Issue link: https://www.e-digitaleditions.com/i/1331851
Tablets & Capsules January/February 2021 35 ulate real process deviation, and sensor data was recorded simultaneously from both NIR probes to evaluate the impact of the probe's position. Results The acquired NIR data was pretreated to smooth interpretation, as shown in Figures 1 and 2. Comparing these graphs, for the through-the-window mounting application, the NIR sensor in the feed frame reported higher variation than the sensor in the loading system pipe. In particular, at 1,391 nanometers Figure 2 clearly shows talc variation during the process. about chemical and physical attributes of the raw mate- rials, intermediate materials, or final products. In a tab- let press, these critical areas include the feeding system, where the raw materials go into the press, and the dis- charge chute, where the tablets exit the press. Achieving accurate NIR control on the tablet press discharge chute is very difficult at production speeds, so several probes should be mounted in the machine to pro- vide 100 percent control of production. It's important to ensure that variations are detected before the entire batch has already been produced. The feeding system has two strategic positions for NIR placement: near the feeding paddles, whose rota- tion could cause variation in the formulation's density or physical characteristics; and in the loading system pipe, through which the blend flows to reach the process area. This article describes a case study evaluating the potential for using NIR spectroscopy to identify blend variations on the feed frame versus on the loading system pipe of a tablet press feeding system. Materials and methods The tests were conducted using a medium-sized, production-scale tablet press (IMA Prexima 300). The press features specially designed seals and protections to ensure complete separation between the processing and mechanical areas. The design allows the processing area to be fully accessible once the external doors are opened, while access to the mechanical area is required only for maintenance. The tablet press' compression support is based on three columns linked together by two cast-iron struc- tures. The compression rollers are incorporated within the two cast-iron structures and supported on both sides. This sturdy structure is essential for producing high-quality tablets and ensures that the press can reli- ably achieve both pre- and main compression forces up to 100 kilonewtons. A Viavi MicroNIR PAT-U probe was positioned on the tablet press feed frame and a Viavi MicroNIR PAT-W probe was positioned on the loading system pipe. The probes were mounted using the standard through-the- window method. The case study used three lactose-based tablet formula- tions that varied only in the percentage of talc content (1 percent, 3 percent, and 5 percent), as shown in Table 1. Formulation 2 represents the manufacturer's desired formulation, while Formulations 1 and 3 are intended to simulate deviations from the correct blending process due to, for example, segregation or demixing. Such devi- ations would result in either lower (1 percent) or higher (5 percent) talc concentrations, which the NIR sensors should detect. The trials were performed by loading all three formu- lations into the tablet press hopper, with the lower talc formulation first, the desired formulation second, and the higher talc formulation last. During tableting, the formu- lations were fed into the tablet press continuously to sim- Figure 1 Data acquired from feed frame NIR sensor (pretreated) First derivative 0.02 0.01 0 -0.01 -0.02 Wavelength (nanometers) 908 982 1057 1131 1205 1280 1354 1428 1503 1577 1651 Ingredient Percentage Formulation 1 Formulation 2 Formulation 3 Lactose 98 96 94 Talc 1 3 5 Magnesium stearate 1 1 1 Table 1 Figure 2 Data acquired from loading system pipe NIR sensor (pretreated) 908 982 1057 1131 1205 1280 1354 1428 1503 1577 1651 First derivative 0.02 0.01 0 -0.01 -0.02 Wavelength (nanometers)