Using Fractionated Natural Organic Matter to Quantitate Organic Byproducts of Ozonation

An improved procedure was used to isolate and fractionate natural organic matter (NOM) in water for subsequent ozonation and disinfection by-product (DBP) and color removal quantisation. Isolated NOM fractions from two different sources, accounting for approximately 50 to 60% of the dissolved organic material and 60 to 75% of the color, were characterized and then ozonated under conditions approximating those encontered during drinking water treatment. The natural waters also were ozonated. Organic DBPs of either health concern or which may contribute to biological instability of finished water were investigated, including aldehydes, oxoacids and low molecular weight carboxylic acids. pH and ozone dosage were the parameters having the greatest effect on DBP formation. On the basis of UV absorbance measurements, the fulvic acid fractions studied taken together accurately represented the natural water and may be the primary sources of precursor material for aldehydes and oxoacid DBPs. However, as yet unidentified NOM fractions contribute significantly to carboxylic acid formation upon ozonation.     

Mechanisms of Phenol Ozonation—Kinetics of Formation of Primary and Secondary Reaction Products

Organic acids are part of assimilable organic carbon (AOC) formed from natural organic matter (NOM) during ozonation for drinking water production. To elucidate the formation of organic acids, phenol as surrogate compound for NOM was ozonated while suppressing reactions of ^·OH radicals by addition of tert-butanol. Batch experiments show benzoquinone (40 and 18%), catechol (33 and 1%), cis,cis-muconic acid (6 and 3%), and hydroquinone (2 and 7%) to be the most abundant primary products at pH 7 and 3, respectively. The tertiary product oxalic acid was obtained in similar yields from phenol ozonation (0.8%) as during ozonation of lake water (1.6%). Together with other low molecular weight organic acids it was formed upon small ozone exposures, as was shown by time-dependent experiments in time ranges of 0.4–23 and 0.07–1.0 seconds for pH 3 and 7.25, respectively.     

Effects of Ozonation on Molecular Weight Distribution of Humic Substances and Coagulation Processes – A Case Study: The Úzquiza Reservoir Water

In this article the influence of preozonation on the effectiveness of NOM removal via coagulation processes will be studied (focusing on the influence of the calcium hardness) as well as changes in MW (molecular weight) distribution of humic substances caused by ozonation. Additionally, THMFP removal in both ozonation and preozonation-coagulation processes is assessed. Three different types of water have been used in this study: a natural water from the Úzquiza Reservoir (Burgos, Spain), a synthetic water prepared using natural fulvic acids extracted from the Úzquiza Reservoir and a synthetic water prepared using a commercially supplied humic acid. Molecular weights of humic substances were determined using high-performance size exclusion chromatography (HPSEC); average molecular weights calculated for the unozonated humic substances are 4500 Da for the commercial humic acids and 1000 Da for the natural fulvic acids extracted from the Úzquiza Reservoir. Preozonation shifted the molecular weight distribution of humic substances (both humic and fulvic acids) towards lower average molecular weight values. For the natural water from the Úzquiza Reservoir (with low levels of calcium hardness and hydrophobic fraction (humic substances) being the main fraction of NOM), preozonation has a negative effect on the effectiveness of the coagulation process for NOM removal: the percentages of TOC removal via coagulation decrease with increasing ozone dosage; the maximum TOC removal (33%) is achieved for the unozonated water. Also for this water, ozonation reduced 5–25% of THMFP with ozone doses varying from 0.25 to 2.5 mg O₃/L. A preferential THMFP removal, that is to say, higher reduction in THMFP (43%) relative to TOC (28%) is achieved by the coagulation-flocculation process; this also occurs when preozonation is used, independently of ozone dosage.     

Ozone Degradation of Cyclophosphamide – Effect of Alkalinity and Key Effluent Organic Matter Constituents

The influence of three effluent organic matter (EfOM) model compounds (alginic acid, peptone and natural organic matter-NOM) and alkalinity on the ozonation of cyclophosphamide (CPD) was investigated. The rate of ozone decay increased with increasing model compounds concentration in the order of peptone > NOM > alginic acid. Increasing alkalinity inhibited ozone decay at all concentrations of alginic acid and at low concentrations of NOM and peptone (DOC < 3 mg/L), while at high NOM and peptone concentrations the effect of alkalinity on ozone decay was minor. Presumably, ozone decay was mainly controlled by direct reaction with these two model compounds, resulting in ?OH (hydroxyl radical) formation yield of 29% and 19%, for peptone and NOM respectively. In the presence of alginic acid ?OH formation through a radical chain reaction resulted in a yield of 30%. Cyclophosphamide (CPD) removal decreased with increasing alkalinity and model compounds concentration (most pronounced for peptone and less pronounced for alginic acid); most likely due to increase in the scavenging effect on ?OH.     

Removal Mechanism of Natural Organic Matter and Organic Acid by Ozone in the Presence of Goethite

The oxidations of natural organic matter (NOM) and a model compound (p-chlorobenzoic acid) were characterized using ozonation and catalytic ozonation processes. In general, the catalytic ozonation showed better performance in the removal of organics tested in the study. The hydrophobic, transphilic, and hydrophilic NOM fractions were isolated using XAD-8 and -4 resins to evaluate the reaction characteristics. The catalytic ozonation in the presence of goethite accomplished the higher removal of NOM with simultaneous reduction of the three fractions than the ozonation which removed the hydrophobic portion only. The analysis of discrete size distributions of NOM revealed that ozonation yielded a removal of >1,000MW and an increase of <1,000MW, whereas all molecular weight fractions reduced after catalytic ozonation. The concentrations of model compound and aqueous and gaseous ozone were monitored during the oxidations, and efficiencies were compared for cases in the absence and presence of iron oxide (FeOOH).     

Investigation of Ozone Reaction in River Waters Causing Instantaneous Ozone Demand

Ozone consumption by water can be characterized by the instantaneous ozone demand (IOD) and a pseudo first order decay constant. Utilizing the flow injection analytical system for measuring IOD, the instantaneous ozone demand characteristics of two river waters (Korea) were investigated, utilizing a ?OH probe compound and ?OH scavenger, and were compared with those of two commercial humic acids (the Suwannee River humic acid and Aldrich humic acid). The major findings were as follows; (1) The IOD in river waters was found to be mainly due to the reaction of the ozone with natural organic matter (NOM), which constituted approximately 0.26–0.29?mg/mg DOC, and was responsible for the consumption of more than 40% of the applied ozone. Whereas, the IOD of the two commercial humic acids were three times more than those of the river waters. (2) The IOD in the river waters was mainly caused by the direct ozone reaction with dissolved organics, not from the ?OH mediated ozone reaction. However, for the two commercial humic acids, more than 40% of the IOD came from the ?OH mediated ozone reaction. (3) The hydrophobic fractions of the dissolved organics in the river waters were mainly responsible for the IOD. The IOD of the hydrophobic organics was approximately ten times larger than that of the hydrophilic organics. Although the exact magnitude of the IOD, and the relative importance of the direct/indirect ozone reaction with river water may vary greatly depending upon the source of the NOM, the characteristics of the IOD compromise a significant fraction of the ozone dose need (especially in achieving good ozone disinfection) in water treatment plants.     

Characteristics of Natural Organic Matter and Formation of Chlorinated Disinfection By-Products from Two Source Waters that Respond Differently to Ozonation

This study investigated the characteristics of natural organic matter (NOM) in two different raw surface water sources that respond differently to ozonation: one for which ozonation decreases the disinfection by-product (DBP) formation potentials (i.e., Capilano Reservoir, Vancouver, Canada), and one for which ozonation does not (i.e., South Thompson River, Kamloops, Canada); and evaluated the effect of ozonation on these characteristics and on the DBP formation potential of the different size and polar fractions of the NOM. Although the South Thompson River and the Capilano Reservoir waters had relatively similar total organic carbon concentrations, the characteristics of the NOM (e.g., size and polar distribution, specific UV absorption), in these water sources differed significantly. In general, no clear and consistent trend was observed with respect to the tendency of different size and polar fractions of NOM to generate DBPs. Nonetheless, the results from the present study suggest that hydrophobic NOM has a higher tendency to form DBPs. In addition, when considering individual size and polar fractions, specific UV absorption was a good overall indicator of the DBP formation potential for a given water source. The effect of ozonation on South Thompson River and Capilano Reservoir waters also differed significantly. For both source waters, ozonation appeared to have a greatest effect on the more hydrophilic fractions, generally increasing the DBP formation potential of the smaller more hydrophilic NOM, while generally decreasing that of the larger more hydrophilic NOM. The beneficial effect of ozonation on reducing haloacetic acid (HAA) formation potentials was due to a reduction in both the dichloroacetic acid (DCAA) and trichloroacetic acid (TCAA) formation potentials, while the negative effect of ozonation on increasing HAA formation potentials was due to an increase in the DCAA formation potentials. The results from the present study clearly indicate that the use of ozone as a primary disinfectant does not necessarily reduce the DBP formation potential of NOM in all water sources, further demonstrating the complex structure of NOM and the fact that NOM from different sources cannot be simply treated as one entity and compared with one another.     

Study of the Catalytic Ozonation of Humic Substances in Water and Their Ozonation Byproducts

It is shown that using transition metals, especially Mn(II) and Ag(I), during ozonation of humic substances in water allows important reductions in the content of organic matter. Characterization of the organic compounds resulting from ozonation was made by concentrating the sample through liquid-liquid extraction or derivation with PFBOA.HCl, along with the GC/MS and GC/ECD techniques. In total, 110 different organic compounds were identified using GC/MS; mainly carboxylic acids, aromatics, hydrocarbons, aldehydes, ketones, and furan-carboxylic acids. The percentages of elimination or formation levels reached during ozonation are discussed.     

Effects Of Ozone On The Production Of Biodegradable Dissolved Organic Carbon (BDOC) During Water Treatment

This article deals with the oxidation effect of ozone on the increasing fraction of biodegradable organic matter with the “ozotest” method, a laboratory technique which simulates the effect of ozonation and allows a complete oxidation assessment. Ozone treatment was performed on river water samples and sand filter effluent samples. Ozone consumption, reduction of UV absorbance and BDOC formation were monitored with applied ozone doses from 0 to 10 mg/L and with contact times from 0 to 60 min. The BDOC formation was optimum at an applied ozone dose of 0.25-0.5 mg O₃ per mg DOC (contact time = 5 min) corresponding to apparition of traces of residual ozone and maximum UV reduction. Maximum ozone consumption, UV reduction and BDOC formation occurred simultaneously during the first two minutes of treatment. Concerning BDOC formation, applied ozone dose showed a greater effectiveness than contact time. For the same quantity of consumed ozone, a short contact time associated with a high ozone dose was preferable to a long contact time and a low ozone dose.     

Changes in NOM Fractionation through Treatment: A Comparison of Ozonation and Chlorination.

NOM isolation and fractionation to provide insight into the effectiveness of ozonation vs. conventional water treatment was done. In this research, the dissolved portion of natural organic matter (NOM) or dissolved organic matter (DOM) at two surface drinking water treatment plants that treat the same source water was fractionated by resin adsorption. The first treatment plant uses conventional treatment (coagulation, sedimentation, and filtration) with intermediate free chlorination and post chlorination while the second plant uses conventional treatment with pre and intermediate ozonation, and multi-media filtration unit operation. Several different sampling locations within each plant were selected for DOM isolation and fractionation into six fractions (hydrophobic acid, neutral and base, and hydrophilic acid, neutral, and base). The effectiveness of each treatment plant on the oxidation and removal of each organic fraction are discussed. Oxidation by ozone leads to better overall performance in the removal of DOM.     

Reaction Mechanisms for DBPS Reduction in Humic Acid Ozonation

In this study, XAD-8 resins were used to extract the natural organic matter (NOM) from samples collected at the intake of Feng Yuan Water Treatment Plant (in central Taiwan) into five groups: humic acids, fulvic acids, hydrophobic neutrals, hydrophobic bases, and hydrophilic fractions. Quantitative results show that hydrophobic and hydrophilic fractions contribute 44.2% and 55.5% of NOM, individually. Ozonation processes will significantly reduce both disinfection by-products formation potential (DBPFP) and average molecular weight of the humic acid sample. Additionally, double carbon bonds are broken up so that hydrophobic fractions were converted to hydrophilic fractions by ozonation leading to the reduction of DBPFP.     

Combined Use of Ozone and Granular Activated Carbon (GAC) in Potable Water Treatment; Effects on GAC Quality After Reactivation

Common processes in potable water treatment regarding dissolved organics removal (natural organic matter, NOM, and organic micropollutants, OMP) include oxidation techniques – like ozonation -and adsorption onto granular activated carbon (GAC). This paper deals with combined ozonation and GAC filtration. Effects on water quality are discussed: partial oxidation of dissolved organics by ozone, leading to changing adsorption behavior and enhanced biodegradation in GAC columns. Special emphasis is laid on long term effects of ozone on GAC quality after several thermal reactivations. Long term experience indicates that ozonation prior to GAC filtration, leading to moderate ozone concentrations in the GAC influent, does not influence GAC quality negatively, even after more than ten reactivation cycles.     

Formaldehyde Formation During Ozonation Of Drinking Water

Partial oxidation of natural organic material during ozonation produces oxygenated by-products of low molecular weight. Formaldehyde, being the most common oxygenated by-product of ozone, is considered to be a problematic compound by the water industry due to its potential adverse health effects. This research attempts to provide specific information on the effects of water quality parameters, specifically, pH and alkalinity, the structure of humic material, and the operational parameters, e.g., ozone dosage and contact time, on generation of formaldehyde. The results showed that ozonation caused almost an immediate formation of formaldehyde, which reached a peak value, and then started to decrease with continued ozonation. Ozonation of aqueous fulvic acid produced higher concentrations of formaldehyde compared to other types of humic material. Formaldehyde formation was suppressed by high bicarbonate levels, and enhanced at higher pH. Formaldehyde accumulation was more dramatic at low ozone dosages.     

Origin and Conditions of Ketoacid Formation During Ozonation of Natural Organic Matter in Water

The batch ozonation of some fractions of aqueous humic substances (humic, fulvic and hydrophilic acids) extracted from natural waters is studied. The reaction leads to oxidation byproducts such as low molecular aldehydes and ketoacids. Formation conditions and origin of some of them (glyoxylic, pyruvic and ketomalonic acids) regarding the extracted fraction of natural organic matter is established.     

Catalytic ozonation of humic acids with Fe/MgO

Humic acids were degraded by ozone at room temperature in a stirred tank reactor and a fixed bed reactor with Fe/MgO catalysts. Experimental results show that the ozonation with Fe/MgO induced a significant reduction in UV absorbance of humic acids as compared to ozone alone. Fe/MgO was the most efficient catalyst to degrade humic acids in the presence of ozone. GPC (gel permeation chromatography) showed that the humic acids with high molecular weight could be severely decomposed into organic compounds with low molecular weight on the Fe/MgO catalyst, indicating that humic acids could be catalytically decomposed. The continuous reaction experiments with the palletized catalysts supported that humic acids can be removed by catalysis as well as adsorption.     

Effects of Ozone on Resin Acids in Thermomechanical Pulp and Paper Mill Circulation Waters

The present study investigated ozonation as means to remove resin acids from pulp and paper mill circulation waters. Ozone selectively oxidizes resin acids in debarking, thermomechanical pulping (TMP), and newsprint machine (PM) circulation waters. The relative ozone dose of 0.2 mgO₃/initial mgCOD eliminated over 90% of total resin acid concentration in all tested waters, despite the initial resin acid or organic matter concentration. With that ozone dose, the removal of organic material (measured as COD) was only about 30%. Because of the different COD concentrations of tested waters, the absolute amount of ozone needed for over 90% resin acid reduction was 2000 mg/1 for debarking water, 600 mg/1 for TMP water, and 300 mg/1 for PM water, when initial COD concentrations were 10,000, 3000, and 1500 mg/1, respectively. In theory, abietane type resin acids should be more susceptible to ozone attack than pimarane type resin acids, because of their conjugated double bond structure. In the experiments, a high oxidizability of pimarane type resin acids, as compared to abietane type resin acids, was found.     

Soil Fulvic Acid Degradation By Ozone In Aqueous Medium

Armadale Armadale soil fulvic acid (SFA) was found to contain several loosely bound organic impurities which could be removed by ethyl acetate extraction. The ozonation of purified Armadale SFA at a variety of dosages was characterized by monitoring the UV absorbance, weight loss, pH changes, total acidity, molecular weight and elemental composition. SFA could only be partially degraded even under ozone dosages as high as SFA/O₃ (w/w) of 1:6. At high ozone dosages (SFA/O3 1:6) mostly aliphatic compounds rich in oxygen were produced, whereas at low ozone dosages (SFA/O₃ < 1:0.5) mainly benzene polycarboxylic acids and polyhydroxy benzene polycarboxylic acids were found. The solid material bisolated from the chlorination (0.5 mg/L residual level) of residues from the ozonation (SFA/O₃ 1:0.2 to 1:0.5) of Contech SFA did not contain any chlorinated products.     

Removal Characteristics of Organic Pollutants in Sewage Treatment by a Pre-Coagulation,Ozonation and Ozone/Hydrogen Peroxide Process

The applicability of a sequential process of ozonation and ozone/hydrogen peroxide process for the removal of soluble organic compounds from a pre-coagulated municipal sewage was examined. 6–25% of initial T-COD_Cr was removed at the early stage of ozonation before the ratio of consumed ozone to removed T-COD_Cr dramatically increased. Until dissolved ozone was detected, 0.3 mgO₃/mgTOC₀ (Initial TOC) of ozone was consumed. When an ozone/hydrogen peroxide process was applied, additional COD_Cr was removed. And we elucidated that two following findings are important for the better performance of ozone/hydrogen peroxide process; those are to remove readily reactive organic compounds with ozone before the application of ozone/hydrogen peroxide process and to avoid the excess addition of hydrogen peroxide. Based on these two findings, we proposed a sequential process of ozonation and multi-stage ozone/hydrogen peroxide process and the appropriate addition of hydrogen peroxide. T-COD_Cr, TOC and ATU-BOD₅ were reduced to less than 7 mg/L, 6 mgC/L and 5 mg/L, respectively after total treatment time of 79 min. Furthermore, we discussed the transformation of organic compounds and the removal of organic compounds. The removal amount of COD_Cr and UV₂₅₄ had good linear relationship until the removal amounts of COD_Cr and UV₂₅₄ were 30 mg/L and 0.11 cm^?1, respectively. Therefore UV₂₅₄ would be useful for an indicator for COD_Cr removal at the beginning of the treatment. The accumulation of carboxylic acids (formic acid, acetic acid and oxalic acid) was observed. The ratio of carbon concentration of carboxylic acids to TOC remaining was getting higher and reached around 0.5 finally. Removal of TOC was observed with the accumulation of carboxylic acids. When unknown organic compounds (organic compounds except for carboxylic acids) were oxidized, 70% was apparently removed as carbon dioxide and 30% was accumulated as carboxylic acids. A portion of biodegradable organic compounds to whole organic compounds was enhanced as shown by the increase ratio of BOD/COD_Cr.     

TiO2-Based Heterogeneous Photocatalytic Water Treatment Combined with Ozonation

The degradations of aliphatic carboxylic acids (formic acid, acetic acid, and propionic acid) were investigated by a combination of TiO₂-based photocatalysis and ozonation at pH = 2. The carboxylic acids were analyzed by high-performance liquid chromatography (HPLC), while the mineralization process was characterized by measuring the dissolved organic carbon (DOC) content. The efficiency of heterogeneous photocatalysis combined with ozonation was compared with that of ozonation and that of heterogeneous photocatalysis alone. In all cases, the combined system proved most effective in the oxidation of carboxylic acids. The degradation of carboxylic acids was accompanied by a continuous decrease in the DOC content, and aldehydes and carboxylic acids containing one carbon atom less than the starting material were formed as intermediates.     

How Bromate and Ozone Concentrations Can be Modeled at Full-Scale Based on Lab-Scale Experiments – A Case Study

This article presents a full-scale modeling study of an industrial ozonation unit for practical application. The modeling framework combines an integrated hydraulic model (systematic network) with a quasi-mechanistic chemical model. Dealing with natural water, the chemical model has to be parameterized, and the parameters calibrated. This was done based on lab-scale experiments. The calibration results showed that the chemical model is able to account for changes in contact time with ozone, pH, temperature, ozone dose, NOM concentration, bromide concentration. Comparison of residence time distributions showed that the hydraulic model accurately reproduces flow conditions. Six sampling points were installed along an industrial ozonation unit of 487 m³ consisting of two baffled tanks in series. Bromate and ozone concentrations were monitored under varying operational process conditions. After the selection of a value for the k_La, simulations were run. Using the lab-scale calibrated models, simulated and experimental data were found in close agreement: 84% of the simulated concentrations for ozone matched measurements (±experimental error), 60 % for bromate. A readjustment of the kinetics of a single reaction (out of 65) showed that seasonal changes in NOM activity may easily be taken into account based on regular concentration measurements (90% of the bromate concentrations were then modeled accurately).