Crystallization has been applied to a broad range of industries such as bulk and fine chemicals and the pharmaceutical and food industries. It is important to strategically control the in situ purification process during crystallization to meet the regulatory and functional specifications of the crystals. While the control of the crystallization−purification process has been widely discussed for batch crystallizers, there has been little focus with the literature on controlling purification for continuous crystallizers. Continuous crystallization is a more intensified approach to crystallization, with lower capital footprint and potentially offering more consistent quality control. This review paper provides an in-depth discussion of the strategies and scientific understanding in controlling the crystallization−purification process in continuous crystallization. In particular, it describes how scientific understanding in the purification process, generated so far for batch crystallization, can be translated to continuous crystallization.

No abstract

No abstract

This study aims to evaluate a suitable operating strategy to produce slow-to-crystallize particles in the oscillatory dynamic baffled crystallizer (DBC) by taking a lactose compound as a case study. Metastable zone width (MSZW) determination experiments were conducted with three different supersaturated concentrations (0.6, 0.73, and 0.9 kg/kg solvent) at fast cooling (0.2 °C/min) and slow cooling (0.05 °C/min). A narrower MSZW was attained in the DBC than that in a stirred tank crystallizer due to the intensive mixing created by the oscillatory flow. The results showed that the DBC conditions evaluated at a higher initial supersaturation would allow fast cooling to approach the behavior of slow cooling. Such convergent behavior was observed in terms of the particle size distribution (PSD) and the resultant apparent MSZW. Despite the slightly lower yield attained with fast cooling, the significantly shorter required crystallization time would effectively allow higher production throughput than slow cooling. A high mixing intensity in the DBC showed that lactose crystallization was prone to secondary nucleation, which produced a broad PSD. A direct nucleation control (DNC) strategy was further adopted to push the boundary of the fast cooling−high initial supersaturation crystallization strategy. While the DNC strategy led to a narrower PSD and a similar high yield as the slow cooling−high initial supersaturation, the overall particle size was unexpectedly shifted to a relatively smaller range. These findings can be used as a basis to improve the lactose crystallization process by further investigating the secondary nucleation threshold of lactose in a DBC to attain larger crystals.

No abstract