Directional amorphous lactose crystallization : a thesis presented in partial fulfilment of the requirements for the degree of Master of Engineering in Bioprocess Engineering at Massey University, Manawatu, New Zealand
It was proposed that during industrial drying of lactose crystals a surface layer of amorphous lactose may be formed in a flash drier and then crystallized during fluid bed drying. This crystallization is hypothesized to occur in one of two directions depending on the conditions, inside-out resulting in a dry product, and outside-in trapping moisture. In the inside-out case the moisture is driven outside the product, in the outside in case this moisture would be contained by a surface layer of crystalline lactose The trapped moisture from the outside-in case is proposed to slowly diffuse through the crystal layer during storage and cause handling problems, explaining observed differences between industrial products.
To investigate this scenario the crystallization of amorphous lactose was modelled, and crystallization trials were conducted to try and achieve inside-out and outside-in crystallization.
William-Landel-Ferry (WLF) and Arrhenius type kinetics were found to fit literature data for amorphous crystallization. Predictions made using these models showed that amorphous lactose crystallization under the high temperature conditions in a fluid bed dryer was possible.
A method for isolating the enthalpy change associated with crystallization of amorphous lactose from simultaneous thermal analysis data was developed. This method shows promise for observing the crystallization process, but it may not be suitable for amorphous lactose quantification.
Two methods were designed to achieve inside-out and outside-in crystallization of amorphous lactose. This required the temperature and water activity conditions to be precisely and independently controlled in lab trials. Simultaneous thermal analysis was used to monitor the crystallization of amorphous lactose samples under these conditions.
Following the simultaneous thermal analysis, the samples were monitored for moisture release. Both the inside-out and outside-in crystallized samples were observed to slowly release moisture, increasing the measured relative humidity above the expected equilibrium value. Afterwards the samples were analysed and found to still contain low levels of amorphous lactose. The source of the rise in relative humidity could not be definitively attributed to either trapped moisture or ongoing crystallization but would not be expected had crystallization not been induced.
Based on these findings it is recommended that a lactose crystal fluid bed drier is operated at conditions which would not allow for amorphous lactose crystallization. These conditions could be determined using the kinetic models fitted here.