The key findings are as follows.
•
The UV-LED with peak emission at 280 nm showed higher time-based inactivation efficiency than the UV-LEDs at 265 and 310 nm. Meanwhile, the UV-LED at 265 nm showed highest fluence-based efficiency among the others.
•
In the batch reactor, the UV-LEDs at 265 and 280 nm achieved over 4 log inactivation of E. coli at adjusted fluences of 10.8 and 13.8 mJ/cm2, respectively, while the UV-LED at 310 nm required 56.9 mJ/cm2 for 0.6 log inactivation of E. coli.
•
The flow-through reactor showed tailing in the inactivation curves and resulted in lower inactivation efficiency than the batch reactor, with lower fluence-based inactivation rate constant of 29% for 265 nm and 32% for 280 nm.
•
Combined emissions at 265/280, 265/310, 280/310 and 265/280/310 nm in the flow-through reactor showed less inactivation efficiency than the component emissions applied separately, and reduced output power for each UV-LED was observed for combined emissions.