Tablets & Capsules

TC0121

Issue link: https://www.e-digitaleditions.com/i/1331851

Contents of this Issue

Navigation

Page 45 of 67

36 January/February 2021 Tablets & Capsules changes over time and if there is some clear variation indicating that the process is moving toward poor tab- let quality. The moving block algorithm allows you to summarize variance in blocks of spectra collected over the time of the process. The graphs show the presence of three different areas corresponding to the three differ- ent formulations. MBM allows you to select a window in which tablet quality is guaranteed. The green line in Fig- ure 4 indicates tablets with optimal content, while the red line indicates tablets with variations. Using the principal component analysis algorithm, it's possible to confirm the presence of the three different formulations, as shown in Figure 5, and to monitor inter- mediate phases (circled in red) where the process drifted. The drifting was caused by mixing between the formula- tions because all three blends were loaded into the hop- per and run continuously. Conclusions This case study demonstrates the suitability of using NIR probes in the Prexima tablet press and the possibil- ity of monitoring formulation characteristics in real time to identify granulation, blending, or handling problems. The study also determined the impact of probe location on data acquisition for through-the-window application of NIR spectroscopy. Selecting a PAT NIR probe in the loading system pipe could be advantageous for detecting process deviations and stopping the machine before the formulation has passed through the process area. Because the graph is smoother in the loading system pipe application than in the feeding frame application, attention can be focused on blend fingerprints without other variables such as pad- dle movement influencing the data. Using the feed frame probe location could be advantageous only for monitor- ing the formulation just before tablet compression. In such a case, using an immersion-mounted probe rather than a through-the-window-mounted probe could be an option to optimize results. Finally, regarding the future, particularly as more tablet- ing processes move toward continuous manufacturing, NIR application can help predict process deviations, for exam- ple by using MBM in real time to fix the window in which quality is guaranteed. After some measurements outside the fixed limits, an alarm can stop the machine to avoid the production of out-of-specification tablets. Manufactur- ers can also implement statistical post-process NIR analysis to redundantly verify tablet quality and ensure that what goes into the tablet press matches what comes out. T&C Federica Giatti is compression technologist, Caterina Funaro is R&D process laboratory manager, Fabrizio Consoli is technical department manager for tablet presses, and Fab- riano Ferrini is product manager for tablet presses at IMA Active, a manufacturer of solid dose processing equipment (+39 051 651 4111, www.ima.it). For easier data interpretation, the case study was led focusing on the loading system pipe position. The pres- ence of feeder paddles and their rotating speed can change the formulation density and, consequently, the fingerprint of the NIR data. The NIR sensor could detect the presence of air generated by the movement of the paddles or empty space if the blend flows poorly and sticks to the paddles. Also, the paddle speed can change during the process, which affects the formulation density. What the NIR sensor might detect as a blend variation may only be an automatic adjustment to ensure accurate dosing for tablet quality. The moving block standard deviation (MBSD) and moving block mean (MBM), as shown in Figures 3 and 4, can help to understand how the blend fingerprint Figure 3 Moving block standard deviation B1 B411 B438 B465 B492 B519 B546 B573 B600 B627 B654 B384 B357 B303 B330 B276 B249 B222 B195 B168 B141 B114 B87 B65 B22 B43 Standard deviation 9e-05 8e-05 7e-05 6e-05 5e-05 4e-05 3e-05 2e-05 1e-05 Block Figure 4 Moving block mean B1 B412 B439 B466 B493 B520 B547 B574 B601 B628 B655 B385 B358 B304 B331 B277 B250 B223 B196 B169 B142 B115 B88 B66 B22 B44 Mean 0.0031 0.0030 Block Figure 5 Principal component analysis PC-2 PC-1 0.002 0.001 0 -0.001 -0.002 -0.001 0 0.001 0.002 0.003

Articles in this issue

Links on this page

Archives of this issue

view archives of Tablets & Capsules - TC0121