A Quantitative Method for Reconstructing Blend Composition Distributions in the Presence of Agglomerates
Marcos Llusa and Fernando J. Muzzio. Chemical and Biochemical Engineering, Rutgers University, 98 Brett Road, Piscataway, NJ 08854
This paper communicates a methodology to reconstruct the composition distribution of a powder blend containing drug agglomerates using experimental data and statistical principles. A statistical model combining multiple distribution functions is used to describe binary mixtures containing a low proportion, partially agglomerated minor component. The local concentration for the non-agglomerated portion of the minor component throughout the entire blend is described by a Gaussian distribution, whereas the population of agglomerates is estimated using (typically sparse) experimental data. The method entails using a simulation of the powder bed in an iterative manner in order to optimize the parameters necessary to describe the population of agglomerates. The simulated blend is extensively sampled and a distribution of RSD values is obtained. The iteration converges when the mode of the distribution of RSD values matches the experimental RSD value, which is the best available estimate of the blend homogeneity. The resulting distributions of composition values closely match experimental observations, and can be used to optimize sampling protocols, compute operating characteristic curves, and estimate process capability for blends containing agglomerates.