Freund Vector

A process analytical technology (PAT) method relying on near infrared (NIR) diffuse reflectance spectrometry was developed to determine the average coating thickness applied and the weight gained by tablets during coating process. The NIR device was installed inside the pan directly in contact with the cores under processing and relevant spectra were recorded in real time during the entire coat-ing process. The method required principal component analysis (PCA) and partial least squares regression (PLS). The primary methods to calibrate PLS are tablet weight gain and coating thickness, they were carried out using the chemometric methods. The PLS-model was employed as the calibration curve for an online process monitoring and was based on the spectra obtained from the tablets during coating, pretreated with SNV and first derivative, and ordered into one object. This way, the method developed could be advantageously used for in-line PAT monitoring. Additional characterization insights were attained by calibrating the spectra versus the weight gain data obtained from at-line tablet weighing. Overall, the possibility to monitor multiple parameters using just one device and one spectrum was demonstrated.