| Precipitation of monodispersed colloids from homogeneous solutions is a complex process and the mechanism has not been fully understood. Early works assumed a short nucleation burst, followed by diffusional growth of the nuclei to form identical larger particles. This mechanism appeared simple, plausible, and worked well for particles up to several tens of nanometers. However, it was found that in many systems nuclei, produced rapidly in a supersaturated solution, grow to nanosize primary particles (singlets), which then coagulate to form much larger final colloids in a process dominated by irreversible capture of these singlets. Many experimental observations have confirmed that these monodispersed colloids consist of small crystalline subunits.; To explain the size selection in such a process, a kinetic model of the two-stage growth process has been developed which combines aggregation with nucleation and growth. Our earlier study has shown the burst-nucleation growth of the primary particles to depend strongly on the value of the effective surface tension (σ) entering the surface term in the free energy of the sub-critical embryos. The model was tested for a system involving formation of uniform spherical gold particles by reduction of auric chloride in aqueous solutions. The calculated average size, the width of the particle size distribution, and the time scale of the process, agree semi quantitatively with the experimental values.; We tried modifying the aggregation rates to account for singlet size and aggregate diffusivity and found that introducing a “bottle-neck” factor (f) in the dimer formation rate has a profound effect on the final size distribution and suggests a possible size-control mechanism. However, the calculated distributions were much narrower than the experimentally measured ones. This probably indicates that cluster-cluster aggregation and detachment and ripening should be considered along with the phenomenological parameter f. We have found that a simple modification of the boundary condition to allow for detachment can remedy the problem caused by ignoring the ripening process. This can be done by introducing only one parameter (β). |
