Abstract
The evolution of the variance and entropy of granule size in the fluidized bed agglomeration process using two different aggregation kernels is examined. The first is a constant kernel (aggregation is independent of both time and granule size) and permits the most unconstrained agglomeration process that can occur where granules in any size class (up to a maximum size) can be formed at any point in time. This gives the fastest and largest increase in the variance and entropy of the resulting granule size distribution. The second kernel is a mechanistic kernel including a granule growth-limiting mechanism, in this case implemented by the consideration that not all collisions result in coalescence. This markedly changes the evolution of the variance and entropy of the distribution and reduces both significantly. Quantifying the entropy of the distribution provides another perspective on the change in the size distribution in an agglomeration process. It is shown that entropy can provide a better measure of size evolution than variance in that it represents the changing shape of the distribution more closely.