Science 340, 720–724 (2013)

Credit: © 2013 AAAS

Looking at an evolving beer head can be fascinating. Indeed, foam dynamics involves cycles of slow draining of the thin-film network of fluid surrounding the gas bubbles, followed by the fast burst of a small section of the network and the subsequent rearrangement of the topology of the bubbles. As these physical processes involve many scales in both space and time, predicting the evolution of foam-like materials accurately is challenging. Now, Robert Saye and James Sethian show that modelling foam dynamics is possible if both space- and timescales are appropriately separated through a multiscale model that combines the Navier–Stokes equations for fluid dynamics, both continuum and mesh-like models of the network, Voronoid tessellation for tracking its structural evolution and numerical schemes for solving the resulting set of coupled partial differential equations. The researchers show that the predictions of their multiscale model (which can be expanded to include other phenomena such as evaporation dynamics or diffusive coarsening) agree with experimental data from the merging of two soap bubbles.