Literature DB >> 26575543

Adaptive switching of interaction potentials in the time domain: an extended Lagrangian approach tailored to transmute force field to QM/MM simulations and back.

Marcus Böckmann1, Nikos L Doltsinis1, Dominik Marx2.   

Abstract

An extended Lagrangian formalism that allows for a smooth transition between two different descriptions of interactions during a molecular dynamics simulation is presented. This time-adaptive method is particularly useful in the context of multiscale simulation as it provides a sound recipe to switch on demand between different hierarchical levels of theory, for instance between ab initio ("QM") and force field ("MM") descriptions of a given (sub)system in the course of a molecular dynamics simulation. The equations of motion can be integrated straightforwardly using the usual propagators, such as the Verlet algorithm. First test cases include a bath of harmonic oscillators, of which a subset is switched to a different force constant and/or equilibrium position, as well as an all-MM to QM/MM transition in a hydrogen-bonded water dimer. The method is then applied to a smectic 8AB8 liquid crystal and is shown to be able to switch dynamically a preselected 8AB8 molecule from an all-MM to a QM/MM description which involves partition boundaries through covalent bonds. These examples show that the extended Lagrangian approach is not only easy to implement into existing code but that it is also efficient and robust. The technique moreover provides easy access to a conserved energy quantity, also in cases when Nosé-Hoover chain thermostatting is used throughout dynamical switching. A simple quadratic driving potential proves to be sufficient to guarantee a smooth transition whose time scale can be easily tuned by varying the fictitious mass parameter associated with the auxiliary variable used to extend the Lagrangian. The method is general and can be applied to time-adaptive switching on demand between two different levels of theory within the framework of hybrid scale-bridging simulations.

Entities:  

Year:  2015        PMID: 26575543     DOI: 10.1021/acs.jctc.5b00142

Source DB:  PubMed          Journal:  J Chem Theory Comput        ISSN: 1549-9618            Impact factor:   6.006


  3 in total

1.  Adaptive-Partitioning Multilayer Dynamics Simulations: 1. On-the-Fly Switch between Two Quantum Levels of Theory.

Authors:  Joani Mato; Adam W Duster; Emilie B Guidez; Hai Lin
Journal:  J Chem Theory Comput       Date:  2021-08-27       Impact factor: 6.578

2.  Explicit Solvation Matters: Performance of QM/MM Solvation Models in Nucleophilic Addition.

Authors:  Jelle M Boereboom; Paul Fleurat-Lessard; Rosa E Bulo
Journal:  J Chem Theory Comput       Date:  2018-03-16       Impact factor: 6.006

3.  Adaptive QM/MM for Molecular Dynamics Simulations: 5. On the Energy-Conserved Permuted Adaptive-Partitioning Schemes.

Authors:  Adam W Duster; Chun-Hung Wang; Hai Lin
Journal:  Molecules       Date:  2018-08-28       Impact factor: 4.411

  3 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.