Caracterisation de la matiere organique dissoute presente dans les eaux en cours d'affinage: Influence des traitements appliques et du climat

Dissolved organic matter in treated surface waters (clarified, possibly ozonized, then GAC-filtered, Fig. 1), is fractionated by ultrafiltration into five molecular classes with MW < 300, 300–1000, 1000–5000, 5000–10000 or > 10,000. Dissolved organic carbon (DOC), oxidizability by KMnO₄ in hot alkaline medium and u.v. absorbances at 240, 254, 280, 300 nm are measured. Fourteen series of samples, distributed on an annual biological cycle are analysed (Figs 2 and 3); multivariate statistical analyses are performed.By PCA (principal component analysis), variations in water supplying the activated carbon units appear to depend for 47% on ozonation and temperature; but river flow rate and quantity of flocculant added are no longer responsible for such variations (Fig. 4). Three groups of water appear (Fig. 5), according to the applied ozone level (zero, medium, high); among the medium ozonized waters, the cold ones differ from temperate ones.Ozonation diminishes molecular size of compounds (Table 1): three major classes with MW < 5000 are present in non- or medium-ozonized waters, but only two, with MW < 1000, remain in highly ozonized waters. This treatment destroys MW > 10,000 and even 1000–5000 ones and yields MW < 300 products; it also minimizes u.v. absorbances and oxidizability. Seasonal variations occur in DOC content of medium ozonized waters, with maxima values in winter or spring and minima in summer or autumn (Fig. 6): occurrence of MW < 300 compounds follows that of DOC, but the presence of 5000–10000 ones is minimal in winter.Quality of GAC-filtered waters varies by 19% with temperature (Fig. 7); ozonation effects are minimized: only previously highly ozonized waters distinguished themselves from the others (Fig. 8). Waters, non or medium ozonized before GAC-filtration, are divided into cold, temperate and warm waters. One, two or three major classes of compounds with MW < 5000, remain in GAC-filtered waters, according to the ozone level applied previously. This filtration reduces DOC by 17%, decreases u.v. absorbances and oxidizability and gives water with the same 0.30 mg O₂ mg⁻¹ C ratio (Table 2): MW 1000–5000 class is much less oxidizable after ozonation-GAC filtration but, on the other hand, MW < 300 class appears rather less oxidizable without ozonation before biological filtration. DOC content in effluent follows that in influent (Fig. 9), but variations are less marked. Total efficiency of the filtration increases with temperature, but behaviour of compounds differs from one class to another: MW 300–1000 and 5000–10000 classes are the most affected; MW 1000–5000 is not really modified. Elimination of MW < 300 or 5000–10000 compounds depends on temperature and may be due to biological phenomena, a but that of 1000–5000 and > 10,000 classes, independent of this parameter, may be related to adsorption mechanisms.     

Dissolved organic matter and the dissipation of chlorine in estuarine water and seawater

Dissolved organic matter in estuarine water and seawater collected in the summers of 1980 and 1981 in the James River, Virginia and the mouth of Chesapeake Bay were separated into fractions according to their nominal molecular weights (NMW) by ultrafiltration. Estuarine waters contained higher concentrations of dissolved organic carbon (DOC). Among the fractions, between 66–89% of the DOC was found in the fraction with NMW below 10,000. Estuarine waters also had higher chlorine demands. At a dose of 5 mg l⁻¹, in 23 h, about 90% of the added chlorine disappeared in estuarine waters, whereas, in seawater, only 60–75% of the chlorine had dissipated. At least two-thirds of the chlorine demand occurred in the first 5 h. About 10–30% of the chlorine demand may be attributed to the fraction with NMW above 10,000. The remaining chlorine demand was distributed almost equally between the fractions with ranges of NMW of 1000–10,000 and below 1000. If reactivity is measured in terms of organic chlorine demand (ΔClo) per unit weight of DOC, the fractions with lower NMW (< 1000 and 1000–10,000) always had a higher reactivity towards chlorine. Between these two fractions, the one with NMW between 1000 and 10,000 exhibited higher reactivity more frequently. The highest reactivity found was 1.4 mg ΔClo mg⁻¹ DOC.     

Reverse osmosis membrane rejection for ersatz space mission wastewaters

Adequate rejection of a variety of inorganic and organic compounds is necessary if reverse osmosis (RO) and nanofiltration (NF) membranes are to be used for space mission wastewater reuse. Three ersatz space mission wastewaters defined by NASA having different pH (2.6-8.9), conductivities (3980-12,640 microS/cm), and amounts of organic compounds (50-2400 mg/L as carbon) were tested to determine the membrane flux and the solute rejection for five RO and two NF membranes that are commercially available. The results show that the rejection of ions depends upon the solution pH which influences electrostatic repulsion. However, the rejection of dissolved organic carbon (DOC) depends upon the composition of the wastewater. The DOC rejection (80-95%) was the highest for the wastewater containing dextran (molecular weight 15-20 k) compared with the other ersatz wastewaters having detergent and urea as the major carbon sources (31-83%). The wastewater having the greatest conductivity (12,640 microS/cm) and DOC (2400 mg/L) showed a greater flux decline (71-96%) than the other ersatz wastewaters (37-82%) having lower conductivities (3980-6980 microS/cm) and DOC (50-660 mg/L) for the RO and NF membranes. The ratio of solute radius (r(i,s)) to effective membrane pore radius (r(p)) was employed to compare ion rejection. For ionic compounds, the rejection is higher than 70% when the r(i,s)/r(p) ratio is greater than 0.5 for both the RO and NF membranes with all wastewaters.     

Simultaneous determination of dissolved organic carbon and phosphorus in waters using flame ionization and photometric detectors

An apparatus with flame ionization and photometric detectors was assembled for the simultaneous determination of dissolved organic carbon (DOC) and phosphorus (DP) in waters. The optimum operating conditions were described. The long-term precision (relative standard deviation) is 5.8% for DOC and 5.2% for DP. The detection limits are 0.09 μg ml⁻¹ for DOC and 0.03 μg ml⁻¹ for DP. The responses for various DOC and DP compounds agreed almost with those obtained by combustion-infrared and persulfate digestion-colorimetric methods, respectively. DOC and DP in several water samples were determined by this method and other methods, and the results obtained by those methods were discussed.     

Dissolved organic carbon during a spring diatom bloom in Lake Mossø, Denmark

The concentration and molecular weight composition of dissolved organic carbon (DOC) was measured during a spring diatom bloom in eutrophic Lake Mossø, Denmark. The concentration fluctuated between 5.2 and 7.2 mg Cl⁻¹. No relationship between DOC concentrations and phytoplankton biomass or primary production was observed. Diurnal variations of a magnitude similar to variations on a weekly basis were found. A decomposition experiment showed the labile fraction to be of a magnitude similar to the diurnal variation, i.e. 0.7 mg Cl⁻¹. By gel filtration (Sephadex G-50 and G-15) it was shown that 70–90% of the DOC pool was <700 daltons and probably <300 daltons. The labile fraction consisted of low molecular weight compounds.     

Ozonation effect on natural organic matter adsorption and biodegradation - Application to a membrane bioreactor containing activated carbon for drinking water production

More stringent legislation on dissolved organic matter (DOM) urges the drinking water industry to improve in DOM removal, especially when applied to water with high dissolved organic carbon (DOC) contents and low turbidity. To improve conventional processes currently used in drinking water treatment plants (DWTPs), the performances of a hybrid membrane bioreactor containing fluidised activated carbon were investigated at the DWTP of Rennes. Preliminary results showed that the residual DOC was the major part of the non-biodegradable fraction. In order to increase the global efficiency, an upstream oxidation step was added to the process. Ozone was chosen to break large molecules and increase their biodegradability. The first step consisted of carrying out lab-scale experiments in order to optimise the necessary ozone dose by measuring the process yield, in terms of biodegradable dissolved organic carbon (BDOC). Secondly, activated carbon adsorption of the DOC present in ozonated water was quantified. The whole process was tested in a pilot unit under field conditions at the DWTP of Rennes (France). Lab-scale experiments confirmed that ozonation increases the BDOC fraction, reduces the aromaticity of the DOC and produces small size organic compounds. Adsorption tests led to the conclusion that activated carbon unexpectedly removes BDOC first. Finally, the pilot unit results revealed an additional BDOC removal (from 0.10 to 0.15 mg L⁻¹) of dissolved organic carbon from the raw water considered.     

Impact of coagulation and adsorption on DOC fractions of secondary effluent and resulting fouling behaviour in ultrafiltration

Membrane fouling by macromolecular dissolved organic compounds is still a fundamental drawback in low-pressure membrane filtration of secondary effluent. In this study, pre-treatment of secondary effluent by coagulation and/or adsorption was investigated in terms of removal of different dissolved organic carbon (DOC) fractions, especially macromolecular substances. DOC fractionation has been characterised by size exclusion chromatography. Adsorption tests using four commercially available activated carbons yielded a removal of small as well as larger organic compounds, revealing differences in the affinity towards macromolecules depending on the type of applied activated carbon. By contrast, coagulation removed predominantly larger molecules, i.e., biopolymers and humic substances. In terms of DOC reduction, the coagulant ferric chloride was superior to aluminium chloride. A combination of coagulation and adsorption resulted in the addition of individual removal efficiencies, suggesting that different fractions of organic compounds were involved in each of the processes. After removal of macromolecular organic compounds either by coagulation or by adsorption, a significant reduction of membrane fouling was observed in tests using two different types of ultrafiltration flat-sheet membranes in 20-h cross-flow filtration tests.     

Umweltrelevanz von natürlichen polyzyklischen aromatischen Kohlenwasserstoffen aus Steinkohlen – eine Übersicht

Releases of PAHs from hard (bituminous) coals into the environment have not been widely recognized. In hard coals, some hundreds of mg/kg of EPA-PAH and some thousands of mg/kg total PAH concentrations can occur and the concentrations vary significantly depending on hard coal maturity and origin. Highest PAH concentrations are present in the maturity range of the so-called “oil-window” (high volatile bituminous coal rank), however there is no general correlation. Only coals from the same origin can be compared. Naphthalene and phenanthrene as well as their alkylated derivatives are the dominant single polyaromatic compounds in hard coals. The PAH distribution patterns change with increasing rank from the low molecular weight, mostly alkylated compounds with linkage to the original organic material to the predominantly higher molecular weight parent compounds. The former are characteristic for coals of petrogenic origin and the latter for those of pyrogenic origin. Currently, PAHs from coals and from crude oil and its products cannot be distinguished by use of common simple diagnostic parameters. In countries holding large coal basins and associated mining activity, coal particles can comprise 10–15?% of the sediment or soil. Until today, little has been known about desorption and bioavailability of native PAH from coals. Literature studies so far show slow and very slow desorption and hardly any or no effects on organisms. This shows that there is a lack of systematic studies using heterogeneous coals from different countries. Bioavailability of PAH from coal particles in soils and sediments has an impact on risk assessment and should be included in sediment and soil standards.     

A comparison of the characteristics of soluble organic nitrogen in untreated and activated sludge treated wastewaters

Soluble organic materials containing nitrogen (SON) are present in effluents from activated sludge treatment of domestic wastewater, but little is known about the sources and characteristics of these materials. The objective of this research was to evaluate the characteristics of SON in untreated wastewaters and activated sludge effluents. Characterization techniques used included low microbial seed biodegradability, molecular weight distribution using gel filtration chromatography, removal by activated carbon and ion exchange, and analysis for free and combined amino acids. Activated sludge effluent SON was more refractory (40–50%; first-order decay rates for the remainder were about 0.014 day⁻¹ than SON in untreated wastewater (18–38%; decay rates for the remainder were 0.08–0.16 day⁻¹). SON produced biologically during treatment had decay rates (about 0.028 day⁻¹) similar to SON in municipal activated sludge effluents, and was from 20 to 100% refractory. Less than 10% of the SON in municipal activated sludge effluent consisted of free or combined amino acids. Approximately 15–30% appeared to nucleic acid degradation products. Fifty to 60% of the SON and SCOD had apparent molecular weights of less than 1800. Apparent molecular weight distributions of treated and untreated wastewaters were similar; however, the excess SON produced during activated sludge start-up contained considerably more SON with molecular weights greater than 1800. The 165–340 molecular weight fraction had a significantly higher SON to SCOD ratio than any other fraction for all wastewaters examined. Activated carbon adsorption efficiently removed SON (72 ± 9%) and SCOD (78 ± 6%) from treated and untreated wastewaters, and from biologically produced organics. Significantly more SON was removed by cationic exchange at pH 2.0 (33–56%) than by anionic exchange at pH 9.5 (10–24%) for all wastewaters tested. Cationic exchange at pH 2.0 selectively removed more biologically produced SON relative to SCOD.     

助剂对煤浆改性的研究

实验指出助剂能增强表面活性剂制浆性能，不同的助剂对提高煤浆浓度、降低粘度、增加流动度、改善煤浆稳定性的作用不同，本文列出三类助剂、五类表面活性剂、四个煤种制浆特性的实验数据。     

Characterisation of organic matter in IX and PACl treated wastewater in relation to the fouling of a hydrophobic polypropylene membrane

Extensive organic characterisation of a wastewater using liquid chromatography with a photodiode array and fluorescence spectroscopy (Method A), and UV₂₅₄ and organic carbon detector (Method B) was undertaken, as well as with fluorescence excitation emission spectroscopy (EEM). Characterisation was performed on the wastewater before and after ion exchange (IX) treatment and polyaluminium chlorohydrate (PACl) coagulation, and following microfiltration of the wastewater and pre-treated wastewaters. Characterisation by EEM was unable to detect biopolymers within the humic rich wastewaters and was not subsequently used to characterise the MF permeates. IX treatment preferentially removed low molecular weight (MW) organic acids and neutrals, and moderate amounts of biopolymers in contrast to a previous report of no biopolymer removal with IX. PACl preferentially removed moderate MW humic and fulvic acids, and large amounts of biopolymers. PACl showed a great preference for removal of proteins from the biopolymer component in comparison to IX. An increase in the fluorescence response of tryptophan-like compounds in the biopolymer fraction following IX treatment suggests that low MW neutrals may influence the structure and/or inhibit aggregation of organic compounds. Fouling rates for IX and PACl treated wastewaters had high initial fouling rates that reduced to lower fouling rates with time, while the untreated Eastern Treatment Plant (ETP) wastewater displayed a consistent, high rate of fouling. The results for the IX and PACl treated wastewaters were consistent with the long-term fouling rate being determined by cake filtration while both pore constriction and cake filtration contributed to the higher initial fouling rates. Higher rejection of biopolymers was observed for PACl and IX waters compared to the untreated ETP water, suggesting increased adhesion of biopolymers to the membrane or cake layer may lead to the higher rejection.     

A non-adsorptive method for the isolation and fractionation of natural dissolved organic carbon

A non-adsorptive method was tested for the concentration and fractionation of dissolved organic carbon (DOC) in groundwater. Reverse osmosis combined with ultrafiltration recovered more than 90% of the DOC and efficiently separated the DOC into three size fractions. Unlike adsorptive methods, this procedure avoids the use of strong acids and bases that may alter the structure of the DOC. A diafiltration method was applied during ultrafiltration that gave good separation of the DOC into high-, medium-, and low-molecular-size fractions. The high- and medium-molecular-size fractions are desalted during this fractionation procedure and can be further analyzed for individual compounds. The inorganic solutes remain in the low-molecular-size fraction. Future advances in ultrafiltration membrane technology may facilitate removal of inorganic ions without significant loss of low-molecular-size organic acids. Studies on a synthetic groundwater showed that the combined reverse osmosis/ultrafiltration method is viable for the concentration and fractionation of a wide range in organic compounds typically found in groundwater.     

Combination of coagulation and ion exchange for the reduction of UF fouling properties of a high DOC content surface water

The treatment of a high DOC content surface water (about 6mg DOC/L) using anion exchange resins (MIEX resin from Orica or IRA958 resin from Rohm and Haas) can remove up to 80% of DOC in less than 45min. The combination of coagulation prior to or after resin treatment only slightly improves the removal of DOC (0.2-0.3mg/L) but eliminates the high MW organic compounds (MW >20kDa) attributed to biopolymers (proteins and polysaccharides) that were not removed using anion exchange resins alone and that were found to be responsible for reversible fouling of UF membranes (YM 100 UF membrane from Millipore with MW cut-off of 100kDa). The combination of treatments then significantly improves the permeability of the UF membrane. Also, the combination of both treatments allows a reduction of the coagulant doses by a factor of 6 with no impact on the DOC removal and the filterability of produced waters.     

Applicability of light absorbance and fluorescence as measures of concentration and molecular size of dissolved organic carbon in humic Lake Tjeukemeer

Ultrafiltration results of humic Lake Tjeukemeer water demonstrated that light absorbance at 250 nm (E250), fluorescence (λ_ex = 365 nm; λ_em = 470 nm) (F), and concentration of dissolved organic carbon (DOC) vary with the molecular size (5–200 nm) of the dissolved organic matter. The ratio of the ultrafiltered organic fractions increased with decreasing molecular size of the DOC. However, under field conditions this ratio failed to predict molecular size distribution.These results limit the applicability of E250 and F as measures of molecular size and concentration of DOC. However, E250 measurements can be made rapidly and easily, and so are useful in estimating (30%) concentrations of DOC in humic waters.     

Granular iron oxide adsorbents to control natural organic matter and membrane fouling in ultrafiltration water treatment

Fine iron oxide particles (IOPs) are effective in removing natural organic matter (NOM) that causes membrane fouling in water treatment, but the separation of used IOPs is problematic. This study focused on the fabrication and use of granular iron oxide adsorbents, in combination with ultrafiltration (UF) membranes while investigating the NOM removal efficiency and fouling control. Sulfonated styrene-divinylbenzene copolymer beads were coated with two types of iron oxides (ferrihydrite and magnetite) and their performances were compared to that of fine IOPs. A significant amount of iron oxide coating (52–63 mg of Fe per g bead) was achieved by means of electrostatic binding and hydrolysis of iron ions. Iron oxide coated polymer (IOCP) beads were able to remove some amounts (～20%) of dissolved organic carbon (DOC) comparable to that achieved by IOPs within a short period of time (<15 min). Regenerated IOCPs exhibited the same sorption capacity as the fresh ones. The integrated IOCP/UF system operation with a 15-min empty bed contact time and 10-h cyclic regeneration maintained the 20% DOC removal with no sign of significant membrane fouling. In contrast, a sharp transmembrane pressure buildup occurred in the UF system when no iron oxide pretreatment was applied, regardless of the types of membranes tested. Iron oxide adsorbed the NOM fraction with molecular weights of >1000 kDa which is believed to be responsible for severe UF fouling.     

Natural versus wastewater derived dissolved organic carbon: implications for the environmental fate of organic micropollutants

The interaction of organic micropollutants with dissolved organic carbon (DOC) can influence their transport, degradation and bioavailability. While this has been well established for natural organic carbon, very little is known regarding the influence of DOC on the fate of micropollutants during wastewater treatment and water recycling. Dissolved organic carbon-water partition coefficients (K_DOC) for wastewater derived and reference DOC were measured for a range of micropollutants using a depletion method with polydimethylsiloxane disks. For micropollutants with an octanol-water partition coefficient (log K_OW) greater than 4 there was a significant difference in K_DOC between reference and wastewater derived DOC, with partitioning to wastewater derived DOC over 1000 times lower for the most hydrophobic micropollutants. The interaction of nonylphenol with wastewater derived DOC from different stages of a wastewater and advanced water treatment train was studied, but little difference in K_DOC was observed. Organic carbon characterisation revealed that reference and wastewater derived DOC had very different properties due to their different origins. Consequently, the reduced sorption capacity of wastewater derived DOC may be related to their microbial origin which led to reduced aromaticity and lower molecular weight. This study suggests that for hydrophobic micropollutants (log K_OW > 4) a higher concentration of freely dissolved and thus bioavailable micropollutants is expected in the presence of wastewater derived DOC than predicted using K_DOC values quantified using reference DOC. The implication is that naturally derived DOC may not be an appropriate surrogate for wastewater derived DOC as a matrix for assessing the fate of micropollutants in engineered systems.     

The effects of organotin on the activated sludge process

Organotin compounds which find increasing use in marine antifouling paints may be present in the discharge from dry dock operations. This investigation was aimed at determining the effect of such wastewater when discharged to a municipal activated sludge treatment plant. Experiments were conducted using a Warburg respirometer and continuous flow bench-scale activated sludge systems. The results showed that unacclimated biological cultures can be inhibited by tributyl tin oxide (TBTO^™) concentrations as low as 25 μgl⁻¹. However, TBTO doses of over 8000 μgl⁻¹ can be tolerated by a well acclimated culture. Continuous loading of up to 1000 μgl⁻¹ TBTO had no effect on organic removal in activated sludge systems. However, an adverse effect on sludge settleability was noticed at 100 μgl⁻¹ TBTO. Shock loadings of 500 and 1000 μgl⁻¹ TBTO had no effect on soluble organic removal but resulted in impaired settling and higher effluent suspended solids. The LC₅₀ of TBTO to the fathead minnow was estimated at 45–200 μgl⁻¹. The toxicity was reduced considerably by activated sludge treatment.     

Soil aquifer treatment of artificial wastewater under saturated conditions

A 2000 mm long saturated laboratory soil column was used to simulate soil aquifer treatment under saturated conditions to assess the removal of chemical and biochemical oxygen demand (COD and BOD), dissolved organic carbon (DOC), nitrogen and phosphate, using high strength artificial wastewater. The removal rates were determined under a combination of constant hydraulic loading rates (HLR) and variable COD concentrations as well as variable HLR under a constant COD. Within the range of COD concentrations considered (42 mg L⁻¹-135 mg L⁻¹) it was found that at fixed hydraulic loading rate, a decrease in the influent concentrations of dissolved organic carbon (DOC), biochemical oxygen demand (BOD), total nitrogen and phosphate improved their removal efficiencies. At the high COD concentrations applied residence times influenced the redox conditions in the soil column. Long residence times were detrimental to the removal process for COD, BOD and DOC as anoxic processes and sulphate reduction played an important role as electron acceptors. It was found that total COD mass loading within the range of 911 mg d⁻¹-1780 mg d⁻¹ applied as low COD wastewater infiltrated coupled with short residence times would provide better effluent quality than the same mass applied as a COD with higher concentration at long residence times. The opposite was true for organic nitrogen where relatively high concentrations coupled with long residence time gave better removal efficiency.     

Effect of pesticide concentration on the degradation process by combined solar photo-Fenton and biological treatment

The influence of pesticide concentration, expressed as dissolved organic carbon (DOC), on combined solar photo-Fenton and biological oxidation treatment was studied using wastewater containing a mixture of five commercial pesticides, Vydate, Metomur, Couraze, Ditumur and Scala. Two initial DOC concentrations, 200 mg L⁻¹ and 500 mg L⁻¹ were assayed. Variation in biodegradability with photocatalytic treatment intensity was tested using Pseudomonas putida. Thus the mineralisation required for combining with biodegradation of intermediates by activated sludge was 33% and 55% at 200 mg L⁻¹ and 500 mg L⁻¹, respectively. Biotreatment was carried out in a stirred tank in sequencing batch reactor (SBR) mode. As revealed by the biodegradation kinetics, intermediates generated at the higher pesticide concentration caused lower carbon removal rates in spite of the longer photo-Fenton treatment time applied. One strategy for treating water with high concentrations of pesticides and overcoming the low biodegradability of photo-Fenton intermediates is to mix it with a biodegradable carbon source before biological oxidation. This combination of photo-Fenton and acclimatized activated sludge in several SBR cycles led to complete biodegradation of a concentrated pesticide solution of 500 mg L⁻¹ DOC in 5 h with a carbon removal efficiency of 90%.     

Characterisation and removal of recalcitrants in reverse osmosis concentrates from water reclamation plants

Water reclamation plants frequently utilise reverse osmosis (RO), generating a concentrated reject stream as a by-product. The concentrate stream contains salts, and dissolved organic compounds, which are recalcitrant to biological treatment, and may have an environmental impact due to colour and embedded nitrogen. In this study, we characterise organic compounds in RO concentrates (ROC) and treated ROC (by coagulation, adsorption, and advanced oxidation) from two full-scale plants, assessing the diversity and treatability of colour and organic compounds containing nitrogen. One of the plants was from a coastal catchment, while the other was inland. Stirred cell membrane fractionation was applied to fractionate the treated ROC, and untreated ROC along with chemical analysis (DOC, DON, COD), colour, and fluorescence excitation-emission matrix (EEM) scans to characterise changes within each fraction. In both streams, the largest fraction contained <1 kDa molecules which were small humic substances, fulvic acids and soluble microbial products (SMPs), as indicated by EEM. Under optimal treatment conditions, alum preferentially removed >10 kDa molecules, with 17-34% of organic compounds as COD. Iron coagulation affected a wider size range, with better removal of organics (41-49% as COD) at the same molar dosage. As with iron, adsorption reduced organics of a broader size range, including organic nitrogen (26-47%). Advanced oxidation (UV/H₂O₂) was superior for complete decolourisation and provided superior organics removal (50-55% as COD).