(304b) Restructuring Phenomena in the Aggregation of a Colloidal Latex in a Turbulent Stirred Tank

Authors: 
Bäbler, M. U., ETH Zurich
Morbidelli, M., Institute of Chemical and Bioengineering, ETH Zurich


Aggregation of colloidal particles in a flowing suspension is accompanied by two important mechanical effects, i.e., breakage and restructuring of the formed clusters caused by hydrodynamic stresses acting on theme [1-3]. Restructuring is thereby understood as the densification of the clusters as they are processed in an agitated vessel. In this work, we investigate the restructuring of colloidal clusters in a turbulent stirred tank both experimentally and theoretically through a population balance equation (PBE) model. In the experiments, a fully destabilized polystyrene latex with a primary particle size of dp=810 nm is aggregated in a stirred tank reactor. The evolution of the cluster mass distribution (CMD) is monitored through small angle static light scattering from which we obtain two moments of the CMD, namely a mean radius of gyration and the zero angle intensity that represents a weighted mean cluster mass. The fractal dimension of the initial clusters is inferred from fitting the initial aggregation kinetics to our PBE model. The latter includes a detailed aggregation rate model [4] that accounts for the specific hydrodynamic and colloidal interactions between colliding clusters. From this we found df ≈ 2.0. As the clusters grow larger the hydrodynamic stresses acting on theme increase and eventually breakup and restructuring set in. As a consequence of this, clusters at steady state assume a relatively compact structure. The fractal dimension at steady state is estimated from microscopic images adopting an image analysis routine and an empirical correlation between the fractal dimension and the perimeter fractal dimension of the projection [5]. From this we found df ≈ 2.6. To gain further insight into this densification process, i.e., the transition from df ≈ 2.0 to 2.6, a PBE model is formulated that combines the previous aggregation rate model with a comprehensive breakage rate model [6]. The evolution of the fractal dimension is then obtained by fitting the PBE model to the experimental data. From this we found a sigmoidal evolution of df with time, which can be understood that the clusters have to grow to a certain size before restructuring sets in [7]. However, whether restructuring is a result of breakage, i.e., formation of dense fragments, or if aggregates compactify as coherent entities cannot be inferred from our data.

[1] Soos M., Sefcik J., Morbidelli M., Chem. Eng. Sci. 61, 2349 (2006)

[2] Selomulya C., Bushell G., Amal R., Waite T. D., Langmuir 18, 1974 (2002)

[3] Selomulya C., Bushell G., Amal R., Waite T. D., Chem. Eng. Sci. 58, 327 (2003)

[4] Bäbler M.U., AIChE J. 54, 1748 (2008)

[5] Lee C., Kramer T. A., Adv. Colloid Interface Sci. 112, 49 (2004)

[6] Bäbler M.U., Morbidelli M., Baldyga J., J. Fluid Mech. 612, 261 (2008)

[7] Becker V., Schlauch E., Behr M., Briesen H., J. Colloid Interface Sci. submitted (2009)