Particle size distribution (PSD) is one of the critical solid-state attributes of crystallization. Its impact can be felt strongly in downstream operations such as filtration, drying, re-dissolution etc. Hence PSD control is of paramount importance in crystallizer operation, especially in pharmaceutical and fine chemical industry where crystallizers are usually operated batchwise and quality inconsistency is a recurring problem due to fluctuations transmitted from upstream as well as the inherent instability of batch processes. Seeding and supersaturation profile have been recognized as two key approaches to address the problem of quality variation [1]. Although simultaneous implementation of these two approaches should lead to better quality control, they were traditionally implemented independently due to a lack of strategy to combine them. Recently, Chung et al. [2], Choong and Smith [3] derived optimal seeding policies and optimal temperature profiles for cooling crystallization through process modeling and optimization. Nonetheless, the robustness of their optimal protocols depends on the accuracy of kinetic parameters and process models. A more straightforward and simpler strategy to combine these two approaches may be more attractive to industry. It has been found that the performance of optimal supersaturation profile is close to that of constant-supersaturation operation for some systems [4], which indicates that process development efforts can be focused on seeding policy once a proper supersaturation level has been determined. In this study, the effects of seed size distribution and loading were investigated for anti-solvent crystallization of paracetamol from a water-acetone mixture operated at constant supersaturation in a one-liter crystallizer. ATR-FTIR technique was employed to implement feedback control of supersaturation. A seeding strategy based on a simple mass balance equation provided a good starting point for the refinement of particle size distribution and adjustment of batch time in different circumstances to obtain a target mean crystal size with narrow size distribution. Results indicate that the ratio of target product to initial seed size should not be too high in order to avoid excessive secondary nucleation. Under constant supersaturation, different seed mass-size couples led to similar PSD of final products when target product size was properly selected. Unimodal PSD can also be attained from undersized seeds, which may have practical implication for industrial operation. References [1] Paul, E. L.; Tung, H. H.; Midler, M. Powder Technol. 2005,150,133-143 [2] Chung, S. H.; Ma, D. L.; Braatz, R. D. Can. J. Chem. Eng. 1999,77,590-595 [3] Choong, K. L.; Smith, R. Chem. Eng. Sci. 2004,59,313-327 [4] Worlitschek, J.; Mazzotti, M. Cryst. Growth Des. 2004,5,891-903