Aggregation and Gelation of Silica Nanoparticles

The gelation mechanism was explored in a comprehensive way both experimentally and numerically. The gelation dynamics of a sol of colloidal silica of approximately 7 nm radius particles is studied using a combination of light scattering and rheometry. By changing the ionic strength (by addition of a salt solution resulting in different ultimate molarities) of the mixture, a stable sol can be destabilized, leading to aggregation and later gelation. The gel time tgel can be varied from hours to weeks, indicating a reaction-limited aggregation process. Static light scattering is used to extract the fractal dimension Df of the aggregates, which is found to be approximately 2. The evolution of cluster size is probed by dynamic light scattering, and follows an exponential growth. Rheometry is used to assess the gelation time and further development of the network strength after gelation. The elastic modulus (G') is found to scale as G' ∼ &phis;3:3, where &phis; is the silica particle volume fraction. It was observed that the gel time (after salt solution addition) depends on both the particle volume fraction and salt concentration, showing a divergence at low volume fraction or low salt concentration. For a single solid fraction, data for the cluster hydrodynamic radius, normalized by the single particle radius, from experiments with a wide range of gel times can be collapsed onto a master curve when the time after the salt addition, t, is scaled as t/t gel; a similar collapse of viscosity and the linear viscoelastic data after gelation can be obtained using the same scaling of time. Salt concentration affects the gel time but not the strength of the gel network, thus allowing very accurate prediction of network formation times and mechanical properties.;The effects of both hydrodynamic and repulsive forces on the rate of aggregation, and on the microstructure and mechanical properties of particle aggregates, are investigated by Brownian dynamics (BD) and Stokesian Dynamics (SD) simulation, over a range of solid volume fraction of 0:04 ≤ &phis; ≤ 0:12. The simulation methods differ in their treatment of the role of the suspending fluid, as SD includes the configuration-dependent hydrodynamic interactions between particles, whereas BD includes the fluid only though a viscous drag independent of configuration. Typical simulations use O(1000) particles in the simulational unit cell, which is periodically replicated in three dimensions. The interparticle potential is parameterized to include a roughly constant primary minimum near contact with U min/kBT > 6, along with a variable repulsive barrier at slightly larger separations; here Umin is the depth of the primary minimum. The structure is characterized through the pair distribution function, g(r), and the static structure factor. The repulsive barrier reduces the rate of aggregation and is also seen to affect the structure, with a large repulsion associated with a more tenuous structure. This is reflected in the potential having a strong effect on the evolution of `bonds' per particle (defined by a pair of particles being inside the repulsive barrier), with larger repulsion leading to smaller numbers of bonds and hence a more loose and open structure. Hydrodynamics was found to reduce the solid fraction required for percolation, with the influence depending upon the form of the potential; the difference in percolation threshold was significant, &phis;c;SD =˙ 0:06 and &phis;c;BD ≥ 0:08 for an intermediate level of repulsive force, where the largest fractional difference is observed. The hydrodynamic interactions were examined through analysis of the statistics of the evolution of numerous independent three-particle aggregates.;The interparticle bonding potential was investigated in a two dimensional system with or without restrictions in the rotational and stretching motion of particles. Two types of non-restrictive bonding force were studied, FENE (finitely extensible nonlinear elastic) and spring potential. Both of the bonding forces are effective in binding two particles together and forming clusters with chain-like structure under the condition that there is no repulsive barrier between particles. However, in the presence of repulsive force, systems with FENE bonding potential form clusters without branched structure; while with spring potential, particles form chain-like clusters which interconnect with each other and evolve into large clusters filling the whole space. (Abstract shortened by UMI.)...