University of Leicester
Browse
JChemPhys_121_108.pdf (114.19 kB)

The canonical ensemble via symplectic integrators using Nosé and Nosé–Poincaré chains

Download (114.19 kB)
journal contribution
posted on 2013-06-26, 15:14 authored by Benedict J. Leimkuhler, Christopher R. Sweet
Simulations that sample from the canonical ensemble can be generated by the addition of a single degree of freedom, provided that the system is ergodic, as described by Nosé with subsequent modifications by Hoover to allow sampling in real time. Nosé–Hoover dynamics is not ergodic for small or stiff systems and the addition of auxiliary thermostats is needed to overcome this deficiency. Nosé–Hoover dynamics, like its derivatives, does not have a Hamiltonian structure, precluding the use of symplectic integrators which are noted for their long term stability and structure preservation. As an alternative to Nosé–Hoover, the Hamiltonian Nosé–Poincaré method was proposed by Bond, Laird, and Leimkuhler [J. Comput. Phys. 151, 114 (1999)], but the straightforward addition of thermostatting chains does not sample from the canonical ensemble. In this paper a method is proposed whereby additional thermostats can be applied to a Hamiltonian system while retaining sampling from the canonical ensemble. This technique has been used to construct thermostatting chains for the Nosé and Nosé–Poincaré methods.

History

Citation

Journal of Chemical Physics, 2004, 121 (1), 08.

Author affiliation

/Organisation/COLLEGE OF SCIENCE AND ENGINEERING/Department of Mathematics

Version

  • VoR (Version of Record)

Published in

Journal of Chemical Physics

Publisher

American Institute of Physics (AIP)

issn

0021-9606

eissn

1089-7690

Copyright date

2004

Available date

2013-06-26

Publisher version

http://jcp.aip.org/resource/1/jcpsa6/v121/i1/p108_s1

Language

en

Usage metrics

    University of Leicester Publications

    Categories

    No categories selected

    Keywords

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC