Evaluation of a laboratory-scale batch crystallizer for lactose isolation from deproteinized whey

Abstract Text: Accurate replication of industrial-scale lactose crystallizations on a laboratory-scale apparatus has been a challenge for researchers wanting to improve lactose yield from dairy streams. The objective of this experiment was to develop a repeatable method for lab-scale batch crystallization of lactose from concentrated deproteinized whey (DPW). Commercial DPW powder was reconstituted to 60% total solids (wt/wt) in 80°C distilled water and held with agitation for 15 min for complete solubilization of lactose. The 80°C reconstituted DPW solution was immersed in an ice water bath to rapidly (15 min) cool to 50°C. Once the temperature reached 50°C, seed crystals were added (0.027 g per 100 g solution) to the DPW solution, mixed thoroughly, and held at 50°C with agitation for 1 h. After 1 h at 50°C, the solution was split into two portions by weight, and poured into lab-scale crystallizers. The crystallization apparatus consisted of a glass beaker with a PTFE magnetic stirrer and vanes for agitation. Stirring rate was maintained at 100 rpm throughout the cooling profile. The immersed crystallizer temperature was maintained by immersing a coil attached to a recirculation waterbath into the bath surrounding the crystallizer. Crystallizers were cooled from 50 to 20°C at a rate of -0.130 ± 0.007 °C per min. At the end of crystallization, the crystallizer was decanted and crystals were harvested from the bottom of the beaker for analysis. Recovered lactose crystals were observed microscopically (40×) to determine mean crystal size. Micrographs of lactose crystals were analyzed using image analysis software provided by the manufacturer of the microscope. A minimum of 10 crystals from each micrograph were measured for height, recorded as the length (µm) of the longest side, and duplicate micrographs were analyzed to improve accuracy. Three crystallizations were carried out using the same lot of DPW reconstituted to 60% total solids. Results indicated that mean lactose crystal length obtained using this method were 21.36 ± 7.42 µm, with some larger (> 90 µm) crystals present in low numbers. Many small (< 10 µm) crystals were observed (not enumerated), which indicates the presence of secondary nucleation during cooling. The crystal sizes obtained using this method were small compared to what is encountered in industry. Small crystals would be easily lost during subsequent refining steps, decreasing yield. An optimized cooling profile, agitation speed during cooling, and crystal isolation scheme should be considered for future developments of this method.

Keywords: lactose, crystallization, lab-scale