Large scale molecular simulations of flow and assembly

SNIC 2020/1-37


SNIC Large Compute

Principal Investigator:

Berk Hess


Kungliga Tekniska högskolan

Start Date:


End Date:


Primary Classification:

20301: Applied Mechanics

Secondary Classification:

10402: Physical Chemistry



The research in my group focuses on algorithms as well as applications for large scale molecular dynamics (MD) simulations. Recently the emphasis of the applications in my group is shifting to the investigation of molecular aspects of flow and assembly of bio-molecules. In flow there are both fundamental aspects that are not well understood, especially at surfaces, as well as questions about particular applications where molecular aspects become more important due to the smaller scales in micro- and nanofluidics. Although even in nanofluidics most of the system is still best described by continuum (or meso-) dynamics, details of molecular interactions can play an essential role. An important case is the the three-phase contact line in wetting. Molecular processes that can not be desrcibed in terms of continuum physics play a crucial role here. Molecular dynamics simulations are the only way to study these effects in detail. We will study the processes at the contact line under non-equilibrium conditions: one case if flow over structured substrates, another is boiling heat transfer. A second project, which will start soon, looks at assembly of cellulose fibrils with the end goal of producing stronger materials. The assembly of nanofibrils is controlled by changes in pH an ion concentrations, which affect protonation states. These states can not be measured experimentally and the only way to get access to these is molecular simulation (with a dynamic protonation method). Ordering of the charged fibrils can be achieved through flow, which will be studied at the meso scale using rod models. The effective forces between the fibrils will be parametrized using molecular dynamics simulations which can take into account the nature of surface groups and the ionic composition of the solution. These simulations will run faster with an efficient long-range electrostatics algorithm, which I am working on together with Rio Yokota at Tokyo Tech. All this work in done using the open-source GROMACS molecular simulation package and all