Fifteen model drugs were quenched from 3:1 (w/w) mixtures with polyethylene glycol 4000 (PEG4000). The resulting solids were characterized using powder X-ray diffraction (PXRD), analysis of pair distribution function-transformed PXRD data (where appropriate), hot-stage polarized light microscopy, and differential scanning calorimetry (DSC). Drug/polymer dispersion behavior was classified using the data from each technique, independent of the others, and limitations to single-method characterization of PEG-based systems are highlighted. The data from all characterization techniques were collectively used to classify dispersion behavior, which was compared with single-technique characterization. Of the 15 combinations, only six resulted in solids whose dispersion behavior was consistently described using each standalone technique. The other nine were misclassified using at least one standalone technique, mainly because the phase behavior was ambiguously interpreted when only the data from one technique were considered. The data indicated that a suite of complementary techniques provided better classifications of the phase behavior. Of all the quenched solids, only cimetidine was fully dispersed in PEG4000, suggesting that it solidified from a completely miscible mixture of molten drug and polymer that did not phase separate upon cooling. In contrast, ibuprofen and PEG4000 completely recrystallized during preparation, whereas the remaining 13 drugs were partially dispersed in PEG4000 at this composition.
Near-infrared chemical imaging (NIR-CI) combines spectroscopy with digital imaging, enabling spatially resolved analysis and characterization of pharmaceutical samples. Hardness and relative density are critical quality attributes (CQA) that affect tablet performance. Intra-sample density or hardness variability can reveal deficiencies in formulation design or the tableting process. This study was designed to develop NIR-CI methods to predict spatially resolved tablet density and hardness. The method was implemented using a two-step procedure. First, NIR-CI was used to develop a relative density/solid fraction (SF) prediction method for pure microcrystalline cellulose (MCC) compacts only. A partial least squares (PLS) model for predicting SF was generated by regressing the spectra of certain representative pixels selected from each image against the compact SF. Pixel selection was accomplished with a threshold based on the Euclidean distance from the median tablet spectrum. Second, micro-indentation was performed on the calibration compacts to obtain hardness values. A univariate model was developed by relating the empirical hardness values to the NIR-CI predicted SF at the micro-indented pixel locations: this model generated spatially resolved hardness predictions for the entire tablet surface.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.