/10.1038/s41524-019-0203-2

Related item:

Other identifier(s):

Contact agent:

Updated at: 2024-11-21 16:31:23 +0900

Published on MDR: 2024-11-21 16:31:24 +0900

Filename Size
Filename s41524-019-0203-2.pdf
application/pdf 		Size 2.25 MB Detail
Filename 41524_2019_203_MOESM2_ESM.pdf
application/pdf 		Size 1.48 MB Detail
Filename 41524_2019_203_MOESM1_ESM.avi (Thumbnail)
video/x-msvideo 		Size 12.7 MB Detail