LAUSR

This study describes the development of an in-line NIR fluid bed granulation monitoring method, evaluating the capacities of a Viavi microNIRTM spectrometer during the process. For this purpose, an experimental design was set up, having binder spraying rate and inlet air temperature as independent inputs. NIR spectra were recorded continuously, the acquired data was pre-processed and principal component analysis was performed in order to develop a multivariate model. Applying the developed model, the evolution of each parameter could be well underlined, moreover the projected score line plots obtained from the data analysis enabled the capturing of the variance related to moisture content and granule growth throughout the granulation process. This allowed the ulterior time optimization by an accurate determination of the process end point. In order to evaluate the sensibility of the monitoring method, additional runs were performed varying the binder spraying rate and evaluating the registered moisture levels. The results showed that the in-line monitoring of moisture content facilitates the process control, increasing its reliability and leading to an optimum product. Furthermore, the NIR process analyser based on linear variable filter technology proved to be sensitive enough to clearly distinguish each step of the process.