Investigating catalyst performance in batch reactive distillation

Abstract

Reactive distillation (RD) combines chemical synthesis with separation by distillation, but this leads to a non-trivial system: the hardware selection, the system components, the mode of operation and the operating conditions all affect the performance of the RD process. A key process development issue is the identification of suitable catalysts that perform well under reactive distillation conditions, as catalysts are crucial for increasing reaction rate when operating temperature range is limited by evaporation. The main goal of this research is to develop a method, utilizing high throughput technology, which can be used to assess many candidate catalysts for batch RD systems. The identification of potentially suitable catalysts should be made as early as possible, but before experimental work begins the only information available is the catalyst composition and structure. The approach taken in this research is to correlate catalyst properties to the performance in RD tests and the outputs from the dynamic simulations. The case study used is a batch reactive distillation for the esterification of a long-chain fatty acid. Potential catalysts are studied at small scale in a high throughput platform, and further investigation if performaed in an experimental batch RD unit. The most active of the screened catalysts, sulfuric acid and MSA also have the highest initial activity under RD. Heteropoly acids appear to have a good activity level, while ferric sulfate gives intermediate but apparently increasing activity. Some outcomes of the RD experiments were unexpected: the strong homogeneous acid catalysts entail low distillate water yield, and some metal acetates had higher activity than anticipated in the RD tests. This demonstrates that pilot scale experiments currently remain necessary for the evaluation of catalyst performance for RD processes. The insights gained from this study lead to key recommendations for future studies: an increased scope of study with a larger number of candidates which preferable have similar structure; evaluation of additional catalyst performance indicators, performed over the full operating temperature range; use of the smallest suitable experimental column; and more focus on physical factors such as solubility. Use of a simulator with an established physical property calculation tool is essential for successful simulations of batch RD.