(344c) Defining a Design Methodology for Conversion of a Batch API Crystallization to a Continuous Crystallization

Crystallization is an essential unit operation for small molecule active pharmaceutical ingredient (API) manufacturing. In recent years there has been an increasing focus on continuous crystallization of such APIs. Advantages of continuous processing include increased control of product attributes, reduced equipment foot print, potential elimination of the requirement for scale-up activities, greater flexibility of equipment infrastructure and reduced expenditure.

This study focuses on creating a defined methodology for converting an existing batch crystallization process to a continuous system. Many previous studies have focused on characterizing continuous crystallization equipment and addressing processing challenges arising from continuous operation such as fouling and attaining steady state operation. The equipment used for this study was a paired CSTR system operated as an MSMPR. The use of this type of system for continuous crystallization has been previously demonstrated in literature [1-4].

The main goal of the design methodology was to transition an existing batch cooling crystallization process to a continuous format, while (at least) maintaining the yield and desired PSD of the existing batch crystallization. The API chosen as the case study for application of the design methodology was produced by a cooling crystallization following a salt formation in solution.

The initial stages of the design methodology consisted of a deeper investigation of the existing batch process, to include solubility screening, solvent selection optimisation, metastable zone width investigation and solid form characterisation. Concurrently, investigations of the existing crystallization process were performed with online monitoring of the process using FBRM and FTIR at 1L scale. From this work an improved understanding of induction times, nucleation temperature and the effect of the cooling rates and hold times on PSD as well as the crystal form was generated.

Temperature selection for continuous cooling crystallization is crucial, and a number of temperature set points were evaluated in batch format to improve confidence for continuous operation. A temperature strategy was selected for first implementation in continuous format, to be sequentially modified as understanding grew.

Following this work a two-step continuous cooling crystallization process was implemented using 2 glass jacketed 1 L reactors. Hot solution was fed to the first crystallizer using a peristaltic pump. Pressure transfer zones were used for intermittent transfer of slurry from the first crystallizer to the second crystallizer and on to a holding vessel for isolation. The effect of start-up strategy on particle size was investigated and was optimized to give the fastest time to steady state for the desired particle size. Various start up strategies were evaluated, including start up with a saturated solution, start up with a seed bed (with various seed loadings), as well as start up with crystal slurry from a previous batch respectively in the first tank. Through systematic trials the initial temperature strategy was sequentially modified until the desired operation and particle size production was achieved. Particle size was measured offline using a Malvern Mastersizer. Crystal form of the product was measured using XRD. The effect of various process parameters such a residence time, agitation and slurry transfer regime on the final PSD were investigated. The product yield was found to compare well with the existing batch process.

This project was made possible by funding from Enterprise Ireland Innovation Partnership Project IP-2014-0356

1. Hou, G., et al., Development and Characterization of a Single Stage Mixed-Suspension, Mixed-Product-Removal Crystallization Process with a Novel Transfer Unit. Crystal Growth & Design, 2014. 14(4): p. 1782-1793.

2. Morris, G., et al., Estimation of Nucleation and Growth Kinetics of Benzoic Acid by Population Balance Modeling of a Continuous Cooling Mixed Suspension, Mixed Product Removal Crystallizer. Organic Process Research & Development, 2015. 19(12): p. 1891-1902.

3. Power, G., et al., Design and optimization of a multistage continuous cooling mixed suspension, mixed product removal crystallizer. Chemical Engineering Science, 2015. 133: p. 125-139.

4. Polster, C.S., et al., Pilot-Scale Continuous Production of LY2886721: Amide Formation and Reactive Crystallization. Organic Process Research & Development, 2014. 18(11): p. 1295-1309.