Issue link: https://www.e-digitaleditions.com/i/446555
the pattern of the textured ribbon. The more accurate average thickness over a wider area will yield lower spatial resolution, but the RI values will be more accurate. Understanding the relationships between process parameters and the critical quality attributes (CQAs) of roller-compacted ribbons is important for small-scale development, scale-up, and commercial manufacturing. Figure 10 shows the results of an experiment to measure how roll speed affects bulk density. The 2D maps (figures 10a, 10b, and 10c) show that RI decreases as roll speed increases. In Figure 10d, the overlaid average cross-section projection line plots of RI are averaged over 40 millimeters in the machine direction (X) versus the ribbon width (Y) for the three roll speeds. In Figure 10e, this average cross-section projection line plot of RI was converted to the line plot of apparent density using the equation for correlating between average RI and apparent density (Figure 10f) that was established independently on MCC compacts. The plot in Figure 10e shows that the apparent density is highest in the middle of the ribbon's width and tapers off toward the edges. Ribbon density decreases as roll speed increases, which corresponds to a shorter dwell time in the nip region of the rollers. Figure 11 shows the results of an experiment to measure how the size of the roller gap affects apparent density. As the average cross-section projection line plots in figures 11c and 11d show, apparent density decreases as the roller gap increases. The terahertz results (red line) are superimposed on the density data that were obtained by cutting the ribbon into sections (blue line). This section-and-weigh method generated fewer points across the ribbon because it was difficult to cut the ribbon into small pieces. More important, the standard deviation is high compared to that of the terahertz results, where the standard deviation is less than the height of the marker on the plot. In-line density measurement Terahertz 2D mapping of roller-compacted ribbons yields detailed quantitative spatial density and porosity information, which enables you to optimize process parameters and ensure different types of equipment provide functional equivalence. Once the process parameters are set, the objective is to monitor the process over time, and this can be done with an in-line terahertz analyzer (Figure 12). In production, ribbons typically travel at linear speeds of 50 to 150 millimeters per second, which is compatible with terahertz sampling intervals of less than 1 second. Different sampling patterns can be used, and figures 12a and 12b illustrate how to use a small spot size to monitor a single track and traverse the ribbon. Figure 12c shows how a cylindrical lens is used to generate a large elliptical spot, which provides efficient sampling over a large area and improves the signal-to-noise ratio. This optical arrangement is preferred for monitoring the approach of steady-state operation at startup and to monitor short- and long-term ribbon density variations during operation. Ideally, this information would be fed back to the machine to provide real-time process control. 32 January 2015 Tablets & Capsules Figure 10 Effect of process parameters on ribbon apparent density cross-section: Variable roller speed and fixed gap (0.95 mm) and 2D RI maps 10 20 30 40 a. 1 rpm Width, Y (mm) 20 10 0 -10 -20 Length, X (mm) 1.7 1.6 1.5 1.4 1.3 1.2 1.1 1 0.9 d. Average RI cross-section projection line plot for 1, 3, and 5 rpm Average RI 10 20 30 40 b. 2 rpm Width, Y (mm) 20 10 0 -10 -20 Length, X (mm) 10 20 30 40 1.8 1.5 1.2 0.9 0.6 0.3 c. 5 rpm Width, Y (mm) 20 10 0 -10 -20 Length, X (mm) -30 -20 -10 0 10 20 30 1 rpm 3 rpm 5 rpm 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 e. Average apparent density cross-section projection line plot for 1, 3, and 5 rpm Apparent density (g/cm 3 ) -30 -20 -10 0 10 20 30 1 rpm 3 rpm 5 rpm 2 1.5 1 0.5 0 f. Calibration curve for average RI vs. apparent density Average RI 0 0.5 1 1.5 Refractive index 5 0.44 3 (Apparent density) 1 1.01 Width, Y (mm) Width, Y (mm) Apparent density (g/cm 3 )