(535i) The Mechanism for Shear Thickening in Viscoelastic Suspensions

Yang, M., Stanford University
Shaqfeh, E. S. G., Stanford University
Suspensions of solid particulates in viscoelastic fluids are ubiquitous in engineering applications, yet their material functions under shear flow are difficult to predict due to the nonlinearity of the stress in the suspending fluid. Previous experiments in the literature on suspensions in Boger fluids have suggested that elasticity in the suspending fluid causes shear thickening in the suspension averaged viscosity and the first normal stress coefficient. Upon examining these results, authors were unable to determine whether this effect arose from particle-particle interactions or particle-fluid interactions. Previously we performed simulations of shear flow past a sphere and identified that the extra fluid stress induced by polymer stretching in the disturbance flow of each particle in a dilute suspension leads to shear-thickening which is at least in qualitative agreement with experimental results. However, quantitative differences with the experimental data suggest a closer examination of the mechanism of shear-thickening is necessary. Thus, we first use numerical simulations to analyze the local disturbance flow field surrounding a single particle as well as its effect on the polymer conformation to determine the regions that most contribute to the average suspension viscosity thickening. We identify the importance of Lagrangian, time-dependent extensional flow in creating large polymer stretching in closed streamlines around the particles. The insights from these analyses imply particle-particle interactions are relatively unimportant for predicting the shear-thickening of the viscosity but are important for accurately predicting the zero-shear viscosity as a function of volume fraction. Moreover, our results suggest that the effective extra viscosity (beyond the zero shear viscosity) of the suspension per particle is correlated with the suspension shear stress. We verify these findings by analyzing literature experiments and new, multi-particle simulations using Immersed Boundary Methods.