Velocimetry Follows the Flow for Clean Water

Hank Hogan

To avoid the fate of Coleridge’s ancient mariner, Fariborz Taghipour wants to make sure that there’s always a drop to drink. Taghipour, an assistant professor in chemical and biological engineering at the University of British Columbia in Vancouver, and fellow researcher D. Angelo Sozzi are working to optimize the operation of ultraviolet photoreactors — devices that remove toxic organics and disinfect water either through direct radiation or by photoinitiated oxidation.

The researchers employ particle image velocimetry to image the hydrodynamics of an ultraviolet reactor used for water treatment. Images courtesy of Fariborz Taghipour, University of British Columbia.

Compared with other approaches to water purification, photoreactors and UV radiation in general offer significant advantages in that there are no potentially hazardous chemicals or residues involved, and the process is relatively fast and the equipment compact.

Photoreactors have not caught on in industrial water treatment applications, however, and Taghipour noted that commercial systems have met with only limited success. “The efficiency is a major issue,” he explained.

Improving that efficiency through a better understanding of the reactors has caught the attention of researchers. In the past, investigators have conducted tests at a reactor’s inlet and outlet to characterize performance. Taghipour and Sozzi instead employed particle image velocimetry to reveal the flow inside the entire reactor.

The efficiency of a photoreactor ultimately comes down to fluence, the amount of UV radiation at a location multiplied by the dwell time there. The energy from a UV source, like all radiation, falls off as the inverse square of the distance, while the dwell time decreases linearly as the velocity increases. An ideal photoreactor would have a fast-moving stream near the source and a slower-moving one far away from it, providing uniform fluence. The hydrodynamics of the chamber, therefore, plays an important role in a reactor’s efficiency.

The university researchers injected a small number of 20-μm particles into the water that was circulating through the photoreactor. They illuminated the movement of the particles using a Dantec Dynamics particle image velocimetry system, which features a dual-head pulsed Nd:YAG laser with frequency-doubled 532-nm beams spread into a 1- to 2-mm sheet through a set of cylindrical lenses. The two lasers can be fired independently with a user-controlled delay between the 10-ns pulses.

“The dual-head design is necessary to permit users to compare subsequent images with sufficiently short time delays to track particle motion,” Taghipour explained.

To do so, the researchers captured the images of the microparticles using a Hamamatsu 1344 × 1024-pixel CCD camera. They used image processing software to track particle movement and thereby reveal fluid flow. By moving the illumination point, they mapped the flow within the photoreactor. To avoid problems caused by the curved external surface of the transparent reactor, they placed it inside a tank of distilled water.

In their investigations, which they describe in the Dec. 21 issue of Industrial & Engineering Chemistry Research, they looked at the performance of a reactor with an inlet parallel to and one with an inlet perpendicular to the main axis under various flow conditions.

They found the former better for producing uniform UV fluence, with a high velocity near the light source and low speeds, even recirculation, farther away.

An L-shaped reactor performs better than a U-shaped one in profiles of microorganism concentrations predicted by an integrated reactor performance model that takes into account kinetics, hydrodynamics and radiation distribution.

Such experimental studies of photoreactor fundamentals will form the basis for models that account for fluid flow and fluence as well as the kinetics involving microorganism inactivation or organic contaminant decomposition through radiation or oxidation. Taghipour said that the goal is to create an integrated model that can be widely used in water and wastewater treatment facilities and products.

“Once validated, the integrated model can be applied to virtual prototyping of UV reactors for design optimization,” he said. “This will eventually lead to a more efficient way of removing contaminants.”