Multiscale simulation of dynamic wetting

Jun Zhang, Matthew Borg, Jason Reese

Research output: Contribution to journalArticlepeer-review


A sequential multiscale strategy that combines molecular dynamics (MD) with volume of fluid (VOF) simulations is proposed to study the spreading of droplets on surfaces. In this hybrid MD/VOF approach, VOF is applied everywhere in the domain with MD pre-simulations distributed along the wetted interface providing the crucial boundary information for the three-phase contact-line dynamics and the solid/liquid interfacial slip. For the latter, molecular shear flow simulations of the liquid in contact with the substrate are used to measure the local slip length to calibrate the Navier slip model. For the contact-line we use MD simulations of nanodroplets spreading on the substrate to calibrate the molecular kinetic theory (MKT) model for both the dynamic contact angle and the slip velocity. We validate this multiscale model for water nanodroplets spreading over a platinum surface by comparing our sequential hybrid simulations with the equivalent dynamics in a full MD treatment. We demonstrate that for nanodroplets spreading on surfaces, applying a dynamic contact angle model is not sufficient to pick up the molecular effects; we need to account for slip across the entire solid/liquid interface, in particular the large slip behaviour at the contact-line. We also demonstrate the application of this multiscale method to larger nanodroplets (up to ~100 nm diameters), where full MD simulation would be computationally intractable. We find that as the droplet size increases, the slip in the contact line region becomes less important. To simulate the full range of nano to macro droplets, an improved way of dealing with the VOF method is needed to reduce the overall number of grid points.
Original languageEnglish
Pages (from-to)886-896
JournalInternational journal of heat and mass transfer
Early online date1 Aug 2017
Publication statusPublished - Dec 2017


Dive into the research topics of 'Multiscale simulation of dynamic wetting'. Together they form a unique fingerprint.

Cite this