(600i) A Finite-Element Based Global Model for Multiphase Flows In a Convective Assembly System

Convective assembly is a promising method for manufacturing self-assembled, colloidal crystals. In this process, a moving meniscus and fluid flow is employed to achieve self-assembly of particles from a suspension --- mono-disperse spheres crystallize near the moving contact line on a stationary substrate immersed in an evaporating medium.  In this process, a thin fcc crystal is deposited on the substrate in a matter of few hours. In order to understand the mechanism behind such fast deposition of ordered arrays of particles, Norris et al. put forth the convective steering hypothesis, which posits that solvent flow through pore spaces of close-packed spheres could preferentially direct advancing spheres into “clear or fcc niches” rather than into the “obstructed niches”. Although calculations based on porous-media network theory suggested that flow is up to 33% higher into the clear niches than into obstructed niches, subsequent calculations based on elaborate three-dimensional finite element analysis of flows into an fcc structure showed that inertial effects of the global flow may overwhelm the local steering effects induced by the particle arrangement.

In this presentation, we focus on the character on the global flows in convective assembly system by employing a two-dimensional, finite element analysis of the temporal flows in the system. We account for time-dependent flows in the solvent as well as the gas phase above it, including a rigorous accounting for evaporation, capillary, and Marangoni forces at the liquid-gas interface. We study the effect of varying gas inflow over the liquid surface, substrate and growth front inclinations, and solute concentration in the suspension on the flow behavior of the system. We validate our model against prior experimental results.