Targeting Particle Size Specification in Pharmaceutical Crystallization: A Review on Recent Process Design and Development Strategies and Particle Size Measurements

The goal of the crystallization process in the pharmaceutical industry is to isolate the drug substances and intermediates as solid materials with targeted product attributes. The physical properties of the solids, e.g., crystallinity, purity, mechanical properties, particle size, and shape, play an important role in the development and production of pharmaceuticals. The U.S. Food and Drug Administration (FDA) defines critical quality attributes (CQA) as physical, chemical, biological, or microbiological properties or characteristics that should be within an appropriate limit, range, or distribution to ensure the desired product quality. CQAs are generally associated with the drug substance (also referred to as active pharmaceutical ingredient, API), excipients, intermediates (in-process materials), and drug product. Particle size is a common CQA of drug substances as it affects downstream processability and drug release characteristics. Achieving the desired particle size can often be challenging owing to the interplay among various process parameters. Substantial efforts have been made to tune the particle size as the targeted product attribute during the process development of crystallization. A series of techniques, including mathematical modeling tools, have been applied for the prediction and control of particle size and distribution.
A large particle size can be the targeted quality attribute for a drug compound. Large particles generally have good flowability and avoid the issues created by small particles, e.g., clogging during a filtration step. The general observation is that the large particles (>250 μm) tend to be free-flowing, whereas fine powders with a high surface area-to-mass ratio become cohesive and tend to stick, especially particles that are less than 10 μm. A small particle size is usually desired for the poorly soluble drug substances with concerns of bioavailability. While it can be challenging to achieve either small or large particle sizes in the crystallization operation, for getting small particles, micronization (e.g., milling) is commonly employed separately for the crystallized solid materials. The current work emphasizes obtaining large particle sizes by crystallization.
For a crystallization process targeting a large particle size, optimized process parameters involve understanding and controlling the underlying mechanisms (e.g., nucleation and growth). For a seeded cooling crystallization in batch, the main parts include optimizing the cooling profile and a seeding strategy. Meanwhile, agitation which renders mixing and mechanical shear is also critical to the final product particle size distribution. The cooling profile significantly impacts the underlying mechanisms, thus affecting a variety of product attributes, such as form, morphology, particle size, etc. Promoting growth mechanisms to achieve a large particle size usually requires a slow cooling rate in a controlled way to maintain the cooling profile within the metastable zone width (MSZW). Seeding is an established technique for improving crystal quality and process robustness, which has been developed for various applications including control of particle size distribution (PSD). Increased particle size can be obtained by using seeds to suppress nucleation and facilitate crystal growth. Mixing by agitation is an important aspect for the crystallization process and significantly impacts particle size distribution as the selection of impeller and agitation rate can affect the process and product attributes. PSD is usually measured using multiple techniques, both inline and offline. Probe-based methods are well-suited for inline “real-time” monitoring of PSD, which has led to improved capability to understand, optimize, and control the crystallization process to achieve target CQAs. However, offline methods are suited for PSD measurement after the downstream operations such as filtration and drying of crystalline material, which could result in agglomeration or breakage of crystals, changing the final PSD obtained from the crystallization process. The discrepancy in PSD measurements arises owing to the physical principles of measurement employed by various methods as well as the sampling location in the process stream.
Targeting particle sizes within certain specifications can be challenging, particularly when trying to produce materials with good flow properties. The inherent thermodynamic and kinetic limitations along with the interplay associated with the critical events, vis-à-vis nucleation, growth, attrition, and agglomeration, leads to convoluted experimental findings. It is beneficial to review the critical factors to be taken into consideration, the relevant strategies used for making large particles by crystallization, as well as the approaches and techniques being applied for various cases. This paper is not on how to grow a large single crystal but rather on increasing the mean particle size in the distribution. Additionally, the current work only summarizes recent approaches focusing on key process parameters including the cooling profile and seeding. Mixing by agitation is also discussed. In addition, a few practical issues such as particle size measurement discrepancies are also addressed.