Chiral symmetry breaking due to impeller size in cooling crystallization of sodium chlorate

The classic experiments of crystallisation of sodium chlorate by Kipping and Pope in 1898 showed that when NaClO 3 , an achiral salt, was crystallized from a tranquil solution, statistically equal numbers of levo and dextro crystals were obtained. Kondepudi et al. in 1990 extended Kipping and Pope's work and revealed that >99% of the crystals having the same handedness, i.e. total chiral symmetry breaking, were achieved when constant stirring was applied in evaporation crystallisation. In cooling crystallisation of NaClO 3 at supercoolings greater than 7 °C and at a stirring rate of 300 rpm, Denk and Botsaris (1972) reported that a 50 : 50 racemic mixture of product crystals was obtained due to primary nucleation that occurred. In this paper we report, for the first time, that the size of impellers, of the same width, the same thickness, the same material of construction and operated at the same rotational speed and at the same degree of supercooling, is the exception to Denk and Botsaris's work where a stirrer of the smallest surface area led to close to 100% deracemization, while the data from the largest impeller agreed with the results of Denk and Botsaris, displaying a racemic mixture. The effects of shear, pre-seeding and surface roughness on chiral symmetry are examined to establish the likely cause of this new finding. We report, for the first time, that the size of impellers is the exception to Denk and Botsaris's work where a stirrer of the smallest surface area led to close to 100% deracemization, while the data from the largest impeller agreed with the results of Denk and Botsaris, displaying a racemic mixture.