Prediction of differential scanning calorimetry curve of theophylline direct compression model tablet using Raman spectra

The aim of this study was to predict Differential Scanning Calorimetry (DSC) curves of pharmaceutical direct compression model tablet by chemometrics using Raman spectroscopy, measured in a nondestructive and contact-free manner. Theophylline anhydrate, theophylline hydrate, and micro-crystalline cellulose were ground and mixed with a mortar and pestle to obtain various formulations. The tablets were then measured with a handheld Raman spectrometer. DSC curves of the ground powder were measured, and the first-order derivative DSC curves were calculated. The Partial Least Squares (PLS) models were constructed to predict the calories at each temperature from the Raman spectra, and the regression vector of the constructed PLS models was plotted with 2D mapping (Raman wavenumber, Temperature, Regression coefficient). This mapping system allowed for the analysis of the correlations between Raman shift and DSC endothermic peaks.