skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Machine learning for quantum dynamics: deep learning of excitation energy transfer properties

Abstract

Understanding the relationship between the structure of light-harvesting systems and their excitation energy transfer properties is of fundamental importance in many applications including the development of next generation photovoltaics. Natural light harvesting in photosynthesis shows remarkable excitation energy transfer properties, which suggests that pigment–protein complexes could serve as blueprints for the design of nature inspired devices. Mechanistic insights into energy transport dynamics can be gained by leveraging numerically involved propagation schemes such as the hierarchical equations of motion (HEOM). Solving these equations, however, is computationally costly due to the adverse scaling with the number of pigments. Therefore virtual high-throughput screening, which has become a powerful tool in material discovery, is less readily applicable for the search of novel excitonic devices. We present the use of artificial neural networks to bypass the computational limitations of established techniques for exploring the structure-dynamics relation in excitonic systems. Once trained, our neural networks reduce computational costs by several orders of magnitudes. Our predicted transfer times and transfer efficiencies demonstrate similar or even higher accuracies than frequently used approximate methods such as secular Redfield theory.

Authors:
 [1];  [1]; ORCiD logo [1]
  1. Harvard Univ., Cambridge, MA (United States)
Publication Date:
Research Org.:
Massachusetts Inst. of Technology (MIT), Cambridge, MA (United States)
Sponsoring Org.:
USDOE Office of Science (SC), Basic Energy Sciences (BES) (SC-22)
OSTI Identifier:
1409485
Alternate Identifier(s):
OSTI ID: 1506094
Grant/Contract Number:  
SC0001088
Resource Type:
Journal Article: Published Article
Journal Name:
Chemical Science
Additional Journal Information:
Journal Volume: 8; Journal Issue: 12; Journal ID: ISSN 2041-6520
Publisher:
Royal Society of Chemistry
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING; 37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CHEMISTRY

Citation Formats

Häse, Florian, Kreisbeck, Christoph, and Aspuru-Guzik, Alán. Machine learning for quantum dynamics: deep learning of excitation energy transfer properties. United States: N. p., 2017. Web. doi:10.1039/c7sc03542j.
Häse, Florian, Kreisbeck, Christoph, & Aspuru-Guzik, Alán. Machine learning for quantum dynamics: deep learning of excitation energy transfer properties. United States. doi:10.1039/c7sc03542j.
Häse, Florian, Kreisbeck, Christoph, and Aspuru-Guzik, Alán. Mon . "Machine learning for quantum dynamics: deep learning of excitation energy transfer properties". United States. doi:10.1039/c7sc03542j.
@article{osti_1409485,
title = {Machine learning for quantum dynamics: deep learning of excitation energy transfer properties},
author = {Häse, Florian and Kreisbeck, Christoph and Aspuru-Guzik, Alán},
abstractNote = {Understanding the relationship between the structure of light-harvesting systems and their excitation energy transfer properties is of fundamental importance in many applications including the development of next generation photovoltaics. Natural light harvesting in photosynthesis shows remarkable excitation energy transfer properties, which suggests that pigment–protein complexes could serve as blueprints for the design of nature inspired devices. Mechanistic insights into energy transport dynamics can be gained by leveraging numerically involved propagation schemes such as the hierarchical equations of motion (HEOM). Solving these equations, however, is computationally costly due to the adverse scaling with the number of pigments. Therefore virtual high-throughput screening, which has become a powerful tool in material discovery, is less readily applicable for the search of novel excitonic devices. We present the use of artificial neural networks to bypass the computational limitations of established techniques for exploring the structure-dynamics relation in excitonic systems. Once trained, our neural networks reduce computational costs by several orders of magnitudes. Our predicted transfer times and transfer efficiencies demonstrate similar or even higher accuracies than frequently used approximate methods such as secular Redfield theory.},
doi = {10.1039/c7sc03542j},
journal = {Chemical Science},
issn = {2041-6520},
number = 12,
volume = 8,
place = {United States},
year = {2017},
month = {10}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record at 10.1039/c7sc03542j

Citation Metrics:
Cited by: 8 works
Citation information provided by
Web of Science

Save / Share: