Design optimization of a cylindrical UV-C LED reactor for effective water disinfection with numerical simulations and test reactor fabrication

Abstract

Ultraviolet C-band light-emitting diodes (UV-C LEDs) provide effective water disinfection and flexible reactor design. In this study, the disinfection performance of the cylindrical UV-C LED reactor was optimized using a new computational fluid dynamics (CFD) technique and validated experimentally with test reactors. The CFD technique provided UV doses for individual particles representing microorganisms, which were integrated along their trajectories using the discrete phase method within the radiation field acquired through the Monte Carlo method. Utilizing the minimum UV dose as a criterion, four key design parameters were evaluated: reactor aspect ratio (L/D) (in the range of 2.5−6), LED power angle (15−55°), wall reflectivity (0.6−0.94), and water flow rate (1−4 l/min). The optimal L/D of 4.2 was determined and subsequently applied to fabricate the test reactors. The polytetrafluoroethylene (PTFE) test reactor exhibited superior disinfection performance for E. coli O157:H7 compared to the one made of SUS304, due to its high wall reflectivity. Furthermore, a lower water flow rate enhanced disinfection performance by extending the residence time. The predicted log reductions exhibited a consistent trend with the measured values, confirming the efficacy of the CFD methodology applicable to various reactor designs and LED configurations.