(714c) Model Aided Scale-up for High Shear Rotor-Stator Wet Milling Process

Particle size specification for an API (Active Pharmaceutical Ingredient) is a Critical Quality Attribute (CQA) that contributes to the bioavailability of the API. The crystallization process is often followed by either wet milling or dry milling. High shear rotor-stator wet milling is used extensively in the industry. However, scaling-up/down of a wet mill is non-trivial¹.

In this work, we demonstrate a new model-aided workflow for Wet Milling scale-up using Population Balance Modelling (PBM) in gFormulate®.The model-aided framework helped us successfully scale-up from the lab-scale (IKA MagicLab <100g API) to the Kilo Lab (IKA 2000/4 - 1.5 kg API) and subsequently to the plant scale (IKA 2000/10 – 40 kg API) with “Right First Time” quality. Traditional high shear wet milling scale-up is based on a single parameter (e.g., Tip Speed, Slot Events², Normalized energy³, & Shear Number¹). The prediction of milling time based on a single scaling factor hasn’t been very successful². The time required for the milling is highly non-linear. Bigger particles break very easily at the start of the milling, however, a large number of batch turnovers are required closer to the grinding limit of the material. A single scale-up parameter wouldn’t be able to characterize this behaviour. The model takes into account the detailed geometry of the rotor-stator, the non-linear change of the particle population during milling (breakage kinetics & breakage probability), and the mill residence time.

A colloidal wet mill (IKA MK) was selected over a traditional toothed rotor/stator mill based on the higher shear required to satisfy product particle size specifications close to the grinding limit. Equipment parameters including the number of rotor teeth, stator teeth, & groove size were measured in this work. The breakage kinetics, selection function, & the grinding limit⁴ for the PBM were estimated based on the Lab experiments in IKA MagicLab. It has been shown in the literature that the breakage kinetics would scale directly with the Shear Number¹. The breakage kinetics was directly proportional to the Shear number induced by the equipment geometry. However, Shear Number alone was not sufficient to predict the performance on scale-up from the MagicLab to the KiloLab scale. The model was refined at the Kilo Lab scale to estimate the breakage kinetics coefficient for Shear Number in the Kilo Lab Scale (IKA 2000/4). All other breakage kinetics & function parameters were kept constant. This model was subsequently transferred to the Plant scale (IKA 2000/10) based on the refined coefficient from the Kilo Lab-scale (IKA 2000/4). The model was used to understand the sensitivity of the Key Process Parameters (Gap width, rpm, residence time in the mill) on the milling time & yield.

References:

1. Luciani, C. V., Conder, E. W. & Seibert, K. D. Modeling-Aided Scale-Up of High-Shear Rotor–Stator Wet Milling for Pharmaceutical Applications. (2015) doi:10.1021/acs.oprd.5b00066.
2. Harter, A., Schenck, L., Lee, I. & Cote, A. High-Shear Rotor–Stator Wet Milling for Drug Substances: Expanding Capability with Improved Scalability. Org. Process Res. Dev. 17, 1335–1344 (2013).
3. Engstrom, J., Wang, C., Lai, C. & Sweeney, J. Introduction of a new scaling approach for particle size reduction in toothed rotor-stator wet mills. Int. J. Pharm. 456, 261–268 (2013).
4. Luciani, C. V. Impact of Process Parameters on the Grinding Limit in High-Shear Wet Milling. Org. Process Res. Dev. 22, 1328–1333 (2018).