(492e) In-Depth Experimental Analysis of Pharmaceutical Continuous Twin-Screw Wet Granulation in View of the Development of a Generic Compartmental PBM

Recently, a mechanistic Population Balanced Model (PBM model) has been developed and calibrated for twin-screw wet granulation for a few specific formulations [1]. In-depth data collection and analysis was performed to investigate the influence of process parameters on granule formation mechanisms along the twin-screw granulator barrel and resulting granule Critical Quality Attributes (CQAs) [2]. A logic next step is to make this model applicable for any formulation by developing a generic PBM for continuous twin-screw wet granulation, predictive for granule size distribution (GSD) in function of both process parameters and raw material properties. Given the potential importance of granule porosity on granule formation and final drug product, information with regard to this property was collected per granule size fraction along the length of the granulator barrel. This information will serve for the development of a generic 2D-PBM model.

MATERIALS AND METHODS

Different commercially available Active Pharmaceutical Ingredients (APIs) with major differences in physicochemical properties, primarily solubility and wettability, were selected. Milled and non-milled APIs were utilized to investigate the effect of API particle size on granule formation and final granule CQAs.

Blends were prepared using a fixed formulation base consisting of microcrystalline cellulose, HPMC, lactose, and 5% or 50% API for the low and high dose formulations, respectively.

APIs, excipients and blends were thoroughly characterized for particle size distribution, bulk and tapped density, dissolution rate, water holding capacity, solubility, residual moisture content, contact angle, ring shear test, true density, angle of repose, dispersion charge, and specific surface area.

Granules were produced using the high shear twin-screw wet

granulation module of the ConSigmaâ„¢-25 system (GEA Pharma systems, Collette, Wommelgem, Belgium).

A fixed screw configuration was used with 2 kneading compartments, each consisting of 6 kneading elements arranged at 60° stagger angle. Both kneading zones were separated by a conveying element and at the end of the granulator barrel, an additional conveying element was added together with three size-control elements.

In order to characterize the granule size distribution and granule porosity along the length of the twin-screw granulator barrel, samples were collected in four well-defined compartments: (i) the wetting zone, (ii) after the first kneading zone, (iii) after the second kneading zone and (iv) at the granulator outlet (after the size control elements) using the method described by Verstraeten et al. [2].

A 5-level Rotatable Central Composite Circumscribed Design of Experiments (CCC) (MODDE 12, MKS Umetrics, Umeå, Sweden) was performed to evaluate the effect of process parameters liquid-to-solid ratio (L/S), material throughput and screw speed on local and final granule size distribution and porosity. The granulator barrel temperature was kept constant at 25 °C. Granules collected at the different locations were dried at ambient temperature and relative humidity (25 °C, 40% RH) for at least 24 hours and granule size distribution (Partan^3D, Microtrac Inc, Montgomeryville PA, USA) and granule porosity by mercury intrusion porosimetry (Porotec, Thermo Finnigan, California, US) were measured. The flowability of the granules at the end of the granulator was evaluated by the Hausner ratio. Also, friability was measured at the end of the barrel to evaluate the effect of granulation process parameters and raw material characteristics on granule strength.

RESULTS AND DISCUSSION

The experimental results not only showed prominent similarities but also differences with what has been observed before due to the different starting material properties [2]. The effect of TSG process parameters material throughput, screw speed and L/S ratio on granule size distribution along the length of the barrel was highly dependent on formulation properties. Furthermore, L/S ratio remained the most important factor dictating the size distribution of granules collected at the outlet of the twin-screw granulator.

However, barrel fill level, dictated by screw speed and material throughput, also had a relevant effect on granule size distribution. While the impact of the narrow chopper section on the granule particle size has been shown to be insignificant at low L/S ratios [2], in this study, the size control elements at the end of the granulator barrel appears to have a major effect on granule size, both at low and high L/S ratios.

CONCLUSION

This experimental study provided valuable information on the granulation mechanisms in function of process parameters and raw material attributes. Interestingly, the results indicate that the size-control elements and raw material properties have an important effect on granule formation and final granule CQAs. This extensive data collection and analysis is a first crucial step in the development of a generic 1D as well as a 2D PBM model for twin-screw wet granulation.

References

[1] Van Hauwermeiren, D., Verstraeten, M., Doshi, P., am Ende, M. T., Turnbull, N., Lee, K., ... Nopens, I. (2018). On the modelling of granule size distributions in twin-screw wet granulation: Calibration of a novel compartmental population balance model. Powder Technology, 341, 116–125. https://doi.org/10.1016/j.powtec.2018.05.025

[2] Verstraeten, M., Van Hauwermeiren, D., Lee, K., Turnbull, N., Wilsdon, D., am Ende, M., De Beer, T. . (2017). In-depth experimental analysis of pharmaceutical twin-screw wet granulation in view of detailed process understanding. International Journal of Pharmaceutics, 529(1–2), 678–693. https://doi.org/10.1016/j.ijpharm.2017.07.045