Linh T. T Vu

Dynamic modelling and optimisation of alpha-lactose monohydrate seeded batch cooling crystallisation

Abstract This work validates the model equations of a seeded batch cooling process to produce α-lactose monohydrate from highly pure syrup. Experiments are conducted in 2L and 20L crystallisers under various seeding and cooling strategies, obtained from solving a dynamic optimisation problem. Using a nonlinear least square method the best curve fitting of experimental data to the model yields k g and n, two constants used in the growth rate equation. The value of n = 2.7 can be fixed regardless of sizes of seeds and crystallisers but k g must be estimated for different seeding weights. The rate of nucleation must be redefined to reflect the variations in crystal median size prediction. The model can be applied to industrial scale crystallisers providing the cooling rate is slow and the seed size range is narrow.

Dynamic modelling and simulation of lactose cooling crystallisation: from batch and semi-batch to continuous operations

Abstract The focus of this work is to revise for semi-batch and continuous operations, the model equations of a seeded batch cooling crystallisation of α-lactose monohydrate. A case study of evaporative semi-batch operation is also included for reference as models of cooling and heating are not much different. In order to achieve the highest yield, dynamic optimisations must be performed on these models to get optimal cooling, feeding and heating (for evaporative mode only) strategies. Applying these strategies in bench scale experiments shows that models of batch and semi-batch work well before the growth becomes slow. Semi-batch is slightly faster than batch and evaporative semi-batch is ten times faster than cooling but is more difficult to control. The performance of a cooling and seeding run in continuous mode is simulated. The system reaches steady state after seven residence times but the predicted particle size could not be stabilised and continued to increase up to one hundred hours. A plug flow reactor is being studied instead of a continuous stirred tank reactor to close the mass and population balances.