Comparing Residence Time Distributions for Ozone Contactors Predicted by Common CFD Codes

Disinfection contactor design and construction can have a significant impact on overall infrastructure costs for water treatment plants (WTP). Experimental tracer studies are used to gain insight into contactor hydraulics to improve volumetric efficiency and to encourage plug-flow behavior and for calculating baffling factors (BF). Often, in the absence of or due to the incompleteness of experimental data, computational fluid dynamics (CFD) is used as a tool to better understand contactor hydraulics. Several approaches and CFD codes are available for modeling the residence time distribution (RTD) of tracers. This paper compares the RTD predictions of three commonly used CFD codes for a full-scale ozone contactor. Multiphase modeling is performed to quantify the effect of volume fraction of ozone gas, method of mixing, and of contactor geometry on the predicted RTD. Results are compared to experimental data for ozone contactors from large water treatment plants. The impacts of poor model prediction can be significant, including failure to achieve target mass transfer efficiency, target dissolved ozone residual and disinfection Ct credits. Since a practical problem was used, the goal of this study is to inspire confidence in different CFD codes and if any discrepancy is observed, provide justifications for features that may cause such differences.