Please use this identifier to cite or link to this item: http://dx.doi.org/10.14279/depositonce-10466
For citation please use:
Main Title: Machine learning enables long time scale molecular photodynamics simulations
Author(s): Westermayr, Julia
Gastegger, Michael
Menger, Maximilian F. S. J.
Mai, Sebastian
González, Leticia
Marquetand, Philipp
Type: Article
Language Code: en
Abstract: Photo-induced processes are fundamental in nature but accurate simulations of their dynamics are seriously limited by the cost of the underlying quantum chemical calculations, hampering their application for long time scales. Here we introduce a method based on machine learning to overcome this bottleneck and enable accurate photodynamics on nanosecond time scales, which are otherwise out of reach with contemporary approaches. Instead of expensive quantum chemistry during molecular dynamics simulations, we use deep neural networks to learn the relationship between a molecular geometry and its high-dimensional electronic properties. As an example, the time evolution of the methylenimmonium cation for one nanosecond is used to demonstrate that machine learning algorithms can outperform standard excited-state molecular dynamics approaches in their computational efficiency while delivering the same accuracy.
URI: https://depositonce.tu-berlin.de/handle/11303/11579
http://dx.doi.org/10.14279/depositonce-10466
Issue Date: 5-Aug-2019
Date Available: 19-Aug-2020
DDC Class: 540 Chemie und zugeordnete Wissenschaften
Subject(s): deep neural networks
machine learning
molecular photodynamics simulations
Sponsor/Funder: EC/H2020/792572/EU/Machine Learning for Catalytic Carbon Dioxide Activation/MachineCat
License: https://creativecommons.org/licenses/by/3.0/
Journal Title: Chemical Science
Publisher: Royal Society of Chemistry (RSC)
Publisher Place: Cambridge
Volume: 10
Publisher DOI: 10.1039/C9SC01742A
Page Start: 8100
Page End: 8107
EISSN: 2041-6539
ISSN: 2041-6520
Appears in Collections:FG Maschinelles Lernen » Publications

Files in This Item:
c9sc01742a.pdf
Format: Adobe PDF | Size: 2.27 MB
DownloadShow Preview
Thumbnail

Item Export Bar

This item is licensed under a Creative Commons License Creative Commons