Mixing and Blending

MIXING AND BLENDING

Mixing and blending may be achieved by rotating or shearing the powder bed. Mixing two or more components that may differ in composition, particle size, or some other physicochemical property is brought about through a sequence of events. Most powders at rest occupy a small volume such that it would be difficult to force two static powder beds to mix. The first step in a mixing process, therefore, is to dilate the powder bed. The second step, which may be concurrent with the first, is to shear the powder bed. Ideally, shearing occurs at the level of planes separated by individual particles. The introduction of large interparticulate spaces is achieved by rotating the bed. A V-blender or barrel roller are classical examples of systems which, by rotating through 3608, dilate the powder bed while, through the influence of gravity, shear planes of particles. A planetary mixer uses blades to mechanically dilate and shear the powder. Each of these systems is a batch process. A ribbon blender uses a screw, or auger, action to rotate and shear the bed from one location to another in a continuous process.

Since the shearing of particles in a bed to achieve a uniform mix of blend is a statistical process, it must be monitored for efficiency. Sample thieves are employed to probe the powder bed, with minimal disturbance, and draw samples for analysis. These samples are then analyzed for the relevant dimen-sion for mixing, for example, particle size, drug, or excipient content. Statistical mixing parameters have been derived based on the mean and standard devia-tion of samples taken from various locations in a blend at various times during processing (Carstensen, 1993). In large-scale mixers random number tables may be employed to dictate sample sites. There is a considerable science of sampling that can be brought to bear on this problem (Thompson, 2002). The sample size for pharmaceutical products is ideally of the scale of the unit dose. This is relevant as it relates to the likely variability in the dose that in turn relates to the therapeutic effect. In the case of small unit doses, the goal should be to sample at a size within the resolution of the sample thief.