Operation and Optimization of Microwave-Heated Continuous-Flow Microfluidics

12 Microwave technology can be powerful for electrification and process intensification, but limited 13 fundamental understanding of scalability and design principles hinders its effective use. In this 14 work, we build a continuous-flow microreactor inside a commercial single-mode microwave 15 applicator and the corresponding computational fluid dynamics (CFD) model to simulate the 16 temperature profile. The model is in good agreement with experiments for various microreactor 17 dimensions and operating conditions. The model indicates that microwave heating is greatly 18 influenced by reactor geometry as well as the operating parameters. We observe a strong 19 correlation between parameters and develop a gradient boost regression tree model to predict the 20 outlet temperature accurately. This model is then applied to optimize the dimensions and operating 21 conditions to maximize the outlet temperature and energy efficiency, resulting in a Pareto optimal. 22 We demonstate computationally and experimentally that it is possible to surpass the Pareto optimal 23 and achieve an energy efficiency of 90% or greater at temperatures relevant for liquid-phase 24 chemistry via salting of the solvent. The present methodology can be applied to other complex 25 microwave reactors. The combined numerical and experimental approach provides insights into 26 and a framework for scale-up and optimization. 27 28