Semi-classical path integral non-adiabatic dynamics: a partial linearized classical mapping Hamiltonian approach
Creators
Abstract
A new partially linearized approximate approach to non-adiabatic quantum dynamics is derived based on linearizing the path difference for nuclear degrees of freedom (DOF) in the classical mapping Hamiltonian while keeping quantum interference effects inherent in the forward and backward propagators for the electronic DOF. With this new approach, the non-adiabatic force that acts on the nuclear DOF is a mean force rather than a state dependent force as found in some alternative approaches. Various benchmark examples are explored to test the accuracy of this new approach, and compare its performance with other approaches for a wide range of physical phenomena including: non-adiabatic scattering, excited state conical intersection dynamics, excited state photoisomerization, and excitation energy transfer in realistic condensed phase model systems. Results indicate that, even though the method is based on a "mean trajectory"-like scheme, it can accurately capture electronic population branching through multiple avoided crossing regions and that the approach offers a robust and reliable way to treat quantum dynamical phenomena in a wide range of condensed phase applications.
Additional Information
© 2012 Taylor & Francis. Received 12 February 2012: final version received 5 April 2012. Accepted author version posted online: 13 Apr 2012. Version of record first published: 11 May 2012. We gratefully acknowledge support for this research from the National Science Foundation under grant CHE-0911635 and support from Science Foundation Ireland under Grant Number 10/IN.1/I3033. DFC acknowledges the support of his Stokes Professorship in Nanobiophysics from Science Foundation Ireland. PH wants to thank Guohua Tao and Jian Liu who, respectively, pointed out the importance of using the mapping Hamiltonian force to get the correct asymptotic behavior and the "running on the inverse potential" problem associated with the original mapping Hamiltonian. We also acknowledge a grant of supercomputer time from Boston University's Office of Information Technology and Scientific Computing and Visualization. Finally, thanks Bill for your inspirational work over all these years.Additional details
Identifiers
- Eprint ID
- 32277
- DOI
- 10.1080/00268976.2012.684896
- Resolver ID
- CaltechAUTHORS:20120706-092822036
Related works
- Describes
- 10.1080/00268976.2012.684896 (DOI)
Funding
- NSF
- CHE-0911635
- Science Foundation, Ireland
- 10/IN.1/I3033
- Boston University
Dates
- Created
-
2012-07-06Created from EPrint's datestamp field
- Updated
-
2021-11-09Created from EPrint's last_modified field