(M1230-08-46) Functionality and Properties Influences of Microcrystalline Cellulose on Wet Granular Processability in a Twin Screw Granulation

Purpose: To investigate the processing window of formulations with microcrystalline celluloses of differing physical properties prepared by wet granulation in two sizes of twin screw extruder important to product development versus production.

Methods: Model formulations were prepared with 80 wt% α-lactose monohydrate (Pharma FlowLac® 100, Meggle) blended with 20 wt% microcrystalline cellulose provided by IFF Nutrition & Biosciences. Blends were pre-dried at 50 ℃ to reach consistent starting moisture of ~1 wt%. The microcrystalline cellulose (MCC) examined included Avicel® PH102, PH103, PH301, and PH302. Continuous granulation was performed with the different formulations in 18 mm (MICRO18) and 27 mm (ZSE 27HP) twin screw extruders (Leistritz Extrusion) at 1 kg/h and 3 kg/h, respectively and constant screw speed of 200 RPM. Identical screw designs were used on the two extruders. An aqueous binder solution of 2% METHOCEL™ E3 LV (IFF Nutrition & Biosciences) was introduced by injection in all cases. Samples were collected at different liquid-to-solid ratios (L/S) up to a tested maximum of 50%, to establish the processing window for granulation of each formulation. For increasing L/S, the lower L/S boundary corresponded to the first time when more than 50% of the tested mass was found on the sieve of 500 mm openings and similarly, the upper L/S boundary corresponded to the first time more than 50% of the tested mass was found on the sieve with 2100 mm openings. Granule characterization involved particle size analysis using a Rotap sieve shaker (W.S. Tyler Inc).

Results: The formulations considered four MCC grades, PH102, PH103, PH301, and PH302 as excipients in the wet granulation process. They are characterized as having different bulk density and particle size. PH102, and PH302 have the same nominal particle size of 100 mm while PH103 and PH301 have a smaller nominal particle size of 50 mm, according to the vendor. The two 300-series MCCs have higher bulk densities (approximately 0.4 g/cc) than the others (approximately 0.3 g/cc). Overall, process performance by twin screw granulation was similar for all four grades, differentiated mostly based on using the 100- versus 300-series of MCCs. Figure 1 shows the granulation ranges for the four MCC in the two extruders. It was found on the smaller extruder (18 mm), formulations could be granulated from ~12% up to ~50% with PH102 and PH103, while ~12% up to 36% with PH301 and PH302. The upper L/S boundaries were lower in this machine when the higher bulk density formulations (PH301 and PH302) were tested. The influence of MCC grade on granulation varied more on the 27 mm extruder than the 18 mm extruder, with a slight rise in the lower L/S boundary (18%) for PH102 and PH103, and a lower upper L/S boundary for PH102, PH103, PH301 and PH302 (46%, 46%, 34% and 32% respectively). The higher bulk density MCCs had narrower granulation ranges on both 18 mm and 27 mm extruders. The particle size distribution (PSD) of these four MCC formulations are shown in Figure 2, given for a L/S ratio of 32% on both extruders. On the 18 mm extruder, samples had similar PSD with a typical bimodal shape peaking at 640 mm and 1750 mm. On the larger extruder, the PSD showed a broader peak between 640 and 1015 mm and a greater fraction of coarse centered at 1750 mm. The 300-series MCC yield few fines and a much higher fraction of coarse on both extruders. The 100-series MCC formulations were much more likely to display significant fines and a narrower peak below 1000 mm. There was also a greater chance of observing highly elongated ‘noodle-like’ granules on 27 mm extruder, which is related to a rolling mechanism between the screw flight and barrel wall.

Conclusion: The processability of model MCC formulations by continuous twin screw granulation were found largely distinguished by the bulk density of the excipient. The selected size of the extruder was also found to be influence though to a lesser degree compared to the bulk density of MCC. These distinctions are limited to the analysis made in this abstract based on the granule size while further characterizations may find other factors of distinction.


Figure 1. Upper and Lower L/S ratio range for MCC on (a) 18mm TSE and (b) 27mm TSE. Dashed line shows the chosen L/S ratio for PSD analysis


Figure 2. Particle size distribution for MCC granules on (a) 18mm TSE and (b) 27mm TSE with L/S ratio of 32%