Formation of halogenated forms of bisphenol A (BPA) in water: Resolving isomers with ion mobility - mass spectrometry and the role of halogenation position in cellular toxicity.

Halogenated BPA (XBPA) forms resulting from water chlorination can lead to increased toxicity and different biological effects. While previous studies have reported the occurrence of different XBPAs, analytical limitation have hindered the analysis and differentiation of the many potential isomeric forms. Using online solid-phase extraction - liquid chromatography - ion-mobility - high-resolution mass spectrometry (OSPE-LC-IM-HRMS), we demonstrated a rapid analysis method for the analysis of XBPA forms after water chlorination, with a total analysis time of less than 10 min including extraction and concentration and low detection limits (∼5-80 ng/L range). A multi in-vitro bioassay testing approach for the identified products revealed that cytotoxicity and bioenergetics impacts were largely associated with the presence of halogen atoms at positions 2 or 2' and the overall number of halogens incorporated into the BPA molecule. Different XBPA also showed distinct impacts on oxidative stress, peroxisome proliferator-activated receptor gamma - PPARγ, and inflammatory response. While increased DNA damage was observed for chlorinated water samples (4.14 ± 1.21-fold change), the additive effect of the selected 20 XBPA studied could not explain the increased DNA damage observed, indicating that additional species or synergistic effects might be at play.

Reactions of Monobromamine and Dibromamine with Phenolic Compounds and Organic Matter: Kinetics and Formation of Bromophenols and Bromoform.

Monobromamine (NH2Br) and dibromamine (NHBr2) produced from reactions of hypobromous acid (HOBr) with ammonia can react with phenolic structures of natural organic matter (NOM) to produce disinfection byproducts such as bromoform (CHBr3). The reactivity of NH2Br was controlled by the reaction of the bromoammonium ion (NH3Br+) with phenolate species, with specific rate constants ranging from 6.32 × 102 for 2,4,6-tribromophenol to 1.22 × 108 M-1 s-1 for phenol. Reactions of NHBr2 with phenol and bromophenols were negligible compared to its self-decomposition; rate constants could be determined only with resorcinol for pH > 7. At pH 8.1-8.2, no formation of CHBr3 was observed from the reaction of NH2Br with phenol while the reaction of NH2Br with resorcinol produced a significant concentration of CHBr3. In contrast to NH2Br, a significant amount of CHBr3 produced with an excess of NHBr2 over phenol was explained by the reactions of HOBr produced from NHBr2 decomposition. A comprehensive kinetic model including the formation and decomposition of bromamines and the reactivity of HOBr and NH2Br with phenolic compounds was developed at pH 8.0-8.3. Furthermore, the kinetic model was used to evaluate the significance of the NH2Br and NHBr2 reactions with the phenolic structures of two NOM isolates.

Kinetic modelling of the bromine-ammonia system: Formation and decomposition of bromamines.

Bromamines i.e. monobromamine (NH2Br), dibromamine (NHBr2), and tribromamine (NBr3) can be formed during oxidative treatment of waters containing bromide and ammonia. The formation and decomposition of bromamines in aqueous solution was investigated and a comprehensive kinetic model of the bromine-ammonia system was developed at 23 ± 1 °C. Determination of rate constants and model validation were primarily performed at pH 8.0 - 8.3 for subsequent application to seawater disinfection. The rate constant of NHBr2 self-decomposition was determined by second-order rate law linearization with k9 = 5.5 (± 0.8) M-1s-1 at pH 8.10. The rate constant of NBr3 self-decomposition increased proportionately to the concentration of hydroxide ions (OH-) according to the equation k10 = 4.4 (± 0.1) × 107. [OH-] over the pH range 6.0 - 8.5, which gave k10 = 56 (± 1) M-1s-1 at pH 8.10. The rate constants of NHBr2 and NBr3 formation were obtained by fitting model-predicted data to the experimental results and were found to be k3 = 2.3 (± 0.2) × 104M-1s-1 and k5 = 4.0 (± 0.6) × 103M-1s-1, respectively at pH 8.10. NBr3 was also found to react with NHBr2 with k11 = 3.4 (± 0.2) × 103M-1s-1 at pH 8.10. A kinetic model was proposed based on these experimental rate constants and literature values, which provided a good prediction of bromamines formation and decomposition for various initial bromine and ammonia concentrations. The kinetic model was also used to accurately predict the total oxidant concentration and the speciation of bromamines during breakpoint bromination. This study provides kinetic data to model more complex oxidative systems such as seawater chlorination in the presence of ammonia.

Genotoxic effects of chlorinated disinfection by-products of 1,3-diphenylguanidine (DPG): Cell-based in-vitro testing and formation potential during water disinfection.

1,3-diphenylguanidine (DPG) is a commonly used rubber and polymer additive, that has been found to be one of the main leachate products of tire wear particles and from HDPE pipes. Its introduction to aquatic environments and potentially water supplies lead to further questions regarding the effects of disinfection by-products potentially formed. Using different bioassay approaches and NGS RNA-sequencing, we show that some of the chlorinated by-products of DPG exert significant toxicity. DPG and its chlorinated by-products also can alter cell bioenergetic processes, affecting cellular basal respiration rates and ATP production, moreover, DPG and its two chlorination products, 1,3-bis-(4-chlorophenyl)guanidine (CC04) and 1-(4-chlorophenyl)-3-(2,4-dichlorophenyl)guanidine (CC11), have an impact on mitochondrial proton leak, which is an indicator of mitochondria damage. Evidence of genotoxic effects in the form of DNA double strand breaks (DSBs) was suggested by RNA-sequencing results and further validated by an increased expression of genes associated with DNA damage response (DDR), specifically the canonical non-homologous end joining (c-NHEJ) pathway, as determined by qPCR analysis of different pathway specific genes (XRCC6, PRKDC, LIG4 and XRCC4). Immunofluorescence analysis of phosphorylated histone H2AX, another DSB biomarker, also confirmed the potential genotoxic effects observed for the chlorinated products. In addition, chlorination of DPG leads to the formation of different chlorinated products (CC04, CC05 and CC15), with analysed compounds representing up to 42% of formed products, monochloramine is not able to effectively react with DPG. These findings indicate that DPG reaction with free chlorine doses commonly applied during drinking water treatment or in water distribution networks (0.2-0.5 mg/L) can lead to the formation of toxic and genotoxic chlorinated products.

Recycling and persistence of iodine 127 and 129 in forested environments: A modelling approach.

Differences in the source and behaviour of 129I compared to 127I isotopes have been described for a variety of surface environments, but little is known about the cycling rates of each isotope in terrestrial ecosystems. We developed a compartment model of the iodine cycle in a forest ecosystem, with a labile and non-labile pool to simplify the complex fate of iodine in the forest floor and soil. Simulations were performed using atmospheric 127I and 129I inputs for sites differing in climate, vegetation, and soil. In general, considering dry deposition in addition to wet deposition improved model simulations. Model results support the view that soil is the sink for atmospheric iodine deposited in forest ecosystems, while tree vegetation has little influence on long-term iodine budgets. Modelling also showed that iodine cycling reaches equilibrium after a period of about 5000 years, mainly due to a gradual incorporation of iodine into the bulk stabilised soil organic matter. At steady state, this pool of non-labile iodine in soil can retain about 20% of total deposition with a mean residence time of 900 years, while the labile iodine pool is renewed after 90 years. The proportions of modern anthropogenic 129I in each modelled pool reflect those of stable 127I at least several decades after input to the forest; this result explains why isotopic disequilibrium is common in field data analysis. Volatilisation plays a central role in regulating iodine storage in soil and, therefore, its residence time, while drainage is a minor export pathway, except at some calcareous sites. Dynamic modelling has been particularly helpful for gaining insight into the long-term response of iodine partitioning to continuous, single or even varying deposition. Our modelling study suggested that better estimates of dry deposition of atmospheric iodine, weathering of parent rock, and volatilisation of the deposited iodine from soil and vegetation will be required for reliable predictions of iodine cycling in specific forests, because these processes remain insufficiently explored.

Brominated trihalamines in chlorinated seawaters: Quantification of tribromamine and identification of bromochloramines by Membrane Introduction Mass Spectrometry.

During chlorination of seawater, the presence of bromide and ammonia alters the speciation of the oxidant and lead to the formation of chlorinated and brominated amines. This can affect the effectiveness of the disinfection treatment and the formation of disinfection by-products released to the environment. In this study, a Membrane Introduction Mass Spectrometry (MIMS) analytical method was developed to differentiate brominated trihalamines (i.e. tribromamine NBr3, dibromochloramine NBr2Cl and bromodichloramine NBrCl2) in synthetic and natural chlorinated seawater. A mass-to-charge ratio of m/z = 253 corresponding to the parent ion was used for the quantification of NBr3 in absence of organic matter and the signal of the fragment at m/z = 177 was chosen in presence of high concentration of organic matter. Limits of detection were 0.23 µM (49 µg Cl2/L) and 0.18 µM (38 µg Cl2/L) for m/z 253 and m/z 177, respectively. Both NBr2Cl and NBrCl2 were monitored in chlorinated seawaters with their respective parent ion at m/z = 207 and m/z = 163 but were not quantified. MIMS results also showed that reaction of brominated trihalamines with natural organic matter (NOM) was a minor pathway for 1-2 mg C/L compared to their auto-decomposition in natural or synthetic seawater. Overall, MIMS was able to unambiguously differentiate and monitor brominated trihalamines for the first time in chlorinated seawater, which was not possible by using UV measurement, titration and colorimetric methods.

Influence of tree species on selenium and iodine partitioning in an experimental forest ecosystem.

Storage of selenium and iodine can greatly vary between forest ecosystems, but the influence of tree species on partitioning and recycling of those elements remains elusive. In this study, contents of Se and I were measured in tree compartments, litterfall, humus, and soil horizons in monospecific stands of Douglas fir, pine, spruce, beech, and oak under identical climatic and edaphic conditions. The cycle of each element was characterized in terms of stocks and fluxes. Lowest concentrations were in wood (Se: 8-13 µg kg-1; I: <16.5 µg kg-1). Senescing organs had higher Se and I content, than the living parts of trees due to direct exposure to atmospheric deposition, with some variation between coniferous and deciduous trees. For all stands, low amounts of Se and I were involved in biological cycle as reflected by low root uptake. In humus, the enrichment of elements greatly increased with the stage of organic matter (OM) degradation with average factors of 10 and 20 for Se and I. OM degradation and element persistence in humus was influenced by tree species. Deciduous trees, with low biomass, and fast degradation of OM stored less Se and I in humus compared to fir and spruce with high humus biomass. Interestingly, tree species did not affect soil reserves of Se and I. Concentration ranges were 331-690 µg Se kg-1 and 4.3-14.5 mg I kg-1. However, the divergent vertical profiles of the elements in the soil column indicated greater mobility of I. Selenium concentrations regularly decreased with depth in correlation with OM and Fe oxides content. For iodine, the maximum iodine concentration at a soil depth of 15 to 35 cm was caused by a parallel precipitation/sorption behavior of aluminium and organic iodine dissolved in the topsoil.

Selenium distribution in French forests: Influence of environmental conditions.

Selenium is a trace element and an essential nutrient. Its long-lived radioisotope, selenium 79 is of potential radio-ecological concern in surface environment of deep geological repository for high-level radioactive waste. In this study, the influence of environmental, climatic and geochemical conditions on stable Se (as a surrogate of 79Se) accumulation was statistically assessed (PCA analysis, Kruskall-Wallis and Spearman tests) based on the analysis of its concentration in litterfall, humus, and soil samples collected at 51 forest sites located in France. Selenium concentrations were in the ranges: 22-369, 57-1608 and 25-1222 µg kg-1 respectively in litterfall, humus, and soil. The proximity of the ocean and oceanic climate promoted Se enrichment of litterfall, likely due to a significant reaction of wet deposits with forest canopy. Se content was enhanced by humification (up to 6 times) suggesting that Se concentrations in humus were affected by atmospheric inputs. Selenium stock in humus decreased in the order of decreasing humus biomass and increasing turnover of organic matter: mor > moder > mull. Positive correlations between Se content and geochemical parameters such as organic carbon content, total Al and total Fe confirmed the important role of organic matter (OM) and mineral Fe/Al oxides in Se retention in soils.

Chlorination and bromination of 1,3-diphenylguanidine and 1,3-di-o-tolylguanidine: Kinetics, transformation products and toxicity assessment.

This works investigates the chlorination and bromination of two rubber and polymer related chemicals, which have emerged as relevant water contaminants, i.e. 1,3-di-o-tolylguanidine (DTG) and 1,3-diphenylguanidine (DPG). Kinetic constants at different pH values were obtained and modelled, taking into account the pKa values of DTG/DPG and HClO, showing that the maximum reaction rate (kapp > 104 M-1 s-1) is obtained at pH values 8.8 for DPG and 9.1 for DTG. Bromination is also very fast, although unlike chlorination, deviation from the model was observed at neutral pH, which was attributed to formation of metastable transformation product (TP). A total of 35 TPs, corresponding to halogenation, hydroxylation, formation of monophenylguanidine derivatives and cyclization reactions, were tentatively identified. Furthermore it was found that chloroform can be formed up to a 25% molar yield, while dichloroacetonitrile was formed into less than a 3% yield. Several ecotoxicological endpoints were predicted by quantitative structure-activity relationship models (QSAR) for the TPs, some of which were predicted to be more toxic than DPG/DTG. Also a chlorinated solution investigated by a Vibrio Fisheri acute toxicity test, confirmed that toxicity increases with chlorination.

Method Development for Quantification of Bromochloramine Using Membrane Introduction Mass Spectrometry.

During chloramination of bromide-containing waters, the main brominated amine formed is bromochloramine (NHBrCl). To date, there is no analytical method, free of interference, allowing its accurate quantification. The major reason is that it is not possible to produce a pure NHBrCl solution. In this study, we report a method allowing the accurate quantification of NHBrCl with membrane introduction mass spectrometry (MIMS). First, the molar absorption coefficient for NHBrCl was determined by quantifying NHBrCl as 2,4,6-tribromophenol by HPLC-UV and comparing the results with the direct UV response at 320 nm. A molar absorption coefficient of 304 M-1cm-1 was obtained. The results obtained by direct UV measurements were compared to the MIMS signal recorded at m/ z 131 corresponding to the mass of the molecular ion and used to establish a calibration curve. A limit of detection of 2.9 µM (378 µg/L) was determined. MIMS is the only method enabling the unambiguous quantification of NHBrCl, as it is based on m/ z 131, while with other analytical techniques, other halamines can interfere, i.e., overlapping peaks with direct UV measurements and reaction of several halamines with colorimetric reagents or phenols. While the detection limit is not quite low enough to measure NHBrCl in actual drinking water, this analytical method will benefit the scientific community by allowing further mechanistic studies on the contribution of NHBrCl to the formation of toxic disinfection by-products.

Adsorption and transformation of the anthelmintic drug niclosamide by manganese oxide.

The manganese oxide birnessite adsorbed and catalyzed the transformation of the anthelminthic drug niclosamide (NIS) into 2-chloro-4-nitroaniline (CNA) and 5-chlorosalicylic acid (CSA) at acidic pH. The adsorption of NIS was fitted using a linear isotherm for all conditions and reaction times. Linear adsorption constant Kd was 103 000 L kg-1 at pH 5.0. The rate of transformation was first order with respect to both MnO2 and NIS. At pH 5.0, the second order rate constant was 3.3 (±0.3) × 10-1 M-1 s-1. The adsorption constants and the rates of transformation decreased when pH increased from 4.0 to 5.5 because of increasing electrostatic repulsions between both negatively charged manganese oxide surface (pHzpc = 2.5) and NIS (pKa = 6.38). The presence of natural organic matter (NOM) extracted from surface water also significantly decreased the adsorption and the rates of transformation of NIS. The rate of transformation decreased by a factor of 20 in presence of 1.6 mgC L-1 even though significant amounts of NIS were adsorbed onto MnO2. The interactions between NOM and NIS were investigated by using the fluorescence quenching method and would explain that NIS adsorbed on the surface of manganese oxide was stable in presence of NOM. Thus, hydrolysis catalyzed by manganese oxide is probably not an important process compared to biodegradation and adsorption because of the presence of organic matter and pH values usually >5.5 in aquatic environment.

Organic matter interactions with natural manganese oxide and synthetic birnessite.

Redox reactions of inorganic and organic contaminants on manganese oxides have been widely studied. However, these reactions are strongly affected by the presence of natural organic matter (NOM) at the surface of the manganese oxide. Interestingly, the mechanism behind NOM adsorption onto manganese oxides remains unclear. Therefore, in this study, the adsorption kinetics and equilibrium of different NOM isolates to synthetic manganese oxide (birnessite) and natural manganese oxide (Mn sand) were investigated. Natural manganese oxide is composed of both amorphous and well-crystallised Mn phases (i.e., lithiophorite, birnessite, and cryptomelane). NOM adsorption on both manganese oxides increased with decreasing pH (from pH7 to 5), in agreement with surface complexation and ligand exchange mechanisms. The presence of calcium enhanced the rate of NOM adsorption by decreasing the electrostatic repulsion between NOM and Mn sand. Also, the adsorption was limited by the diffusion of NOM macromolecules through the Mn sand pores. At equilibrium, a preferential adsorption of high molecular weight molecules enriched in aromatic moieties was observed for both the synthetic and natural manganese oxide. Hydrophobic interactions may explain the adsorption of organic matter on manganese oxides. The formation of low molecular weight UV absorbing molecules was detected with the synthetic birnessite, suggesting oxidation and reduction processes occurring during NOM adsorption. This study provides a deep insight for both environmental and engineered systems to better understand the impact of NOM adsorption on the biogeochemical cycle of manganese.

Mind the Gap: Persistent and Mobile Organic Compounds-Water Contaminants That Slip Through.

The discharge of persistent and mobile organic chemicals (PMOCs) into the aquatic environment is a threat to the quality of our water resources. PMOCs are highly polar (mobile in water) and can pass through wastewater treatment plants, subsurface environments and potentially also drinking water treatment processes. While a few such compounds are known, we infer that their number is actually much larger. This Feature highlights the issue of PMOCs from an environmental perspective and assesses the gaps that appear to exist in terms of analysis, monitoring, water treatment and regulation. On this basis we elaborate strategies on how to narrow these gaps with the intention to better protect our water resources.

Chlorination of the ß-triketone herbicides tembotrione and sulcotrione: Kinetic and mechanistic study, transformation products identification and toxicity.

The fate of tembotrione (TBT) and sulcotrione (SCT) during chlorination was investigated in this work. Chlorination kinetics of TBT and SCT were studied by using a continuous-flow reactor in the pH range 2-12 with an excess of total chlorine. Second-order reaction was observed and rate constants of 9.69 (±0.15) × 10(3) M(-1)s(-1) for TBT and 9.48 (±0.62) × 10(3) M(-1)s(-1) for SCT were obtained at pH 7. Intrinsic rate constants for the elementary reactions of chlorine species with neutral and deprotonated forms of TBT and SCT were also calculated, leading to the conclusion that the reaction between hypochlorous acid and the deprotonated form of the pesticide is predominant at neutral pH. Several degradation products during chlorination of TBT and SCT were identified by LC-MS/MS and a reaction pathway was proposed. Chlorine initially reacted on the α-carbon of the three carbonyl functional groups. This reaction initiated the well-known haloform reaction and produced chloroform as end-product. Molar yields of 0.99 mol CHCl3/mol and 0.91 mol CHCl3/mol were obtained for TBT and SCT, respectively at pH 7. Moreover, a toxicity evaluation using Vibrio fischeri was carried out to study the toxicity pattern during TBT and SCT chlorination. An increase in toxicity was observed but it could not be clearly assigned to the identified byproducts.

Effects of charging on the chromophores of dissolved organic matter from the Rio Negro basin.

This study demonstrates that the deprotonation of dissolved organic matter (DOM) originating from a small creek characteristic for DOM-rich waters located in the Rio Negro basin can be quantified based on measurements of pH effects on its absorbance spectra. The method was ascertained by the data of Near-Edge X-Ray Absorbance Spectroscopy (NEXAFS), potentiometric titration to quantify the structural and compositional differences between the colloidal and hydrophobic fractions that contribute 91% of black-water creek DOM. Changes in the absorbance spectra of the DOM fractions caused by deprotonation quantified via numeric deconvolution which indicated the presence of six well-resolved Gaussian bands in the differential spectra. The emergence of these bands was determined to be associated with the engagement of carboxylic and phenolic functionalities and changes of inter-chromophore interactions in DOM molecules. Interpretation of the data based on the NICA-Donnan approach showed that behavior of DOM chromophores was consistent with results of potentiometric titrations. Similar trends were observed for changes of the spectral slope of the DOM absorbance spectra in the range of wavelengths 325-375 nm (DSlope325-375). The behavior of DSlope325-375 values was modeled based on the NICA-Donnan approach and correlated with potentiometrically-estimated charges attributed to the carboxylic and phenolic groups. The correlations between DSlope325-375 and charges of low- and high-affinity protonation-active groups in DOM were monotonic but not linear and had important differences between the colloidal and hydrophobic fractions.

Transformation of paracetamol into 1,4-benzoquinone by a manganese oxide bed filter.

This study investigates the transformation of paracetamol (PRC) by a granular manganeseoxide in a column bed reactor. Paracetamol was quantitatively transformed into p-benzoquinone(BZQ) for empty bed residence times (EBRT) <5 min at pH 6.0. For 5mM MOPS (3-morpholinopropane-1-sulfonic acid) and pH 7.0, the mean removal yield of PRC was 77% for initial PRC concentrations ranging from 0.1 to 50 µM. Conversion of PRC and formation of BZQ decreased when pH increased from 6 to 8. Dimer of PRC was observed at pH 7.0, which could explain the lower conversion into BZQ when pH increased. The presence of organic buffer MOPS and natural organic matter (NOM) reduced the oxidation of PRC because of competition reactions for active sites. The formation of the toxic BZQ metabolite was reduced in presence of NOM because of cross-coupling reactions between phenoxyl radicals and NOM. Results suggest that manganese oxide bed filter can be used to remove pharmaceutical compounds including phenolic moiety in their structure.

Oxidative decarboxylation of diclofenac by manganese oxide bed filter.

Diclofenac (DCF) was eliminated by fast chemical oxidation on natural manganese oxide in a column reactor. Identification of transformation by-products of DCF by HPLC-UV-MS(n) gave evidence of decarboxylation, iminoquinone formation and dimerization. The fast oxidation of DCF is also accompanied by a strong adsorption of organic carbon that was explained by the sorption of dimer products on the surface of manganese oxide. Decarboxylation and dimerization increased the hydrophobic interactions with manganese oxide and reduced the presence of potentially toxic by-products in the effluent. The rate of oxidation was first order with respect to DCF and was slowed down by the presence of organic buffer MOPS (3-morpholinopropane-1-sulfonic acid). The first order rate constant in absence of MOPS was extrapolated by considering a surface site-binding model and MOPS as a co-adsorbate. The rate constant was 0.818 min(-1) at pH 7 and 10 mM NaCl corresponding to empty bed residence time of 50 s only for 50% removal of DCF. Rate constants increased when pH decreased from pH 8.0 to 6.5 and when ionic strength increased. Manganese oxide bed filter can be considered as an alternative treatment for polishing waste water effluent or for remediation of contaminated groundwater.

Abiotic formation of methyl iodide on synthetic birnessite: a mechanistic study.

Methyl iodide is a well-known volatile halogenated organic compound that contributes to the iodine content in the troposphere, potentially resulting in damage to the ozone layer. Most methyl iodide sources derive from biological activity in oceans and soils with very few abiotic mechanisms proposed in the literature. In this study we report that synthetic manganese oxide (birnessite δ-MnO2) can catalyze the formation of methyl iodide in the presence of natural organic matter (NOM) and iodide. Methyl iodide formation was only observed at acidic pH (4-5) where iodide is oxidized to iodine and NOM is adsorbed on δ-MnO2. The effect of δ-MnO2, iodide and NOM concentrations, nature of NOM and ionic strength was investigated. High concentrations of methyl iodide were formed in experiments conducted with the model compound pyruvate. The Lewis acid property of δ-MnO2 leads to a polarization of the iodine molecule, and catalyzes the reaction with natural organic matter. As manganese oxides are strong oxidants and are ubiquitous in the environment, this mechanism could significantly contribute to the global atmospheric input of iodine.

NDMA formation by chloramination of ranitidine: kinetics and mechanism.

The kinetics of decomposition of the pharmaceutical ranitidine (a major precursor of NDMA) during chloramination was investigated and some decomposition byproducts were identified by using high performance liquid chromatography coupled with mass spectrometry (HPLC-MS). The reaction between monochloramine and ranitidine followed second order kinetics and was acid-catalyzed. Decomposition of ranitidine formed different byproducts depending on the applied monochloramine concentration. Most identified products were chlorinated and hydroxylated analogues of ranitidine. In excess of monochloramine, nucleophilic substitution between ranitidine and monochloramine led to byproducts that are critical intermediates involved in the formation of NDMA, for example, a carbocation formed from the decomposition of the methylfuran moiety of ranitidine. A complete mechanism is proposed to explain the high formation yield of NDMA from chloramination of ranitidine. These results are of great importance to understand the formation of NDMA by chloramination of tertiary amines.

Identification of effluent organic matter fractions responsible for low-pressure membrane fouling.

Anion exchange resin (AER), powder activated carbon (PAC) adsorption and ozonation treatments were applied on biologically treated wastewater effluent with the objective to modify the effluent organic matter (EfOM) matrix. Both AER and PAC led to significant total organic carbon (TOC) removal, while the TOC remained nearly constant after ozonation. Liquid Chromatography-Organic Carbon Detection (LC-OCD) analysis showed that the AER treatment preferentially removed high and intermediate molecular weight (MW) humic-like structures while PAC removed low MW compounds. Only a small reduction of the high MW colloids (i.e. biopolymers) was observed for AER and PAC treatments. Ozonation induced a large reduction of the biopolymers and an important increase of the low MW humic substances (i.e. building blocks). Single-cycle microfiltration (MF) and ultrafiltration (UF) tests were conducted using commercially available hollow fibres at a constant flux. After reconcentration to their original organic carbon content, the EfOM matrix modified by AER and PAC treatments exhibited higher UF membrane fouling compared to untreated effluent; result that correlated with the higher concentration of biopolymers. On the contrary, ozonation which induced a significant degradation of the biopolymers led to a minor flux reduction for both UF and MF filtration tests. Based on a single filtration, results indicate that biopolymers play a major role in low pressure membrane fouling and that intermediate and low MW compounds have minor impact. Thus, this approach has shown to be a valid methodology to identify the foulant fractions of EfOM.