Engineering reaction chemistry with flow technology : a discussion on method development and multi-step organic synthesis

Abstract

The adoption of flow technology for the manufacture of chemical entities, and in particular pharmaceuticals, has seen rapid growth over the past two decades with the technology now blurring the lines between chemistry and chemical engineering. Current indications point to a future in which flow chemistry and related technologies will be a major player in modern chemical manufacturing and the 4th industrial revolution. In this thesis we have highlighted the role of new reactor configurations and designs in the context of either bespoke or commercial flow apparatus specifically related to photochemical transformations and multi-phasic synthetic reactions pertaining to ozonolysis-based reactions. In addition, we have highlighted the relevance of the bespoke chemical engineering application towards multi-step organic synthesis. We have shown the relevance of integrating new innovative engineering concepts along with existing flow technology to synthesize Bupropion hydrochloride as well as benzo-fused heterocycles. We have further highlighted the relevance of the field by discussing engineering applications pertaining to microwave-based reactions, 3D printing, electrochemical applications and computer-aided automation. The areas highlighted in this text have showed rapid growth within the field.

The adoption of flow technology to novel applicator devices to perform ozonolysis reactions as well as photochemical reactions, have been described in this text. In the case of the ozonolysis applicator we have successfully demonstrated several ozonolysis reactions in a safe, yet continuous manner under flow conditions with the removal of excess ozone on the fly monitored by a bubble counting technique developed with Python coding. A model reaction showing the oxidation of α‐methylstyrene to acetophenone was conducted ultimately affording an optimised yield of 86 %, at 1 sL/h, 0.25 mL/min, 0 ºC and 0.5 M. A further study into the possibility of challenging selective ozonolysis was also investigated. With substrates including 1,3-di(prop-1-en-2-yl)benzene as well as 1,5-cyclooctadiene. An 18 % conversion was obtained for selectively ozonolysing 1,3-di(prop-1-en-2-yl)benzene under flow (0.5 M, 1.0 sL/h, 0.5 mL/min and 0 ºC) compared to 39 % achieved under batch, however, in the case of 1,5-cyclooctadiene selective ozonolysis was significantly better with the desired mono-ozonolysed product obtained in 66 % (0.5 M, 0.3 sL/h, 1.0 mL/min and -30 ºC) compared to 35 % under batch conditions. Furthermore, an expanded study investigating the application of the reactor to various substrates was also completed with several substrates showing moderate to high yields, these included the formation of alcohols, acetals and aldehydes.

Two photocatalytic applicator devices were designed and tested under photocatalytic conditions. The first system consisted of fibre optic cabling delivering photons from a 30 W blue LED light source, while the second reactor designed consisted of a 15 W, Laser unit with 445 nm wavelength. The second system transmitted light directly into the reaction stream. Both reactors proved successful in transforming N-phenyltetrahydroisoquinoline to 1-(nitromethyl)-2-phenyl-1,2,3,4- tetrahydroisoquinoline, with the laser system providing more desirable product conversions of 89 %, compared to the LED system which afforded a best conversion of only 58 %. Further, investigation of the laser system, into a dual photocatalytic cross-coupling reaction of 1-(4- trifluoromethyl)phenylpyrrolidine had limited success with an optimal yield of 16 % obtained under flow conditions.

Investigation into multi-step organic synthesis was conducted with a green approach to the translation of Bupropion hydrochloride under flow conditions. The process consisted of an α-bromination of 3- chloropropiophenone followed by a nucleophilic substitution of the bromine with tert-butylamine. The first step was successfully translated to flow utilising polymer bound pyridinium tribromide negating the need for liquid bromine. Reasonable residence times and high yields (81 % isolated) were achieved under flow conditions with little to no post reaction clean-up. The nucleophilic substitution reaction was achieved swopping the reported NMP solvent system for a greener alternative in the form of ACN/DMSO under flow conditions with 20-minute residence time at 90 °C, permitting the production of the free base Bupropion at 3.08 g/h equating to a space-time productivity of 123.2 g/h/L. The telescoped continuous process proved even more advantageous, eliminating the need for reaction processing of both stages. The total synthesis of Bupropion free base under flow proved more efficient in terms of the production rate (0.36 g/h in batch vs. 1.44 g/h in flow), however the overall total yield of the free base in batch was slightly higher (73 % vs 64 %) but with a significantly longer reaction time (16 hours vs 80 min residence time in flow).

Further multi-step organic synthesis was investigated synthesizing benzo-fused heterocycles under a continuous fashion. The synthetic approach translated to flow conditions involved five stages, each of which were demonstrated under flow conditions. These included: two allylation reactions, an aldehyde reduction, a Claisen rearrangement and finally ending with ring closing metathesis to afford the final benzo-fused heterocycle of interest. In the case of the first stage a significant improvement was observed under flow. When employing the use of a column reactor packed with potassium carbonate an optimised yield of 98 % was obtained. The second stage involving a Claisen rearrangement was completed under flow obtaining the intermediary alcohol, with a significant increase in yield (27 to 100 %) and time reduction from 64 hours in batch to 45 minutes in flow. The third stage was completed under flow with polymer bound sodium borohydride to reduce an aldehyde functionality to an alcohol in 85 % yield. The fourth stage was a repeat of the first allylation transformation and the final RCM reaction was completed under flow with an increase in the yield from 52 % in batch to 60 % in flow with a reduction in reaction time from 1 hour to 30 minutes.

For the final chapter of this thesis we described the mild and selective reduction of aldehydes utilising sodium dithionite under flow conditions. The work followed our recently reported novel hybrid batch-flow synthesis of the anti-psychotic Clozapine in which the reduction of an aryl nitro group is described under flow conditions using sodium dithionite. In this chapter we report the expansion of the method to include the reduction of aldehydes. The method developed affords yields which are comparable to those under batch conditions, with reduced reaction times and improved space-time productivity. Furthermore, the approach allows the selective reduction of aldehydes in the presence of ketones and has been demonstrated as a continuous process.