(678d) Pervaporation: An Alternative Process for the Dehydration of Pharmaceutical Streams

Pervaporation is a process that has been used successfully the last two decades for the dehydration of organic solvents at large scales and for improving distillation efficiency of azeotropic systems like ethanol-water. Similarly to other membrane processes, pervaporation offers several advantages. It often results in lower costs, it is modular and therefore easy to scale up, offers much higher mass transfer rates per unit volume compared to conventional separation equipment and can effect separations difficult to achieve otherwise. Despite the growing importance of pervaporation in the chemical industry, there are no reported efforts to apply the technology in the dehydration of pharmaceutical streams. The latter is important in several cases: removal of moisture during condensation reactions, generation of supersaturation during crystallization of intermediates or Active Pharmaceutical Ingredients (APIs) and recovery of solvent at a low energy cost. In this paper we report for the first time our efforts to dehydrate a stream containing a pharmaceutical intermediate. This intermediate is crystallized from a 2-MeTHF solution saturated with water. Moisture must be removed to levels lower than 0.3% wt. to avoid excessive yield losses during the crystallization process. The dehydrated solution is then distilled, cooled to below its saturation temperature, seeded and isolated. Prior to this study water removal was achieved by a distillation process. The latter required the consumption of significant amounts of solvent (6X) as well as prolonged processing time (about 30 hrs) at commercial scale. We have performed process development in a two-tier approach. We first evaluated the feasibility of the process at the laboratory scale by using ceramic modified silica membranes. Our laboratory runs clearly illustrated the feasibility of the process; water levels were reduced routinely to 0.15-0.25% wt. after only 2-3 hours of operation. Contrary to the previously used distillation process, no additional solvent was required to achieve the separation. We also assessed the long term performance of the membrane by operating the membrane system over an extended period of time. Continuous monitoring of moisture levels in the stream by an in-line KF instrument showed that the test was successful. Second, we successfully scaled up the process at our pilot plant. The KF specification was met and no significant operation issues were observed. The process is currently evaluated for implementation at the commercial scale. Our calculations indicate that significant savings can be realized. First, the savings by the reduced solvent usage can pay for the capital investment of purchasing the membrane modules after only 4-5 batches are performed. Second, operation time can be reduced to 18-24 hrs resulting in additional savings in operational costs and cycle time and throughput. Third, the process requires a much lower energy input, since the process stream is never heated to its reflux temperature as in the conventional distillation process. In conclusion, the developed pervaporation process can deliver the required separation goal in a robust manner, at a lower cost and reduce solvent usage resulting in a greener process.