Effect of solids retention time on membrane fouling intensity in two-stage submerged anaerobic membrane bioreactors treating palm oil mill effluent.

Submerged anaerobic membrane bioreactors (SAnMBRs) treating palm oil mill effluent were analysed in terms of membrane fouling dynamics when working at three different sludge retention times (SRTs of 15, 30 and 60 d). The average permeate flux was fixed at 2.4 L x m(-2) x h(-1). During operation, the membrane was regenerated by using two steps: membrane wiping during each experiment as soon as trans-membrane pressure reached 125-130 mbars, and complete membrane cleaning including backwash and chemical cleaning at the end of each experiment when analysing the membrane surface and foulant material. Whatever the SRT, the cake formation was the dominant effect on membrane fouling dynamics. The concentration of suspended solids in the SAnMBRs, depending on the SRT, was then a determining criterion. Scanning electron microscopy, energy dispersive X-ray spectroscopy, atomic force microscopy, Fourier transform infrared spectroscopy indicated that fouled membrane surfaces were covered with a cake layer containing organic and inorganic elements whose concentrations were higher when working at a higher SRT; the higher concentrations of such elements gave to the cake layer a denser and more compact structure. In these experiments, the soluble fractions played a secondary role because of the dominant effect of cake layer structuring.

Deposit membrane fouling: influence of specific cake layer resistance and tangential shear stresses.

Cake fouling is the leading cause of membrane permeability decrease when filtering mixed liquor suspension containing high suspended solid concentrations. A simple model is proposed to simulate the cake resistance evolution with time by considering a macro-scale fouling linked only to the accumulation of particles on the membrane surface. This accumulation appears as the difference between the flux of deposited particles due to the filtration and the flux of particles detached from the membrane surface due to the tangential shear stresses caused by recirculation flow in the sidestream membrane bioreactor (MBR) or gas sparging close to the membrane surface for submerged MBR configuration. Two determining parameters were then highlighted: the specific cake resistance and the 'shear parameter'. Based on these parameters it is possible to predict model outputs as cake resistance and permeate flux evolution for short-time filtration periods.

New urban wastewater treatment with autotrophic membrane bioreactor at low chemical oxygen demand/N substrate ratio.

The potential for total nitrogen removal from municipal wastewater has been evaluated in an autotrophic membrane bioreactor running with a low chemical oxygen demand (COD)/N ratio to simulate its combination with an upstream physicochemical process that retains a large proportion of organic matter. The tests were conducted in a laboratory scale submerged membrane bioreactor loaded with a synthetic influent. Nitrogen loading rate was 0.16 kgN-NH4+.m(-3).d(-1) and sodium acetate was added as a carbon source. Results have shown that nitrogen elimination can reach 85% for a COD/N ratio of 5, with COD removal exceeding 97%. However, a COD/N ratio of 3.5 was found to be the limiting factor for successfully reaching the overall target value of 10 mgN.L(-1) in the effluent. Nevertheless, low COD/N ratios make it possible to work with low total suspended solid concentrations in the bioreactor, which greatly facilitates membrane fouling control by a simple aeration and backwashing strategy.

How to optimize hollow-fiber submerged membrane bioreactors.

Membrane fouling is linked to reversible or irreversible accumulation of macromolecules and solids on membrane surfaces and to the irreversible adsorption inside pores. If reversible accumulation can be controlled by filtering in subcritical conditions, then adsorption could also be minimized by reducing the soluble organic matter [extracellular polymeric substances, soluble microbial products (SMP)]. This research shows how the choice of operating parameters related to biological reaction (solid retention time and the organic loading rate) can influence the process rate and the by-product (SMP) production. It also illustrates how suspension characteristics and membrane aeration can influence membrane fouling control according to the hollow fiber configuration and to the different scales of observation. The investigations were based on the definition of different fouling level and fine-tuning of a model to better understand the effects of operating parameters on membrane bioreactor filtration.

Effect of imposed flux on fouling behavior in high rate membrane bioreactor.

The influence of imposed flux and aeration rates on membrane fouling in a submerged membrane bioreactor was studied. The experiments were conducted at four imposed fluxes and three aeration rates. The effect of flux on the reduction of membrane fouling was much higher than that caused by aeration rate. A lower flux of 20 L/m(2) h produced 75 times more water than a higher flux of 40 L/m(2) h with an aeration rate of 2 L/min. Low flux showed slightly higher removal of NH(4)-N and 93-96% removal of dissolved organic matter and chemical oxygen demand. Imposed flux also had a significant effect on the composition of organics present in the soluble microbial product (SMP) and extracellular polymeric substances (EPS). At a higher flux, both SMP and EPS had organics of high molecular weight (MW) of around 48 kDa as well as lower MW organics below 200 Da.

Performances and fouling control of a flat sheet membrane in a MBR pilot-plant.

This paper deals with the performance and the optimisation of the hydraulic operating conditions of the A3 Water Solutions flat sheet membrane technology in a MBR pilot-plant to achieve a satisfying fouling control and also a reduction in the required aeration. Two vertically stacked modules were tested at pilot-scale at Anjou Recherche under typical biological operating conditions (mixed liquor suspended solids concentration (MLSS) =10 g/l; sludge retention time (SRT) =28 days; food to microorganism ratio (F/M)=0.12 kg COD/kg MLSS/d). The use of a double-deck and of specific backwashes for this membrane technology enabled to achieve satisfying membrane performances for a net flux of 25 L h(-1) m(-2), 20 degrees C at a low specific aeration demand per membrane surface (SADm = 0.2 Nm(3) h(-1) m(-2)) which corresponds to a specific aeration demand per permeate volume unit (SADp) of 8 Nm(3) air/m(3) permeate, which is lower than reported for many commercial membrane systems. The mixed liquor characteristics (foaming, MLSS concentration) appeared to influence the fouling behaviour of the membranes but no correlation was found with the fouling rate. However, with the new operating conditions, the system is robust and can cope with fouling resulting from biological stress and daily peak flows for MLSS concentrations in the membrane tank up to 18 g/l.

Influence of water compositions on fouling of plane organic membrane in frontal filtration: application to water and wastewater clarification.

The aim of this study was to evaluate and quantify the filterability of suspended/soluble organic and suspended inorganic matter in a condition without and with chemical conditioning on membrane fouling using cake filtration model. The experiments were conducted with different feed water concentrations under a given TMP (0.2 to 0.5 bar). The fouling potential was examined and described in terms of resistance coefficient (alpha x W) and specific resistance (alpha). The results showed an increase of alpha x W and alpha within the concentration of wastewater samples tested. The soluble fractions in wastewater induced fouling and its mechanism was due both to the interaction of soluble organic components and also some of the particular colloids in MLSS, causing irreversible fouling, followed by thin film formation on membrane surfaces with low porosity, dense structure and also internal fouling. This phenomenon promoted the values of alpha x W and alpha from final treated wastewater 5-20 times higher than in bentonite suspension and on reservoir surface water. Higher pressure than 0.2 bar induced greater hydraulic resistance values than lower applied pressure. The pore size of the porous membrane did not show any difference in the values of alpha x W and alpha obtained, but they mostly depended on the water composition tested. The hydraulic resistance values appeared largely to minimise when using chemical conditioning because of cake forming as a dynamic membrane that reduced the irreversible fouling phenomena giving a constant filtration rate.

In the search of alternative cleaning solutions for MBR plants.

The objective of the study was to identify alternative cleaning reagents to chlorine for membrane permeability regeneration in MBR applications. Indeed, chlorine is prohibited in some countries because of the formation of by-products such as THM. The study was focused on the comparison of ten cleaning reagents performances and in particular on their ability to remove irreversible fouling. The tests were carried on with the A3 Water Solutions' Maxflow membrane (flat sheet membrane). A specific experimental protocol was defined at lab scale to develop an irreversible fouling by filtering sludge supernatant. The more promising reagents at lab scale were then tested on the A3 membrane continuously immersed in a MBR pilot plant functioning under typical biological conditions (MLSS=11 g/l; SRT=28 days). A full scale test was finally performed with hydrogen peroxide, one of the best reagents. Chlorine was taken as reference for all performed tests. The cleaning performances of the selected reagents were different at the different scales, probably due to the difficulty to obtain an irreversible membrane fouling at larger scales. This testing procedure will be reproduced with other membrane materials to have a better understanding of interactions between irreversible fouling, material nature and chemical reagents.

Biomass adaptation to complex substrate degradation in membrane bioreactors: appropriated operating conditions.

The aim of this work is to analyse the biological performances of two immersed membranes bioreactors focusing on the biomass adaptation to complex substrate degradation and the performance in term of permeate quality. Two influents were selected: a synthetic complex influent (acetate/Viandox, MBR1) and a real seafood processing wastewater (surimi product, MBR2). The MBR systems were operated for long periods without any sludge extraction except for sampling. Organic matter removal, sludge production and quality of the treated wastewater were analysed and studied. COD removal efficiencies after a period of biomass adaptation were higher than 97% and 95% for the synthetic and real wastewater respectively. In both cases, the COD of the treated wastewater was lower than 50 mg.L(-1). In spite of salt concentration in the real wastewater a biomass adaptation process occurs. In the overall operational period, a 0.058 gCOD P.gCOD T(-1) and a 0.12 gCOD P.gCOD T(-1) observed sludge yields were obtained for the MBR1 and MBR2 respectively. These values are approximately 5 to 10 times lower than those measured in conventional activated sludge process. These results showed that the presence of particular and some of non-easily degradable compounds in the influent of MBR2 didn't limit the performance of MBR in term of COD removal achieved. The results have also confirmed the excellent permeate quality for water reuse from MBRs systems.

Continuous and sequencing membrane bioreactors applied to food industry effluent treatment.

This work focuses on the performances of two immersed membrane bioreactors used for the treatment of easily biodegradable organic matter present in food industry effluents, for the purpose of water reuse. Two reactor functioning modes (continuous and sequencing) were compared in terms of organic carbon removal and of membrane permeability. For each working mode, pollutant removal was very high, treated water quality presented a low COD concentration (< 125 mg x L(-1)), no solids in suspension and low turbidity (< 0.5 NTU). The quality of the treated water (including germ removal) enabled its reuse on site. Moreover, by developing high biomass concentrations in the reactor, excess sludge production remained very low (< 0.1 gVSS x gCOD(-1)). The performances appeared slightly better for the continuous system (lower COD concentration in the effluent, < 50 mg x L(-1), and lower sludge production). In terms of filtration, a distinct difference was observed between continuous and sequencing systems; transmembrane pressure showed a small and constant evolution rate in continuous membrane bioreactor (CMBR) although it appeared more difficult to control in sequencing membrane bioreactor (SMBR) probably due to punctually higher permeate flow rate and modified suspension properties. The rapid evolution of membrane permeability observed in SMBR was such that more frequent chemical cleaning of the membrane system was required.

Comparison of textile dye treatment by biosorption and membrane bioreactor.

The Cassulfon CMR is a sulphuric textile dye mainly used to colour "jeans". It has a dark black-blue colour, with high intensity of colour and high mineral compounds (71% of dry matter). Direct filtration experiments were carried out to quantity the capacity of macro porous membranes (1.2, 0.2 or 0.1 microm) to separate organic matter and colour from the effluent. The results show that no direct membrane filtration was efficient. To evaluate the capacity of a biological way for the elimination of this dye, batch experiments were performed to quantify the dye sorption capacity on activated sludge. Results show the high capacity of the biomass to adsorb colour (more than 4gCOD gMLVSS(-1)) while 15% of COD remain in the soluble fraction. To evaluate the biodegradability potential of the sludge, continuous operations were carried out in a membrane bioreactor (MBR). Results confirm the very high MBR potential to treat such dye effluents. During operations, the organic load was progressively increased from 0.33 to 1.33 kg m(-3) d(-1) and the permeate quality was always free of suspended solids or turbidity. Moreover, the permeate COD values were always lower than 60 mg l(-1) and small permeate coloration only appeared during malfunctioning periods.

Effect of reeds and feeding operations on hydraulic behaviour of vertical flow constructed wetlands under hydraulic overloads.

Vertical flow constructed wetlands (VFCWs) have been very successful in France over the last 5 years. The sizing of VFCWs is still roughly based on organic load acceptance with slight clear water intrusion into the sewerage system which is often wrong in the context of small communities. To specify the hydraulic limits would be of great help to Water Authorities in deciding at what point is it preferable to build separate sewers rather than adapt the wastewater treatment plant. The study of the hydraulic limits of reed beds, based on the knowledge of hydrodynamics in unsaturated porous media, shows the ability of the system to accept flow overloads. Measuring different parameters (flow, pollutant removal, infiltration rate (IR), pressure head profiles) in pilot and full-scale studies, we explain the hydraulic behaviour of the filter, and the role of reeds and batch frequency on the IRs. Consequently, new hydraulic limits with accompanying sizing rules and operational recommendations according to the level of deposit on the filter surface are suggested. The study shows the robustness of reed beds systems as designed in France to accept hydraulic overloads. Overloads up to ten times the dry weather flow are possible whilst still complying with the European standards.

Apatite as an interesting seed to remove phosphorus from wastewater in constructed wetlands.

Intensive use of phosphates has resulted in high P levels in surface waters and therefore eutrophication problems. Over the last decade many studies have revealed the advantage of using specific materials with efficient phosphorus retention capacities. Recent studies state that Ca materials are of particular interest for long-term retention of P, but can induce negative effects. To improve P retention and avoid negative counter-effects we tested the potential of natural apatites. Apatite sorption was evaluated using batch and open reactor experiments. Batch experiments identify sorption mechanisms and the influence of the ionic characteristics of the solution; open reactor experiments evaluate sorption capacities in relation to the ionic composition of the solution and biomass development. In parallel, observation of the material by electron microscopy was used to give more precision information about the mechanisms involved. This work reveals the strong chemical affinity between apatites and phosphorus. Compared to other calcareous materials apatite is better able to maintain low outlet P levels. After more than 550 days feeding, sorption was still present and low P outlet levels were still being obtained when sufficient contact time and calcium content in the solution were ensured. This work demonstrates the advantages of using apatites for phosphorus removal in constructed wetlands. The behaviour of apatite in phosphorus retention is explained and its suitability for use in such extensive systems defined.

Effects of starvation conditions on biomass behaviour for minimization of sludge production in membrane bioreactors.

The behaviour of an activated sludge system in starvation conditions was examined in batch according to substrate impulses defined by different S0/X0 ratio. The answer was characterised by an exogenous phase followed by a starvation one. If at high S0/X0 ratio, bacterial cell multiplication was the main synthesis process during exogenous phase, at low S0/X0 ratio the observed phenomenon was compound storage. In starvation conditions, for the lowest S0/X0 ratio, a rapid decrease in the MLVSS without soluble proteins production was observed. No bacterial lysis occurred and this phenomenon was due to consumption of the storage compounds with a decrease rate equal to 0.74d(-1). For high S0/X0 ratio, as soon as the exogenous phase was completed, a decrease of the MLVSS simultaneously to a soluble protein production was observed. An immediate bacterial lysis occurred with a decay rate equal to 0.53 d(-1). Because MBR systems work generally in low F/M conditions, the activity of the present microbial population is close to the one observed in starvation phase. This work points out that these conditions do not allow net bacterial growth and cells just use lysis products to satisfy their maintenance requirements. These assumptions confirm the feasibility of a decrease of the net biomass production in a MBR when high sludge retention time is operated.

Influence of module configuration and hydrodynamics in water clarification by immersed membrane systems.

Different immersed membrane systems were compared according to the module configuration. Filtering concentrated aqueous suspensions under constant permeate flux, the hydraulic performances of the systems were evaluated and compared through parameters such as critical permeate flux notion and trans-membrane pressure variation rates. Operational variables were membrane size and module fibre density, aeration inside or outside the fibre network, suspension concentration and physico-chemical conditioning. When using hollow fibres including a possible air injection inside the fibre network, results pointed out the positive role of the aeration on the fouling control. But too high a fibre density did not allow an optimal control when the aqueous suspension was very concentrated. On the other hand, when working with capillary membranes showing sufficient space between fibres, the major parameters were the transversal suspension flow circulation through the fibre network and the FeCl3 conditioning of the suspension. Experimental results show a possible working at a 0.07 m3 x m2 x h(1) permeate flow rate under low TMP evolutions, 0.02 Pa/s, even if the filtration was operated under high concentrated suspension, 5kgSS/m3.

Phosphorus retention in subsurface constructed wetlands: investigations focused on calcareous materials and their chemical reactions.

Phosphorus removal from wastewater has been of growing interest for some decades to avoid eutrophication in surface water. In subsurface constructed wetlands precipitation and adsorption are the main mechanisms responsible for P uptake. Two media (calcite and recycled crushed concrete (RCC)) were examined in batch and continuous systems. Batch experiments show attractive sorption capacities, however experiments carried out in open reactors pointed out some limitation in retention capacities and effluent quality. RCC is sensitive to a strong dissolution leading to a quick phosphorus precipitation but induces high conductivity and pH values in the treated water. Calcite efficiency depends on the carbonate equilibrium of the solution. Microscopic observations of the calcite surface show crystal growth of phosphorus precipitate. Crystallisation seems to be the main P uptake once a material's surface is covered.

Critical evaluation of 4-week incubation for fungal cultures: is the fourth week useful?

To determine the benefit of a 4-week incubation for mycology cultures, we evaluated all positive cultures during the fourth week of incubation in a 1-year period. Of 3,855 positive mycology cultures (yeast, 82%; molds, 18%), 62 (1.6%) were positive during the fourth week (yeast, 42%; molds, 58%). Only 15 of the 62 cultures (24%) were considered clinically relevant (2 isolates from invasive fungal infection and 13 isolates from cutaneous mycosis). With the exception of those from skin samples, isolates recovered during the fourth week are rarely important for patient care.

Comparison of rapid identification method and conventional substrates for identification of Corynebacterium group JK isolates.

In an effort to identify Corynebacterium group JK isolates rapidly, Rapid Identification Method (RIM series; Austin Biological Laboratories, Inc., Austin, Tex.) substrates were tested in parallel with conventional substrates. RIM reactions agreed with conventional substrate results, respectively, as follows: urea, 38 of 38; nitrate, 35 of 38; glucose, 35 of 38; maltose, 28 of 38; sucrose, 37 of 38; and o-nitrophenyl-beta-D-galactopyranoside, 24 of 26. As a supplement to initial screening tests, the RIM tests offer a rapid method for identifying group JK isolates.

Clinical comparison of the AutoMicrobic system gram-positive identification card, API Staph-Ident, and conventional methods in the identification of coagulase-negative Staphylococcus spp.

In an effort to rapidly identify coagulase-negative staphylococci (CNS), a clinical comparison was conducted with the AutoMicrobic system Gram-Positive Identification Card (GPI) (Vitek Systems, Inc.), the API Staph-Ident (Analytab Products), and the conventional methods of W. E. Kloos and K. H. Schleifer (W. E. Kloos and K. H. Schleifer, J. Clin. Microbiol. 1:82-88, 1975). CNS isolates tested included 157 from blood and 33 from urine in pure culture at greater than 10(5) CFU/ml. S. epidermidis accounted for 79.6 and 60.6% of the isolates from blood and urine, respectively. S. saprophyticus was the next most frequent urine isolate (27.4%). Other CNS species were isolated from blood and urine specimens with frequencies of less than 5%. Overall, the GPI correctly identified 158 (83.2%) of the 190 CNS, whereas the Staph-Ident identified 124 (65.3%) without further testing. This resulted in the GPI and Staph-Ident correctly identifying 95.9 and 74.5% of the S. epidermidis and 100 and 33% of the S. saprophyticus, respectively. The GPI misidentified 8 (47%) of the S. hominis and S. warneri isolates as S. saprophyticus, indicating the need for novobiocin testing. These data suggest that the GPI is a more definitive method for the rapid identification of S. epidermidis than the Staph-Ident and that both systems require additional testing to identify S. saprophyticus.