EUROPEAN CONGRESS AND EXHIBITION ON ADVANCED MATERIALS AND PROCESSES - EUROMAT 2021

In cocrystalization-based processing of pharmaceuticals, Raman spectra is utilized as a standard tool for quality control. However, such analysis suffers from the lack of relevant tools for realtime analysis of spectra. Here we report an automation method to facilitate such analysis. For this purpose, we coupled the nonlinear multivariable functional series approximation with molecular fingerprints as derived from density functional theory calculations. As the case study, we considered the cocrystalization process of ibuprofen and nicotinamide. We showed that for any mixture, it is possible to identify the contribution (probability) of different molecular fingerprints by just reading the offline and/or online Raman spectra, with significant level of confidence. Therefore, a realtime monitoring of continuous cocrystalization-based pharmaceuticals processing is achieved. Using the model, it is possible to identify and adopt the operating conditions of system on realtime toward fingerprint (molecular structure) of interest i.e., cocrystals.