Department of Chemical & Biotechnological Engineering, Université de Sherbrooke, Sherbrooke, Québec, Canada.
Pharm Dev Technol. 2022 Apr;27(4):448-458. doi: 10.1080/10837450.2022.2078348. Epub 2022 May 25.
Concentration monitoring inside a tablet press feed frame is important not only to assess the composition of the powder blend compressed into tablets but also to detect quality affecting phenomena such as powder segregation. Near infrared spectroscopy has been successfully used to monitor powder concentration inside the feed frame; however, so far, this methodology does not provide information on local spatial variability, since it probes a very small area of powder sample. Near infrared chemical imaging (NIR CI) has the potential to improve process monitoring because it can simultaneously acquire a plurality of spectra covering nearly the entire width of the feed frame, thereby making it possible to detect local variations in powder concentration. The present work uses both NIRS and NIR CI to monitor the concentration of Ibuprofen and Ascorbic acid in multi-component mock pharmaceutical blends flowing through the feed frame of an industrial tablet press. The concentrations of Ibuprofen and Ascorbic acid were successfully monitored in multi-component powder blends. NIR spectral wavelength ranges and pre-treatments were simultaneously optimized via a genetic algorithm. N-way PLS approach for concentration monitoring was found to be more suitable than regular PLS when analyzing spectral images and provided the ability to visualize spatial segregation.
压片机料斗内的浓度监测不仅对于评估压制成片剂的粉末混合物的组成很重要，而且对于检测影响质量的现象（如粉末分离）也很重要。近红外光谱法已成功用于监测料斗内的粉末浓度；然而，到目前为止，由于该方法仅探测到粉末样品的一个非常小的区域，因此无法提供有关局部空间变化的信息。近红外化学成像（NIR CI）具有改善过程监测的潜力，因为它可以同时采集覆盖料斗几乎整个宽度的多个光谱，从而可以检测到粉末浓度的局部变化。本工作使用近红外光谱和近红外化学成像来监测通过工业压片机料斗流动的多组分模拟药物混合物中布洛芬和抗坏血酸的浓度。成功地监测了布洛芬和抗坏血酸在多组分粉末混合物中的浓度。通过遗传算法同时优化了近红外光谱波长范围和预处理。与分析光谱图像时的常规 PLS 相比，N 路 PLS 方法更适合浓度监测，并能够可视化空间分离。
