Mixing Times of Miscible Liquid Systems in Agitated Vessels

A better understanding of mixing times for mixed solvent systems in laboratory-scale vessels is crucial for improving mixing-sensitive processes such as antisolvent crystallisation. Whilst mixing in agitated vessels has been extensively studied using solutions of additives in the same solvent, there is very limited literature on the mixing of different miscible solvents and none which would be relevant to antisolvent crystallisation processes. In this work, the mixing times of water–ethanol systems in a 1 litre vessel, agitated by a pitched blade impeller with probes used as baffles, were investigated in the transitional flow regime using both experimental and computational fluid dynamics (CFD) approaches. We studied two scenarios: adding sodium chloride tracer to premixed water–ethanol solutions and adding ethanol containing a tracer to water. Mixing was investigated experimentally through conductivity measurements and computationally using large eddy simulations with the M-Star CFD software package. Empirical correlations from the Dynochem engineering toolbox were also used for comparison. The results showed significant run-to-run variability in the mixing times from both experiments and CFD simulations, with experimental ranges being notably wider than CFD ones under the given conditions. While the CFD simulations showed consistent mixing times across different solvent compositions, the experimental mixing times decreased with increasing ethanol content. The mixing times were approximately inversely proportional to the impeller speed. The CFD simulations indicated that 25–40 impeller rotations were required for homogenisation, while the experiments needed 25–100 rotations. The Dynochem correlations predicted 40 rotations, independent of speed, but could not capture the inherent variability of the mixing times.