Process behavior of vertical ultraviolet disinfection systems
Abstract
The use of ultraviolet (UV) radiation for disinfection has been frequently chosen as an alternative to chlorination in water and wastewater disinfection because of concerns of disinfection by-products and associated toxicity in treated water. UV has shown to be effective for inactivating pathogenic microorganisms. The kinetics of microbial inactivation as a result of exposure to UV radiation are a function of UV dose, expressed as the product of intensity of radiation and exposure time. In continuous-flow UV systems, due to the complex characteristics of the configuration, UV intensity is a function of position, and exposure time is dependent of the hydrodynamic characteristics of the system. A number of techniques have been used to estimate dose in continuous flow UV systems. However, due to the lack of information regarding hydrodynamic behavior in UV systems, no theory was available that included the information of both non-uniform intensity and velocity fields. In this research, a dose-distribution model is developed by integrating the complex velocity field (through a random-walk model) and intensity field (through a point-source summation method) to develop estimates of a dose-distribution function for vertical open-channel ultraviolet (UV) disinfection systems. Then the dose-distribution was combined with a dose-response relationship, as measured using a collimated-beam system, to predict the process performance of a pilot-scale system. The velocity field information necessary for the random walk model was obtained from laboratory measurements of the turbulent flow using laser Doppler velocimetry. This approach provides a useful tool for understanding the hydraulic behavior of UV disinfection systems and its effects on process performance. Results of the dose-distribution model revealed critical paths through which particles moved and experienced low UV doses. In vertical UV systems, these paths are generally found near the channel walls with characteristics of high velocity and low turbulence intensity. Moreover, these paths generally coincide with low UV intensity areas and appear to represent the primary limitation of process performance. Process modifications have been designed to eliminate or modify these trajectories, thereby improving process performance. Two geometric modifications of channel walls with “wave” and “baffles” shapes have been developed and examined in a pilot-scale open-channel system. Similar to the case of the unmodified UV system, the dose-distribution model was used to predict process performance of the pilot-scale UV system with the wave-shaped modification. The results showed model predictions agree well with the observed data at higher approach velocities. Both geometric modifications in pilot-scale systems improved in microbial inactivation compared to the unmodified system. Model comparison of the behavior of the modified and unmodified systems suggested that the improvements could be attributed to changes in the particle trajectories and mixing patterns. Head loss in the UV system, simply defined as the drop of water surface elevation across the UV lamp module(s), is a factor in process design. Head loss data of a few vertical and horizontal UV systems, including the modified vertical systems, were collected. Because head loss was caused by the lamp arrays and also the support structures, it was effected by the configuration of the system (i.e., head loss was dimension-specific).
Degree
Ph.D.
Advisors
Blatchley, Purdue University.
Subject Area
Civil engineering|Environmental science
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.