Chemical imaging method of vibrational spectroscopy, which provides both spectral and spatial information, creates a three-dimensional (3D) dataset with a huge amount of data. When the components of the sample are unknown or their reference spectra are not available, the classical least squares (CLS) method cannot be applied to create visualized distribution maps. Raman image datasets can be evaluated even in such cases using multivariate (chemometric) methods for extracting the needed hidden information. The capability of chemometrics-assisted Raman mapping is evaluated through the analysis of pharmaceutical tablets (considered as unknown) with the aim of estimating the pure component spectra based on the collected Raman image. Six chemometric methods, namely, principal component analysis (PCA), maximum autocorrelation factors (MAF), sample-sample 2D correlation spectroscopy (SS2D), self-modeling mixture analysis (SMMA), multivariate curve resolution-alternating least squares (MCR-ALS), and positive matrix factorization (PMF), were compared. SMMA was found to be the best choice to determine the number of components. MCR-ALS and PMF provided the pure component spectra with the highest quality. MCR-ALS was found to be superior to PMF in the estimation of Raman scores (which correspond to the concentrations) and yielded almost the same results as CLS (using the real reference spectra). Thus, the combination of Raman mapping and chemometrics could be successfully used to characterize unknown pharmaceuticals, identify their ingredients, and obtain information about their structures. This may be useful in the struggles against illegal and counterfeit products and also in the field of pharmaceutical industry when contaminants are to be identified.
Crystalline and amorphous interfaces were formed and modified in multiphase composite and pharmaceutical systems. Nanosize drug of antacid effect was prepared in continuous extrusion process. The structural features were analyzed using (micro-) thermal analytical and (micro-) Raman spectrometric methods combined with chemometric evaluation.
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.