Search

An examination of shallow GAC filter caps to promote biological filtration without requiring costly full filter bed media replacements.

Microfiltration (MF) and ultrafiltration (UF) consistently remove suspended material and pathogens from drinking water; however, membrane fouling inhibits their application by increasing operation and maintenance costs. Coagulation/flocculation is a commonly used pretreatment method for the reduction of membrane fouling; in this review it has been grouped into three typical configuration types: Type 1: coagulation + no/incidental flocculation, Type 2: coagulation + flocculation, and Type 3: conventional coagulation, based on operational conditions. The impact of each configuration on floc properties, membrane fouling, and organics removal has been reviewed in detail. Due to relatively high membrane resistance and low NOM reductions, configuration Type 1 may not be optimal for fouling control and organics removal when compared to Types 2 and 3. Configuration Type 2 led to the lowest cake layer and specific cake layer resistance for both MF and UF, while there is evidence that Type 3 results in the greatest reduction in fouling rate by reducing mass flux towards the membrane surface. As expected, with no coagulant results indicate that UF achieves greater organics removal when compared to MF, but with the addition of coagulant performance is similar for all configuration types. By highlighting the connection between coagulation/flocculation configuration types and membrane performance, the review provides insight for the design and operation of pretreatment for low pressure membrane filtration. In addition, understanding the impact of configuration types on floc properties aids in revealing the fouling mechanisms that dictate membrane performance. Knowledge gaps have been identified for guidance on future research.

Biofiltration has been shown to be effective for disinfection by-product (DBP) precursor control, however few studies have considered its role in the potential formation of DBPs. Biofilm is composed of heterogeneous bacteria as well as extracellular polymeric substances (EPS). The objective of this study was to determine the contribution of biofilm-related materials such as EPS to form nitrogen-containing DBPs upon chloramination, and to determine the influence of cyclical (scheduled on-off) biofilter operation on DBP precursor removal. Biologically active media was sampled from a full-scale biofilter operating under cold-water conditions (3.6 ± 0.5 °C) and extracted using a cation exchange resin into a phosphate buffer solution. Biomass concentrations, as determined using adenosine triphosphate (ATP) measurements, remained stable at 298 ± 55 ng ATP/g media over the trial period. N-nitrosodimethylamine (NDMA) and haloacetonitrile (HAN4) formation potential (FP) tests conducted under uniform formation conditions (UFC) using extracted biofilm yielded 0.80 ± 0.27 ng NDMA/g media and 18.7 ± 3.3 ng dichloroacetonitrile (DCAN)/g media. Further analyses of extracted biofilm using fluorescence spectroscopy and liquid chromatography-organic carbon detection indicated the presence of proteins above 20 kDa and humic-like substances. Extracted proteins (93.5 ± 8.1 μg/g media) correlated well (R = 0.90) with UV 280 measurements, indicating that spectrophotometry may serve as a valuable tool to quantify proteins in extracted biofilms. While substances in biofilms can serve as NDMA and DCAN precursors, the full-scale cyclically-operated biofilter that was examined did not show release of NDMA precursors during start-up following stagnation periods of 6 h or more. These biofilters consistently removed 6.9 ± 4.3 ng/L of NDMA precursors; typical NDMA UFC-FP of biofilter effluent was 8.5 ± 2.6 ng/L.

The objective of this pilot study was to evaluate the impact of cyclical (operated 8–12 h per day) and continuous biofilter operation with respect to biomass development, biological enzyme activity and treated water quality (in terms of organics, nutrients and disinfection by-product (DBP) formation potential). Continuously operated biofilters developed greater densities of biomass, as measured by ATP, when compared to cyclically operated filters; reducing the empty bed contact time (EBCT) increased biomass density under continuous flow conditions. However, once normalized to biomass, it was shown that cyclically operated filters exhibited higher enzyme activity, indicating that this method of operation may improve bacterial function. Reduction of organics was generally similar for both continuous and cyclical filters with the same EBCT, however, cyclical filters demonstrated higher variability during the first 4 h following start-up. Overall, HAA formation potential was better controlled by continuously operated filters, due to poor performance by the cyclical filters upon start-up while THM precursors were removed equally well by all filters. To understand the removal capacity for NDMA precursors through biological filters, both naturally occurring NDMA FP and NDMA FP resulting from spiked anthropogenic precursors was monitored through the filter depth. All the filters removed 90% of the naturally occurring NDMA FP within the first 45 cm; cyclical operation resulted in higher reduction of spiked anthropogenic NDMA precursors (50% higher than continuously operated) demonstrating the advantage of routine shut down on overall microbial activity. Tools to monitor and predict biofilter performance are in high demand. Here we present an “effective activity” term which combines enzyme activity with contact time (EBCT). Effective esterase activity was strongly correlated to DOC reduction as a function of filter operation (cyclical or continuous) and EBCT; effective phosphatase activity was indicative of phosphate removal. The results of this study indicate that routine shut down of the filters as this location improved enzyme activity without compromising control of chlorinated DBPs (THMs and HAAs) or NDMA derived from natural and anthropogenic precursors.

Many water providers monitor adenosine triphosphate (ATP) as an indicator of biological acclimation of their biofilters; however, strong correlations between ATP concentration and filter performance (e.g., organic matter or disinfection by-product precursor removal) are not typically observed. As an alternative, this study evaluated the use of enzyme activity for monitoring biological processes within filters. Recent studies have proposed that enzyme activity may be used as an indicator of biofilter function as it provides a means to quantify biodegradation which may allow for a more accurate measure of degradation potential and to gain a better understanding of biofilter performance. Sampling was completed from full- and pilot-scale biofilters to assess impacts associated with pre-treatments, varying sources waters, as well as pre-treatment and operating conditions. Enzyme activity (carboxylic esterase, phosphatase, s-glucosidase, α-glucosidase, s-xylosidase, chitinase, and cellulase) and ATP were measured from the top 5 cm of biofilter media representative of typical full-scale sampling; water quality parameters included dissolved organic carbon (DOC) and disinfection by-products (DBPs): trihalomethane (THM) formation potential (FP), and haloacetic acid FP (HAA FP). Results confirmed that ATP was not a reliable monitoring tool for DOC and DBP FP reduction in biofilters. A strong relationship was observed between esterase activity and DOC reduction; chitinase activity significantly correlated to THM FP reduction for filters treating three different source waters and HAA FP reduction achieved by filters treating the same source water with a range of pre-treatment and backwash conditions. This study showed that enzyme activity may be appropriate for monitoring biological processes within drinking water filters and may act as a surrogate for the removal of organic compounds.

Application of real-time fluorescence excitation emission matrices (EEM) as a tool for water quality assessment was investigated. A bench-scale fluorescence system with on-line monitoring capabilities was used to quantify several polycyclic aromatic hydrocarbons (PAHs) and pesticides in surface waters of Ontario, Canada. Parallel factors analysis (PARAFAC), an unsupervised multi-way analysis approach, was compared to a supervised peak-picking approach. Both approaches produced sensitive regression models capable of predicting contaminant concentration with mean absolute errors (MAE) that ranged from 0.032 to 0.293 μg L−1 for lake water, and 0.052 to 0.921 μg L−1 for river water. In addition, results showed accurate detection of contaminant concentrations exceeding 0.25 μg L−1. However, concentration and variability of natural organic matter (NOM) in natural waters presented unique challenges to peak-picking, PARAFAC, and regression analysis, which were mitigated by expert supervision as well as site-specific and timely calibration.

MX precursor removal was compared at pilot-scale in continuously and cyclically operated biofilters with empty bed contact times (EBCTs) of 15 and 30 minutes. Water and filter media samples were collected under ambient water quality conditions and during a spike of 7 mg L−1 humic acid (HA). MX formation potential (FP) removals ranged from 26 to 49% and 45 to 61% under ambient precursor and spiked conditions, respectively, suggesting that biofiltration was capable of removing or altering MX precursors. Removal was very strongly correlated (R = 0.90) with ATP (a measure of biomass density) for all trials. MX FP reduction was significantly higher (p < 0.05) with respect to biomass density when comparing cyclically vs. continuously operated filters for all trials. Although lower average ATP concentrations were observed for the cyclically operated filters, these filters achieved a greater MX FP reduction per unit biomass which implies a higher rate of biological function. EBCT did not impact biofilter performance when considering MX FP reduction with respect to either filter depth or biomass density.

Quantitative microbial risk assessments (QMRA) was conducted for 10 drinking water facilities located in Canada. Monthly risk estimates were based upon a suite of reference pathogens for the raw water, as well as detailed process assessments that considered the variability treatment at individual treatment facilities. The majority of the facilities (9 out of 10) maintained risk levels significantly below the World Health Organization's target of 10−6 DALY per person per year. Cryptosporidium was shown to be the most significant contributor for microbial risk at all locations. Potential treatment specific failure scenarios were evaluated; most treatment facilities were capable of withstanding partial filtration and disinfection failures as a result of the multi-barriers employed.

The use of fluorescence data coupled with neural networks for improved predictability of drinking water disinfection by-products (DBPs) was investigated. Novel application of autoencoders to process high-dimensional fluorescence data was related to common dimensionality reduction techniques of parallel factors analysis (PARAFAC) and principal component analysis (PCA). The proposed method was assessed based on component interpretability as well as for prediction of organic matter reactivity to formation of DBPs. Optimal prediction accuracies on a validation dataset were observed with an autoencoder-neural network approach or by utilizing the full spectrum without pre-processing. Latent representation by an autoencoder appeared to mitigate overfitting when compared to other methods. Although DBP prediction error was minimized by other pre-processing techniques, PARAFAC yielded interpretable components which resemble fluorescence expected from individual organic fluorophores. Through analysis of the network weights, fluorescence regions associated with DBP formation can be identified, representing a potential method to distinguish reactivity between fluorophore groupings. However, distinct results due to the applied dimensionality reduction approaches were observed, dictating a need for considering the role of data pre-processing in the interpretability of the results. In comparison to common organic measures currently used for DBP formation prediction, fluorescence was shown to improve prediction accuracies, with improvements to DBP prediction best realized when appropriate pre-processing and regression techniques were applied. The results of this study show promise for the potential application of neural networks to best utilize fluorescence EEM data for prediction of organic matter reactivity.

Surface waters may contain varying levels of wastewater effluent associated with de facto reuse, which may influence their toxicological properties both prior to and following treatment. This study examined the genotoxic response of three surface waters containing a range of wastewater effluent (5%, 10%, and 25% by volume). The SOS Chromotest™ was used to assay the genotoxicity of both chlorinated and unchlorinated mixtures. Chlorinated mixtures were also analyzed for trihalomethanes (THMs), haloacetonitriles (HANs), and halonitromethanes (HNMs); their concentrations were used to calculate a relative toxicity index for each sample, based on published potencies in the comet assay and subsequently referred to as predicted genotoxicity. Wastewater effluents were observed to be reactive in the genotoxicity assay, whereas raw and chlorinated surface waters were not. Upon chlorination, surface waters containing 5% or 10% wastewater did not elicit a response and only modest effects were observed for higher wastewater ratios (25%). The measured SOS responses correlated well with predicted genotoxicity (R = 0.92) and THM concentrations (R = 0.92). This is important since THMs themselves are non-reactive in either the SOS or comet genotoxic assays, but their formation may serve as surrogates for non-regulated DBPs which drive toxic effects.

N-Nitrosodimethylamine (NDMA) is a disinfection by-product (DBP) with many known precursors such as amine-containing pharmaceuticals that can enter the environment via treated wastewater. Reverse osmosis and tight nanofiltration membranes (MW cutoff200 Da) are treatment technologies that demonstrate high removal of many compounds, but at relatively high energy costs. Looser membranes (200 Da) may provide sufficient removal of a wide range of contaminants with lower energy costs. This study examined the rejection of pharmaceuticals that are known NDMA precursors (∼300 Da) using nanofiltration (MW cutoff ∼350 Da). MQ water was compared to two raw water sources, and results illustrated that NDMA precursors (as estimated by formation potential testing) were effectively rejected in all water matrices (84%). Mixtures of pharmaceuticals vs. single-spiked compounds were found to have no impact on rejection from the membranes used. The use of MQ water vs. surface waters illustrated that natural organic matter, colloids, and inorganic ions present did not significantly impact the rejection of the amine-containing pharmaceuticals. This study illustrates that NDMA formation potential testing can be effectively used for assessing NDMA precursor rejection from more complex samples with multiple and/or unknown NDMA precursors present, such as wastewater matrices.

“Effective Enzyme Activity”, or simply “Effective Activity”, is proposed as a biofiltration monitoring tool which combines enzyme activity with empty bed contact time (EBCT) to quantify biodegradation potential. The primary objective of this study was to evaluate the applicability of the Effective Activity concept for predicting water quality in biofiltration systems. This pilot-scale study evaluated eight different biofilter configurations in order to quantify impacts associated with filter media (anthracite/sand or granular activated carbon), pre-treatment (settled water with or without ozonation) and operating conditions (15- and 30-min EBCT, and backwash with or without chlorine). Microbial characterization included biomass concentration, as measured by adenosine triphosphate (ATP), in addition to esterase and phosphatase activity. Water quality parameters included dissolved organic carbon (DOC), trihalomethane (THM) formation potential (FP), haloacetic acid (HAA) FP, haloacetonitrile (HAN) FP, iodinated DBP FP (THMs and HAAs) and inorganic nutrients (phosphorus and nitrogen). Results confirmed the benefits to treated water quality associated with the application of an ozone residual of 0.5 mg/L, utilization of GAC filter media, eliminating chlorinated backwash, and extending EBCT. This study demonstrated a good relationship between effective esterase activity and reductions in DOC and THM FP, including those systems which incorporate pre-ozonation. As such, this study showed that Effective Activity may be appropriate for relating biomass characterization to treated water quality and highlights the importance of quantifying biomass activity in addition to quantity.

Previous studies have reported that biofilm extracted from full-scale biofilters can serve as nitrogenous disinfection by-product (N-DBP) precursors. Detached biofilm materials could escape during filter ripening and form N-DBP upon chloramination. This study examined the potential breakthrough of biofilm and N-DBP precursors during filter ripening at two water treatment plants (WTPs). The presence of biofilm material in aqueous samples was estimated by total adenosine triphosphate (tATP) levels; N-DBP formation potential (FP) tests were conducted under uniform formation conditions to quantify N-nitrosodimethylamine (NDMA) and haloacetonitrile (HAN4) precursors. While tATP peaks in filter effluent were observed post backwash at both WTPs, temporary increases of effluent NDMA FP were only observed during filter ripening where particle-associated NDMA precursors served as the dominant contributor. Overall, biofilters examined in this study demonstrated a consistent removal of NDMA FP regardless of the filter ripening process.

To provide information for the design and improvement of full-scale biofilters, pilot-scale biofiltration studies are the current industry standard because they utilize the same filter media size and loading rate as the full-scale biofilters. In the current study, bench-scale biofilters were designed according to a biofilter scaling model from the literature, and the ability of the bench-scale biofilters to accurately represent the organics removal of pilot-scale biofilters was tested. To ensure similarity in effluent water quality between bench- and pilot- or full-scale biofilters at the same influent substrate concentration, the tested model requires that either mass transport resistance or biofilm shear loss takes primacy over the other. The potential primacy of mass transport resistance or biofilm shear loss was evaluated via water quality testing (dissolved organic carbon, specific ultraviolet absorbance, liquid chromatography – organic carbon detection, trihalomethane formation potential, and haloacetic acid formation potential). The biofilters also were characterized for adenosine triphosphate (ATP) content, enzyme activity, extracellular polymeric substances, and microbial community structure. The results of this study indicate that biofilm shear loss takes primacy over mass transport resistance for bench-scale biofilter design in this system; thus, bench-scale biofilters designed in this manner accurately represent organics removal in pilot-scale biofilters. Applying this scaling procedure can reduce filter media requirements from many kilograms to just a few grams and daily water requirements from thousands of liters to less than 10 L. This scaling procedure will allow future researchers to test alternative treatment designs and operating conditions without the need for expensive pilot-scale studies.

Legacy lead and copper components are ubiquitous in plumbing of large buildings including schools that serve children most vulnerable to lead exposure. Lead and copper samples must be collected after varying stagnation times and interpreted in reference to different thresholds. A total of 130 outlets (fountains, bathroom and kitchen taps) were sampled for dissolved and particulate lead as well as copper. Sampling was conducted at 8 schools and 3 institutional (non-residential) buildings served by municipal water of varying corrosivity, with and without corrosion control (CC), and without a lead service line. Samples included first draw following overnight stagnation (8h), partial (30 s) and fully (5 min) flushed, and first draw after 30 min of stagnation. Total lead concentrations in first draw samples after overnight stagnation varied widely from 0.07 to 19.9 μg Pb/L (median: 1.7 μg Pb/L) for large buildings served with non-corrosive water. Higher concentrations were observed in schools with corrosive water without CC (0.9-201 μg Pb/L, median: 14.3 μg Pb/L), while levels in schools with CC ranged from 0.2 to 45.1 μg Pb/L (median: 2.1 μg Pb/L). Partial flushing (30 s) and full flushing (5 min) reduced concentrations by 88% and 92% respectively for corrosive waters without CC. Lead concentrations were10 μg Pb/L in all samples following 5 min of flushing. However, after only 30 min of stagnation, first draw concentrations increased back to45% than values in 1st draw samples collected after overnight stagnation. Concentrations of particulate Pb varied widely (≥0.02-846 μg Pb/L) and was found to be the cause of very high total Pb concentrations in the 2% of samples exceeding 50 μg Pb/L. Pb levels across outlets within the same building varied widely (up to 1000X) especially in corrosive water (0.85-851 μg Pb/L after 30MS) confirming the need to sample at each outlet to identify high risk taps. Based on the much higher concentrations observed in first draw samples, even after a short stagnation, the first 250mL should be discarded unless no sources of lead are present. Results question the cost-benefit of daily or weekly flushing as a remediation strategy. As such, current regulatory requirements may fail to protect children as they may not identify problematic taps and effective mitigation measures.

A pilot-scale study was performed to evaluate a coagulant dose which had been optimized for biopolymer (i.e., foulant) removal on subsequent ultrafiltration (UF) fouling, as well as disinfection by-product (DBP) precursor removal. Polyaluminum chloride (PACl) dosages were selected based on a point of diminishing returns for biopolymer removal (0.5 mg/L) and directly compared to that applied at full-scale (6 mg/L). Membrane fouling (reversible and irreversible) was measured as resistance increase over a 48 hour filtration period. DBP formation potential (total trihalomethanes (TTHMs), haloacetic acids (HAA9) and total adsorbable organic halides (AOX)) were measured in both raw and treated waters. Results of the study indicate that application of a PACl dose optimized for biopolymer reduction (0.5 mg/L) resulted in 65% less irreversible UF fouling when compared to 6 mg/L. The addition of PACl prior to the membrane resulted in up to a 14% reduction in DBP precursors relative to the UF membrane alone. A similar level of DBP precursor reduction was achieved for both 0.5 and 6 mg/L dosages. The results have implications for cost savings, which may be realized due to decreased chemical use, as well as increased membrane life associated with lower irreversible fouling rates.

Disinfection by-products (DBPs) are formed when naturally occurring organic matter reacts with chlorine used in drinking water treatment, and DBPs formed in chlorinated drinking water samples have been shown to cause a genotoxic response. The objective of the current study was to further understand the principles of biofiltration and the resulting impacts on the formation of DBPs and genotoxicity. Pilot-scale systems were utilized to assess the performance of engineered biofilters enhanced with hydrogen peroxide, in-line coagulants, and nutrients when compared to passively operated biofilters and conventional treatment (coagulation, flocculation, sedimentation, non-biological filtration). Organic fractionation was completed using liquid chromatography-organic carbon detection (LC-OCD). Water samples were chlorinated after collection and examined for the removal of trihalomethane (THM), haloacetic acid (HAA), and adsorbable organic halide (AOX) precursors. Additionally, the formation potential of two halogenated furanones, 3-chloro-4(dichloromethyl)-2(5H)-furanone (MX) and mucochloric acid (MCA), and genotoxicity was determined. Biofiltration was shown to preferentially remove more DBP precursors than dissolved organic carbon (DOC). Formation potential of the unregulated DBPs, including MX and MCA, and genotoxic response was shown to be correlated to THM formation. These results infer that monitoring for THMs and HAAs provide insight to the formation of more mutagenic DBPs such as halogenated furanones, and that biofiltration may preferentially remove precursors to DBPs at a rate exceeding the removal of DOC.

Effective coagulation is essential to achieving drinking water treatment objectives when considering surface water. To minimize settled water turbidity, artificial neural networks (ANNs) have been adopted to predict optimum alum and carbon dioxide dosages at the Elgin Area Water Treatment Plant. ANNs were applied to predict both optimum carbon dioxide and alum dosages with correlation (R2) values of 0.68 and 0.90, respectively. ANNs were also used to developed surface response plots to ease optimum selection of dosage. Trained ANNs were used to predict turbidity outcomes for a range of alum and carbon dioxide dosages and these were compared to historical data. Point-wise confidence intervals were obtained based on error and squared error values during the training process. The probability of the true value falling within the predicted interval ranged from 0.25 to 0.81 and the average interval width ranged from 0.15 to 0.62 NTU. Training an ANN using the squared error produced a larger average interval width, but better probability of a true prediction interval.

Natural organic matter (NOM) in drinking water can react with disinfectants to form disinfection by-products (DBPs). Halogenated furanones are a group of emerging DBPs that can account for 20–60% of the total mutagenicity observed in drinking water. This study examined the impacts of bench-scale coagulation and subsequent chlorination on DBP formation as well as genotoxicity using three source waters located in Ontario, Canada. Two halogenated furanones 3-chloro-4-(dichloromethyl)-2(5H)-furanone (MX) and mucochloric acid (MCA) were analyzed; along with trihalomethanes (THMs), haloacetic acids (HAAs), and absorbable organic halides (AOX). NOM was quantified using liquid chromatography–organic carbon detection (LC–OCD). Measured MX and MCA formation was 6.9–15.3 ng/L and 43.2–315 ng/L following optimized coagulation and subsequent chlorination of the three waters tested. DBP formation and speciation were evaluated as a function of the specific NOM fractions present in the source waters. Humics, building blocks, and biopolymers were highly correlated with DBP formation. Correlations between DBPs were also investigated and a potential relationship between MCA and/or MX vs. HAAs was observed. MX was the only measured DBP that contributed to genotoxicity, representing less than 0.001% of AOX by mass but responsible for 40–67% of the genotoxic response in chlorinated Ottawa River water samples. Genotoxic potential decreased with alum dosages, signifying that coagulation was effective at removing genotoxic DBP precursors.

Surface shear stress induced by different air sparging regimes on a submerged hollow fiber ultrafiltration module with horizontally-oriented, densely packed fibers was characterized. Continuous and intermittent (cycling on and off) coarse bubbles (0.75–2.5 mL), as well as large pulse bubble (150 and 500 mL) sparging were considered for a range of air flow rates. The power required to induce surface shear stress on the surface of the hollow fibers was substantially lower when using large pulse bubble sparging compared to both continuous and intermittent coarse bubble sparging. Results indicated that the air flow required for pulse bubble sparging was more than 80% lower than that required for coarse bubble sparging to induce comparable surface shear stress (and corresponding fouling control). This study demonstrates the potential value and efficiency of pulse bubble air sparging as a fouling control option in densely packed hollow fiber membrane systems.

The potential application of fluorescence spectroscopy for monitoring of organic matter concentration and character at four water treatment facilities was investigated. Results are presented showing impacts on natural organic matter (NOM) due to intake location on the same water body and from individual unit processes including ozonation, granular-activated carbon filtration, and coagulation/flocculation. For validation and comparison of fluorescence methods, organic matter was quantified and characterized using liquid chromatography-organic carbon detection (LC-OCD). Principal component analysis (PCA) and parallel factors analysis were used for dimensionality reduction and to represent individual organic components observed through fluorescence excitation-emission matrices. Fluorescence results generally agreed with LC-OCD characterization, indicating that complete treatment reduced organic concentrations and preferential removal of humic-like material was associated with coagulation/flocculation. PCA results indicated higher concentrations of humic-like material at the Island water treatment plant intake that was not well reduced by inline polyaluminum chloride coagulation and direct filtration. Through fluorescence spectroscopy, ozonation increased Rayleigh scattering, which is correlated to small colloidal/particulate concentrations. Full-scale results from four water treatment plants presented demonstrate that fluorescence methods can characterize NOM, providing similar identification of trends to LC-OCD, with possible online application and use in real-time water treatment process control.

Fluorescence spectroscopy was used as a characterization method to examine organic fouling of single ultrafiltration (UF) fibres at bench-scale. Low doses of coagulant were applied to modify organic properties, without significant formation of precipitates. This approach compliments previous studies investigating coagulation as a pre-treatment method for UF fouling control, which have principally focused on reduction of foulant concentrations. Using a continuous system, short time-scale fluorescence results demonstrated significant adsorption of humic components to virgin membrane fibres. Following an initial adsorption phase, protein-like material was the only organic component to be significantly removed by UF. Low doses of coagulant (

Media type is a critical design consideration when implementing biofiltration for drinking water treatment. Granular activated carbon (GAC) has been shown to provide superior performance when compared to a wide range of media types, largely due to its higher surface area. Engineered ceramic media is an attractive alternative to GAC as it has a similar surface area but at a lower cost. This pilot-scale biofiltration study compared the performance of GAC, anthracite and two different effective sizes of ceramic (CER) media (1.0 mm and 1.2 mm), in terms of dissolved organic carbon (DOC), head loss, turbidity, and disinfection by-product formation potential (DBPFP). Biological acclimation was monitored using adenosine tri-phosphate (ATP) measurements; biomass was further examined using laccase and esterase enzyme activity assays. When compared to other media types examined, biological GAC had higher (p > 0.05) removals of DOC (9.8 ± 3.8%), trihalomethane formation potential (THMFP, 26.3 ± 10.2%), and haloacetic acid formation potential (HAAFP, 27.2 ± 14.0%). CER media required 6–7 months to biologically acclimate, while filters containing GAC and anthracite were biologically active (>100 ng of ATP/g media) following 30–45 days of operation. Once acclimated, ATP values of 243 and 208 ng/g attained for CER 1.0 and 1.2, respectively, were statistically comparable to GAC (244 ng/g) and higher than anthracite (110 ng/g), however this did not translate into greater organics removal. Esterase and laccase enzyme kinetics were highest for GAC, while CER was shown to have greater biodegradation potential than anthracite. The four media types attained similar turbidity reduction (p > 0.05), however ceramic media filters were observed to have run times which were 1.5–2.3 times longer when compared to anthracite, which could represent potential cost savings in terms of energy for pumping and backwash requirements. Overall, ceramic media was shown to be a potential alternative to anthracite when considering biofiltration, especially during cold water conditions (T

A pilot-scale study was conducted to evaluate the impact of several biofiltration enhancement strategies in terms of organic removal to reduce disinfection by-product (DBP) formation potential and mitigate ultrafiltration (UF) fouling. Strategies included nutrient addition (nitrogen and phosphorus) to optimize metabolic degradation of organics, use of hydrogen peroxide (H2O2, peroxide) to improve filter run times, and the application of in-line aluminum sulphate (alum) for biopolymer removal. The impact of media type on performance was also examined (anthracite versus granular activated carbon (GAC)). Passive biofiltration (without enhancement) reduced dissolved organic carbon (∼5%), biopolymers (∼20%), and trihalomethane and haloacetic acid precursors (∼20% and ∼12%, respectively) while mitigating UF irreversible fouling (∼60%). Nutrient addition was not observed to enhance biological performance. Addition of 0.5 mg/L hydrogen peroxide decreased head loss by up to 45% without affecting organic removal; however at a dosage of 1 mg/L, it negatively impacted both UF fouling and DBP precursor removal. In-line alum addition prior to biofiltration (

A 14 month pilot-scale coagulation optimization study was conducted at the Peterborough Water Treatment Plant (Ontario, Canada) to compare treatment performance resulting from the application of aluminum sulfate vs. polyaluminum chloride (PACl). This paper describes results obtained from applying a statistical analysis approach to evaluate impacts on pH, turbidity, total organic carbon (TOC), ultraviolet absorbance (UVA), particle counts, chlorine residuals, filter head loss, flow rate, trihalomethanes (THMs), and nine haloacetic acids (HAA 9 ). To allow a direct comparison, parallel pilot trains were operated such that they achieved equal settled water TOC by adjusting PACl dose, (52–74% of the alum dose by weight). Settled water turbidity was significantly higher (on average 0.23 NTU) in the alum treated water when compared to PACl. For equivalent filter run-times, head loss was greater by 0.002–0.011 m h −1 when applying alum. An increase in pH by approximately 0.7 units when using PACl was observed to cause a significant increase in THM formation (10–30%).

The effect of surface shear stress on membrane fouling during submerged hollow fiber ultrafiltration of three different surface waters (two lakes, one river) was investigated. Surface shear stresses that mimicked those induced when applying continuous and intermittent coarse bubble air sparging, large pulse bubble air sparging, as well as no air sparging were considered. The results suggest that fouling was mainly due to the accumulation of the biopolymer fraction of the natural organic matter present in the raw water. Inducing shear stresses onto the membrane surface significantly decreased the rate of membrane fouling (relative to no shear stress applied) in all waters tested. Of the shear stress conditions studied, that which mimicked large pulse bubble sparging had the greatest effect, reducing fouling by up to 80% when compared to conditions with no sparging applied. Conditions that mimicked intermittent and continuous coarse bubble fouling reduced the rate of fouling by up to 77 and 49%, respectively. These results suggest that the shear stresses induced by sparging can promote back transport of soluble organic material from the membrane surface. A semi-empirical relationship was developed to estimate the effect of raw water characteristics and applied sparging conditions on membrane fouling.

This study assessed the impact of chemical coagulation using alum on the removal of three endocrine-disrupting compounds (EDCs; bisphenol A, clofibric acid and estriol) and nine pharmaceutically active compounds (PhACs; acetaminophen, carbamazepine, diclofenac, gemfibrozil, ketoprofen, naproxen, pentoxifylline, sulfamethoxazole and sulfachloropyridazine). The impact on natural organic matter (NOM) fractions as determined using liquid chromatography–organic carbon detection (LC–OCD; total dissolved organic carbon (DOC), hydrophobic DOC, biopolymers, humic substances, building blocks, low molecular weight neutrals and acids) was also examined. Three test surface waters were included: Lake Ontario, Grand River and Otonabee River water (Ontario, Canada). Gemfibrozil concentrations were reduced in both Otonabee and Grand River waters. Reductions were noted for carbamazepine and (inconsistently) for acetaminophen, and estrone appeared to increase in concentration in Grand River water with increasing alum doses. NOM removal was primarily attributed to the humic fraction, with small reductions in biopolymers in all of the waters studied.

Halo-benzoquinones (HBQs) have been previously detected as disinfection by-products in chlorinated drinking water. The current work investigates the link between natural organic matter (NOM) characteristics and HBQ formation during bench-scale coagulation of raw water. Three source waters (Lake Ontario, Otonabee River and Grand River) were subjected to jar testing using alum followed by chlorination. NOM fractions were analyzed via liquid chromatography-organic carbon detection (LC-OCD), while HBQs were quantified using liquid chromatography-triple quadrupole mass spectrometry. One HBQ, 2,6-dichloro-(1,4)benzoquinone (2,6-DCBQ), was identified in all waters after chlorination, and appeared to decrease with increased applied alum dose. 2,6-DCBQ exhibited high correlations with some humic NOM indicators: humic substance concentration (in Grand and Otonabee River waters only), UV absorbance at 254 nm, UV absorbance at 254 nm of the humic peak, and specific UV absorbance of humics (humic SUVA). With data pooled from the three waters, the biopolymer fraction of NOM was most strongly correlated with 2,6-DCBQ formation (R(2) = 0.78, p < 0.001); this may be due to co-removal of biopolymers with HBQ precursors during coagulation. These results indicate that coagulation processes can be effective for reduction, but not elimination, of HBQ precursors.

Bench-scale tests were conducted to evaluate enhanced coagulation as a method for removing natural organic matter (NOM) from a surface water to reduce the formation of disinfection by-products (DBPs). Aluminium sulphate (alum) and two polyaluminium chloride (PACl) coagulants were used, as well as alum with pH depression. Using a PACl coagulant alone or alum with pH depression was shown to attain 35% removal of TOC at lower dosages (31 and 29 mg/L, respectively) when compared to the use of alum alone (43 mg/L). In addition to TOC and UV254, a fluorescence excitation–emission matrix (FEEM) approach and liquid chromatography–organic carbon detection (LC-OCD) were used to further characterize the removal of NOM in both untreated and filtered waters. Principal component analysis of FEEM was able to identify the presence of humic-like substances (HS), protein-like substances (PS), and colloidal/particulate matter (CPM); HS were found to have a close correlation with TOC and UV254. LC-OCD enabled the quantitative detection of hydrophobic and hydrophilic DOC; the latter was further separated into five components, the largest of which was HS. Strong linear correlations were calculated between TOC, UV254, HS, and hydrophilic DOC (r2 > 0.96); these parameters were also found to be closely correlated with the formation of trihalomethanes (THMs, r2 > 0.78) and haloacetic acids (HAAs, r2 > 0.92). Linear correlations with THMs and HAAs indicated that FEEM and LC-OCD provide good measures of DBP precursors when compared with TOC and UV254.

Filtration is the final physical barrier preventing the passage of microbial pathogens into public drinking water. Proper pre-treatment via coagulation is essential for maintaining good particle removal during filtration. To improve filter performance at the Elgin Area WTP, artificial neural network (ANN) models were applied to optimize pre-filtration processes in terms of settled water turbidity and alum dosage. ANNs were successfully developed to predict future settled water turbidity based on seasonal raw water variables and chemical dosages, with correlation (R2) values ranging from 0.63 to 0.79. Additionally, inverse-process ANNs were developed to predict the optimal alum dosage required to achieve desired settled water turbidity, with correlation (R2) values ranging from 0.78 to 0.89.

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.

The overall objective of this study was to investigate the impact of blending membrane-treated water treatment plant (WTP) residuals with plant-filtered water on finished water quality in terms of biostability and disinfectant by-product (DBP) formation. Filter backwash water (FBWW) was treated with a pilot-scale ultrafiltration (UF) membrane to produce permeate that was blended with plant-finished water. The batch studies involved storing samples for a specified time with a disinfectant residual to simulate residence time in the distribution system. Both chlorinated and non-chlorinated FBWW streams were evaluated, and the experimental design incorporated free chlorine, monochloramine, and chlorine dioxide in parallel to a model system that did not receive a disinfectant dose. The results of the study found that blending 10% UF-treated FBWW with plant-filtered water did not have an impact on water biostability as monitored with heterotrophic plate counts (HPCs) or DBP concentrations as monitored by TTHM and HAA5 concentrations. However, the presence of preformed THM and HAA species found in chlorinated FBWW streams may result in higher levels of initial DBP concentrations in blended water matrices, and could have a significant impact on finished water quality in terms of meeting specific DBP guidelines or regulations.

Reference cultures from culture collections are often used in disinfection experiments instead of indigenous environmental bacteria, but they may not accurately represent the organisms found in drinking water distribution systems due to physiological differences. This may explain why disinfectant concentrations and contact times, determined under typical laboratory conditions, are not always sufficient to control microbial growth or survival in distribution systems. The objective of this study was to investigate the effect of chlorine or monochloramine disinfection on Escherichia coli O157:H7 and other heterotrophic bacteria obtained from a culture collection and strains isolated from various environments, including disinfected water. It was hypothesized that previous exposure of the environmental strains to sublethal concentrations of disinfectant may allow them to develop greater resistance. Accordingly, it was observed that environmental strains of E. coli O157:H7 were either equally or less susceptible to chlorine and monochloramine than the reference strain. In contrast however, environmental strains of Brevundimonas vesicularis, Pseudomonas fluorescens, and Sphingomonas paucimobilis were either equally or more susceptible to free chlorine or monochloramine than their corresponding reference strains. This was counterintuitive because the environmental strains were able to survive disinfection in the pipes from which they were isolated. It is hypothesized that upon culturing in the laboratory the environmental strains may have become susceptible to low levels of disinfectant. Other researchers have suggested that changes in culture conditions may impact disinfectant sensitivity by affecting cell permeability, cell composition, or growth rate. This emphasizes the importance of designing laboratory studies to mimic environmental conditions as closely as possible to accurately represent environmental inactivation kinetics, including proper handling of organisms (subculturing, media transfers, etc.) before their use in inactivation studies.

Lead results from 78,971 water samples collected in four Canadian provinces from elementary schools, daycares, and other large buildings using regulatory and investigative sampling protocols were analyzed to provide lead concentration distributions. Maximum concentrations reached 13,200 and 3890 μg/L following long and short stagnation periods respectively. High lead levels were persistent in some large buildings, reflected by high median values considering all taps, or specific to a few taps in the building. Simulations using the Integrated Uptake Biokinetic (IEUBK) model and lead concentrations after 30 min of stagnation in the dataset showed that, for most buildings, exposure to lead at the tap does not increase children's blood lead levels (BLLs). However, buildings or taps with extreme concentrations represent a significant health risk to young children attending school or daycare, as the estimated BLL far exceeded the 5 μg/dL threshold. Ingestion of water from specific taps could lead to acute exposure. Finally, for a few taps, the total daily lead intake reached the former World Health Organization (WHO) tolerable level for adults, suggesting potential health risks.

Ammonia can be used to minimize bromate concentrations by blocking two of three potential bromate formation pathways. It was theorized that (bi)carbonate alkalinity in the presence of ammonia would inhibit bromate formation since the pathway that ammonia does not block requires hydroxyl radicals (OH ), and (bi)carbonate alkalinity is an OH scavenger. Experiments where (bi)carbonate alkalinity was increased from 50 to 119 mg/L (as CaCO 3 ) in the presence of excess ammonia resulted in up to 50% reduction in bromate formation, providing evidence in support of the theory. While OH is scavenged by (bi)carbonate alkalinity, it is promoted by hydrogen peroxide (H 2 O 2 ). When ozone reacts with natural organic matter the H 2 O 2 that is formed may therefore render ammonia less effective. Experiments conducted in this study demonstrated this principle.

North American drinking water utilities are currently re-evaluating their disinfection strategies for controlling microbiological growth in distribution systems. Most water systems in North America use free chlorine as a secondary disinfectant. Since chlorine is known to form potentially carcinogenic byproducts in drinking water, utilities are looking for an alternative to maintain a disinfectant residual in the distribution system. The objective of this study was to evaluate the response of model drinking water distribution systems to a change in disinfectant from either free chlorine or chloramines to chlorine dioxide, in terms of its impact on microbiological water quality (bulk water and biofilm). Switching from a chlorine residual of 0.5 mg/L to a chlorine dioxide residual of 0.25 mg/L did not impact (negatively or positively) microbial water quality as quantified by heterotrophic and total cell counts (sample size = 8 data points). Thus, on the basis of the mass of disinfectant applied, chlorine dioxide was more efficient than free chlorine at controlling microbiological growth in the model distribution system. Similarly, chlorine dioxide was more efficient than chloramines, as a chlorine dioxide residual of 0.25 mg/L inactivated 0.75–1 log more suspended organisms than 1.0 mg/L residual of chloramines. Therefore, under the tested conditions, chlorine showed similar or better disinfection efficiency than free chlorine and chloramines, respectively.

Synthetic water matrices containing ammonia, bicarbonate, hydrogen peroxide, and natural organic matter were studied to identify the effects of these compounds (individually and in combination) on bromate formation. Ammonia alone was observed to significantly reduce bromate formation through the sequestering of brominated intermediates as bromamines. Natural organic matter reacted quickly with bromamines, which could impair the ability of ammonia to block bromate formation. Bicarbonate was observed to generally promote bromate formation in otherwise pure water, but bicarbonate worked synergistically with ammonia to reduce bromate formation by a greater factor than ammonia alone, due to OH radical scavenging. Experiments showed that hydrogen peroxide lowered the effectiveness of ammonia to block bromate formation.

Chlorine dioxide (ClO2) inactivation of Bacillus subtilis ATCC 19659 spores was examined at pilot-scale during periods representative of winter and summer temperature extremes at the Britannia Water Treatment Plant in Ottawa, Canada. In addition, bench-scale experiments using the same source water (Ottawa River, Ontario, Canada), as well as buffered and unbuffered laboratory waters were conducted using B. subtilis spores. Bench-scale inactivation of B. subtilis spores by ClO2 was similar to reported values for Cryptosporidium parvum (both organisms being more resistant to ClO2 than Giardia lamblia), suggesting the possibility that these spores may be used as potential indicators for protozoan parasites. Additionally, spore inactivation was observed to be influenced by pH in laboratory (distilled deionised water) water but not in Ottawa River water. At pilot-scale, spore inactivation was influenced by water temperature: a ClO2 dose of 2.5 mg/L resulted in a spore inactivation of approximately 2.0 log10 and 0.5 log10 at water temperatures of 23.2d egrees C and 5.2 degrees C, respectively. Chlorite concentrations remained below both the US EPA maximum contaminant level of 1.0 mg/L and the maximum contaminant level goal of 0.8 mg/L for up to 2.0log10 B. subtilis inactivation.

Factorial experiments were conducted using source waters from seven drinking water treatment plants in Ontario, Canada to develop statistically based model equations capable of predicting chlorine dioxide consumption and chlorite and chlorate formation upon chlorine dioxide application. The equations address raw water quality and operational parameters including pH, temperature, chlorine dioxide concentration, reaction time and water organic content (as described by non-purgeable organic carbon x ultraviolet absorbance measured at 254 nm, NPOC x UV254). Terms describing two-factor interaction effects were also included, improving the accuracy of the predictive equations in fitting measured response concentrations as evaluated through internal and external validations. Nearly 80% of the predictions for chlorine dioxide consumption and chlorite formation were observed to be within 20% of the measured levels. Over 90% of the predicted chlorate levels were within +/- 0.1 mg/L of the measured levels. Chlorine dioxide concentration and NPOC x UV254 were key parameters when developing the predictive models.

Chlorite, chlorate, and trihalomethane (THM) formation was measured following the disinfection of drinking water using chlorine dioxide either alone, or followed by chlorine or monochloramine. Chlorine dioxide doses were selected to provide a range of Giardia inactivation levels as calculated using CT values (disinfectant concentration × contact time) presented in the US Surface Water Treatment Rule. Experiments were conducted at bench-scale in batch reactors, using water obtained from two river sources. Dose requirements did not vary greatly when disinfecting at 1°C versus 25°C, since the higher CT requirements in the cold (due to slower disinfection kinetics) were offset by greater chlorine dioxide stability at low temperature. Higher doses were required at pH 9 compared to pH 6, due to chlorine dioxide disproportionation at the higher pH. Greater formation of chlorite and chlorate was observed at the higher pH. Temperature exerted a relatively small effect on chlorite and chlorate formation, following ...

The recognition that chlorine disinfection of drinking water may not be effective in controlling pathogens such as Cryptosporidium may lead to the greater use of stronger alternative disinfectants, such as chlorine dioxide. Typical chlorine dioxide residual concentration requirements for disinfection may extend to less than 0.1 mg l−1, thus requiring very good quantitation methods for optimal process control. Traditional methods have been cumbersome and sometimes inaccurate. This study examined three spectrophotometric methods for measuring chlorine dioxide in the

On-line particle count and turbidity data was used to monitor the impacts of transient particulate water quality events on the inactivation of indigenous total coliform bacteria by a pilot-scale ultraviolet (UV) disinfection unit at two drinking water treatment plants (referred to as WTPs ‘A’ and ‘B’). The intent of the study was to assess the performance of UV disinfection when exposed to temporary ‘worst case’ particulate water quality conditions in a drinking water treatment context. The pilot unit was installed downstream of a poorly operating filter at WTP ‘A’ and downstream of an intentionally destabilized Actiflo® system at WTP ‘B’. In each case, elevated particle content (turbidity up to 2.9 NTU) did not cause observable adverse effects on the UV disinfection of total coliform bacteria over the test periods, even when a particle extraction technique was applied to count particle-associated bacteria in the water samples.

Studies have shown that viruses are frequently attached onto or enmeshed within suspended particles in natural water and wastewater and that this viral–particle association can impede disinfection processes in some circumstances. This article summarizes the current state of knowledge of the occurrence of particle-associated viruses in the aquatic environment and the reported subsequent impacts on disinfection processes. The intention is that this article will help guide future research and aid regulators attempting to address the disinfection of particle-laden waters.

Haloacetic acids (HAAs) are produced by the reaction of chlorine with natural organic matter and are regulated disinfection by-products of health concern. Biofilms in drinking water distribution systems and in filter beds have been associated with the removal of some HAAs, however the removal of all six routinely monitored species (HAA6) has not been previously reported. In this study, bench-scale glass bead columns were used to investigate the ability of a drinking water biofilm to degrade HAA6. Monochloroacetic acid (MCAA) and monobromoacetic acid (MBAA) were the most readily degraded of the halogenated acetic acids. Trichloroacetic acid (TCAA) was not removed biologically when examined at a 90% confidence level. In general, di-halogenated species were removed to a lesser extent than the mono-halogenated compounds. The order of biodegradability by the biofilm was found to be monobromo > monochloro > bromochloro > dichloro > dibromo > trichloroacetic acid.