کد مقاله | کد نشریه | سال انتشار | مقاله انگلیسی | نسخه تمام متن |
---|---|---|---|---|
236577 | 465680 | 2013 | 10 صفحه PDF | دانلود رایگان |

• A lab-scale process was monitored at several positions with NIR.
• Multivariate methods were used for spectral interpretation.
• Calibration was done on moving samples.
• Differences in blending behavior inside the vessel were identified.
The production of pharmaceutical dosage forms relies on efficient blending of granular components. By monitoring in real-time the blending processes via non-invasive techniques, such as near-infrared (NIR) spectroscopy, the blending end-point can be determined and homogeneity can be guaranteed for a given material blend. In this study an experimental setup with multiple NIR probes attached to a lab-scale blending vessel is presented. The probes were connected to a Fourier-transform NIR spectrometer via an optical fiber switch. Thereby, the blending processes could be “quasi-simultaneously” monitored at multiple positions. Differences in blending behavior in the monitored positions inside the vessel were identified, which would be impossible using a single probe system.The interpretation of the NIR-spectra was done using a chemometric model based on partial least squares regression. Premixed samples with varied content of an active pharmaceutical ingredient were used for calibration. To prevent segregation problems during calibration, it was performed on a rotating disk outside the vessel.Using this multiple-probe setup in combination with the developed chemometric model, segregation and blending inside the vessel were analyzed by observing the temporal and spatial concentration changes, without interruptions caused by manual sampling and/or pausing of the process.
A blending process of pharmaceutical powders was monitored with near-infrared spectroscopy. Positioning multiple fiber optic probes enabled to temporally and spatially resolve the powder dynamics during blending.Figure optionsDownload as PowerPoint slide
Journal: Powder Technology - Volume 243, July 2013, Pages 161–170