A Multiscale Algorithm to Drastically Reduce Computational Times when Simulating Liquid/Solid Interaction at Atomic Resolution with Realistic Hydrodynamics Effects

Authors

DOI:

https://doi.org/10.4208/cicp.OA-2024-0013

Keywords:

Multiscale modelling, fluctuating hydrodynamics, molecular dynamics, Atomic Force Microscopy, interfacial water

Abstract

A hybrid multiscale model that incorporates both continuum fluid dynamics effects including thermal fluctuations and molecular dynamics is presented for the challenging 3D liquid-solid interface problem of gap water flows between the moving tip of the Atomic Force Microscope (AFM) cantilever and a material substrate surface. Highlights of the method include all-atom resolution in the vicinity of material surfaces, thereby avoiding the empiricism of interface boundary treatments in continuum mechanics approaches, with considerable computational savings compared with direct single-scale approaches such as Non-Equilibrium Molecular Dynamics (NEMD) methods. All components of the multiscale method are systematically described and verified. To optimize the coupling of the continuum fluid dynamics solution with the molecular dynamics solution, a simple iterative method is used. Predictions of the multiscale method are compared with reference equilibrium molecular dynamics solutions and discussed in the context of available AFM measurements in the literature. Computational performance of the proposed multiscale method is analysed in comparison with that of the NEMD method for a relevant range of AFM problem parameters. In particular, compared with the single-resolution MD method, the implemented multiscale model leads to an eight-order-of-magnitude acceleration of the solution for a slowly moving AFM tip typical of existing experiments. Perspectives on the further development and application of the proposed multiscale method are discussed. The current model is implemented in the open-source GROMACS software, and the corresponding source code is also provided.

Author Biographies

  • Fan Li

    School of Physical Science and Technology, Northwestern Polytechnical University, Xi’an, 710072, China

  • Ivan Korotkin

    The School of Engineering and Materials Science, Queen Mary University of London, Mile End Road, E1 4NS London, United Kingdom

    Mathematical Sciences, University of Southampton, University Road, SO17 1BJ Southampton, United Kingdom

  • Sergey Karabasov

    The School of Engineering and Materials Science, Queen Mary University of London, Mile End Road, E1 4NS London, United Kingdom

Published

2025-11-28

Abstract View

  • 738

Pdf View

  • 21

Issue

Section

Articles

How to Cite

A Multiscale Algorithm to Drastically Reduce Computational Times when Simulating Liquid/Solid Interaction at Atomic Resolution with Realistic Hydrodynamics Effects. (2025). Communications in Computational Physics, 39(2), 475-512. https://doi.org/10.4208/cicp.OA-2024-0013