Model-Based Scale-up and Design Space Determination for a Batch Reactive Distillation with a Dean–Stark Trap

Batch reactive distillations are commonly used unit operations in the pharmaceutical industry to drive chemical equilibrium reactions to completion. Their scale-up and transfer to manufacturing is not straightforward due to the interplay of scale dependent and scale independent phenomena. The increased process knowledge as required by the Quality by Design (QbD) approach calls for a first-principles-based design, scale-up and transfer of such processes. This paper presents a systematic approach consisting of a combination of first principles modeling and experimentation for the scale-up from bench to pilot-plant scale. The model is then used to estimate the process performance at different scales and study the sensitivity of the process to operational parameters such as heat transfer driving force, solvent recycle, removed fraction of volatiles. This approach is capable of robustly predicting process outcomes at lab and pilot-plant scale and delivers a better understanding of the underlying physics governing the process. The model is used further to map the design space (a region in the space of operating parameters where given quality and/or performance constraints are met) taking into account both model parameter uncertainty and routine operational variability.