Determination of an empirical correlation for the mass transfer coefficients of pharmaceutical powders in a fluid bed layer

Abstract

The pharmaceutical industry has four basic operations to produce tablets: mixing, drying, milling, and tablet pressing. Fluidized bed drying of the pharmaceutical powders is one of the most efficient methods, which is finding ever-growing applications in diverse industries, especially in the pharmaceutical and the food industries. However, one of the problems that exist in the pharmaceutical industry is to determine when the formulation reaches the desired moisture content. This final value can be either determined with a Loss on Drying (LOD) technique on estimated using a model of the process. To obtain this model, it is necessary to know the parameters that control drying process as well as the characteristics of the granulation used in this stage. Unfortunately the modeling of drying is still uncertain. The objective of this research was to determine an empirical correlation for the mass transfer coefficient (kga) of pharmaceutical powders in a Fluid Bed Dryer (FBD) which will then be used a long with fundamental models to predict moisture content. Granulations of blends consisting of lactose monohydrate and pregelatinized starch combined with distilled water were dried and samples were withdrawn at different intervals of time to generate drying curves. These curves were divided in two regions, the constant and the falling rate periods. Correlations based on dimensionless numbers such as Reynolds, Schmidt, and Sherwood were developed. Predicted values were compared to values obtained from Loss on Drying (LOD). Good correlation was obtained between predicted values of kga with the experimental ones. Additionally, an average residual of 0.18 % was obtained. It was found that the air flow velocity is the most significant factor in the drying curve behavior. Other phenomena such as bed porosity and changing the composition of the granulation seem not to affect the prediction of moisture content (MC) for the granulation used. The study indicates that prediction of MC curves of some pharmaceutical powders in a FBD can be achieved by establishing a correlation of kga based just on air flow and temperature.