Large-Scale Quantum-Mechanical Molecular Dynamics Simulations Using Density-Functional Tight-Binding Combined with the Fragment Molecular Orbital Method

Yoshio Nishimoto, Hiroya Nakata, Dmitri G. Fedorov, Stephan Irle

Research output: Contribution to journalArticlepeer-review

44 Scopus citations

Abstract

The fully analytic gradient is developed for density-functional tight-binding (DFTB) combined with the fragment molecular orbital (FMO) method (FMO-DFTB). The response terms arising from the coupling of the electronic state to the embedding potential are derived, and the gradient accuracy is demonstrated on water clusters and a polypeptide. The radial distribution functions (RDFs) obtained with FMO-DFTB are found to be similar to those from conventional DFTB, while the computational cost is greatly reduced; for 256 water molecules one molecular dynamics (MD) step takes 73.26 and 0.68 s with full DFTB and FMO-DFTB, respectively, showing a speed-up factor of 108. FMO-DFTB/MD is applied to 100 ps MD simulations of liquid hydrogen halides and is found to reproduce experimental RDFs reasonably well.

Original languageEnglish
Pages (from-to)5034-5039
Number of pages6
JournalJournal of Physical Chemistry Letters
Volume6
Issue number24
DOIs
StatePublished - Dec 1 2015
Externally publishedYes

Funding

Y.N. and H.N. were supported by a Research Fellowship of the Japan Society for Promotion of Science for Young Scientists. This work is partially supported by JSPS KAKENHI Grant Number 15H06316 (Grant-in-Aid for Research Activity Startup). D.G.F. acknowledges support of the Next Generation Super Computing Project, Nanoscience Program (MEXT, Japan) and D.G.F. and S.I. thank Computational MaterialsScience Initiative (CMSI, Japan) for financial support. D.G.F. thanks Prof. Kazuo Kitaura for many fruitful discussions about the FMO method.

FundersFunder number
S.I.
Japan Society for the Promotion of Science15H06316
Japan Society for the Promotion of Science
Ministry of Education, Culture, Sports, Science and Technology

    Fingerprint

    Dive into the research topics of 'Large-Scale Quantum-Mechanical Molecular Dynamics Simulations Using Density-Functional Tight-Binding Combined with the Fragment Molecular Orbital Method'. Together they form a unique fingerprint.

    Cite this