Comparison of the granulation behavior of three different excipients in a laboratory fluidized bed granulator using statistical methods.

The granulation process and, subsequently, the tableting behavior of the resulting granules of alpha-lactose-monohydrate, dicalcium phosphate anhydrous, and potato starch were investigated using statistical designs. The three substances were chosen due to their differences in granulation and tableting behavior, such as water solubility, swelling, and compressional properties. Granulation process variables, namely the inlet air temperature, spray rate, binder concentration of granulating solution, and inlet air flow rate were investigated. A central composite design was applied to study the granulation of alpha-lactose-monohydrate. Granulations of dicalcium phosphate and potato starch were investigated using a 2(3) factorial design, in which the effects of the inlet air temperature, spray rate, and binder concentration were considered. To compare the granulation behavior of theses substances the particle size distribution, angle of repose, and yield of the granules were used as responses for the statistical designs. The granules were compressed into tablets and the tensile strength was used as an additional statistical response. Based on the process parameters, models were developed using multiple regression modeling for each examined response. These models were then used to optimize the granulation process that provides granules with a Sauter mean diameter of D32 between 300 and 500 microm, an angle of repose smaller than 36 degrees and a granule yield above 90%. Moreover, the values of the tensile strength of the tablets should be between 1.6 and 2.5 N/mm2. The intersections of the response surfaces of each examined substance were compared using contour plots. To achieve the largest "satisfactory zone," the granulations of milled alpha-lactose-monohydrate, dicalcium phosphate, or potato starch should be performed at low inlet air temperatures.