Raman Spectroscopy Combined with Small and Wide Angle X-Ray Scattering as a Non Destructive Quality Control Tool for Powder Compression Analytics in Pharmaceutical Applications

This study evaluates the feasibility of a combined use of small-wide angle X-ray scattering (SWAXS) and Raman spectroscopy as a non-invasive process analytical tool (PAT) for the analysis of pharmaceutical granulation and compression procedures. Placebo formulations containing Lactose monohydrate (LM, 49 wt %), microcrystalline cellulose (MCC, 49 wt %), silicone dioxide (SiO2, 1 wt %) and magnesium stearate (MgSt) were subjected to iterative treatment cycles comprising compression, sieving and reaming. The morphology of the raw powders, and characteristics of the respective compressed pellets, was examined after every treatment cycle. Correlation analysis among compression force (F) and tablet hardness (H) revealed an increasing linearity, indicated by the correlation coefficient (r), with continuative granulation from 0.95 to 0.98. The SAXS spectra showed incremental particle-particle interfaces by multiple granulations whereas the WAXS data indicated no changes in morphology during the treatment procedures. Generally, with increasing powder compaction force (i) the particles consolidated and (ii) the surface smoothened. The Raman findings were in good agreement with the SWAXS data, meaning that a decrease in Raman backscattering intensity with enhanced powder compaction force was observed. Multivariate data analysis (MVDA) modeling in form of principal component analysis (PCA) and partial least squares (PLS) regression was used to qualitatively study and quantitatively obtain information from the Raman spectra. The multivariate correlation analysis among the X-variables (Raman spectra) and Y-variables (tablet hardness) showed the same tendencies as the linear correlation analysis, i.e. the validation set (v-set) r constantly increased from 0.88 to 0.96. The final PLS model for determining the hardness of unknown samples showed a root mean squared error (RMSE) of calibration (RMSEC) of 10.12 and prediction (RMSEP) of 12.88, respectively. These findings indicate that by combining SWAXS and Raman spectroscopy into one approach a very powerful and useful new process analytical tool would be available for future pharmaceutical quality control implementations.