Peters Johanna, Bartscher Kathrin, Döscher Claas, Taute Wolfgang, Höft Michael, Knöchel Reinhard, Breitkreutz Jörg
Heinrich-Heine-University Düsseldorf, Institute of Pharmaceutics and Biopharmaceutics, Düsseldorf, Germany; NextPharma Waltrop, Pharbil Waltrop GmbH, Waltrop, Germany.
NextPharma Waltrop, Pharbil Waltrop GmbH, Waltrop, Germany.
Talanta. 2017 Aug 1;170:369-376. doi: 10.1016/j.talanta.2017.03.105. Epub 2017 Apr 5.
Microwave resonance technology (MRT) is known as a process analytical technology (PAT) tool for moisture measurements in fluid-bed granulation. It offers a great potential for wet granulation processes even where the suitability of near-infrared (NIR) spectroscopy is limited, e.g. colored granules, large variations in bulk density. However, previous sensor systems operating around a single resonance frequency showed limitations above approx. 7.5% granule moisture. This paper describes the application of a novel sensor working with four resonance frequencies. In-line data of all four resonance frequencies were collected and further processed. Based on calculation of density-independent microwave moisture values multiple linear regression (MLR) models using Karl-Fischer titration (KF) as well as loss on drying (LOD) as reference methods were build. Rapid, reliable in-process moisture control (RMSEP≤0.5%) even at higher moisture contents was achieved.
微波共振技术（MRT）是一种用于流化床制粒中水分测量的过程分析技术（PAT）工具。即使在近红外（NIR）光谱适用性有限的情况下，例如有色颗粒、堆积密度变化较大时，它在湿法制粒过程中也具有很大潜力。然而，以前围绕单个共振频率运行的传感器系统在颗粒水分约高于7.5%时显示出局限性。本文描述了一种工作在四个共振频率的新型传感器的应用。收集并进一步处理了所有四个共振频率的在线数据。基于与密度无关的微波水分值的计算，使用卡尔费休滴定法（KF）以及干燥失重法（LOD）作为参考方法建立了多元线性回归（MLR）模型。即使在较高水分含量下，也实现了快速、可靠的过程中水分控制（RMSEP≤0.5%）。
