Micromechanics and rheology of colloidal gels via dynamic simulation
Colloidal gels are soft solids comprising a viscoelastic, networked structure embedded in solvent. This network forms from microscopically small particles initially dispersed in a solvent which self-assemble into a hierarchical, space-spanning network of particles connected by physical bonds. When subjected to external forces, colloidal gels exhibit a solid-to-liquid transition yet regain elastic character when forcing is removed. Their tunable mechanical properties and ability to flow enable colloidal gels to serve as the foundation of a multitude of applications ranging from everyday products, like yogurt, to biomedical applications, such as injectable therapeutics. The nonlinear rheology of colloidal gels underlies their utility in nearly every application, for example, spreading, injecting, or pouring. The transition from rest to steady flow of colloidal gels is characterized by one or more stress overshoots indicative of gel yield. In strongly-bonded, dilute colloidal gels, yield is hypothesized to result from the catastrophic loss of the network structure. Solid-like fracture leading to fluidization of strongly bonded gels may not be relevant where particle strands are not single-particle thick chains but rather bicontinuous and time-evolving due to reversible bonds. The connections between gel yield and the structural evolution of dense, bicontinuous gels remains poorly understood due to the difficulty of imaging of the internal structure of dense particulate gels with sufficient time resolution in experiments and due to the large system size required in computational studies. Here we report large-scale dynamic simulation to study reversible colloidal gels to elucidate the micromechanical underpinnings of non-Newtonian behavior of soft materials and to understand ongoing phase separation. First, we show that the startup of a fixed strain rate reveals that colloidal gel yield, separating the short-time solid-like response from the long-time liquid-like response, can be framed as a transition in energy storage. Contrary to prior hypotheses connecting yield to loss of network connectivity, the network persists after flow startup and a predictive model connecting hierarchical structure to early-time stress growth is presented. We devised a novel approach to monitor bond stretching, compression, formation, and loss alongside macroscopic deformation. We find that changes in structure that underlie the stress growth and post-yield relaxation, as monitored by bond dynamics, indicate the switch from energy storage to release. After rheological yield, energy release continues if flow is sufficiently strong; however, when imposed flow is weak, energy release reverses after yield, and the gel densifies. This gives the important result that yield under weak flow can be viewed as a release from kinetic arrest, permitting the gel to evolve toward more complete phase separation. This supports our view that yield of weakly sheared gels is a `non-equilibrium phase transition'. Second, we compare our simulations to experimental measurements of colloidal gel rheology to study the influence of bond strength, volume fraction, and network morphology on the viscoelastic moduli. Strong agreement is found between linear viscoelasticity from simulation and experiment. Details of reversible colloidal gel morphology depend on the volume fraction of colloids. The growth of the dominant length scale predicts age-stiffening of the elastic modulus of reversible colloidal gels as found as found in our previous study. Finally, we provide a unifying framework for the ongoing phase separation of colloidal gels and to account for the rapid progress towards more complete phase separation induced by external flow, which we classify as a `non-equilibrium phase transition'. We show that an interplay between external perturbation and osmotic pressure temporarily releases the gel from kinetic arrest, advancing it toward more complete phase separation. Overall, this study connects weak external forcing to the reactivation of phase separation of a colloidal gel comprising a non-equilibrium network structure. The specific cases examined here have multiple real-world applications, for example, demonstrating that even weak forces may impact the stability or shelf-life of particulate network structures.
Simulation; Chemical engineering; Rheology; Colloidal gels
Zia, Roseanna N.
Kirby, Brian; Clancy, Paulette
Ph. D., Chemical Engineering
Doctor of Philosophy
dissertation or thesis