Online Estimation and Monitoring of Diastereomeric Resolution Using FBRM, ATR-FTIR, and Raman Spectroscopy

This paper first demonstrates that the estimation of fractional solid composition of two diastereomers is possible by incorporating Raman spectra, slurry density, and temperature into a partial least squares (PLS) model. Since slurry density measurement is not readily available, the online estimation is obtained through another PLS model that utilizes online Raman spectroscopy, focus beam reflectance measurement (FBRM), attenuated total reflection Fourier transform infrared (ATR-FTIR), and/or additional process information such as temperature and agitation rate to infer slurry density online. It is argued that the model, which infers slurry density from IR spectra, gives more accurate estimation of the fractional composition of two diastereomers. Furthermore, the paper also successfully demonstrates through a third and a more general PLS model the ability of real time monitoring several crystallization process variables by the use of the aforementioned online measurements. Besides monitoring the changing diastereomeric composition and slurry density, one can also monitor the solute concentration of each diastereomer in a complex crystallization system.