Evaluation of Risk and Benefit in the Implementation of Near-Infrared Spectroscopy for Monitoring of Lubricant Mixing

Abstract

On-line near-infrared spectroscopy (NIRS) was used to monitor lubricant blending to ensure the quality of the final dosage form. A quantitative multivariate NIR model was developed using different lubricant concentration levels. Real-time model predictions correlated well with the expected lubricant concentration during blending, which allowed determination of blend quality. The significance of sensor location on the blender at different fill levels was evaluated. The capability of this application was further assessed by real-time study of blending dynamics under varying process conditions and raw material attributes. The response of the developed NIR method to sudden spikes in analyte concentration, changes in raw material attributes, and perturbations to standard mixing procedures was evaluated. This study allows an understanding of risk factors associated with the implemented technology, and its ability to accurately monitor the process events. Furthermore, it highlights the importance of proper selection of processing conditions and raw material attributes to improve process robustness.

Keywords:

• Process Analytical Technology (PAT)
• process analyzers
• near-infrared spectroscopy
• process variation
• raw material characterization
• mixing
• magnesium stearate