AN APPLICATION OF THE COMPUTER OPTIMIZATION TECHNIQUE TO WET GRANULATION PROCESS INVOLVING EXPLOSIVE GROWTH OF PARTICLES

The computer optimization technique based on well-designed experiments is applied to optimize the wet granulation process in which an explos ive growth of particles is involved. Total volume of binder solution u sed in the process (X-1), power consumption (X-2) of the blade and the interaction between X-1 and the blade rotation speed (X-3) were found to have strong influence on the granulation process based on the anal ysis of variance: Further, scraping off of powder adhered to the vesse l wall by the blade could play a critical role in such explosive growt h of particles. In scale-up studies (1.2 kg and 3.0 kg scale), an opti mization incorporating scale effect of three response variables (geome trical mean size of granules, the portion of granules passed through 7 5 mu m, and granulation time) which have high correlation coefficients in regression analysis was possible by using an integrated optimizati on function, T(X). A universal optimal condition unaffected by manufac turing scale is obtained by the minimization of T(X). Experimental val ues of these three response variables of granules prepared under the u niversal optimal condition agreed well with the predicted values in tw o manufacturing scales, suggesting that the computer optimization tech nique would be useful to optimize and visualize the process of wet gra nulation involving the explosive growth of particles. (C) 1997 Elsevie r Science B.V.