Combination of ferric and MIEX for the treatment of a humic rich water

Seasonal periods of high rainfall have been shown to cause elevated natural organic matter (NOM) loadings at treatment works. These high levels lead to difficulties in removing sufficient NOM to meet trihalomethane (THM) standards, and hence better alternative treatments are required. Here the removal of NOM was investigated by a new ion exchange process (MIEX) using both bulk and fractionated NOM. Initial results showed that in excess of 80% of the raw water dissolved organic carbon (DOC) and greater than 85% of the UV absorbance from the bulk raw water could be removed by the use of MIEX alone. It was also seen that the removal of the more recalcitrant isolated fractions was increased. When MIEX was combined with a significantly reduced dose of coagulant a slight improvement on the overall DOC and UV removals was observed, however a significant decrease in the amount of THM formation potential (THMFP) in the final water was seen. This combined with the reduction in coagulant would imply a more efficient process during the times when the water becomes increasingly difficult to treat.     

The effect of ozonation on natural organic matter removal by alum coagulation

Natural organic matter (NOM) was extracted from a moderately colored, eutrophic surface water source (Forge Pond, Granby, MA), and fractionated into quasi-homogeneous fractions. Fulvic acid (FA) and hydrophilic neutrals (HN) were the two most abundant NOM fractions that were isolated. Adsorption affinity of the isolated NOM fractions on preformed aluminum hydroxide flocs increased with increase in specific organic charge of the fractions, except for the two most highly charged fractions, FA and hydrophilic acids (HAA), which showed less adsorption affinity than expected based on their specific organic charge. Prior ozonation of FA and HN fractions resulted in a decline and an increase, respectively, in their adsorption affinity on aluminum hydroxide surface. Prior ozonation of Forge Pond raw water resulted in a progressive decline in dissolved organic carbon (DOC) removal by alum coagulation with increase in ozone dose. It appeared that ozone applied to raw water reacted preferentially with the humic fraction of NOM, resulting in the detrimental effects of ozonation on subsequent NOM removal by alum coagulation being magnified. Forge Pond raw water was pre-coagulated to remove humic substances. Ozonation of the pre-coagulated water demonstrated the beneficial effects of ozonation on the removal of non-humic NOM through alum coagulation. A strategy for staged coagulation with intermediate ozonation was proposed for waters containing both humic and non-humic NOM for maximum DOC and specific UV absorbance at 254nm (SUVA) removal.     

The effects of combined ozonation and filtration on disinfection by-product formation

The effects of combined ozonation and membrane filtration on the removal of the natural organic matter (NOM) and the formation of disinfection by-products (DBPs) were investigated. Ozonation/filtration resulted in a reduction of up to 50% in the dissolved organic carbon (DOC) concentration. Furthermore, humic substances were converted to non-humic substances, with changes in the humic and non-humic substance concentrations of up to −50% and +20%, respectively. Ozonation/filtration resulted in the formation of partially oxidized compounds from NOM that were less reactive with chlorine, decreasing the concentration of simulated distribution system total trihalomethanes (SDS TTHMs) and simulated distribution system halo acetic acids (SDS HAAs) by up to 80% and 65%, respectively. Reducing the molecular weight cut-off (MWCO) of the membranes resulted in reductions in the concentrations of SDS TTHMs and SDS HAAs. Using a membrane with a 5 kD MWCO, the minimum gaseous ozone concentration required to bring about effective NOM degradation and meet regulatory requirements for chlorinated DBPs was 2.5 g/m³.     

Disinfection byproduct formation in reverse-osmosis concentrated and lyophilized natural organic matter from a drinking water source

Drinking water treatment and disinfection byproduct (DBP) research can be complicated by natural organic matter (NOM) temporal variability. NOM preservation by lyophilization (freeze-drying) has been long practiced to address this issue; however, its applicability for drinking water research has been limited because the selected NOM sources are atypical of most drinking water sources. The purpose of this research was to demonstrate that reconstituted NOM from a lyophilized reverse-osmosis (RO) concentrate of a typical drinking water source closely represents DBP formation in the original NOM. A preliminary experiment assessed DBP formation kinetics and yields in concentrated NOM, which demonstrated that chlorine decays faster in concentrate, in some cases leading to altered DBP speciation. Potential changes in NOM reactivity caused by lyophilization were evaluated by chlorination of lyophilized and reconstituted NOM, its parent RO concentrate, and the source water. Bromide lost during RO concentration was replaced by adding potassium bromide prior to chlorination. Although total measured DBP formation tended to decrease slightly and unidentified halogenated organic formation tended to increase slightly as a result of RO concentration, the changes associated with lyophilization were minor. In lyophilized NOM reconstituted back to source water TOC levels and then chlorinated, the concentrations of 19 of 21 measured DBPs, constituting 96% of the total identified DBP mass, were statistically indistinguishable from those in the chlorinated source water. Furthermore, the concentrations of 16 of 21 DBPs in lyophilized NOM reconstituted back to the RO concentrate TOC levels, constituting 86% DBP mass, were statistically indistinguishable from those in the RO concentrate. This study suggests that lyophilization can be used to preserve concentrated NOM without substantially altering the precursors to DBP formation.     

The use of MSSV pyrolysis to assist the molecular characterisation of aquatic natural organic matter

Microscale sealed vessel pyrolysis (MSSVpy) with online gas chromatography mass spectrometry (GC-MS) was used with several other established and complementary analytical methods to robustly characterize the structure of aquatic natural organic matter (NOM) and to practically assess the analytical value of MSSVpy. The NOM used in the study was from North Pine (NP) reservoir, which is one of the major source waters supplying Brisbane, the capital city of the Australian state of Queensland. The reservoir has moderate dissolved organic carbon (DOC; 5 mg L⁻¹) levels and is impacted by algae which periodically occur in bloom proportions. The hydrophobic (HPO; 65% initial DOC) and transphilic (TPI; 12%) fractions showed H/C values >1, low UV_abs and low aryl-C measured by NMR which are all indicative of low aromaticity. MSSVpy produced distinctly higher product concentrations than traditional flash pyrolysis and the molecular profile of the HPO and TPI fractions revealed by MSSVpy was correlated with the other analytical data to help establish their structural relevance. Prolific distributions of alkyl substituted aromatic (e.g., benzenes, naphthalenes) and hydroaromatic (e.g., tetralins) products detected in the HPO fraction were attributed to the aromatisation of terpanes and other aliphatic compounds from algal, and possibly also plant sources. Alkyl phenols also detected in HPO in high abundance, are probably from algal biopolymers, but may also reflect a contribution from non-methoxylated lignin units of catchment grasses. There was no analytical evidence of the dihydroxy or methoxy aromatic structures typical of wood lignin or tannin. N-organic pyrolysates (e.g., alkyl pyrroles, pyridines, indoles) of diagenetically altered proteins were concentrated in the TPI fraction. The quantitative importance of the N-organic moiety of the TPI fraction was corroborated by a low C/N ratio and distinctive amide and amine signals detected by ¹³C NMR and FTIR. This integrated study demonstrates that the qualitative speciation provided by MSSVpy can make a significant contribution to the structural characterisation and source recognition of aquatic NOM.     

Characterization of natural organic matter in conventional water treatment processes for selection of treatment processes focused on DBPs control

Natural organic matter (NOM) from raw and process waters at a conventional water treatment plant was isolated into hydrophobic and hydrophilic fractions by physicochemical fractionation methods to investigate its characteristics. Formation potential of trihalomethanes (THMs) was highly influenced by the hydrophobic fraction, whereas haloacetic acids formation potential (HAAFP) depended more on the hydrophilic fraction. However the hydrophobic fraction was removed more than the hydrophilic fraction through conventional water treatment. Therefore residual hydrophilic NOM after conventional treatment needs to be removed to reduce HAAFP. Feasible additional processes are required to be evaluated by comparing preferential removal efficiency of hydrophilic NOM through pilot tests. The structural and chemical characteristics of hydrophobic NOM (i.e., humic substances (HS)) were further investigated to know how they are influenced by conventional treatment. The phenolic fraction in the hydrophobic NOM was mainly removed compared to the carboxylic fraction through water treatment, and a higher formation potential of THMs resulted from NOM with a higher phenolic content. The Fourier-transform infrared (FT-IR) and proton nuclear magnetic resonance ((1)H-NMR) employed for characterization of NOM through water treatment were insightful revealing that their results were quite close to each other. Decreases of ratio of UV absorbance at 253 and 203 nm, respectively (A(253)/A(203) ratio) and trihalomethane formation potential/dissolved organic carbon (THMFP/DOC) showed consistent trends; therefore, the A(253)/A(203) ratio may be a good indicator of tendency for the formation potential of disinfection by-products (DBPs).     

Influences of solution chemistry and polymeric natural organic matter on the removal of aquatic pharmaceutical residuals by nanofiltration

This study demonstrates the removal efficiency and the permeate flux behavior during cross-flow nanofiltration (NF) of aqueous solutions of five pharmaceutically active compounds (PhACs). Cephalexin, tetracycline, acetaminophen, indomethacin and amoxicillin were used as models of PhACs, and alginate was selected as model of natural organic matter (NOM). Two commercial composite NF membranes (SR2 and SR3) with different characteristics were used. The highest rejection was observed for tetracycline, i.e., 75-95% for membrane SR 2 and 95-100% for membrane SR 3, while the rejection was least for acetaminophen (32-36% for SR2 and 52-59% for SR3). As the pH of acetaminophen solution was increased (from 6 to 9) the rejection would increase. Changes of ionic content (from 10 to 20 mM) lead to increase (from 89 to 93% for SR 3) or decrease (from 100 to 91% for SR2) of cephalexin rejection depending on the membrane used. The permeate flux would decrease with decreasing the pH solution and increasing ionic strength. The addition of alginate in the feed stream decreased the permeate flux, with lower reduction for SR3, and increased the PhAC rejection except for acetaminophen and amoxicillin. Both size and Donnan exclusions seemed to occur, and the effect of Donnan exclusion was more pronounced for the NF membrane having larger effective pore size (SR2).     

Modeling dichloroacetic acid formation from the reaction of monochloramine with natural organic matter

A kinetic model was developed to predict dichloroacetic acid (DCAA) formation in chloraminated systems. Equations describing DCAA formation were incorporated into an established comprehensive monochloramine-natural organic matter (NOM) reaction model. DCAA formation was theorized to be proportional to the amount of NOM oxidized by monochloramine and described by a single dimensionless DCAA formation coefficient, theta(DCAA) (M(DCAA)/M(DOC(ox)). The applicability of the model to describe DCAA formation in the presence of six different NOM sources was evaluated. DCAA formation could be described by considering a single NOM source-specific value for theta(DCAA) over a wide range of experimental conditions (i.e., pH, NOM, free ammonia, and monochloramine concentrations). DCAA formation appears to be directly proportional to the amount of active chlorine (monochloramine and free chlorine) that reacted with the NOM under these experimental conditions. Values of theta(DCAA) for all six NOM sources, determined by nonlinear regression analysis, varied from 6.51 x 10(-3) to 1.15 x 10(-2) and were linearly correlated with specific ultraviolet absorbance at 280 nm (SUVA(280)). The ability to model monochloramine loss and DCAA formation in the presence of NOM provides insight into disinfection by-product (DBP) formation pathways under chloramination conditions. The subsequent model and correlations to SUVA has the potential to aid the water treatment industry as a tool in developing strategies that minimize DBP formation while maintaining the microbial integrity of the water distribution system.     

Multi-wavelength spectroscopic and chromatography study on the photocatalytic oxidation of natural organic matter

The effect of TiO₂ photocatalytic oxidation on the natural organic matter (NOM) properties of two Australian surface waters were quantified using UV-vis spectroscopy, high performance size exclusion chromatography (HPSEC) with a multi-wavelength UV detector, liquid chromatography with organic carbon detector (LC-OCD), and trihalomethane formation potential (THMFP) analyses. Both the UV absorbance at wavelengths greater than 250 nm and dissolved organic carbon (DOC) content decreased significantly with treatment, although complete mineralization of NOM could not be achieved. Multi-wavelength UV detection of HPSEC analysis was shown to be useful to display further changes to NOM composition and molecular weight profiles because the organic molecules was transformed into compounds that absorb weakly at the typical detection wavelength of 250-260 nm. The multi-wavelength HPSEC results also revealed that photocatalytic oxidation yields by-products with a low aromaticity and low molecular weight. The LC-OCD chromatograms indicated that low molecular acids and neutral compounds remained after photocatalytic oxidation. Those groups of compounds did not seem to contribute significantly to the formation of trihalomethanes.     

Efficacy and environmental acceptability of two ballast water treatment chemicals and an alkylamine based-biocide

Ship's ballast waters transport large numbers of organisms which may become invasive in coastal regions. One option to address this problem is the use of biocides as ballast water treatment (BWT). Efficacy and environmental acceptability of three commercial active substances (the BWT biocides Peraclean^® Ocean and Seakleen^®, and alkylamine-based biocide Mexel^® 432/336) were tested against three bacteria species, two vegetative microalgae and one zooplanktonic larva, in 10 and 30 Practical Salinity Unit (PSU) waters. In both salinities, PeraClean^® Ocean was the most effective biocide against bacteria causing > 90% mortality at 20 mg/l, compared with 50 mg/l for Mexel^® 432/336 and > 500 mg/l for Seakleen^®. Regarding zooplankton, Seakleen^® was the most effective chemical causing 90% mortality in 24 h at concentrations < 6 mg/l (LC90_24 h) in both salinities, compared with 23 and 26 mg/l for Mexel^® 432/336 and 370 and 480 mg/l for PeraClean^® Ocean in 10 and 30 PSU, respectively. Similar pattern of efficacy was obtained for microalgae in 30 PSU: effective concentrations inducing 50% growth inhibition in 4 days were ≤ 1.6 mg/l for Seakleen^®, ≤ 10.1 mg/l for Mexel^® 432/336 and ≤ 30.9 mg/l for PeraClean^® Ocean. Our work highlighted that treated waters displayed residual toxicity after 24 h still inducing mortality depending on the organism and biocide. However Mexel^® 432/336 is the only biocide which had no impact on oyster larvae development at effective concentration. Altogether our data showed that Mexel^® 432/336 was the only biocide displaying a broad spectrum efficacy in concentrations < 50 mg/l and not toxic for oyster larvae development at this concentration. However residual toxicity of treated waters for any organism should be taken into account in BWT systems utilising biocides.     

The influence of natural organic matter and cations on the rejection of endocrine disrupting and pharmaceutically active compounds by nanofiltration

The impact of natural organic matter (NOM) and cations on the rejection of five endocrine disrupting compounds (EDCs) and pharmaceutically active compounds (PhACs) (acetaminophen, carbamazepine, estrone, gemfibrozil, oxybenzone) by nanofiltration (NF) was examined. The water matrices included membrane bioreactor (MBR) effluent, Lake Ontario water and laboratory-prepared waters modelled to represent the characteristics of the Lake Ontario water. The impact of cations in natural waters on compound rejection was also examined by doubling the natural cation concentration (calcium, magnesium, sodium) in both the Lake Ontario water and the MBR effluent. The presence of Suwannee River NOM spiked into laboratory-grade water was found to cause an increase in compound NF rejection. In addition, the presence of cations alone in laboratory-grade water did not have a significant impact on rejection with the exception of the polar compound gemfibrozil. However, when cation concentration in natural waters was increased, a significant decrease in the rejection of EDCs and PhACs was observed. This suggests that the presence of cations may result in a reduction in the association of EDCs and PhACs with NOM.     

Comparison of advanced oxidation processes for the removal of natural organic matter

This study examined the impact of UV, ozone (O₃), advanced oxidation processes (AOPs) including O₃/UV, H₂O₂/UV H₂O₂/O₃ in the change of molecular weight distribution (MWD) and disinfection by-product formation potential (DBPFP). Bench-scale experiments were conducted with surface river water and changes in the UV absorbance at 254 nm (UV₂₅₄), total organic carbon (TOC), trihalomethane and haloacetic acid formation potential (THMFP, HAAFP) and MWD of the raw and oxidized water were analyzed to evaluate treatment performance. Combination of O₃ and UV with H₂O₂ was found to result in more TOC and UV₂₅₄ reduction than the individual processes. The O₃/UV process was found to be the most effective AOP for NOM reduction, with TOC and UV₂₅₄ reduced by 31 and 88%, respectively. Application of O₃/UV and H₂O₂/UV treatments to the source waters organics with 190-1500 Da molecular weight resulted in the near complete alteration of the molecular weight of NOM from >900 Da to <300 Da H₂O₂/UV was found to be the most effective treatment for the reduction of THM and HAA formation under uniform formation conditions. These results could hold particular significance for drinking water utilities with low alkalinity source waters that are investigating AOPs, as there are limited published studies that have evaluated the treatment efficacy of five different oxidation processes in parallel.     

Surface-retained organic matter of Microcystis aeruginosa inhibiting coagulation with polyaluminum chloride in drinking water treatment

Algogenic organic matter produced by the excess growth of cyanobacteria in semi-closed water areas causes coagulation inhibition in drinking water production. In this study, hydrophilic substances of Microcystis aeruginosa, which were mainly composed of lipopolysaccharide (LPS) and RNA, were prepared, and the involvement of these cyanobacterial hydrophilic substances in coagulation inhibition was investigated. As a result, it was found that the negatively charged hydrophilic substances with a molecular weight higher than 10 kDa have a significant role in coagulation inhibition. Further fractionation of cyanobacterial hydrophilic substances revealed that surface-retained organic matter (SOM), including LPS, could exhibit a potent inhibitory effect on the coagulation using polyaluminum chloride (PACl), presumably because of the direct interaction of hydrophilic SOM with cations originated from PACl, which could impede the hydrolysis of the coagulant.     

Mechanism of natural organic matter removal by polyaluminum chloride: effect of coagulant particle size and hydrolysis kinetics

The mechanism of natural organic matter (NOM) removal by AlCl(3) and polyaluminum chloride (PACl) was investigated through bench-scale tests. The fraction distributions of NOM and residual Al after coagulation in solution, colloid and sediment were analyzed as changes of coagulant dosage and pH. The influence of NOM, coagulant dose and pH on coagulation kinetics of AlCl(3) was investigated using photometric dispersion analyzer compared with PACl. Monomeric Al species (Al(a)) shows high ability to satisfy some unsaturated coordinate bonds of NOM to facilitate particle and NOM removal, while most of the flocs formed by Al(a) are small and difficult to settle. Medium polymerized Al species (Al(b)) can destabilize particle and NOM efficiently, while some flocs formed by Al(b) are not large and not easy to precipitate as compared to those formed by colloidal or solid Al species (Al(c)). Thus, Al(c) could adsorb and remove NOM efficiently. The removal of contaminant by species of Al(a), Al(b) and Al(c) follows mechanisms of complexation, neutralization and adsorption, respectively. Unlike preformed Al(b) in PACl, in-situ-formed Al(b) can remove NOM and particle more efficiently via the mechanism of further hydrolysis and transfer into Al(c) during coagulation. While the presence of NOM would reduce Al(b) formed in-situ due to the complexation of NOM and Al(a).     

Effects of operational parameters on cake formation of CaSO4 in nanofiltration

Inorganic fouling is one of the major limitations of nanofiltration (NF) application in drinking water treatment. In this study, we report the effect of operating parameters on the fouling of NF membranes by CaSO4 in terms of cake growth at membrane surface under various operating conditions. Increasing operating pressure and decreasing crossflow velocity promote CaSO4 nucleation and crystallization, thus enhance membrane fouling. The flux decline and its dynamics due to the change of operating parameters are assessed quantitatively based on a resistance model. Applied filtration pressure is found to have the most dominant impact on the inorganic fouling. It is shown that the filtration resistance caused by the formation and growth of CaSO4 cake on membrane surface is dependent on operating parameters.     

Removal of natural organic matter and THM formation potential by ultra- and nanofiltration of surface water

Natural organic matter (NOM) and trihalomethane formation potential (THMFP) removal were evaluated by ultrafiltration (UF) and nanofiltration (NF). Ten different raw water sources in Alicante province (SE Spain) were analysed. Five types of membranes of different materials were tested with a dead-end-type stirred UF cell. Additional measurements, such as dissolved organic carbon, ultraviolet absorbance (254nm), THMFP, ion concentration, pH, conductivity, etc. were made on raw water, permeates and concentrates. The SUVA value was used to determine the hydrophobicity of the water analysed. The elimination of NOM and THMFP is correlated with the molecular weight (MW) of NOM determined by size exclusion chromatography (SEC). The flux decline trends were correlated with cation concentration. NOM removal by UF is low, which correlates with the average MW determined by SEC with an average value of 922g/mol (between 833 and 1031g/mol). However, the NOM removal obtained with the NF90 and NF270 NF membranes for all water sources is almost complete (90%). THMFP removal is related to hydrophobicity and permeability of membrane. The NFT50 membrane removes almost 100% of the THMFP of more hydrophobic waters.     

Characterization of natural organic matter fractions separated by ultrafiltration using flow field-flow fractionation

Natural organic matter (NOM) from two sites in South Australia were separated by Amicon YM and YC ultrafiltration (UF) membranes into five nominal fractions (<0.5, 0.5-3, 3-10, 10-30 and >30 kDa). These nominal fractions were then characterized for size and molecular weight (MW) distributions using flow field-flow fractionation. The results show that separation by UF did not produce fractions with the expected MW and size. Electrophoretic mobility measurements of the NOM fractions adsorbed to colloidal goethite showed no significant difference between the fractions. However solid-state (13)C NMR of the NOM fractions showed that the separation was influenced by molecular structure as well as molecular size. The results suggest that great caution needs to be exercised when interpreting molecular size and speciation results for humic substances obtained by membrane UF.     

The role of extracellular polymeric substances on the sorption of natural organic matter

In this study, the influence of extracellular polymeric substances (EPS) composition and quantity was explored for biosorption of natural organic matter (NOM), using variants of Pseudomonas aeruginosa and Pseudomonas putida. Model EPS (sodium alginate beads) were tested and sorption capacity for NOM was also elucidated. In the absence of divalent ions, minimal NOM biosorption was observed and differences among strains were negligible. Under presence of divalent ions, biosorption of NOM was proportional to the amount of EPS secreted by P. aeruginosa variants. For sorption tests with model EPS, divalent ions also promoted biosorption of tested NOM, and total biosorption was also proportional to alginate quantity. Carboxyl group content in both alginate EPS and NOM appeared to be linked to increased biosorption via bridging with divalent ions. The alginate overproducing strain possessed more potential NOM biosorption sites, while the wild-type and alginate deficient strains possessed fewer potential binding sites. In comparison, P. putida, secreting protein-based EPS, behaved differently for NOM biosorption, due to its hydrophobicity and the structural characteristics of proteins. Hydrophobic interactions appeared to enhance the biosorption of more hydrophobic Suwannee River humic acid by P. putida, whose biosorption of more hydrophilic NOM variants was similar to the alginate deficient strain. Mechanistically, the presence of a diffuse electrical double layer will present potential energy barriers limiting biosorption; however, divalent ion concentrations in the aquatic environment will promote biosorption processes, permitting functional group interactions between EPS and NOM. Bridging between hydrophilic carboxyl groups on alginate EPS and NOM appeared to be the dominant form of biosorption, while hydrophobic interactions enhanced biosorption for protein-based EPS.     

Bioanalytical and chemical assessment of the disinfection by-product formation potential: Role of organic matter

Disinfection by-products (DBP) formed from natural organic matter and disinfectants like chlorine and chloramine may cause adverse health effects. Here, we evaluate how the quantity and quality of natural organic matter and other precursors influence the formation of DBPs during chlorination and chloramination using a comprehensive approach including chemical analysis of regulated and emerging DBPs, total organic halogen quantification, organic matter characterisation and bioanalytical tools. In vitro bioassays allow us to assess the hazard potential of DBPs early in the chain of cellular events, when the DBPs react with their molecular target(s) and activate stress response and defence mechanisms. Given the reactive properties of known DBPs, a suite of bioassays targeting reactive modes of toxic action including genotoxicity and sensitive early warning endpoints such as protein damage and oxidative stress were evaluated in addition to cytotoxicity. Coagulated surface water was collected from three different drinking water treatment plants, along with reverse osmosis permeate from a desalination plant, and DBP formation potential was assessed after chlorination and chloramination. While effects were low or below the limit of detection before disinfection, the observed effects and DBP levels increased after disinfection and were generally higher after chlorination than after chloramination, indicating that chlorination forms higher concentrations of DBPs or more potent DBPs in the studied waters. Bacterial cytotoxicity, assessed using the bioluminescence inhibition assay, and induction of the oxidative stress response were the most sensitive endpoints, followed by genotoxicity. Source waters with higher dissolved organic carbon levels induced increased DBP formation and caused greater effects in the endpoints related to DNA damage repair, glutathione conjugation/protein damage and the Nrf2 oxidative stress response pathway after disinfection. Fractionation studies indicated that all molecular weight fractions of organic carbon contributed to the DBP formation potential, with the humic rich fractions forming the greatest amount of DBPs, while the low molecular weight fractions formed more brominated DBPs due to the high bromide to organic carbon ratio. The presence of higher bromide concentrations also led to a higher fraction of brominated DBPs as well as proportionally higher effects. This study demonstrates how a suite of analytical and bioanalytical tools can be used to effectively characterise the precursors and formation potential of DBPs.     

Disinfection by-product formation following chlorination of drinking water: Artificial neural network models and changes in speciation with treatment

Artificial neural network (ANN) models were developed to predict disinfection by-product (DBP) formation during municipal drinking water treatment using the Information Collection Rule Treatment Studies database complied by the United States Environmental Protection Agency. The formation of trihalomethanes (THMs), haloacetic acids (HAAs), and total organic halide (TOX) upon chlorination of untreated water, and after conventional treatment, granular activated carbon treatment, and nanofiltration were quantified using ANNs. Highly accurate predictions of DBP concentrations were possible using physically meaningful water quality parameters as ANN inputs including dissolved organic carbon (DOC) concentration, ultraviolet absorbance at 254 nm and one cm path length (UV₂₅₄), bromide ion concentration (Br⁻), chlorine dose, chlorination pH, contact time, and reaction temperature. This highlights the ability of ANNs to closely capture the highly complex and non-linear relationships underlying DBP formation. Accurate simulations suggest the potential use of ANNs for process control and optimization, comparison of treatment alternatives for DBP control prior to piloting, and even to reduce the number of experiments to evaluate water quality variations when operating conditions are changed. Changes in THM and HAA speciation and bromine substitution patterns following treatment are also discussed.