Process development using oscillatory baffled mesoreactors

Abstract

The mesoscale oscillatory baffled reactor (meso-OBR) is a flow chemistry platform whose niche is the ability to convert long residence time batch processes to continuous processes. This reactor can rapidly screen reaction kinetics or optimise a reaction in flow with minimal waste. In this work, several areas were identified that could be addressed to broaden the applicability of this platform. Four main research themes were subsequently formulated and explored: (I) development of deeper understanding of the fluid mechanics in meso-OBRs, (II) development of a new hybrid heat pipe meso-OBR for improved thermal management, (III) further improvement of continuous screening using meso-OBRs by removing the solvent and employing better experiment design methodologies, and (IV) exploration of 3D printing for rapid reactor development. I. The flow structures in a meso-OBR containing different helical baffle geometries were studied using computational fluid dynamics simulations, validated by particle image velocimetry (PIV) experiments for the first time. It was demonstrated, using new quantification methods for the meso-OBR, that when using helical baffles swirling is responsible for providing a wider operating window for plug flow than other baffle designs. Further, a new flow regime resembling a Taylor-Couette flow was discovered that further improved the plug flow response. This new double vortex regime could conceivably improve multiphase mixing and enable flow measurements (e.g. using thermocouples inside the reactor) to be conducted without degrading the mixing condition. This work also provides a new framework for validating simulated OBR flows using PIV, by quantitatively comparing turbulent flow features instead of qualitatively comparing average velocity fields. II. A new hybrid heat pipe meso-OBR (HPOBR) was prototyped to provide better thermal control of the meso-OBR by exploiting the rapid and isothermal properties of the heat pipe. This new HPOBR was compared with a jacketed meso-OBR (JOBR) for the thermal control of an exothermic imination reaction conducted without a solvent. Without a solvent or thermal control scheme, this reaction exceeded the boiling point of one of the reactants. A central composite experiment design explored the effects of reactant net flow rate, oscillation intensity and cooling capacity on the thermal and chemical response of the reaction. The HPOBR was able to passively control the temperature below the boiling point of the reactant at all conditions through heat spreading. Overall, a combined 260-fold improvement in throughput was demonstrated compared to a reactor requiring the use of a solvent. Thus, this ii wholly new reactor design provides a new approach to achieving green chemistry that could be theoretically easily adapted to other reactions. III. Analysis of in situ Fourier transform infrared (FTIR) spectroscopic data also suggested that the reaction kinetics of this solventless imination case study could be screened for the first time using the HPOBR and JOBR. This was tested by applying flow-screening protocols that adjusted the reactant molar ratio, residence time, and temperature in a single flow experiment. Both reactor configurations were able to screen the Arrhenius kinetics parameters (pre-exponential factors, activation energies, and equilibrium constants) of both steps of the imination reaction. By defining experiment conditions using design of experiments (DoE) methodologies, a theoretical 70+% reduction in material usage/time requirement for screening was achieved compared to the previous state-of-the-art screening using meso-OBRs in the literature. Additionally, it was discovered that thermal effects on the reaction could be inferred by changing other operating conditions such as molar ratio and residence time. This further simplifies the screening protocols by eliminating the need for active temperature control strategies (such as a jacket). IV. Finally, potential application areas for further development of the meso-OBR platform using 3D printing were devised. These areas conformed to different “hierarchies” of complexity, from new baffle structures (simplest) to entirely new methods for achieving mixing (most complex). This latter option was adopted as a case study, where the passively generated pulsatile flows of fluidic oscillators were tested for the first time as a means for improving plug flow. Improved plug flow behaviour was indeed demonstrated in three different standard reactor geometries (plain, baffled and coiled tubes), where it could be inferred that axial dispersion was decoupled from the secondary flows in an analogous manner to the OBR. The results indicate that these devices could be the basis for a new flow chemistry platform that requires no moving parts, which would be appealing for various industrial applications. It is concluded that, for the meso-OBR platform to remain relevant in the next era of tailor-made reactors (with rapid uptake of 3D printing), the identified areas where 3D printing could benefit the meso-OBR should be further explored.