In-line analysis of a fluid bed pellet coating process using a combination of near infrared and Raman spectroscopy

Abstract

Near infrared (NIR) and Raman spectroscopic analyzers applied through an immersion Lighthouse Probe (LHP) were used for simultaneous in-line monitoring of a fluid bed pellet coating process. Multivariate curve resolution analysis of data, collected from four pilot-scale batches, has shown that the two techniques deliver complementary information about the process and their combination may be synergistic. This data analysis enabled a much better understanding of some of the process observations and also gave some interesting insights into the best way to use the techniques themselves. PLS regression analysis of the product moisture and the quantity of coating material sprayed was performed using NIR and Raman data blocks both separately and in combination. The performance of method combination compared to individual techniques is analyzed and discussed. Copyright © 2010 John Wiley & Sons, Ltd.