Design of Continuous Reactor Systems for API Production

The pharmaceutical industry has experienced many changes over the last few decades. Continuous production has been promoted as one of the more promising methods for making the industry more efficient and sustainable. The primary focus of this thesis is on the performance of Grignard chemistry in continuous reactor setups. Grignard chemistry encompasses a very powerful reaction type frequently applied in the pharmaceutical industry, for the formation of new carbon-carbon bonds. Three Grignard addition reactions have been studied, all having very different behaviors related to aspects of reaction engineering. A double Grignard addition (two different Grignard reagents) to a lactone was studied with continuous production in mind. The complexity of the reaction was investigated kinetically in order to optimize a potential flow setup. The investigation indicated that reaction temperatures below -40 °C could suppress the formation of an undesired bis-addition product by stabilizing the mono-addition adduct. A Grignard addition to a poorly soluble tricyclic ketone, previously studied in the laboratory, was transferred to full-scale production. Successful upscaling of the laboratory setup to full-scale production equipment enabled complete replacement of the existing batch production of this intermediate. The crowning achievement in this work was the realization of continuous laboratory reactor setups capable of manufacturing the entire GMP portion of the synthesis of melitracen HCl at H. Lundbeck A/S. The formation of a carbon-carbon bond between a tricyclic ketone and a Grignard reagent was the primary objective, this being the first step in GMP synthesis. The process was optimized to include one-step hydrolysis and dehydration, followed by phase separation of the product-containing organic phase, which was then precipitated with hydrogen chloride to obtain the final API. The Grignard reagent was also produced in a continuous laboratory setup involving handling of solid magnesium turnings. Likewise, the alkyl halide used in the formation of the Grignard reagent was produced continuously. The three segmented units were able to be coupled to construct a single continuous reactor facility for manufacturing melitracen HCl. The study of Grignard addition reactions to the three different substrates investigated in this thesis has culminated in a methodology by which reaction engineering decisions can be guided. The methodology provides suggestions on when and how decisions should be made on continuous production methods for Grignard chemistry within pharmaceutical manufacturing. Physicochemical properties, such as solubility, were found to be critical. However, from a business perspective, issues such as the current lifecycle of the API and GMP can make a potential reactor setup non-feasible. If the pharmaceutical industry is to adapt to recent trends towards end-to-end and on-demand pharmaceutical production, access to standard reactor units for commonly-used chemical transformations and methods for timely decision-making are essential. The methodology described herein provides an approach to fulfilling this need for Grignard chemistry in flow reactors.