Uncertainty Analysis of Conventional Water Treatment Plant Design for Suspended Solids Removal

An approach that links a Monte-Carlo simulation based reliability program with the water treatment process behavior and performance model is presented for uncertainty analysis of the conventional water treatment plant (WTP) design. The expected mass concentration of suspended solids (SS) in the effluent water is employed as a measurement of system reliability. The approach for uncertainty analysis of a WTP is illustrated with a hypothetical case study. The parameters contributing significantly to the variability of SS concentration in the effluent water are identified. From an operational viewpoint, the variability in effluent water quality resulting from uncertainty in the system parameters is investigated. Also, improvement in the reliability of WTP using modified design parameters is investigated along with its cost implications. From the results, a method to calculate a set of safety factors corresponding to various performance reliability levels is proposed.     

Reliability-Constrained Optimization of Water Treatment Plant Design Using Genetic Algorithm

A new approach that links genetic algorithm (GA) as an optimization tool with Monte Carlo simulation (MCS)-based reliability program is presented for reliability-constrained optimal design of water treatment plant (WTP). The reliability of a WTP is defined as the probability that it can achieve the desired effluent water quality standard (WQS). The objective function minimizes the treatment cost, subjected to design and performance constraints, and to achieve desired reliability level for meeting the given effluent WQS. The random variables used to generate the reliability estimates are suspended solids (SS) concentration, flow rate, specific gravity of floc particle, temperature of raw water, sedimentation basin performance index, and model coefficients. The application of GA-MCS approach for design of a WTP is illustrated with a hypothetical case study. The annualized cost of WTP is affected by the number of uncertain parameters included in the analysis, coefficient of variation of uncertain parameters, effluent WQS, and target reliability level. Analysis suggests that higher reliability at lower annual cost of treatment can be achieved by limiting the fluctuation of uncertain parameters. Results show that distribution of effluent SS is also affected by the uncertainty. The suggested GA-MCS approach is efficient to evaluate treatment cost-reliability tradeoff for WTP. Results demonstrate that the combination of GA with MCS is an effective approach to obtain the reliability-constrained optimal/near-optimal solution of WTP design problem consistently.     

Drinking Water Production from Well Water with High Sulfur and Sulfur Bacteria Content

Problems with treatment of Ahaste well water (Audru Parish, P?rnu County, Estonia) were studied in field conditions. The groundwater contains a significant amount of sulfur, from which at pH = 7.5–8.0 about 70–90% is in the form of hydrogen sulfide ion and 30–10% in the form of dissolved H2S. After aeration in the summer, this water formed an unknown slimy microbial mass with a distinct sulfuric odor, which clogged the aeration tanks, catalytic filters, and even the distribution network up to the consumers. After laboratory analyses, it was established that this unknown microbial mass is obviously the filiform sulfur bacteria Thiothrix, together with particles of free sulfur. It was found that this mass formed during enrichment under warm conditions using air as the oxidizer. Under these conditions, elemental sulfur was liberated and created favorable conditions for the Thiothrix to thrive using the elemental sulfur as their food source. A new treatment technology for this well water was developed using ozone as a strong oxidant/disinfectant, followed by filtration through an Everzit-Special, a material, which besides filtration, partially behaves as an adsorbent. The capital and operation costs of the new technology were estimated.     

Chromium Redox Chemistry in Drinking Water Systems

In order to understand the redox chemistry of chromium at low concentrations (100?μg/L) under conditions typically found in drinking water systems, three reductants and four oxidants were tested in three different waters at pH 5, 7, and 9. In the absence of any oxidant or reductant, Cr(VI) was stable at all three pHs, while Cr(III) precipitated out of solution at pH 9 and greatly impacted the reduction reactions. Stannous chloride was more effective than sodium sulfite or sodium sulfide for reducing Cr(VI) to Cr(III). Sulfide is not likely to be used as a reductant due to the long reaction time (120?h) to achieve the same reduction as SnCl2, while sulfite may be effective at higher doses. The oxidation of Cr(III) by dissolved oxygen and chloramine was very slow, while Cl2 and KMnO4 were effective oxidants under many conditions. A Cl2 residual in a drinking water distribution system may oxidize any soluble Cr(III) to Cr(VI) because of the long contact time, so Cr treatment strategies will need to remove both Cr(III) and Cr(VI).     

Reduction of Fulvic Acid in Drinking Water by Ferrate

Acting as a powerful oxidant, efficient coagulant, and effective disinfectant, ferrate is a good multifunctional agent for water and wastewater purification. In this study, some of these key aspects were studied for treating water containing fulvic acid, including the reduction efficiency of fulvic acid (FA) in drinking water and the optimum conditions for reducing FA using ferrate prepared in the laboratory. Jar tests and pilot tests were employed in the studies and the results of FA reduction were analyzed spectrophotometrically. Ferrate in these tests exhibited good oxidation capacity on FA. The ultraviolet absorption (UVA) of 2 mg/L FA in water can be reduced about 90% when the weight ratio of ferrate to FA was 12:1. Adsorption to and coprecipitation with Fe(OH)3 precipitate produced by ferrate decomposition can also result in additional FA reduction. In the presence of turbidity, ferrate can oxidize and adsorb simultaneously and over 95% of UVA was reduced with the same weight ratio of 12:1. Combining the application of ferrate and polyaluminum chloride (PAC) or iron chloride (FeCl3) was more effective for FA reduction. The UVA of 2 mg/L FA was reduced 100% in synthetic drinking water by a joint treatment with 8 mg/L ferrate (FeO42? by weight) and 0.8 mg/L PAC (Al by weight) or 6 mg/L ferrate (FeO42? by weight) and 0.8 mg/L Fe3+ (Fe by weight). The pilot test performed more effectively for FA reduction than did the jar test.     

Modified Larsons Ratio Incorporating Temperature, Water Age, and Electroneutrality Effects on Red Water Release

Corrosion indices have a historical as well as practical relevance in drinking water treatment. The development of reliable indicators of corrosion related problems, like red water, is an ongoing process in the drinking water industry. Due to the complexity of interaction among the physical, chemical, and biological reactions taking place within a typical distribution system, mechanistic models are difficult to formulate. Even if such a model was available, fitting it to actual field conditions would still be an empirical process. Corrosion indices give simplistic generalizations to complex corrosion phenomena. A modified form of the Larson Ratio that includes the effects of temperature and hydraulic retention time is proposed based on apparent color release data available from a 2 year pilot distribution system study.     

Numerical Modeling of Helical Static Mixers for Water Treatment

To improve the understanding of how static mixers work and how to better utilize them in environmental engineering (or, specifically, drinking water treatment), a numerical model for simulating turbulent flows in helical static mixers is developed. The model solves the three-dimensional, Reynolds-averaged Navier-Stokes equations, closed with the k-ω turbulence model, using a second-order-accurate finite-volume numerical method. Numerical simulations are carried out for a two-element helical static mixer, and the computed results are analyzed to elucidate the complex, three-dimensional features of the flow. The results show that the flow field within the mixer is characterized by the presence of pockets of reversed flow and the growth and interaction of strong longitudinal vortices. As an example of the kind of practical insights that can be gained from such detailed three-dimensional computations, the simulated flow field is used to investigate two quantities that are often used to characterize mixing within a static mixer and to discuss the merits of these quantities for coagulant mixing in drinking water treatment.     

Spatial Diversity of Chlorine Residual in a Drinking Water Distribution System

Chlorine residuals of drinking water have long been recognized as an excellent indicator for studying water quality in the distribution network. This research applied factor analysis and cluster analysis to determine the spatial diversity of chlorine residual in the distribution system of Feng-Yuan city. Thirteen sampling sites were established. From the results of factor analysis, the sampling sites of the study area could be classified into three groups: residential zone, mixed zone, and commercial zone. The spatial diversity of chlorine residual was found to correlate with the daily lifestyle of the inhabitants and the commercial activities. From the results of cluster analysis, the sampling sites of the study area could also be classified into three groups: high-chlorine-level zone, medium-chlorine-level zone, and low-chlorine-level zone. By combining the results of factor analysis and cluster analysis, the worst case scenarios for drinking water quality in the distribution network also could be determined.     

Synthetic Musk Fragrances in a Conventional Drinking Water Treatment Plant with Lime Softening

Synthetic musk fragrances are common personal care product additives and wastewater contaminants that are routinely detected in the environment. This study examines the presence of eight synthetic musk fragrances [7-acetyl-1,1,3,4,4,6-hexamethyl-1,2,3,4-tetrahydronaphthalene (AHTN), 1,3,4,6,7,8-hexahydro-4,6,6,7,8-hexamethylcyclopenta-γ-2-benzopyran (HHCB), 5-acetyl-1,1,2,6-tetramethyl-3-iso-propylindane (ATII), 4-acetyl-1,1-dimethyl-6-tert-butylindane (ADBI), 6-acetyl-1,1,2,3,3,5-hexamethylindane (AHMI), 6,7-dihydro-1,1,2,3,3,-pentamethyl-4-(5H)-indanone (DPMI), 1-tert-butyl-3,5-dimethyl-2,4,6-trinitrobenzene (musk xylene), and 4-tert-butyl-3,5-dinitro-2,6-dimethylacetophenone (musk ketone)] in source water and the removal of these compounds as they flow through a Midwestern conventional drinking water plant with lime softening. The compounds were measured in water, waste sludge, and air throughout the plant. HHCB and AHTN were detected in 100% of the samples and at the highest concentrations. A mass balance on HHCB and AHTN was performed under warm and cold weather conditions. The total removal efficiency for HHCB and AHTN, which averaged between 67–89%, is dominated by adsorption to water softener sludge and its consequent removal by sludge wasting and media filtration. Volatilization, chlorine disinfection, and the disposal of backwash water play a minor role in the removal of both compounds. As a result of inefficient overall removal, HHCB and AHTN are a constant presence at low levels in finished drinking water.     

Modeling and Control of Vinyl Chloride in Drinking Water Distribution Systems

In water distribution systems containing PVC pipe manufactured in the “early era” (prior to 1977), vinyl chloride can leach into drinking water resulting in vinyl chloride concentrations exceeding the 2 μg?L?1 maximum contaminant level. Field testing of dead-end segments of water distribution systems consisting of early-era PVC pipe was conducted to examine their initial intrapipe vinyl chloride monomer (VCM) concentrations based on a Fickian-diffusion-based leaching model. The experiments showed a wide range of VCM concentrations within early-era PVC pipe ranging from less than 50 to more than 600 mg?kg?1. Based on the diffusion modeling approach, a protocol was designed that provides a means for utility managers to calibrate the model for specific dead-end lines. The paper delineates procedures to determine which dead ends require flushing to control vinyl chloride, examines the effects of system parameters such as temperature on vinyl chloride leaching, and provides a method to devise flush schedules and volumes. Through a properly designed, tested, and maintained flush protocol such as that developed in this research, public water systems with dead-end lines consisting of early-era PVC pipe can control vinyl chloride concentrations using either manual or automatic flush valves.     

Design and Operation of Point-of-Use Treatment System for Arsenic Removal

A point-of-use (POU) system was designed and constructed using commercially available activated alumina to remove arsenic from drinking water. Testing with City of Albuquerque chlorinated tap water containing an average of 23 ug/L arsenic found that 1 L of adsorbent would provide water for direct consumption by a family of four for 435 days. It was estimated that the POU system constructed for this study could be sold for $162, and the arsenic adsorption columns were estimated to cost $4. A monthly cost to the customer of $10/month was estimated to purchase, install, and operate this POU system, assuming annual replacement of adsorption media cartridges. The implications of relying upon POU systems to comply with a new drinking water standard for arsenic are discussed.     

Hydrophobicity and Molecular Size Distribution of Unknown TOX in Drinking Water

The analysis of total organic halogen (TOX) in drinking water indicates that a substantial amount of the halogenated compounds cannot be accounted for by known specific disinfection by-products (DBPs). The primary aim of this research was to characterize the hydrophobicity and molecular size distribution of the unknown halogenated DBPs using XAD resins and ultrafiltration membranes. The impact of membrane rejection on the size analysis of unknown TOX was also investigated using chlorinated fulvic acid. Six finished waters from different locations and treatment processes were collected and fractionated into various hydrophobicity and molecular size groups. The results showed that most unknown TOX was in the size range between 0.5?kDa and 10?kDa, but it could have a wide spectrum of hydrophobicities. Simple ultrafiltration was not always reliable as a characterization tool, as it was shown to reject a significant fraction of DBPs with molecular weight (MW) lower than the membrane cutoffs. Flushing with deionized water was effective in removing these low MW compounds from the ultrafiltration cell. A significant reduction in the apparent size of unknown TOX resulted when low MW DBPs were flushed out of the cell (comparing with classic parallel ultrafiltration). Coagulation of fulvic acid also significantly reduced the apparent size of unknown TOX formed by chlorine.     

Removal of Antibiotics from Surface and Distilled Water in Conventional Water Treatment Processes

Conventional drinking water treatment processes were evaluated under typical water treatment plant conditions to determine their effectiveness in the removal of seven common antibiotics: carbadox, sulfachlorpyridazine, sulfadimethoxine, sulfamerazine, sulfamethazine, sulfathiazole, and trimethoprim. Experiments were conducted using synthetic solutions prepared by spiking both distilled/deionized water and Missouri River water with the studied compounds. Sorption on Calgon WPH powdered activated carbon, reverse osmosis, and oxidation with chlorine and ozone under typical plant conditions were all shown to be effective in removing the studied antibiotics. Conversely, coagulation/flocculation/sedimentation with alum and iron salts, excess lime/soda ash softening, ultraviolet irradiation at disinfection dosages, and ion exchange were all relatively ineffective methods of antibiotic removal. This study shows that the studied antibiotics could be effectively removed using processes already in use in many water treatment plants. Additional work is needed on by-product formation and the removal of other classes of antibiotics.     

Incorporating Reservoir Transfer into Treatment Optimization Decisions

The Massachusetts Water Resources Authority (MWRA) supplies unfiltered water from two large surface water reservoirs to the metropolitan Boston area, as well as to three smaller communities in central Massachusetts [the Chicopee Valley Aqueduct (CVA) communities]. Quabbin Reservoir is larger than Wachusett Reservoir, and has traditionally been used to supplement the Wachusett during the summer period. Quabbin water is also of better quality, with lower reactive natural organic matter (NOM). The MWRA began to add chlorine at Wachusett in 1997, and a new facility for adding chlorine at Quabbin for the CVA was also started up in 2000 to meet primary disinfection regulations to meet pathogen inactivation. The reaction of chlorine with NOM produces undesirable disinfection by-products (DBPs). The absorption of ultraviolet light at a wavelength of 254 nm was identified in chlorine decay studies to be the most important raw water quality parameter for predicting chlorine decay and DBP formation. This technical note summarizes the chlorine decay model for Wachusett and Quabbin water. The model is extended to ozonation of Wachusett water for the future Walnut Hill treatment plant. The models allowed the development of a trigger using UV-254 to time the Quabbin transfer to optimize treatment results. It is believed that the model for disinfectant decay and the use of UV-254 as a trigger for water treatment decisions are generalized and applicable to other water utilities.     

Effect of ClO2 Pretreatment on Subsequent Water Treatment Processes

The effect of chlorine dioxide (ClO2) pretreatment on subsequent treatment processes (coagulation, flocculation, sedimentation, filtration, and ozonation) was studied at pilot-scale at the Upper San Leandro Water Treatment Plant near Oakland, Calif. Potential impacts of ClO2 on the distribution system were also studied at bench scale using simulated distribution system (SDS) tests. Pilot trials were conducted with one train operating without ClO2 pretreatment (Train 1) and the other with a ClO2 dose of between 0.6 and 1.0?mg/L (Train 2). Comparison between Trains 1 and 2 showed that ClO2 pretreatment resulted in a 0.1–0.2 NTU decrease in settled water turbidity when compared to no pretreatment. ClO2 pretreatment also resulted in a small (0.01?cm?1) decrease in ultraviolet absorbance at 254?nm. Following sedimentation, about 60% of the applied ClO2 formed chlorite (ClO2?), with 10–20% forming chlorate (ClO3?). Ozonation immediately converted all residual ClO2 and ClO2? to ClO3?. There was no significant difference in the performance of the filters between the two trains in terms of headloss, particle count, and turbidity. Bench-scale SDS tests indicated that chlorine dioxide preoxidation did not affect subsequent chloramine stability or concentrations of trihalomethanes, haloacetic acids, or adsorbable organic halides in the distribution system.     

Inactivation of Adenovirus Types 2, 5, and 41 in Drinking Water by UV Light, Free Chlorine, and Monochloramine

A bench-scale study was conducted to determine the inactivation of adenovirus (Ad) types 2, 5, and 41 by ultraviolet (UV) light, chlorine, and monochloramine. The motivation for this study was to determine whether UV disinfection followed by chlorine or monochloramine for a very short contact time (e.g., a minute) could satisfy regulatory requirements for four-log virus inactivation. In order to overcome the difficulty Ad 41 presents for enumeration of the virus in cell culture, a technique was used that combined immunofluorescent staining of viral antigen with traditional scoring of cytopathic effect. A UV dose of 40?mJ/cm2 (millijoules per square centimeter) (applied using a collimated beam apparatus) achieved approximately one-log inactivation of adenovirus types 2, 5 and 41, confirming previous research. Ad 41 was found to be more UV resistant to UV light than Ad 2 or Ad 5 at UV doses >70?mJ/cm2 to a statistically significant degree (95% confidence); however, at lower UV doses there were no statistically significant differences. Experiments with Ad 5 and Ad 41 at 5°C and pH 8.5 showed that chlorine was very effective against Ad 5 and Ad 41, with a product of disinfectant concentration and contact time (CT) of 0.22?mg min/L providing four-log inactivation. Monochloramine was less effective against these adenoviruses, with a CT of 350?mg min/L required to achieve 2.5-log inactivation of Ad 5 and 41 at 5°C and pH 8.5.     

Condition Assessment of Water Treatment Plant Components

Potable water treatment is one of the most challenging and complex systems that municipalities need to deal with considering limited resources. A study from mid-90s showed that the continuously deteriorating Canadian water supply system would require $3.1 billion to bring the system at satisfactory level. Drinking water treatment plants (WTP) include several components, such as tanks, basin, and pumps. Operators are able to spend a small portion of the available resources or their plant’s infrastructure and equipment compared to water quality and day-to-day operational activities. The research presented in this technical paper aims at developing condition assessment model(s) for the WTP components. Essential condition parameters of WTP include technical, physical, environmental, and operational aspects. To determine the condition index of a WTP component, value additive multi-attribute theory has been used where average weights and scores are considered for the model parameters. Data on WTP conditions are collected from experts and consultants across Canada and the United States. It is concluded from the model results that the average condition index for settling basins, ranges from 9.6 (best scenarios) to 1.9 (worst scenarios) and from 9.6 to 3.4 for pumps. Analysis reveals that, for tank and basins, design and construction parameter is the most important parameter for WTP condition, while the operational parameter is the most important one for pumps. The developed models are expected to benefit academics and practitioners (municipal engineers, consultants, and contractors) to prioritize inspection and rehabilitation planning for existing water treatment plants.     

Modeling the Impact of Microbial Intrusion on Secondary Disinfection in a Drinking Water Distribution System

The purpose of this study was to quantify the potential level of protection that secondary disinfection may provide in response to an intrusion event. Although several uncertainties exist regarding intrusion events, this study presents an analysis of the inactivation provided by disinfectant residuals by using a distribution system model, inactivation and disinfectant decay models, and conservative assumptions based on available data. A variety of conditions were modeled, including a range of water quality parameters (pH, temperature); inactivation of two microorganisms, Giardia and E. coli O157:H7; and intrusion water dilution ratios. Despite the assumptions inherent in the model, several generalizations were derived from the study. A free chlorine residual of 0.5?mg/L may be insufficient to provide adequate control of disinfectant-resistant Giardia even at low pH (6.5) and high temperature (25°C) conditions that enhance chlorine effectiveness. For E. coli, an organism of “average” disinfectant resistance relative to others, a residual of 0.5?mg/L may provide ample protection against intrusion even assuming that the chlorine residual is reduced within several minutes, such as would be predicted to occur with sewage intrusion at levels below 1% of the total flow. Importantly, chloramines may have a negligible benefit in terms of protecting against intrusion for even relatively susceptible organisms such as E. coli. Consequently, systems should consider protection against intrusion when choosing their secondary disinfectant.     

Evaluation of a New Field Measurement Method for Arsenic in Drinking Water Samples

The recently lowered arsenic maximum contaminant level will require numerous U.S. water utilities and agencies to monitor and treat for arsenic. This paper describes a new method that measures arsenic in drinking water samples by generating arsine gas from the water and detecting the arsine using a paper-tape instrument. Laboratory tests indicated the method is capable of accurately detecting arsenic in water samples at the microgram per liter level (method detection limit of 0.5?μg/L and practical quantification limit of 2.5?μg/L). The only significant interferences were hydrogen sulfide and antimony. Using the paper-tape instrument, it is also possible to detect As(III) and As(V) that have been separated by either selective arsine generation or ion exchange. While the method proved accurate in the lab, difficulties were encountered during preliminary field testing on 18 different real samples. This technique of converting aqueous arsenic to arsine gas for analysis shows great promise, but the method needs to be refined for use in the field.     

Estimating the Removal of Anthropogenic Organic Chemicals from Raw Drinking Water by Coagulation Flocculation

A mathematical model was developed to estimate the efficacy of coagulation–flocculation treatment for removing neutral hydrophobic organic chemicals from raw drinking water. The model assumed that the only significant removal mechanism was the destabilization and settling of organic matter containing sorbed anthropogenic organic compounds. The model was validated with standard jar tests using compounds with a range of hydrophobicities (log?Kow = 1.89?to?5.48), including contaminant candidate list chemicals, pesticides, pharmaceuticals, and endocrine disrupting chemicals. Final concentrations of test compounds after coagulation and flocculation were in good agreement with model estimations for synthetic waters composed of Aldrich (Milwaukee, WI) humic acid solutions. The final compound concentrations in coagulated natural waters from two drinking water reservoirs were about 80% lower than those estimated with the model. Overestimations of treated water concentrations by the model were attributed to an increase in sorption by natural organic matter when coiled in aluminum hydroxide flocs, compared to sorption to dispersed natural organic matter in untreated water.