In-situ monitoring of pharmaceutical and specialty chemicals crystallization processes using endoscopy-stroboscopy and multivariate image analysis

This contribution presents the proof of concept of endoscopy-stroboscopy based in situ low-cost imaging of crystallization processes. This low-cost sensor currently is widely spread in the field of medical diagnosis of human vocal chords and this work presents its application in the context of pharmaceutical and chemical crystallization process monitoring. The model compounds used in this study are the active pharmaceutical ingredient (API) flufenamic acid and citric acid. Since the acquired images are colored, the second aim of the paper is to evaluate the principal component (PCA) based multivariate image analysis (MIA) as a color to gray scale transformation method, and to compare it to the National Television System Committee (NTSC) standard, which uses fixed weights. It was found that particle color, transparency, size and shape related information based on visual inspection is feasible using the endoscope-stroboscope system. The MIA results show that in the case of transparent particles the red, green, blue channels contribute equally to the total information content of color images. The acquisition price of the ATMOS endoscope is similar to that of a laboratory turbidity probe, feature which is relevant for the widespread use of this sensor. (C) 2012 The Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.