Calibration-free quantitative application of in situ Raman spectroscopy to a crystallization process

In this work, in situ Raman spectroscopy is applied in a quantitative manner to crystal suspensions during crystallization in order to monitor a solvent-mediated polymorph transformation without the use of a calibration model. Assuming a linear dependency of the Raman signal intensity on the solute and on the solid concentrations of both solid-state forms, the measured time-resolved Raman spectra are fitted directly using a detailed model that describes the time evolution of the process. The applicability of this novel method is demonstrated thoroughly through the application to synthetic data of unseeded and seeded transformations as well as to various seeded polymorph transformation experiments. The resulting concentration profiles show a good agreement with the concentrations obtained by a multivariate calibration model. Additionally, the estimated kinetic parameters are compared to parameters obtained through fitting the solid-phase composition profiles that result from the calibration. The discrepancy between the estimated model parameters is small, and essentially the same descriptive process model is obtained. However, by fitting the time-resolved Raman spectra directly, a significant amount of calibration effort can be avoided.