Hybrid simulations of flow properties using atomistic – fluctuating hydrodynamics modelling for nanoconfined water

Project Information

HIPPOGRIFFE

Grant agreement ID: 700276

DOI

10.3030/700276

Project closed

EC signature date 10 March 2016

Start date 1 July 2016

End date 30 June 2018

Funded under

• EXCELLENT SCIENCE - Marie Skłodowska-Curie Actions

Total cost

€ 195 454,80

EU contribution

€ 195 454,80

Coordinated by

QUEEN MARY UNIVERSITY OF LONDON
United Kingdom

Periodic Reporting for period 1 - HIPPOGRIFFE (Hybrid simulations of flow properties using atomistic – fluctuating hydrodynamics modelling for nanoconfined water)

Reporting period: 2016-07-01 to 2018-06-30

Summary of the context and overall objectives of the project

The project is primarily of computational modelling character and aims to develop the universal framework of hybrid multiscale modelling methods with atomistic resolution coupled with hydrodynamics for a range of science and engineering problems which involve fluid/solid interfaces. This is as an interdisciplinary project that brings together hydrodynamics, molecular dynamics, statistical physics, applied mathematics, computational modelling, numerical methods, material science, experimental nanofluidics, and Atomic Force Microscopy (AFM) measurement technologies.

To date, no existing modelling framework can address the role and behaviour of nanoconfined water in situations when the nanoscale region is an intrinsic part of the macroscopic flow, e.g. the nanoscale gap between the moving tip of the atomic microscope and the substrate or the fluid behaviour in the vicinity of a protein molecule during its diffusion in water. The main problem is that the water as a chemical substance has to be considered on atomistic and hydrodynamic levels simultaneously. The two descriptions correspond to very different scales, both spatial and temporal. On the one hand, Molecular Dynamics (MD) modelling is a mature tool for the description of dynamics of complex molecular systems at the atomic scale. On the other hand, collective motions of molecules can be described by means of statistical physics, stochastic hydrodynamics methods, e.g. based on Langevin and Landau-Lifshitz Fluctuating Hydrodynamics (FH) equations. There is still a conceptual gap in smoothly combining the two descriptions within a single multi-space-time-scale model: while the stochastic fluctuations slowly converge to macroscopic laws at the macroscopic limit they do not become atoms in the microscopic limit. Various hybrid multiscale methods have been suggested in the literature that combine adaptive resolution at the atomistic level with hydrodynamics, e.g. combining molecular dynamics with adaptive resolution coarse-grained particle methods and coupling MD with FH based on the analogy with two-phase flows, which address the problem of space-time scale interconnection with some success for liquids. However, no such adaptive resolution multiscale method currently exists for liquid-liquid/solid interface problems. The HIPPOGRIFFE project responds to challenge to develop such a method and apply it for a state-of-the art AFM experiment.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

As a result of the work a new Generalised Landau-Lifshitz Fluctuating Hydrodynamics (GLL-FH) equations were derived and implemented into open source MD code Gromacs. Modified version of Gromacs software is published on GitHub (https://github.com/ikorotkin/gromacs_fhmd). The new two-way coupling model was tested for preserving correct RDF, VACF functions, standard deviations, on stationary problems such as shear (Couette) flow and non-stationary acoustic wave propagation problem.

One of the most useful improvement of the new model in comparison with the previous one-way coupling method is that the pure FH region (without particles) was implemented into the hybrid model. This means the pure FH approach is now smoothly coupled with the hybrid and pure atomistic approaches closing the multiscale model. Thanks to this, an order of magnitude reduction in computational cost has been achieved in comparison with the all-atom molecular dynamics solution in several numerical test examples of liquid argon simulations in Gromacs.

The theory of GLL-FH, computational scheme, and some numerical examples are described in (Korotkin, I., & Karabasov, S. (2018). A Generalised Landau-Lifshitz Fluctuating Hydrodynamics model for concurrent simulations of liquids at atomistic and continuum resolution. https://doi.org/10.13140/RG.2.2.19830.45123).

Another journal article where a new extension of the hybrid MD-FH method to include the shear flow boundary condition induced by the moving boundary is in preparation (authors: J. Hu, I. Korotkin, S. Karabasov). In this paper we apply the new MD-FH method to analyse the self-diffussion property of water and the molecular diffusion of a small peptide in water under shear flow conditions.

Another paper entitled “A multi-resolution particle/fluctuating hydrodynamics model for hybrid simulations of liquids based on the two-phase flow analogy” (Hu, Korotkin, & Karabasov, 2018) has been accepted in the Journal of Chemical Physics.

There are also two journal articles in preparation (authors: F. Li, I. Korotkin, and S. Karabasov). In the first one we analyse the material properties of nano-confined water between graphene sheets under shear flow conditions. In the second article we develop the first all-atom model of an AFM tip moving in a drop of water near the mica substrate including realistic surface interaction effects. The final part of the model with taking into account a large volume of the surrounding water molecules will be developed using the new hybrid MD/FH model, which work is underway.

In addition to the AFM tip simulation, MD simulations of the experiment with artificial ELP-IK24 protein and Graphene Oxide sheets were performed to complement a set of state-of-the-art experiments carried out by Prof. Alvaro Mata and his research group in Queen Mary University. A joint paper about the experiment and its simulation was prepared and submitted to Nature Chemistry journal.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

The project has a wide range of impacts to the individual researcher and to public.

Multiscale modelling of nanoconfined water at liquid/solid interface opens a link to manipulation of liquid properties at the nanoscale. This technology is one of the several Key Enabling Technologies that the EU has identified as strategic in terms of market size and growth: Nanotechnologies, Advanced Materials, Advanced Manufacturing, and Biotechnologies. The proposed research into multiscale modelling will significantly advance the fundamental understanding of macroscopic water properties at the nanoscale. These will contribute to the research excellence of the Europe, and benefit researchers in a wide range of disciplines. The efficient high-fidelity multiscale model with all-atom resolution in the region of interest would contribute to a paradigm shift from experiment-based to simulation-based confidence in design and decision-making, and will, therefore, benefit EU companies in manufacturing, materials, healthcare, environment etc. and raise the EU economic competitiveness.

The project delivered an efficient and promising multiscale model described in (Korotkin, I., & Karabasov, S. (2018). A Generalised Landau-Lifshitz Fluctuating Hydrodynamics model for concurrent simulations of liquids at atomistic and continuum resolution. https://doi.org/10.13140/RG.2.2.19830.45123). The model was implemented as an effective boundary condition in open-source software Gromacs, which is extremely popular in the computational bio and material science communities. We presented the model at workshops and conferences, which were attended by CFD companies who are interested in multiphysics and multiscale simulations, and arranged one-to-one discussions with potential end-users.

Schematic of the AFM experiment simulation - Molecular Dynamics and Hydrodynamics regions