Modeling Chloramines Formation in Turbulent Flow
No Thumbnail Available
Files
Date
2004-12-03
Authors
Journal Title
Series/Report No.
Journal ISSN
Volume Title
Publisher
Abstract
A study was performed to investigate the use of Computational Fluid Dynamics (CFD) for the analysis of ammonia injection methods to produce chloramines. Three types of ammonia diffusers were simulated and tested in the present work. In this study, CFD fluid flow and turbulence models were combined with chloramines kinetic models to predict downstream spatial distribution of residual free chlorine. As part of this study, two chemical species transport models, single fluid model (SFM) and multi-fluid model (MFM), were evaluated. In addition, several turbulence models were combined with the MFM approach to investigate the impact of turbulent model selection on the free chlorine residual. The turbulence models examined in this study include the standard k-e, RNG k-e, and k-ω models. All model predictions were compared with experimental measurements of the free chlorine residual at different upstream and downstream locations from the ammonia injection point. The experimental tests include the placement of a perforated baffle to investigate the impact of background turbulence on the overall mixing process. A simple flow visualization experiment using an inert tracer was also performed to qualitatively evaluate the fluid flow pattern in the vicinity of the ammonia injection point. Furthermore, a sensitivity analysis was performed to investigate the impact of several turbulent mixing time scales (eddy-dissipation time scale, Kolmogorov time scale, and Corrsin time scale) on the residual free chlorine concentration.
Results showed that the flow field with the standard turbulence model and RNG turbulence model reasonably matched the fluid flow pattern characterized by the flow visualization test. The CFD/MFM modeling approach was found to better predict the free chlorine spatial distribution than the CFD/SFM approach. Moreover, results showed that the CFD chloramines model with the standard or RNG turbulence model was able to predict the downstream chloramines formation for a cone-shape diffuser and a three-bar diffuser. Results also showed that the CFD chloramines model with the EDT or KOLM time scale was able to predict the downstream chloramines formation for the cone-shape diffuser and the three-bar diffuser. However, the CFD chloramines model with different turbulence models consistently over-predicted the residual free chlorine for a T-bar diffuser. This could be explained by the complex turbulent structure, produced by a strong inlet jet, was not properly characterized using these two-equation turbulence models. The CFD/MFM chloramines model was found to improve the prediction of the free chlorine spatial distribution for the T-bar diffuser when the Corrsin time scale was used. In addition, a baffle, placed upstream from the injection point to reduce the strong inlet jet, was found to have significant impact on the mixing and chloramines formation for the T-bar diffuser mixer case. Overall, the CFD/MFM chloramines models of the baffle configuration better predicted residual free chlorine for all three diffuser configurations than the un-baffled reactor cases.
Description
Keywords
MFM, Chloramines, CFD
Citation
Degree
PhD
Discipline
Civil Engineering
