Quantification of a cocrystal and its dissociated compounds in solid dosage form using transmission Raman spectroscopy.

The performance of transmission Raman spectroscopy (TRS) for quantifying a cocrystal and its dissociation in solid dosage form was investigated. Some tablets containing 0%-20% (w/w) of a cocrystal of carbamazepine (CBZ)/succinic acid (SUC), 0%-4% of CBZ, 0%-4% of SUC, and 75%-99% of D-mannitol were prepared. The Raman spectra of these tablets were preprocessed using the standard normal variate (SNV) or multiplicative scatter correction (MSC) as well as the Savitzky Golay second derivative, and then, these were used to generate calibration models using partial least squares (PLS) regression. The performance of the model was superior when the MSC preprocessing spectra of the cocrystal between 200 and 1800 cm-1 were used for calibration. The determination coefficient of the PLS calibration curve for the CBZ/SUC cocrystal between 200 and 1800 cm-1 with MSC was 0.97, root mean square error of cross validation (RMSECV) was 1.16, and root mean square error of prediction (RMSEP) was 1.10. As in the case of the CBZ/SUC cocrystal, the performance of the model was superior when the MSC preprocessing spectra of CBZ and SUC between 200 and 1800 cm-1 were used for calibration. These data suggest that TRS is useful for quantifying a cocrystal and its dissociation compounds in solid dosage forms.

Pharmaceutical quantification with univariate analysis using transmission Raman spectroscopy.

The purpose of this study was to investigate the quantification performance of transmission Raman spectroscopy with univariate analysis. Model dosage forms containing acetaminophen and an excipient, lactose monohydrate, were prepared. The Raman spectra of the tablets were obtained using the modes of transmission, backscattering micro-spectroscopy, and wide area illumination. Calibration curves for quantification of acetaminophen in the tablets were created using peak heights of the Raman spectra. Of the three modes of measurement, the quantitative results by transmission had the highest correlation with those by conventional UV-vis methods. In the validation of quantification by the transmission mode with univariate analysis, a certain degree of daily variation was confirmed. Additionally, quantitative results using peak heights were compared with those of partial least squares (PLSs) multivariate analysis. The root mean square error of prediction (RMSEP) suggested that quantification using PLS provided better precision than the peak height method as expected. However, content uniformity test using large sample sizes by the Raman spectra is not required to be very highly predictive because they usually employ non-parametric criteria and include wide specification ranges. Therefore, univariate analysis using transmission Raman spectroscopy was a suitable quantitative method for conducting content uniformity tests of large sample sizes.