Microbial inactivation kinetics of UV LEDs and effect of operating conditions: A methodological critical analysis

Tiny ultraviolet (UV) light-emitting diodes (LED)s that are replacing the conventional energy-intensive mercury UV lamps have gained interest since the early 2000's because of their promising advantages. In the context of microbial inactivation (MI) of waterborne microbes, disinfection kinetics of those LEDs exhibited variations among studies, in terms of varying the UV wavelength, the exposure time, power, and dose (UV fluence) as well as other operational conditions. While reported results may appear contradictory when examined separately, they probably are not when analyzed collectively. As such, in this study, we carry out a quantitative collective regression analysis of the reported data to shed light on the kinetics of MI by the emerging UV LEDs technology alongside the effects of varying operational conditions. The main goal is to identify dose response requirements for UV LEDs and to compare them to traditional UV lamps in addition to ascertaining optimal settings that could help in achieving the optimal inactivation outcome for comparable UV doses. The analysis showed that kinetically, UV LEDs are as effective as conventional mercury lamps for water disinfection, and at times more effective, especially for UV resistant microbes. We defined the maximal efficiency at two wavelengths, 260–265 nm and 280 nm, among a wide range of available LED wavelengths. We also defined the UV fluence per log inactivation of tested microbes. At the operational level, we identified existing gaps and developed a framework for a comprehensive analysis program for future needs. [Display omitted] •We examined the kinetics of microbial inactivation by UV LED under varying operational conditions.•We identified dose response, compared with traditional UV lamps, and ascertained optimal settings.•Kinetically, UV LEDs are as effective as conventional mercury lamps and at times more effective.•We identified maximal efficiency at two wavelengths, 260–265 nm & 280 nm.•Operationally, we developed a framework to fill in existing gaps and guide future work.