Can we predict uranium bioavailability based on soil parameters? Part 2: soil solution uranium concentration is not a good bioavailability index

The present study aimed to quantify the influence of soil parameters on uranium uptake by ryegrass. Ryegrass was established on eighteen distinct soils, spiked with (238)U. Uranium soil-to-plant transfer factors (TF) ranged from 0.0003 to 0.0340kgkg(-1). There was no significant relation between the U soil-to-plant transfer (or total U uptake or flux) and the uranium concentration in the soil solution or any other soil factor measured, nor with the U recovered following selective soil extractions. Multiple linear regression analysis resulted in a significant though complex model explaining up to 99% of variation in TF. The influence of uranium speciation on uranium uptake observed was featured: UO(2)(+2), uranyl carbonate complexes and UO(2)PO(4)(-) seem the U species being preferentially taken up by the roots and transferred to the shoots. Improved correlations were obtained when relating the uranium TF with the summed soil solution concentrations of mentioned uranium species.     

Phosphate-induced metal immobilization in a contaminated site

To assess the efficiency of P-induced metal immobilization in soils, a pilot-scale field experiment was conducted at a metal contaminated site located in central Florida. Phosphate was applied at a P/Pb molar ratio of 4.0 with three treatments: 100% of P from H3PO4, 50% of P from H3PO4+ 50% of P from Ca(H2PO4)2, and 50% of P from H3PO4+5% phosphate rock in the soil. Approximately 1 year after P application, soil and plant samples were collected to determine mobility and bioavailability of selected metals (Pb, Zn, and Cu) using sequential extraction procedure and mineralogical characterization using X-ray diffraction (XRD) and scanning electron microscope-energy dispersive X-ray (SEM-EDX) analysis. Phosphorus distribution and soil pH effects were also evaluated. Phosphate was more effective in transforming soil Pb (to 53%) from the non-residual to the residual phase than soil Zn (to 15%) and soil Cu (to 13%). This was because Pb was immobilized by P via formation of an insoluble pyromorphite-like mineral in the surface and subsurface of the soil, whereas no phosphate mineral Zn or Cu was identified. While P amendment enhanced metal uptake in the roots of St. Augustine grass (Stenotaphrum secundatum), it significantly reduced metal translocation from root to shoot, especially Pb via formation of a pyromorphite-like mineral on the membrane surface of the root. A mixture of H3PO4 and phosphate rock was effective in metal immobilization, with less soil pH reduction and less soluble P. Although H3PO4 was effective in immobilizing Pb, its use should be limited to minimize soil pH reduction and potential eutrophication risk.     

Mechanisms of lead, copper, and zinc retention by phosphate rock

The solid-liquid interface reaction between phosphate rock (PR) and metals (Pb, Cu, and Zn) was studied. Phosphate rock has the highest affinity for Pb, followed by Cu and Zn, with sorption capacities of 138, 114, and 83.2 mmol/kg PR, respectively. In the Pb-Cu-Zn ternary system, competitive metal sorption occurred with sorption capacity reduction of 15.2%, 48.3%, and 75.6% for Pb, Cu, and Zn, respectively compared to the mono-metal systems. A fractional factorial design showed the interfering effect in the order of Pb>Cu>Zn. Desorption of Cu and Zn was sensitive to pH change, increasing with pH decline, whereas Pb desorption was decreased with a strongly acidic TCLP extracting solution (pH = 2.93). The greatest stability of Pb retention by PR can be attributed to the formation of insoluble fluoropyromorphite [Pb(10)(PO(4))(6)F(2)], which was primarily responsible for Pb immobilization (up to 78.3%), with less contribution from the surface adsorption or complexation (21.7%), compared to 74.5% for Cu and 95.7% for Zn. Solution pH reduction during metal retention and flow calorimetry analysis both supported the hypothesis of retention of Pb, Cu, and Zn by surface adsorption or complexation. Flow calorimetry indicated that Pb and Cu adsorption onto PR was exothermic, while Zn sorption was endothermic. Our research demonstrated that PR can effectively remove Pb from solutions, even in the presence of other heavy metals (e.g. Cu, Zn).     

Phosphate-induced lead immobilization from different lead minerals in soils under varying pH conditions

This study investigated phosphate-induced lead immobilization from different Pb minerals in soils under varying pHs. Four soils were used, including one Pb-contaminated soil (NC-Soil) and three soils spiked with litharge (PbO), cerrusite (PbCO3), or anglesite (PbSO4), referred to as PbO-soil, PbCO3-soil, and PbSO4-soil, respectively. The soils were equilibrated with KCl and Ca(H2PO4)(2).H2O under pH of 3-7. At low pH (3 and 5), Pb solubility followed PbO-soil>PbCO3-soil>PbSO4-soil; while at pH=7, it was PbSO4-soil>PbO-soil>PbCO3-soil. Phosphate decreased Pb dissolution time from >180 to <60 min and reduced soluble Pb by 67-100%. This was mostly via transformation of Pb minerals into chloropyromorphite [Pb(5)(PO(4))(3)Cl]. Our results indicated that P addition can effectively transform various Pb minerals into insoluble chloropyromorphite in soils. This transformation was more significant at acidic condition (e.g., pH相似文献   

Adsorption of uranium(VI) on amorphous ferric oxyhydroxide at high concentrations of dissolved carbon(IV) and sulfur(VI)

Amorphous ferric oxyhydroxide is being used to treat groundwater contaminated with uranium(VI); the compound also has potential for use as a component in in situ chemical barriers. To quantitatively evaluate its effectiveness in such applications, adsorption of uranium(VI) onto amorphous ferric oxyhydroxide was experimentally investigated under a wide range of uranium(VI) (8.40·10⁻⁷−2.10·10⁻³ mol L⁻¹; 0.2–500 mg L⁻¹), sulfur(VI) (0–0.07 mol L⁻¹; 0–2240 mg L⁻¹) and carbon(IV) (0–0.0195 mol L⁻¹; 0–234 mg L⁻¹) concentrations, and pH-values (4–9.6). The adsorption behavior of uranium(VI) (uranyl ion and its complexes) is similar to that of other cations; it exhibits a sharp rise in the extent of adsorption with increasing pH. Interactions among uranyl complexes and surface sites are interpreted using a site complexation model. Although the model does not incorporate electrostatics and includes only a single type of adsorption site, it provides a reasonable match to measured adsorption and proton exchange data. The simplicity of the model and the resulting reduction in computational demand allow its efficient incorporation into coupled reaction-transport models.     

Estrogenic activity removal of 17beta-estradiol by ozonation and identification of by-products

This work investigated the degradation of a natural estrogen (17beta-estradiol) and the removal of estrogenic activity by the ozonation process in three different pHs (3, 7 and 11). A recombinant yeast assay (YES assay) was employed to determine estrogenic activity of the ozonized samples and of the by-products formed during the ozonation. Ozonation was very efficient for the removal of 17beta-estradiol in aqueous solutions. High removals (>99%) were achieved with low ozone dosages in the three different pHs. Several by-products were formed during the ozonation of 17beta-estradiol. However, only a few compounds could be identified and confirmed. Different by-products are formed at different pHs, which is probably due to different chemical pathways and different oxidants (O(3) and OH radical). The by-products formed at pH 11 were 10epsilon-17beta-dihydroxy-1, 4-estradieno-3-one (DEO) and 2-hydroxyestradiol, which were not formed in pH 3. Only testosterone could be observed in pH 3, whereas at pH 7 all three by-products were found. At pH 7 and 11 the applied ozone dosages were not enough to remove all the estrogenicity from samples, even though the 17beta-estradiol residual concentration for these two pHs was lower than at pH 3. Higher estrogenicity was detected at pH 11. An explanation to this fact may be that oxidation via OH radical forms more by-products with estrogenic activity. Probably, the formation of 2-hydroxyestradiol at pHs 7 and 11 is contributing to the residual estrogenicity of samples ozonized at these pHs. In this work, complete removal of estrogenic activity was only obtained at pH 3.     

Retention of Cd, Zn and Co on hydroxyapatite filters

This study qualifies and quantifies the immobilization of Cd, Zn and Co, (used as models of bivalent metal ions due to their relevant toxicity) in filters of synthetic hydroxyapatite (HAP) [Ca5(PO4)3OH]. They were flushed with solutions containing Cd (1 x 10(-5)M), Zn and Co (1 x 10(-4)M) at constant pH (8.6) and ionic strength (0.01 M). The concentration of these metal ions in the outlet was measured by ICP-OEM spectroscopy. The software PHREEQC (version 2.4.2) was used to model sorption process and the potential effect of salinity (KCl), pH, alkalinity (NaHCO3) and hardness (CaCl2) over the efficiency of the treatment. Results showed an excellent retention capacity of HAP for Cd, Zn and Co. Sorption data were successfully described considering a mix model of surface complexation onto phosphate surface groups, ionic exchange in surface calcium sites and the precipitation of ZnO. Co exchange and surface complexation constants (Kex and Kc) were taken from previous experiments, while KexCd=0.32 and KcCd=0.63 were estimated from our modeling results. Predictive values of metal ion sorption show that: (a) an increase in hardness does not play a significant role in the retention capacity of these metals on HAP; (b) an increase in alkalinity promotes the precipitation of MeCO3 which could alter the hydrodynamic of the column; (c) a decrease in pH and an increase in salinity inhibit ZnO precipitation enhancing Zn and Cd adsorption and decreasing Co retention on HAP.     

Modeling the competitive effect of phosphate, sulfate, silicate, and tungstate anions on the adsorption of molybdate onto goethite

The mobility of Mo in soils and sediments depends on several factors including soil mineralogy and the presence of other oxyanions that compete with Mo for the adsorbent's retention sites. Batch experiments addressing Mo adsorption onto goethite were conducted with phosphate, sulfate, silicate, and tungstate as competing anions in order to produce competitive two anions adsorption envelopes, as well as competitive two anions adsorption isotherms. Tungstate and phosphate appear to be the strongest competitors of Mo for the adsorption sites of goethite, whereas little competitive effects were observed in the case of silicate and sulfate. Mo adsorption isotherm from a phosphate solution was similar to the one from a tungstate solution. The charge distribution multi-site complexation (CD-MUSIC) model was used to predict competitive adsorption between MoO(4)(2-) and other anions (i.e., phosphate, sulfate, silicate and tungstate) using model parameters obtained from the fitting of single ion adsorption envelopes. CD-MUSIC results strongly agree with the experimental adsorption envelopes of molybdate over the pH range from 3.5 to 10. Furthermore, CD-MUSIC prediction of the molybdate adsorption isotherm show a satisfactory fit of the experimental results. Modeling results suggest that the diprotonated monodentate complexes, FeOW(OH)(5)(-0.5) and FeOMo(OH)(5)(-0.5), were respectively the dominant complexes of adsorbed W and Mo on goethite 110 faces at low pH. The model suggests that Mo and W are retained mainly by the formation of monodentate complexes on the goethite surface. Our results indicate that surface complexation modeling may have applications in predicting competitive adsorption in more complex systems containing multiple competing ions.     

Can we predict uranium bioavailability based on soil parameters? Part 1: effect of soil parameters on soil solution uranium concentration

Present study aims to quantify the influence of soil parameters on soil solution uranium concentration for (238)U spiked soils. Eighteen soils collected under pasture were selected such that they covered a wide range for those parameters hypothesised as being potentially important in determining U sorption. Maximum soil solution uranium concentrations were observed at alkaline pH, high inorganic carbon content and low cation exchange capacity, organic matter content, clay content, amorphous Fe and phosphate levels. Except for the significant correlation between the solid-liquid distribution coefficients (K(d), L kg(-1)) and the organic matter content (R(2)=0.70) and amorphous Fe content (R(2)=0.63), there was no single soil parameter significantly explaining the soil solution uranium concentration (which varied 100-fold). Above pH=6, log(K(d)) was linearly related with pH [log(K(d))=-1.18 pH+10.8, R(2)=0.65]. Multiple linear regression analysis did result in improved predictions of the soil solution uranium concentration but the model was complex.     

Humic acid complexation of basic and neutral polycyclic aromatic compounds

Complexation by humic acid (HA) of basic (quinoline) and neutral (naphthalene) polycyclic aromatic compounds (PACs) was compared using fluorescence spectroscopy and equilibrium dialysis (ED). These compounds sorb to HA via cation exchange and hydrophobic interactions, respectively. Ionization of quinoline strongly affects its sorption to HA; maximum sorption is observed at pH close to logKb (4.92), and competition with H+ and electrolyte cation (Li+) is evident. Spectroscopic experiments indicate that quinolinium (QH+) cation fluorescence is quenched via a static mechanism (i.e., a dark complex is formed) when the protonated form is adsorbed via ion exchange to HA. The extent of sorption, calculated from fluorescence data using the Stern-Volmer equation, was compared to independent ED measurements. Although both methods indicated the same trends with solution chemistry, fluorescence quenching data suggested more extensive complexation than that measured using ED. In contrast to ionizable PACs, studied here and previously, interaction of naphthalene with HA is unaffected by changes in solution conditions (pH, ionic strength).     

Preparation of acidic and alkaline macrocapsules for pH control

A series of experiments was performed to prepare acidic macroencapsulated buffers composed of 20% Ca(H2PO4)(2) and 80% Eudragit S 100 polymer and alkaline macrocapsules composed of 65% K2HPO4 and 35% Eudragit E PO polymer (the powdered form of Eudragit E 100). Eudragit S 100 was shown to be soluble at a pH greater than 7.0, while Eudragit E 100 was soluble at a pH less than 7.0. Both polymers did not impart significant biochemical oxygen demand. The Eudragit E PO polymer solution showed low toxicity (EC50=91%) based on the Microtox Acute Toxicity Test compared to the 0.1mM background phosphate buffer solution (EC50=100%) while the Eudragit S 100 polymer solution showed higher toxicity (EC50=53%). Batch tests showed that the acidic macrocapsules reduced the pH of a 0.1mM phosphate solution from 11 to neutral, while the alkaline macrocapsules increased the pH of a 0.1mM phosphate solution from 3 to neutral. The macrocapsules could potentially be used as an in situ proportional pH controller for groundwater remediation.     

Mechanisms of phosphorus solubilisation in a limed soil as a function of pH

Phosphorus (P) quantity-intensity relationships are central to the solubility and release of P from soil to water. Relationships between P extractable by 0.5 M NaHCO extractable P (Olsen P; quantity, Q) and P extractable by 0.01 M CaCl(2) (CaCl(2)-P; possible predictor of soil solution or drainage water P; intensity, I) are curvilinear: above a certain Olsen P concentration, CaCl(2)-P becomes much more soluble than when below it. Aluminium-, Fe- and Ca-P forms (extractable by Olsen's reagent) are thought to control P solubility. Thus, our objectives were to identify P forms in equilibrium with CaCl(2)-P via solubility equilibrium experiments, and the behaviour of CaCl(2)-P in relation to Al, Fe and Ca associated P, determined with 31P high power decoupling magic angle spinning nuclear magnetic resonance spectroscopy (31P HPDec/MAS NMR). Results indicated that two Q-I relationships occurred, one for soils above pH 5.8, and the other for soils below pH 5.8. Above pH 5.8, soils were saturated with respect to hydroxyapatite (Ca(5)(PO(4))(3)OH) and undersaturated with respect to beta-tricalcium phosphate (beta-Ca(3)(PO(4))(2)), while log ion-activity products showed that all soils and pHs were either saturated or in equilibrium with variscite (AlPO(4).2H(2)O) or its amorphous analogue. Using 31P HPDec/MAS NMR, Ca-P was best correlated with CaCl(2)-P in soils above pH 5.8, and with Al-P in soils below this pH. This study demonstrates the value of solid-state NMR in conjunction with wet chemical techniques for the study of labile P and P loss from pasture soils with a wide range of managements.     

原位水解生成的羟基氧化铁凝聚吸附除磷效能与机制

将不同摩尔比Fe3＋与OH-（[Fe3＋]：[OH-]=1：0、1：1、1：2和1：3）反应获得原位水解生成的羟基氧化铁（insituFeOxHy）,研究了具有不同水解程度的羟基氧化铁对凝聚吸附除磷效能与机制。研究显示,InsituFeOxHy对磷的去除率随铁投量增大而升高,且均在中性pH范围内具有最佳除磷效果;在相同铁投量条件下,磷去除率随着[OH-]：[Fe3＋]的升高而降低;当体系碱度较低时（pH〈6）,引入OH-可促进Fe3＋水解而提高除磷效果。4种羟基氧化铁均可在15s内可快速吸附磷,且吸附过程符合准二级动力学模型;Freundlich模型均可很好地描述磷在4种羟基氧化铁表面的吸附行为。磷酸盐吸附后,InsituFeOxHy表面Zeta电位明显降低,且[Fe3＋]：[OH-]为1：0的羟基氧化铁降低最为显著。结合MINITEQ计算软件磷酸盐、铁盐形态分析结果显示,对于碱度较低的体系,通过投加一定量OH-可促进Fe3＋水解,进而使得其更易与水中H2PO4-与HPO2 4-结合,生成具有多核羟基的磷酸铁络合物,进而提高除磷效果。     

Sorption behavior of U(VI) on phyllite: experiments and modeling

The sorption of U(VI) onto low-grade metamorphic rock phyllite was modeled with the diffuse double layer model (DDLM) using the primary mineralogical constituents of phyllite, i.e. quartz, chlorite, muscovite, and albite, as input components, and as additional component, the poorly ordered Fe oxide hydroxide mineral, ferrihydrite. Ferrihydrite forms during the batch sorption experiment as a weathering product of chlorite. In this process, Fe(II), leached from the chlorite, oxidizes to Fe(III), hydrolyses and precipitates as ferrihydrite. The formation of ferrihydrite during the batch sorption experiment was identified by M?ssbauer spectroscopy, showing a 2.8% increase of Fe(III) in the phyllite powder. The ferrihydrite was present as Fe nanoparticles or agglomerates with diameters ranging from 6 to 25 nm, with indications for even smaller particles. These Fe colloids were detected in centrifugation experiments of a ground phyllite suspension using various centrifugal forces. The basis for the successful interpretation of the experimental sorption data of uranyl(VI) on phyllite were: (1) the determination of surface complex formation constants of uranyl with quartz, chlorite, muscovite, albite, and ferrihydrite in individual batch sorption experiments, (2) the determination of surface acidity constants of quartz, chlorite, muscovite, and albite obtained from separate acid-base titration, (3) the determination of surface site densities of quartz, chlorite, muscovite, and albite evaluated independently of each other with adsorption isotherms, and (4) the quantification of the secondary phase ferrihydrite, which formed during the batch sorption experiments with phyllite. The surface complex formation constants and the protolysis constants were optimized by using the experimentally obtained data sets and the computer code FITEQL. Surface site densities were evaluated from adsorption isotherms at pH 6.5. The uranyl(VI) sorption onto phyllite was accurately modeled with these newly determined constants and parameters of the main mineralogical constituents of phyllite and the secondary mineralization phase ferrihydrite. The modeling indicated that uranyl sorption to ferrihydrite clearly dominates uranyl sorption, showing the great importance of secondary iron phases for sorption studies.     

In situ immobilization of Cu(II) in soils using a new class of iron phosphate nanoparticles

This study tested the feasibility of using a new class of iron phosphate (vivianite) nanoparticles synthesized using sodium carboxymethyl cellulose (NaCMC) as a stabilizer for in situ immobilization of Cu(II) in soils. Transmission electron microscopy measurements demonstrated that the particle size was about 8.4+/-2.9 nm. Batch tests showed that nano-sized vivianite particles can effectively reduce the leachability and in vitro bioaccessibility of Cu(II) in three representative soils (calcareous, neutral, and acidic) at the low doses of 0.61 and 3.01 mg PO(4) g(-1) soil. The Cu leachability was evaluated by the toxicity characteristic leaching procedure and in vitro bioaccessibility was evaluated by the physiological based extraction test. In the case of soil amendment with nanoparticles in 3.01 mg PO(4) g(-1) soil, Cu leachability reduced 63-87% and Cu concentrations in TCLP extract decreased from 1.74-13.33 mg l(-1) to 0.23-2.55 mg l(-1) after those soils were amended for 56 d. Meanwhile, the bioaccessibility of Cu was reduced by 54-69%. Sequential extraction procedures showed the significant decrease of water soluble/exchangeable Cu(II) and carbonate bound fractions and concomitant increase of Cu residual fraction after the soils were amended with the nanoparticles, suggesting that the formation of copper phosphate minerals through precipitation and adsorption was probably responsible for the decrease of Cu availability in soils. Visual MINTEQ modeling further revealed that Cu(3)(PO(4))(2) and Cu(5)(PO(4))(3)OH were formed in the vivianite-solid Cu(II) system, resulting in the decreased solubility of the Cu(II) in the acidic pH range.     

The role of Mn oxide doping in phosphate removal by Al-based bimetal oxides: adsorption behaviors and mechanisms

This study investigated the behaviors and mechanisms of phosphate adsorbed onto manganese (Mn) oxide-doped aluminum (Al) oxide (MODAO). The isotherm results demonstrated that the maximum amount of phosphorus (P) adsorbed onto MODAO was 59.8 mg/g at T?=?298 K (pH 6.0). This value was nearly twice the amount of singular AlOOH and could increase with rising temperatures. The kinetic results illustrated that most of the P was adsorbed onto MODAO within 5 h, which was shorter than the equilibrium time of phosphate adsorption onto AlOOH. The Elovich model effectively described the adsorption kinetic data of MODAO because of its heterogeneous surface. The optimal solution pH for phosphate removal was approximately 5.0 because of electrostatic interaction effects. Meanwhile, the decrease in P uptake with increasing ion strength suggested that phosphate adsorption occurred through an outer-sphere complex. Phosphates would compete for adsorption sites on the surface of MODAO in the presence of fluoride ion or sulfate. In addition, the spectroscopic analysis results of Fourier transform infrared spectroscopy and X-ray photoemission spectroscopy indicated that removal mechanisms of phosphate primarily include adhesion to surface hydroxyl groups and ligand exchange.     

Lead phosphate formation in soils

Pyromorphite (Pb5(PO4)3Cl) is one of the most thermodynamically stable lead minerals under the geochemical conditions prevailing in the surface environment. It is therefore expected to form in soils contaminated with Pb if sufficient phosphorus is available. Pyromorphite has previously been identified in mine-waste and industrially contaminated soils but has not previously been identified in urban soils. This paper reports on the presence of a Pb phosphate in urban and roadside soils. This phase has formed in the soil as a weathering product of Pb-bearing grains. Quantitative EDX analyses indicated that the Pb phosphate phase is pyromorphite with Ca frequently substituting for Pb between 21-31 atomic percent. However, positive identification of this phase by XRD was hindered by the deviation from pure end-member and possibly also by the poorly crystalline nature of the phase. Pyromorphite accounted for less than 2% of the total Pb in these soils. However, phosphate amendments to the soil could induce further formation of pyromorphite. As pyromorphite is a highly insoluble mineral, this may be effective in reducing the bioavailability of Pb in urban soils.     

Phosphate recovery from greenhouse wastewater

A study was conducted to investigate the suitability of phosphate recovery from greenhouse wastewaters by using precipitation/crystallization process. More than 90% of the phosphate could be removed from the greenhouse wastewater. Various calcium phosphate salts were obtained in the process; hydroxyapatite [Ca5(PO4)3OH] could be the main product from the precipitates. Phosphate removal was affected by the presence of magnesium ion in wastewaters. An increase of magnesium concentrations in wastewaters decreased phosphate removal rates. The chemical contents of precipitates in terms of calcium, magnesium and phosphate were affected by calcium to magnesium molar ratio. Higher calcium contents were obtained at wastewaters with high calcium to magnesium molar ratio. An addition of magnesium did not affect the potassium contents in the precipitates. K-struvite, MgKPO4 x 6H2O, was not the major product in the precipitate, even with addition of a large quantity of magnesium.     

Formation of volatile halogenated by-products during chlorination of isoproturon aqueous solutions

The present paper deals with the formation of volatile halogenated by-products (POX) during the chlorination (160 mg/l) of aqueous solutions of the herbicide isoproturon (40 mg/l). Chlorination reactions have been carried out over 48 h, at ambient temperature, at two pHs (6 and 9) and in the presence or not of bromide ions (80 mg/l). The main results obtained have been as follows: (1) in the presence of bromide, isoproturon degradation is rather fast and it results affected by pH, complete isoproturon degradation is achieved within 1 and 15 min at pH 6 and 9, respectively; (2) in the absence of bromide herbicide degradation is slow (complete degradation is achieved within 180 min) and it is not affected by pH; (3) at pH 6, regardless of the presence of bromide, the maximum amount of POX formed is low (approximately 15 micromol X-/l) and remains constant during the reaction; (4) at pH 9 the amount of POX formed is far greater and continuously increases during the reaction, reaching a value of about 110 micromol X-/l after 48 h; (5) two different groups of by-products have been identified by solid phase micro extraction (SPME)-gas chromatography (GC)-mass spectrometry (MS) for the reactions carried out with or without bromide; among them, aliphatic as well as aromatic by-products containing chlorine, bromine or both halogens are present even though the most abundant are halogenated-methane derivatives (haloforms); pH value affects the amount of these by-products but does not modify their chemical nature.