Wastewater Disinfection With Ozone - Process Control And Operating Results

The City of Indianapolis, Indiana operates two 125 mgd advanced wastewater treatment plants with ozone disinfection. The rated capacity of the oxygen-fed ozone generators is 6,380 Ib/day, which is used to meet geometric mean weekly and monthly disinfection permit limits for decal conforms of 400 and 200 per 100 mL, respectively. Since 1989, a disciplined process monitoring and control program was initiated. Records indicate a significant effect on process performance due to wastewater flow, contactor influent fecal coliform concentration, and ozone demand. Previously, ozone demand information was unknown. Several tasks/studies were performed in order to better control the ozone disinfection process. These include the recent installation of a pilot-scale ozone contactor to allow the plant staff to measure ozone demand on a daily basis. Also, tracer tests were conducted to measure contactor short-circuiting potential. Results demonstrated a noticeable benefit of adding additional baffles. Results also indicated operating strategies that could maximize fecal coliform removal, such as reducing the number of contactors in service at low and moderate flow conditions.     

Designing High Concentration Ozone Contactors for Drinking Water Treatment Plants

The design of high concentration ozone contacting systems requires special attention to ensure that mass transfer, mixing and disinfection objectives are met when low volumetric gas flow rates are used. This becomes especially critical when the ozone contactor is designed for primary disinfection and must meet CT (ozone concentration multiplied by contact time] disinfection values. This paper presents design guidelines for high concentration ozone contactors. These guidelines then are applied to the design of a new 55-mgd two-stage ozonation potable water plant for Henrico County, Virginia. Finally, a new type of in–line ozone injection contactor is presented as an alternative to fine-bubble diffusion contactors to provide primary disinfection.     

Optimizing an Intermediate Ozone System Used for Primary Disinfection at a 55 MGD Surface Water Treatment Plant

The Henrico County VA Water Treatment Facility has been in operation since April 2004, making use of intermediate ozone for primary disinfection to take advantage of both the disinfection and DBP benefits. Since the summer of 2007 the staff at the Henrico WTF has embarked on an optimizing journey with their ozone system. This paper overviews the significant milestones accomplished on this journey as well as ongoing improvements. Optimizing efforts to reduce operating costs while achieving disinfection goals included: Staff training workshops; Automating ozone data collection and access; Operating system at optimum ozone concentration; Selecting optimum contactor sampling locations and number of contactors; Overcoming minimum gas flow limitations; Solenoid contactor sampling system issues and improvements; Disinfection calculation constraints; Excessive buildup of liquid oxygen (LOX) tank pressure. Each of these items is discussed following an overview of this 55 MGD surface water treatment facility.     

Ozone Contactor Hydrodynamics

Tracer tests were conducted at the 6,000 pounds of ozone per day Tucson, CAP Water Treatment Plant in Tucson, Arizona. The tests were designed to determine T₁₀ values through the contactors at various operating conditions. The tests were modeled using three techniques. Peclet Number was calculated for each of the runs, which would indicate the hydrodynamic conditions inside the ozone contactor. The results indicated that the increase in water flow rate and the number of cells with gas flow increased Peclet Number. The flow rate of liquid seemed to impact the Peclet Number more than gas flow. The headloss in each cell appeared to be important in controlling the distribution of liquid and gas through the cell. A correlation was developed between the product of gas and liquid phase Reynolds Number and Peclet Number.     

Comparing Different Designs and Scales of Bubble Columns for Their Effectiveness in Treating Kraft Pulp Mill Effluents

The experimental results obtained in three different types of ozone contactors were analyzed to study the effects of the ozone contactor design, configuration, operating conditions, and scale-up on the: (1) ozonation process induced reduction efficiencies of color, AOX, COD, and TOC from biologically treated Kraft pulp mill effluents; (2) the increase in biodegradability of this type of wastewater; and (3) the dynamics of the ozone gas absorption process. The three types of ozone contactors included: (1) an extra-coarse-bubble diffuser ozone contactor; (2) an impinging-jet ozone contactor; and (3) a fine-bubble diffuser ozone contactor. Similar treatment levels were achieved in those ozone contactors although the impinging-jet bubble column was more effective in treating Kraft pulp mill effluents due to its smaller reactor volume and lower off-gas ozone concentrations. Consequently, the operating costs of an ozonation process and ozone off-gas destruction facilities will be greatly reduced when using the impinging-jet bubble column design for treating Kraft pulp mill effluents.     

Ozone Side-Stream Design Options and Operating Considerations

Disinfection with ozone is achieved by CT product, where “C” is residual ozone concentration and “T” is reaction time. Ozone residual is developed after ozone gas is dissolved into the water. Ozone dissolution is most often achieved by either bubble-diffuser or side-stream ozone systems. Side-stream ozone systems are different from bubble-diffuser ozone systems in many ways: 1) Ozone addition occurs outside of the ozone contactor; 2) additional energy is used due to side-stream pump operation; 3) equipment maintenance can be completed independent of ozone contactor operation; 4) ozone contactor size and shape can have unique design features, such as pipeline contacting or shallow-depth basins. Side-stream ozonation is a treatment technique that is gaining in popularity at larger and larger plants. Ozone generators can now operate at elevated ozone concentration, which improves the economics for installing and operating side-stream processes. Two different side-stream arrangements are discussed in this paper. The arrangements differ with the presence or absence of a de-gas vessel. The de-gas vessel option is more expensive in terms of capital and operating cost, but might be the chosen option for particular contactor configurations. Issues and considerations for installing a de-gas vessel are discussed in this paper.     

Pilot-Scale Evaluation Of The Effects Of Mixing On Ozone Disinfection Of Escherichia Coli In A Semi-Batch,Stirred Tank Reactor

A research program was undertaken to examine the effects of mixing intensity and post-ozonation conditions on the survival of Escherichia coli ATCC 11775 in a high quality secondary wastewater effluent with a total organic carbon content of 8 mg/L and a chemical oxygen demand of 26 mg/L. The study was conducted using a pilot-scale, semi-batch stirred tank reactor with a 6-blade Rushton turbine. Two power conditions were investigated: 255 W/m³ and 870 W/m³. Two post-ozonation conditions also were investigated. In the first, a headspaceless sample was withdrawn and stored in the dark, in a quiescent condition. For the second, the gas flow to the reactor was stopped, but mixing was continued.

For equivalent contact times, it was found that two orders of magnitude more E. coli survived under the second post-ozonation condition when compared with the first condition. There was also a significant difference associated with the mixing intensity in the contactor, with the higher power input resulting in less efficient inactivation of E. coli.

The results confirm the importance of designing an ozone contactor to promote the maintenance of aqueous ozone in the contactor. In addition, optimum ozone mass-transfer may require different contacting conditions than those required for optimum disinfection performance. It was concluded that the design of ozone contactors should consider the use of at least two-stages: one optimized for ozone mass-transfer and one optimized for disinfection contacting.     

Mass Balance Analysis of Ozone in Conventional Bubble Contactors

Ozone transfer into potable water was studied in a commercial scale contactor. Ozone mass balances have been calculated to determine ozone utilization in the contactor. Gas and liquid flowrates, as well as inlet ozone concentrations in the gas were varied. From these data it was possible to determine the mass transfer coefficient, the ozone consumption and the transfer efficiency. Procedures for calculating the design of contactors and the optimal operating conditions are proposed.     

Modeling and Validating the Effective Hydraulic Detention Time for a 10 mgd Ozone Contactor at the Lake Washington Surface Water Treatment Plant,Melbourne, Florida

Ozonation has been added to the treatment process at the 20 mgd Lake Washington Surface Water Treatment Plant (SWTP), Melbourne, Florida to improve finished water quality, control taste and odor, and provide greater than 0.5-log credit for inactivation of Cryptosporidium. The new ozone contactors (2 parallel 10 mgd trains) are located between the existing Actiflo^TM clarifiers and the filters on a relatively congested site. Prior to construction, a Computational Fluid Dynamics (CFD) study indicated that the proposed contactor design was not optimized to provide the targeted Cryptosporidium log inactivation. Based on the CFD modeling, the design was modified to reduce the baffle heights (increasing end gaps at turns) and the baffle ends were chamfered to improve flow patterns within the contactor. Internal access platforms were rotated from parallel to the flow (along outside walls) to perpendicular to the flow (along baffle walls). These small design changes increased the modeled effective detention time (t₁₀) for disinfection and the corresponding hydraulic efficiency (baffling factor) by 22 %. In addition, the dosing location for the ozone quenching chemical (hydrogen peroxide) was modified to provide extra ozone detention. The ozone contactors were then constructed to incorporate the design changes optimized from the modeling. During the start up of the ozone system, fluoride tracer tests were completed to validate the effective detention time (and baffling factor) calculated by CFD. Furthermore, a series of modifications are planned for the ozonated water sampling regime to take greater advantage of Ct₁₀ credits and reduce the required ozone dose, thereby saving energy and operating costs. This paper discusses several calculation methods for Cryptosporidium inactivation Ct₁₀ reporting in compliance with the Long-Term 2 Enhanced Surface Water Treatment Rule (LT2ESWTR) and the LT2ESWTR Pre-proposal Draft Regulatory Language for Stakeholder Review (USEPA, 2002). The selection of the method will depend upon ease of implementation, modifications required to the ozone-in-water residual sampling system, long-term results of bromate sampling and actual Cryptosporidium bin classification under the LT2ESWTR. The article describes the CFD studies, improvements made to the contactor design, the tracer validation of the design, Ct₁₀ reporting methods and modifications to the ozone residual sampling system that the City is considering.     

Engineering and Economic Aspects of Wastewater Disinfection with Ozone Under Stringent Bacteriological Srandards

Successful wastewater disinfection was achieved under stringent bacteriological standards with a 480 g/hr generator pilot facility. Optimization of the facility required an applied dose of 30 mg/1 (13 to 19 W–hr/g ozone produced) to obtain a 2.2 total coliform per 100 ml disinfection level of nitrified effluent. An applied dose of 20 mg/1 (18 W–hr/g) is optimal for a 70 total coliform per 100 ml disinfection level of secondary effluent. The approach to ozonation system optimization, including generator mapping and contactor evaluation, is presented.     

Ozone Contactor Hydraulic Considerations in Meeting “CT” Disinfection Requirements

Tracer studies were performed in bench and pilot scale ozone diffusion contactors to determine actual contact times for the bench and pilot scale units and to characterize the flow pattern through these reactors. It was recognized that the flow pattern through the ozone contactors approached plug flow conditions. Due to the short circuiting and dispersion, the actual detention time in the reactors was one-half of the theoretically calculated detention time. Actual values of observed detention time were used in calculations of the “CT” values for ozone in the water. Optimization of ozone dose and contact time for “CT” calculations was performed in the pilot plant. Operational guidelines for the application of ozone in Union Hills pilot plant were developed based on the data obtained in the pilot plant runs.     

Modelling Hydraulics of Ozone Contactors

The U.S. EPA published the surface water treatment rule which imposed stringent requirements for disinfection. The rule is based on the C × T concept. The detention time T₁₀ must be determined for each type of contactor by performing tracer tests. Those tests have been performed for various ozone contactors. The results were evaluated using two types of models. For a classical diffuser-bubble contactor the ratio T₁₀/τ is about 0.5. For an industrial deep U-tube this ratio is equal to 0.55, but it reaches 0.9 for a pilot system. These results are indicating that a deep U–tube can provide a greater T₁₀/τ ratio than a diffuser–bubble contactor.     

A Two-Phase Computational Fluid Dynamics Model for Ozone Tank Design and Troubleshooting in Water Treatment

A computational fluid dynamics (CFD) tool was employed to design and study ozone contactors. The emphasis was to achieve the desired flow distribution. The Eulerian-Eulerian multiphase model was used with the standard k-? turbulent model. The water surface was slip wall boundary and was specified as a sink to remove ozone bubbles. For a single-column contactor with side entry, the flow pattern was found to be crucially dependent on both the direction and magnitude of the entry velocity from the inlet pipe. It was difficult to achieve uniform gas concentration over the contactor volume. In a multicompartment contactor, the countercurrent flow resulted in a mixed flow condition and the mixing increased with a higher gas rate. For the cocurrent flow, water was accelerated by the gas and the plug flow pattern was achieved. The flow distribution in each compartment can be significantly different even though the overall residence time distribution curves are similar.     

Design And Efficiency Of Ozone Contactors For Disinfection

One commercial contact chamber which is designed for ozone disinfection has been tested for the following : determination of residence time distribution (RTD), measurement of dissolved ozone concentrations in water, and measurement total plate counts. Disinfection is achieved but this type of reactor is not ideal, due to dispersion of the residence time distribution. As these bubble contactors do not behave like plug flow reactors, it is possible to improve the hydrodynamics of ozone contactors with the use of static mixers. Experiments with ozone static mixers have been conducted.     

Mathematical Modeling and Simulation of Ozonation Processes in a Downstream Static Mixer with Sieve Plates

New standards for drinking water disinfection require better optimization of the ozonation stage on the basis of the concentration×contact time (CT) concept, and production of ozone from pure oxygen at higher concentrations presumes application of the new type of contactors operating efficiently at lower gas/liquid volumetric ratios. One possible construction to meet these requirements is a downstream static mixer with sieve plates. At higher flow rates of liquid in this mixer, the interfacial area may reach 10,000m²/m³ at energy dissipation 1–5kW/m³. Due to the very intensive hydrodynamic regime the ozone utilization degree in the gas phase reaches 98–100% in natural lake water ozonation. Mathematical simulation of lake water ozonation in this mixer indicated that the process proceeds mostly in the diffusion or kinetic regime depending on the operating parameters. The dominating parameters besides the sieve geometry are the liquid flow rate in the holes of the sieves and the volumetric liquid/gas ratio.     

A Computational Fluid Dynamics Based Integrated Disinfection Design Approach for Improvement of Full-scale Ozone Contactor Performance

A three-dimensional multiphase computational fluid dynamics (CFD) model has been developed, to address all the major components of ozone disinfection processes at the Charles DesBaillets Water Treatment Plant in Montréal, Canada. Good agreement was observed between the numerical results and full-scale tracer data. The CFD predicted flow fields showed that recirculation zones and short circuiting existed in the DesBaillets contactors. Installation of four additional baffles in the second chamber would significantly improve contactor mixing performance. The modeling results also indicated that ozone residuals at the cross section of the outlet of each chamber were very sensitive to the locations of monitoring points.     

Ozone Plant Operations Under Drought Conditions

The Metropolitan Water District of Southern California (Metropolitan) owns and operates five potable water treatment plants. Currently four of the treatment plants utilize ozone as the primary disinfectant. The Metropolitan ozone contactors are over-under baffled and include 10 chambers, an afterbay, and an effluent channel. Ozone is generally added in either the first countercurrent gas/liquid flow chamber (Chamber 1) or the first and second (co-current gas/liquid flow) chambers. Disinfection credit is received from ozone residuals in Chambers 2 through 6. An ozone destruct system collects and destroys ozone off-gases from Chambers 1 through 9 and an ozone quenching system is used to minimize dissolved ozone from leaving the contactors. Severe drought in California, and resulting water supply limitations, have created significant operational issues ranging from reduced plant flows to source water changes and water quality challenges. The combination of reduced flow and different source water conditions resulted in the ozone systems being operated occasionally under unfamiliar conditions. This article discusses some of the challenges faced in operating the ozone plants at low flows and low ozone production rates. Further, water quality issues required prudent management of operations to meet regulatory requirements. From the lessons learned, several recommendations are offered in preparation for similar conditions in the future.     

Practical Design Of A New Ozone Contactor: The Deep U-Tube

This paper presents a practical design of a new ozone contactor (the deep U tube). It is a based on the modeling of upward and downward bubbly flow of air and water, derived from a balanced equation of continuity and momentum. Two hydrodynamic parameters are estimated: the void fraction and the frictional pressure drop. Experimental data carried out on full-scale plant and pilot plant are compared to theoretical results predicted by the model. Flow of the liquid phase is performed from tracer tests in order to determine the detention time which is introduced into the CT concept for disinfection. The influence of flow model shows that a plug flow reactor is more efficient than a CSTR; the DUT is close to plug flow.     

Ozone Disinfection of Sewage in a Static Mixer Contacting Reactor System on a Plant Scale

Results of our earlier laboratory study on ozone contacting systems in a continuous flow mode identified that the ozone disinfection process is limited by the mass transfer rate (7). The main controlling factor is the mass transfer efficiency rather than the contact time of the contactor in determining the effect of disinfection. By applying these concepts, we suggested a new ozone disinfection technique of using a static mixer as the contactor to substitute for a conventional bubble column designed with contact time.     

Ozonation Dynamics and Its Implication for Off-Gas Ozone Control in Treating Pulp Mill Wastewaters

Ozonation of biologically pretreated pulp mill wastewaters was studied using both bench and pilot scale fine bubble contactors to determine the oxidation efficiencies, mass transfer coefficients (k_La) and enhancement factors (E) due to the occurrence of chemical reactions. A sensitivity analysis based on the measured process parameters was then used to reveal the interrelated effects of key factors on off-gas ozone concentrations. It was shown that the removal efficiencies of color and AOX were simply related to the amount of utilized ozone, regardless of variation of other operating conditions. Furthermore, the rate of absorption fell within the fast or instantaneous kinetics regimes due to the occurrence of rapid chemical reactions. The Ek_La values were found to vary substantially during the course of ozonation, indicating that the enhancement factors were not only affected not only by operating conditions but also by wastewater characteristics. To effectively control the off-gas ozone emission, measures should be taken to minimize the backmixing, use a counter-current flow arrangement and provide adequate contact time.