(401j) Using Brownian Dynamics to Model Nanoparticle Aggregation under Shear

Light-scattering data for the aggregation of colloidal nanoparticles shows that hybrid aggregates with two different fractal dimensions can be formed depending on the magnitude and duration of the applied shear (C. M. Sorensen's group, Kansas State University). Formation of such super-aggregate structures with D_f = 2.55, and their reverting to D_f = 1.8 depending on the shear condition is not completely explained yet. Simulations can provide insight into this behavior, provided they can accurately capture the essential aggregation physics. Brownian Dynamics (BD) simulations are used to study aggregation in this work. First, the stability and accuracy requirements of BD simulations to study aggregation are established. To satisfy these requirements, the computational time step needs to be small enough to resolve two characteristic times: one corresponds to the frictional coefficient that is related to the dissipation of energy in the system, and second characteristic time corresponds to the systematic force that acts between nanoparticles. To establish convergence of the BD simulations, a numerical test is performed to determine the time step and multiple independent simulations (MIS) required for accurate solution of simple one and two-particle systems. Our tests reveal that, even for such simple systems, a large number of MIS is needed to accurately estimate the probability of particles forming aggregates. Estimates for a system of N particles indicate the need for computational speed-up. Strategies to speed-up accurate BD calculations of aggregation are explored.