Elucidating competitive adsorption mechanisms of atrazine and NOM using model compounds

Based on the relative adsorbability of natural organic matter (NOM) fractions with different molecular weights (MWs), two model compounds, poly(styrene sulfonate) (PSS) (nominal MW=1800 Dalton) and p-dichlorobenzene (DCB), were chosen to study the competitive effect of large and small NOM molecules on atrazine adsorption by two powdered activated carbons (PACs) with different pore size distributions. Both isotherm and kinetic tests of atrazine adsorption were conducted using fresh PAC and PAC preloaded with the model compounds. The model compounds were found to affect atrazine adsorption through two different mechanisms due to their size difference: direct competition for sites by p-DCB and pore constriction/blockage by PSS-1.8k. p-DCB was found to significantly reduce atrazine adsorption capacity but to have no effect on atrazine adsorption kinetics. In contrast, the effect of PSS-1.8k on atrazine adsorption capacity was very small. Furthermore, during simultaneous adsorption, PSS-1.8k had no effect on atrazine surface diffusion. However, preloading PAC with PSS-1.8k lowered the atrazine surface diffusion coefficient, D(s), by more than three orders of magnitude; D(s) decreased with increasing solid phase PSS-1.8k concentration. The pore size distribution of the PAC was found to play an important role in competitive adsorption. A high mesopore surface area could alleviate pore blockage significantly.     

Natural organic matter (NOM) and pesticides removal using a combination of ion exchange resin and powdered activated carbon (PAC)

The combination of anion exchange resins (AERs) and powdered activated carbon (PAC) was studied to remove both natural organic matter (NOM) and pesticides. Experiments were conducted with high dissolved organic carbon (DOC) surface water (about 6.0mg DOC/L) spiked with both atrazine and isoproturon. AERs, like MIEX and IRA938, showed up to 75% removal of DOC after 30min contact time. The addition of PAC after treatment with these AERs only slightly decreased the residual DOC from 1.4 to 1.2mg/L. Experiments conducted with high (200microg/L) and low (1microg/L) initial pesticide concentrations showed that simultaneous and successive combinations of AER and PAC significantly improve the removal of both pesticides compared with PAC treatment on raw water. The improvement of short-term adsorption kinetics was explained by the adsorption of pesticides on AERs (about 5%) and the removal of high molecular weight (MW) NOM structures by AERs that reduce pore blockage phenomena. For 24h contact time with PAC (adsorption isotherms), the benefit of AER treatment was lower, which indicates that the refractory DOC to AER treatment still competes through direct site competition mechanism. MIEX resin had a distinct behavior since the simultaneous treatment with PAC showed no benefit on pesticide adsorption. The presence of fine residues of MIEX was shown to interfere with PAC adsorption.     

Comparison of natural organic matter adsorption capacities of super-powdered activated carbon and powdered activated Carbon

We examined the natural organic matter (NOM) adsorption characteristics of super-powdered activated carbon (S-PAC) produced by pulverizing commercially available, normal PAC to a submicron particle size range. The adsorption capacities of S-PAC for NOM and polystyrene sulfonates (PSS) with molecular weights (MWs) of 1.1, 1.8, and 4.6 kDa, which we used as model compounds, were considerably higher than those of PAC. The adsorption capacity increases were observed for all five types of carbon tested (two wood-based, two coconut-based, and one coal-based carbon). The adsorption capacities of S-PAC and PAC for polyethylene glycols (PEGs) with MWs of 0.3 and 1.0 were the same. The adsorption capacities of S-PAC for PEGs with MWs of 3.0 and 8.0 kDa were slightly higher than the adsorption capacities of PAC, but the difference in adsorption capacity was not as large as that observed for NOM and the PSSs, even though the MW ranges of the adsorbates were similar. We concluded that the adsorption capacity differences between S-PAC and PAC observed for NOM and PSSs were due to the difference in particle size between the two carbons, rather than to differences in internal pore size or structure, to differences in activation, or to non-attainment of equilibrium that resulted from the change in particle size. The difference in adsorption capacity between S-PAC and PAC was larger for NOM with a high specific UV absorbance (SUVA) value than for low-SUVA NOM. The larger adsorption capacities of S-PAC compared with PAC were explained by the larger specific external surface area per unit mass. We hypothesize that a larger fraction of the internal pore volume is accessible with carbon of smaller particle size because the NOM and PSS molecules preferentially adsorb near the outer surface of the particle and therefore do not completely penetrate the adsorbent particle.     

Effect of NOM, turbidity and floc size on the PAC adsorption of MIB during alum coagulation

The effect of natural organic material (NOM) and turbidity on the powdered activated carbon (PAC) adsorption of the odour compound 2-methylisoborneol (MIB) was evaluated during alum coagulation. The character of the flocs, in terms of their size and fractal dimensions (Df), was used to interpret the observed adsorption behaviour of MIB during the coagulation process. As the alum dose was increased, the adsorption of MIB decreased. This was determined to be due to the size of the flocs, with larger flocs incorporating PAC into their structure, reducing the efficiency of mixing, and the bulk diffusion kinetics for the MIB molecule. The presence of turbidity also reduced MIB adsorption due to the formation of larger flocs. The character of NOM was found to have a greater influence on the adsorption of MIB than the floc structure.     

Pore distribution effect of activated carbon in adsorbing organic micropollutants from natural water

Adsorption isotherms of organic micropollutants in coexistence with natural organic matter (NOM) were analyzed to evaluate the impacts of pore size distribution of activated carbon (AC) on the competition effects of the NOM. Single solute adsorption experiments and simultaneous adsorption experiments with NOM contained in a coagulation-pretreated surface water were performed for four agricultural chemicals and three coal-based activated carbons (ACs) having different pore distributions. The results showed that, for all the carbons used, the adsorption capacity of the chemicals was reduced distinctly in the presence of NOM. Such a reduction was more apparent for AC with a larger portion of small pores suitable for the adsorption of small organic molecules and for the agricultural chemicals with a more hydrophilic nature. Ideal adsorbed solution theory (IAST) incorporated with the Freundlich isotherm expression (IAST-Freundlich model) could not interpret the impact of NOM on the adsorption capacity of the chemicals unless a pore blockage effect caused by the adsorption of NOM was also considered. By taking into account this effect, the adsorption isotherm of the chemicals in the presence of NOM was well described, and the capacity reduction caused by the NOM was quantitatively assessed from the viewpoints of the site competition and the pore blockage. Analytical results clearly indicated that pore blockage was an important competition mechanism that contributed to 10-99% of the total capacity reductions of the chemicals, the level depended greatly on the ACs, the chemicals and the equilibrium concentrations, and could possibly be alleviated by broadening the pore size distributions of the ACs to provide a large volume percentage for pores with sizes above 30 A.     

Assessing PAC contribution to the NOM fouling control in PAC/UF systems

This paper investigates the powdered activated carbon (PAC) contribution to the fouling control by natural organic matter (NOM) in PAC/UF hybrid process, as well as the foulant behaviour of the PAC itself. Solutions of NOM surrogates (humic acids, AHA, and tannic acid, TA) and AOM/EOM (algogenic organic matter/extracellular organic matter) fractions from a Microcystis aeruginosa culture were permeated through an ultrafiltration (UF) hollow-fibre cellulose acetate membrane (100 kDa cut-off). The greatest impairment on flux and the poorest rejection were associated with polysaccharide-like EOM substances combined with mono and multivalent ions. PAC, either in the absence or in the presence of NOM, did not affect the permeate flux nor the reversible membrane fouling, regardless of the NOM characteristics (hydrophobicity and protein content) and water inorganics. However, PAC controlled the irreversible membrane fouling, minimising the chemical cleaning frequency. Furthermore, PAC enhanced AHA and TA rejections and the overall removal of AOM, although it was apparently ineffective for the highly hydrophilic EOM compounds.     

Characteristics of competitive adsorption between 2-methylisoborneol and natural organic matter on superfine and conventionally sized powdered activated carbons

When treating water with activated carbon, natural organic matter (NOM) is not only a target for adsorptive removal but also an inhibitory substance that reduces the removal efficiency of trace compounds, such as 2-methylisoborneol (MIB), through adsorption competition. Recently, superfine (submicron-sized) activated carbon (SPAC) was developed by wet-milling commercially available powdered activated carbon (PAC) to a smaller particle size. It was reported that SPAC has a larger NOM adsorption capacity than PAC because NOM mainly adsorbs close to the external adsorbent particle surface (shell adsorption mechanism). Thus, SPAC with its larger specific external surface area can adsorb more NOM than PAC. The effect of higher NOM uptake on the adsorptive removal of MIB has, however, not been investigated. Results of this study show that adsorption competition between NOM and MIB did not increase when NOM uptake increased due to carbon size reduction; i.e., the increased NOM uptake by SPAC did not result in a decrease in MIB adsorption capacity beyond that obtained as a result of NOM adsorption by PAC. A simple estimation method for determining the adsorbed amount of competing NOM (NOM that reduces MIB adsorption) is presented based on the simplified equivalent background compound (EBC) method. Furthermore, the mechanism of adsorption competition is discussed based on results obtained with the simplified EBC method and the shell adsorption mechanism. Competing NOM, which likely comprises a small portion of NOM, adsorbs in internal pores of activated carbon particles as MIB does, thereby reducing the MIB adsorption capacity to a similar extent regardless of adsorbent particle size. SPAC application can be advantageous because enhanced NOM removal does not translate into less effective removal of MIB. Molecular size distribution data of NOM suggest that the competing NOM has a molecular weight similar to that of the target compound.     

Identification and understanding of fouling in low-pressure membrane (MF/UF) filtration by natural organic matter (NOM)

An understanding of natural organic matter (NOM) as a membrane foulant and the behavior of NOM components in low-pressure membrane fouling are needed to provide a basis for appropriate selection and operation of membrane technology for drinking water treatment. Fouling by NOM was investigated by employing several innovative chemical and morphological analyses.

Source (feed) waters with a high hydrophilic (HPI) fraction content of NOM resulted in significant flux decline. Macromolecules of a relatively hydrophilic character (e.g. polysaccharides) were effectively rejected by low-pressure membranes, suggesting that macromolecular compounds and/or colloidal organic matter in the hydrophilic NOM fraction may be a problematic foulant of low-pressure membranes. Moreover, the significant organic fouling that is contributed by polysaccharides and/or proteins in macromolecular and/or colloidal forms depends on molecular shape (structure) as well as size (i.e. molecular weight). More significant flux decline was observed in microfiltration (MF) compared to ultrafiltration (UF) membrane filtration. MF membrane fouling may be caused by pore blockage associated with large (macromolecular) hydrophilic molecules and/or organic colloids. In the case of UF membranes, the flux decline may be caused by sequential or simultaneous processes of surface (gel layer) coverage during filtration. Morphological analyses support the notion that membrane roughness may be considered as a more important factor in membrane fouling by controlling interaction between molecules and the membrane surface, compared to the hydrophobic/hydrophilic character of membranes. Membrane fouling mechanisms are not only a function of membrane type (MF versus UF) but also depend on source (feed) water characteristics.     

Effect of NOM on arsenic adsorption by TiO(2) in simulated As(III)-contaminated raw waters

The effect of natural organic matter (NOM) on arsenic adsorption by a commercial available TiO(2) (Degussa P25) in various simulated As(III)-contaminated raw waters was examined. Five types of NOM that represent different environmental origins were tested. Batch adsorption experiments were conducted under anaerobic conditions and in the absence of light. Either with or without the presence of NOM, the arsenic adsorption reached steady-state within 1h. The presence of 8 mg/L NOM as C in the simulated raw water, however, significantly reduced the amount of arsenic adsorbed at the steady-state. Without NOM, the arsenic adsorption increased with increasing solution pH within the pH range of 4.0-9.4. With four of the NOMs tested, the arsenic adsorption firstly increased with increasing pH and then decreased after the adsorption reached the maximum at pH 7.4-8.7. An appreciable amount of arsenate (As(V)) was detected in the filtrate after the TiO(2) adsorption in the simulated raw waters that contained NOM. The absolute amount of As(V) in the filtrate after TiO(2) adsorption was pH dependent: more As(V) was presented at pH>7 than that at pH<7. The arsenic adsorption in the simulated raw waters with and without NOM were modelled by both Langmuir and Frendlich adsorption equations, with Frendlich adsorption equation giving a better fit for the water without NOM and Langmuir adsorption equation giving a better fit for the waters with NOM. The modelling implies that NOM can occupy some available binding sites for arsenic adsorption on TiO(2) surface. This study suggests that in an As(III)-contaminated raw water, NOM can hinder the uptake of arsenic by TiO(2), but can facilitate the As(III) oxidation to As(V) at TiO(2) surface under alkaline conditions and in the absence of O(2) and light. TiO(2) thus can be used in situ to convert As(III) to the less toxic As(V) in NOM-rich groundwaters.     

Effects of activated carbon characteristics on the simultaneous adsorption of aqueous organic micropollutants and natural organic matter

The overall objective of this research was to determine the effects of physical and chemical activated carbon characteristics on the simultaneous adsorption of trace organic contaminants and natural organic matter (NOM). A matrix of 12 activated carbon fibers (ACFs) with three activation levels and four surface chemistry levels (acid-washed, oxidized, hydrogen-treated, and ammonia-treated) was studied to systematically evaluate pore structure and surface chemistry phenomena. Also, three commercially available granular activated carbons (GACs) were tested. The relatively hydrophilic fuel additive methyl tertiary-butyl ether (MTBE) and the relatively hydrophobic solvent trichloroethene (TCE) served as micropollutant probes. A comparison of adsorption isotherm data collected in the presence and absence of NOM showed that percent reductions of single-solute TCE and MTBE adsorption capacities that resulted from the presence of co-adsorbing NOM were not strongly affected by the chemical characteristics of activated carbons. However, hydrophobic carbons were more effective adsorbents for both TCE and MTBE than hydrophilic carbons because enhanced water adsorption on the latter interfered with the adsorption of micropollutants from solutions containing NOM. With respect to pore structure, activated carbons should exhibit a large volume of micropores with widths that are about 1.5 times the kinetic diameter of the target adsorbate. Furthermore, an effective adsorbent should possess a micropore size distribution that extends to widths that are approximately twice the kinetic diameter of the target adsorbate to prevent pore blockage/constriction as a result of NOM adsorption.     

The effects of dissolved natural organic matter on the adsorption of synthetic organic chemicals by activated carbons and carbon nanotubes

Understanding the influence of natural organic matter (NOM) on synthetic organic contaminant (SOC) adsorption by carbon nanotubes (CNTs) is important for assessing the environmental implications of accidental CNT release and spill to natural waters, and their potential use as adsorbents in engineered systems. In this study, adsorption of two SOCs by three single-walled carbon nanotubes (SWNTs), one multi-walled carbon nanotube (MWNT), a microporous activated carbon fiber (ACF) [i.e., ACF10] and a bimodal porous granular activated carbon (GAC) [i.e., HD4000] was compared in the presence and absence of NOM. The NOM effect was found to depend strongly on the pore size distribution of carbons. Minimal NOM effect occurred on the macroporous MWNT, whereas severe NOM effects were observed on the microporous HD4000 and ACF10. Although the single-solute adsorption capacities of the SWNTs were much lower than those of HD4000, in the presence of NOM the SWNTs exhibited adsorption capacities similar to those of HD4000. Therefore, if released into natural waters, SWNTs can behave like an activated carbon, and will be able to adsorb, carry, and transfer SOCs to other systems. However, from an engineering application perspective, CNTs did not exhibit a major advantage, in terms of adsorption capacities, over the GAC and ACF. The NOM effect was also found to depend on molecular properties of SOCs. NOM competition was more severe on the adsorption of 2-phenylphenol, a nonplanar and hydrophilic SOC, than phenanthrene, a planar and hydrophobic SOC, tested in this study. In terms of surface chemistry, both adsorption affinity to SOCs and NOM effect on SOC adsorption were enhanced with increasing hydrophobicity of the SWNTs.     

受典型除草剂污染原水的应急处理工艺研究

以水中阿特拉津和莠灭净浓度突增为背景,研究了混凝、PAC吸附和PAC吸附＋混凝等工艺对它们的去除效率,同时根据原水水质的变化和水厂的常用工艺,分别考察了混凝剂投加量、pH值、预氧化对混凝去除阿特拉津和莠灭净的影响,以及目标物初始浓度、天然有机物浓度和预氧化对PAC吸附的影响。结果表明,调节pH值及采取预氧化措施均能改善混凝对阿特拉津和莠灭净的去除效果,但其出水浓度仍不能达标;天然有机物浓度对PAC吸附的影响并非是简单的线性关系,同时投加氧化剂和PAC会相互削弱其作用,PAC吸附＋混凝才是去除阿特拉津和莠灭净最简单、有效的方法。     

Effect of natural organic matter on powdered activated carbon adsorption of trace contaminants: characteristics and mechanism of competitive adsorption

Batch adsorption experiments using powdered activated carbon (PAC) to remove trace synthetic organic chemicals (SOCs) from water containing natural organic matter (NOM) were conducted. The percentage of SOC removed at any contact time and at any PAC dose was observed to be independent of the initial SOC concentration. Equations derived from the ideal adsorbed solution theory and the pore surface diffusion model validated this observation. For the strongly adsorbing SOCs (simazine and simetryn), the percentage of SOC removed was independent only at low initial SOC concentrations. The NOM fraction competing with the weakly adsorbing SOC (asulam) constituted a larger percentage of the total NOM than that competing with the strongly adsorbing SOCs. Although the adsorptive capacities of the SOCs were greatly reduced in water containing NOM compared with those in pure water, the change in the pore diffusion coefficient was insignificant. Therefore, NOM competed with the SOCs for adsorption sites, reducing the adsorptive capacity, but the amount of NOM loading was not so severe that it blocked or filled the pores, hindering the internal diffusion of the SOCs.     

Influence of the character of NOM on the ozonation of MIB and geosmin

Tastes and odours (T&Os) are a major concern in drinking water as they are not efficiently removed by conventional water treatment. Ozonation has been effective for their destruction in some studies. However, the natural organic matter (NOM) in waters can affect the ozonation process and subsequently affect the destruction of T&Os. Five NOM fractions were isolated and ozonated in synthetic waters. The fraction containing the more highly coloured, higher molecular weight compounds exhibited the highest ozone (O3) demand, whereas the low aromatic fraction exhibited the lowest O3 demand. The character of the NOM fractions influenced the ozonation of MIB and geosmin. The destruction of MIB and geosmin was significantly higher in the fraction with the highest colour and UV/visible absorbance at all O3 doses. The destruction of the compounds in the other fractions showed the same trends, increasing MIB and geosmin destruction with increasing UV/visible absorbing character of the NOM. MIB was also ozonated in two real waters. with results showing a competing effect between NOM concentration and NOM character. The O3 reaction time was shown to be important for the destruction of both compounds.     

Removal of 2,4,6-trichlorophenol and natural organic matter from water supplies using PAC in floc-blanket reactors

The objective of this study was to evaluate the addition of powdered activated carbon (PAC) to upflow floc-blanket reactors for the adsorption of natural and synthetic organic chemicals. A 15.5-1. bench-scale floc-blanket reactor was operated with PAC addition for the adsorption of 2,4,6-trichlorophenol (TCP) and natural organic matter from one groundwater and two surface waters under laboratory and field conditions, respectively. Influent TCP concentrations ranged from 21 to 415 μg/l. The PAC doses ranged from 2 to 12 mg/l. While the hydraulic residence time in the floc-blanket reactor varied from 15 to 30 min, the carbon residence time ranged from 9 to 34 h. This is due to the high solids concentration in the floc blanket, which ranged from 1200 to 8700 mg/l. Comparison between the extent of TCP adsorption through the floc-blanket reactor and the equilibrium adsorption isotherms of TCP on PAC showed that the maximum adsorption capacity of PAC for TCP was utilized in the reactor. However, this study showed that the maximum adsorptive capacity of the carbon in a continuous process is dependent on the influent adsorbate concentration. This was in agreement with isotherm studies conducted with varying initial TCP concentration. The maximum PAC adsorption capacity for natural organic matter was also achieved in the floc-blanket reactor.     

Adsorption of geosmin and 2-methylisoborneol onto powdered activated carbon at non-equilibrium conditions: influence of NOM and process modelling

The adsorption of the taste and odour (T&O) compounds geosmin and 2-methylisoborneol (2-MIB) onto powdered activated carbon (PAC) has been studied under conditions which are typical for a drinking water treatment plant that uses reservoir water for drinking water production. The reservoir water as well as the pre-treated water (after flocculation) contains NOM that competes with the trace compounds for the adsorption sites on the carbon surface. Although the DOC concentrations in the reservoir water and in the pre-treated water were different, no differences in the competitive adsorption could be seen. By using two special characterisation methods for NOM (adsorption analysis, LC/OCD) it could be proved that flocculation removes only NOM fractions which are irrelevant for competitive adsorption.Different model approaches were applied to describe the competitive adsorption of the T&O compounds and NOM, the tracer model, the equivalent background compound model, and the simplified equivalent background compound model. All these models are equilibrium models but in practice the contact time in flow-through reactors is typically shorter than the time needed to establish the adsorption equilibrium. In this paper it is demonstrated that the established model approaches can be used to describe competitive adsorption of T&O compounds and NOM also under non-equilibrium conditions.The results of the model applications showed that in particular the simplified equivalent background compound model is a useful tool to determine the PAC dosage required to reduce the T&O compounds below the threshold concentration.     

Long term case study of MIEX pre-treatment in drinking water; understanding NOM removal

Removal of natural organic matter (NOM) is a key requirement to improve drinking water quality. This study compared the removal of NOM with, and without, the patented magnetic ion exchange process for removal of dissolved organic carbon (MIEX DOC) as a pre-treatment to microfiltration or conventional coagulation treatment over a 2 year period. A range of techniques were used to characterise the NOM of the raw and treated waters. MIEX pre-treatment produced water with lower concentration of dissolved organic carbon (DOC) and lower specific UV absorbance (SUVA). The processes incorporating MIEX also produced more consistent water quality and were less affected by changes in the concentration and character of the raw water DOC. The very hydrophobic acid fraction (VHA) was the dominant NOM component in the raw water and was best removed by MIEX pre-treatment, regardless of the raw water VHA concentration. MIEX pre-treatment also produced water with lower weight average apparent molecular weight (AMW) and with the greatest reduction in complexity and range of NOM. A strong correlation was found between the VHA content and weight average AMW confirming that the VHA fraction was a major component of the NOM for both the raw water and treated waters.     

Comparison of NOM character in selected Australian and Norwegian drinking waters

Observations from many countries around the world during the past 10-20 years indicate increasing natural organic matter (NOM) concentration levels in water sources, due to issues such as global warming, changes in soil acidification, increased drought severity and more intensive rain events. In addition to the trend towards increasing NOM concentration, the character of NOM can vary with source and time (season). The great seasonal variability and the trend towards elevated NOM concentration levels impose challenges to the water industry and the water treatment facilities in terms of operational optimisation and proper process control. The aim of this investigation was to compare selected raw and conventionally treated drinking water sources from different hemispheres with regard to NOM character which may lead to better understanding of the impact of source water on water treatment. Results from the analyses of selected Norwegian and Australian water samples showed that Norwegian NOM exhibited greater humic nature, indicating a stronger bias of allochthonous versus autochthonous organic origin. Similarly, Norwegian source waters had higher average molecular weights than Australian waters. Following coagulation treatment, the organic character of the recalcitrant NOM in both countries was similar. Differences in organic character of these source waters after treatment were found to be related to treatment practice rather than origin of the source water. The characterisation techniques employed also enabled identification of the coagulation processes which were not necessarily optimised for dissolved organic carbon (DOC) removal. The reactivity with chlorine as well as trihalomethane formation potential (THMFP) of the treated waters showed differences in behaviour between Norwegian and Australian sources that appeared to be related to residual higher molecular weight organic material. By evaluation of changes in specific molecular weight regions and disinfection parameters before and after treatment, correlations were found that relate treatment strategy to chlorine demand and DBP formation.     

AFM study on the sorbed NOM and its fractions isolated from River Songhua

Natural organic matter (NOM) and its fractions from River Songhua were concentrated and fractionated with the reverse osmosis technology and XAD resin. With the purpose of exploring the adsorbed morphology of NOM and its fractions from a specific water source, the tapping-mode atomic force microscopy (TMAFM) was employed. The effect of solution properties (pH, Ca(2+)) on NOM microtopography was also verified. The results showed that NOM and its fractions displayed obviously diverse morphologies due to their different characteristics. The macromolecular compounds with large SUVA value were prone to adsorb on the mica surface. The particle-shaped NOM macromolecules were loosely adsorbed on the mica surface, and the molecular aggregation phenomena were proved to be ubiquitous in this experimental condition. With the increase of pH value, the surface coverage of mica decreased evidently. Spherical conformation with relatively smaller diameters was observed in the acidic condition compared with the neutral and basic condition. In the presence of 0.1 mol L(-1) NaCl, aggregation phenomena were observed in all pH conditions and obvious aggregates were presented with the addition of only 5 mmol L(-1) CaCl(2).     

Effect of membrane character and solution chemistry on microfiltration performance

To help understand and predict the role of natural organic matter (NOM) in the fouling of low-pressure membranes, experiments were carried out with an apparatus that incorporates automatic backwashing and long filtration runs. Three hollow fibre membranes of varying character were included in the study, and the filtration of two different surface waters was compared. The hydrophilic membrane had greater flux recovery after backwashing than the hydrophobic membranes, but the efficiency of backwashing decreased at extended filtration times. NOM concentration of these waters (7.9 and 9.1mg/L) had little effect on the flux of the membranes at extended filtration times, as backwashing of the membrane restored the flux to similar values regardless of the NOM concentration. The solution pH also had little effect at extended filtration times. The backwashing efficiency of the hydrophilic membrane was dramatically different for the two waters, and the presence of colloid NOM alone could not explain these differences. It is proposed that colloidal NOM forms a filter cake on the surface of the membranes and that small molecular weight organics that have an adsorption peak at 220nm but not 254nm were responsible for "gluing" the colloids to the membrane surface. Alum coagulation improved membrane performance in all instances, and this was suggested to be because coagulation reduced the concentration of "glue" that holds the organic colloids to the membrane surface.