Study on the removal efficiency of UF membranes using bacteriophages in bench-scale and semi-technical scale

To determine the removal efficiency of ultrafiltration (UF) membranes for nano-particles in the size range of viruses the state of the art uses challenge tests with virus-spiked water. This work focuses on bench-scale and semi-technical scale experiments. Different experimental parameters influencing the removal efficiency of the tested UF membrane modules were analyzed and evaluated for bench- and semi-technical scale experiments. Organic matter in the water matrix highly influenced the removal of the tested bacteriophages MS2 and phiX174. Less membrane fouling (low ΔTMP) led to a reduced phage reduction. Increased flux positively affected phage removal in natural waters. The tested bacteriophages MS2 and phiX174 revealed different removal properties. MS2, which is widely used as a model organism to determine virus removal efficiencies of membranes, mostly showed a better removal than phiX174 for the natural water qualities tested. It seems that MS2 is possibly a less conservative surrogate for human enteric virus removal than phiX174. In bench-scale experiments log removal values (LRV) for MS2 of 2.5-6.0 and of 2.5-4.5 for phiX174 were obtained for the examined range of parameters. Phage removal obtained with differently fabricated semi-technical modules was quite variable for comparable parameter settings, indicating that module fabrication can lead to differing results. Potting temperature and module size were identified as influencing factors. In conclusion, careful attention has to be paid to the choice of experimental settings and module potting when using bench-scale or semi-technical scale experiments for UF membrane challenge tests.     

Mg/Al layered double hydroxide for bacteriophage removal in aqueous solution

The objective of this study was to investigate the removal of bacteriophages in Mg/Al layered double hydroxide (LDH). Batch experiments were performed with bacteriophage MS2 in a powder form of Mg/Al LDH under various LDH doses. Column experiments were also performed under flow-through condition with bacteriophages MS2 and phiX174 in Mg/Al LDH immobilized on sand surfaces. Batch tests demonstrated that the powder form of Mg/Al LDH was effective in removing MS2 with the removal capacity of 2.2 × 10(8) plaque forming unit (pfu)/g under the given experimental conditions (LDH dose = 2 g/L; initial MS2 concentration = 4.61 × 10(5) pfu/mL). Column experiments showed that the log removal of phiX174 was 4.40 in columns containing 100% Mg/Al LDH-coated sand while it was 0.05 in 100% quartz sand. These findings indicated that Mg/Al LDH-coated sand was effective in removing bacteriophages compared with sand. A more than 4 log removal (=5.44) of MS2 was achieved in 100% Mg/Al LDH-coated sand. This study demonstrates the potential application of Mg/Al LDH for virus removal in water treatment.     

不同孔径平板膜对景观水体水质控制效果的试验研究

为研究平板膜处理景观水体的特性,采用孔径为0.1 μm、0.2μm和0.3μm平板膜处理景观水体用水.结果表明,三种膜对悬浮物和藻类有很好的去除效果,去除率都在95％以上,对于溶解态的N、P去除效果不好;三种膜处理效果相比较,孔径最小的0.1μm膜对各种污染物去除效果最好;在过滤不同污染物配水时,0.1μm膜的膜通量下降速率最慢,稳定后较0.2μm和0.3μm膜的膜通量大;通过对污染物的粒径分析和膜表面污染物分析认为,景观水体中,主要污染物对膜通量的影响顺序为:腐殖酸＞藻＞悬浮物;150 nm是景观水体中污染物质粒径的分界点,污染物质粒径主要集中在150 nm以上;与0.2μm、0.3μm膜相比,0.1μm膜抗污染能力更强,更适合用来处理景观水体用水.     

Optimization study of a hybrid alum coagulation-membrane filtration system for virus removal

Due to the intrinsically small sizes of enteric viruses (20-100 nm) and their relatively high resistance to most disinfectants, detection of viruses in treated drinking water is not a rare phenomenon. This study therefore evaluates various aspects involved in a hybrid alum coagulation-ultrafiltration (UF) system for virus removal. Coagulant doses (0, 1 and 10 mg Al(3+)/L) and pH conditions relevant to drinking water (pH 6-8) were investigated. With this hybrid system, removal was not attributable merely to MS2 adsorption to flocs and subsequent retention by UF membranes. MS2 removal comprises of inactivation by the effect of pH and coagulant and subsequently, rejection of virus-associated flocs by UF membrane. Coagulation with 1 mg Al(3+)/L at pH 6 and 7 resulted in an overall reduction brought about by an average of 0.62 log inactivation via the pH effect, 1.2 log inactivation by alum coagulant, and >5.4 log rejection by the 100 kDa polyethersulfone UF membrane. In contrast, negligible upstream inactivation was noted with a coagulant dose of 1 mg Al(3+)/L at pH 8, but 5.8 log rejection was attained with downstream UF filtration. By optimizing the conditions appropriate for upstream inactivation and subsequent membrane rejection, virus removal efficiencies can be enhanced.     

Occurrence and removal of microbial indicators from municipal wastewaters by nine different MBR systems

Nine different membrane bioreactor (MBR) systems with different process configurations (submerged and external), membrane geometries (hollow-fiber, flat-sheet, and tubular), membrane materials (polyethersulfone (PES), polyvinylidene fluoride (PVDF), and polytetrafluoroethylene (PTFE)) and membrane nominal pore sizes (0.03-0.2 μm) were evaluated to assess the impact of influent microbial concentration, membrane pore size and membrane material and geometries on removal of microbial indicators by MBR technology. The log removal values (LRVs) for microbial indicators increased as the influent concentrations increased. Among the wide range of MBR systems evaluated, the total and fecal coliform bacteria and indigenous MS-2 coliphage were detected in 32, 9 and 15% of the samples, respectively; the 50th percentile LRVs were measured at 6.6, 5.9 and 4.5 logs, respectively. The nominal pore sizes of the membranes, membrane materials and geometries did not show a strong correlation with the LRVs.     

Membrane separation of indigenous noroviruses from sewage sludge and treated wastewater.

In this study, feasibility of membrane separation for the removal of indigenous noroviruses (NVs) is evaluated. The indigenous NV gene was never detected from ultrafiltration (UF) permeates of sewage sludge and treated wastewater. Indigenous NV gene was also not detected from permeates of sewage sludge and treated wastewater by microfiltration (MF) with a pore size of 0.1 microm (MF0.1). Even though the pore size of MF (0.1 microm) was much larger than the diameter of virus particle (approximately 30-40nm), more than 4-log10 reduction value (LRV) at maximum was achieved by membrane separation with MF0.1. NV genes were often detected from permeates of sewage sludge and treated wastewater by MF with a pore size of 0.45 microm (MF0.45), although the maximum log10 reduction values were more than 3.59 for sewage sludge and more than 2.90 for treated wastewater. It is important to verify factors determining the removal efficiency of viruses with MF membranes.     

Elimination of viruses, bacteria and protozoan oocysts by slow sand filtration.

The decimal elimination capacity (DEC) of slow sand filters (SSF) for viruses, bacteria and oocysts of Cryptosporidium has been assessed from full-scale data and pilot plant and laboratory experiments. DEC for viruses calculated from experimental data with MS2-bacteriophages in the pilot plant filters was 1.5-2 log10. E. coli and thermotolerant coliforms (Coli44) were removed at full-scale and in the pilot plant with 2-3 log10. At full-scale, Campylobacter bacteria removal was 1 log10 more than removal of Coli44, which indicated that Coli44 was a conservative surrogate for these pathogenic bacteria. Laboratory experiments with sand columns showed 2-3 and >5-6 log10 removal of spiked spores of sulphite-reducing clostridia (SSRC; C. perfringens) and oocysts of Cryptosporidium respectively. Consequently, SSRC was not a good surrogate to quantify oocyst removal by SSF. Removal of indigenous SSRC by full-scale filters was less efficient than observed in the laboratory columns, probably due to continuous loading of these filter beds with spores, accumulation and retarded transport. It remains to be investigated if this also applies to oocyst removal by SSF. The results additionally showed that the schmutzdecke and accumulation of (in)organic charged compounds in the sand increased the elimination of microorganisms. Removal of the schmutzdecke reduced DEC for bacteria by +/-2 log10, but did not affect removal of phages. This clearly indicated that, besides biological activity, both straining and adsorption were important removal mechanisms in the filter bed for microorganisms larger than viruses.     

Adsorption of viruses to soil: impact of anaerobic treatment.

The adsorption of viruses in untreated flushed dairy manure wastewater (FDMW), anaerobically digested flushed dairy manure wastewater (ADFDMW) and groundwater to sandy soil was investigated. Batch adsorption studies showed differential adsorption of viruses in groundwater to soil. Less than 75% of PRD1 and MS2 added to groundwater adsorbed after 1 h, but greater than 95% of phiX174 and poliovirus 1 adsorbed to the soil. Adsorption differences in groundwater were related to the isoelectric points of the viruses. Suspending phages in untreated and treated wastewater reduced adsorption compared with groundwater. For MS2, more phages were adsorbed using ADFDMW than with FDMW. Adsorption of poliovirus 1 was not affected by FDMW and ADFDMW. Small column studies (6 x 2.5 cm) produced a similar trend in that adsorption was observed with groundwater and both FDMW and ADFDMW reduced virus adsorption. Groundwater, FDMW or ADFDMW did not affect the adsorption of poliovirus 1 in column studies. The major difference between FDMW and ADFDMW was in mobilisation of adsorbed viruses. The application of FDMW to soil columns with adsorbed viruses caused significantly more viruses to be mobilised than did the application of rainwater or ADFDMW. These results showed that treating FDMW by anaerobic digestion increased the adsorption of viruses to soil and decreased detachment of adsorbed viruses. As the potential for new zoonotic pathogens becomes known, the treatment of animal wastes may become mandatory. The assessment and management of viruses in manure for addressing possible risk to animal and human health is of interest.     

Addition of a magnetite layer onto a polysulfone water treatment membrane to enhance virus removal

The applicability of low-pressure membranes systems in distributed (point of use) water treatment is hindered by, among other things, their inability to remove potentially harmful viruses and ions via size exclusion. According to the USEPA and the Safe Drinking Water Act, drinking water treatment processes must be designed for 4-log virus removal. Batch experiments using magnetite nanoparticle (nano-Fe3O4) suspensions and water filtration experiments with polysulfone membranes coated with nano-Fe3O4 were conducted to assess the removal of a model virus (bacteriophage MS2). The membranes were coated via a simple filtration protocol. Unmodified membranes were a poor adsorbent for MS2 bacteriophage with less than 0.5-log removal, whereas membranes coated with magnetite nanoparticles exhibited a removal efficiency exceeding 99.99% (4-log). Thus, a cartridge of PSf membranes coated with nano-Fe3O4 particles could be used to remove viruses from water. Such membranes showed negligible iron leaching into the filtrate, thus obviating concern about coloured water. Further research is needed to reduce the loss of water flux caused by coating.     

Seasonal profiles of human noroviruses and indicator bacteria in a wastewater treatment plant in Tokyo, Japan.

The seasonal profiles of microorganisms in raw sewage, secondary-treated sewage, and final effluent at a wastewater treatment plant in Tokyo, Japan, were quantitatively determined each month for one year, from July 2003 to June 2004. Human noroviruses, which were determined by real-time PCR, in raw sewage varied from 0.17-260 copies/mL for genotype 1 and from 2.4-1900 copies/mL for genotype 2, showing much higher values in winter, the epidemic season. The concentration of total coliforms, Escherichia coli, or F-specific phages in raw sewage was almost constant throughout the year. Human noroviruses of genotype 2 were removed most effectively (3.69 log10 on average) at the wastewater treatment plant, followed by E. coli (3.37 log10), total coliforms (3.05 loglo), F-specific phages (2.81 log10), and human noroviruses of genotype 1 (2.27 log10). The removal ratio of human noroviruses was almost constant, independent of the initial concentration of the viruses in raw sewage, which led to the increasing concentration of human noroviruses in final effluent in winter. None of the tested bacteria was judged to be a reliable indicator of human noroviruses in final effluent.     

Effect of organic loading rate on a wastewater treatment process combining moving bed biofilm and membrane reactors.

The effect of moving bed biofilm reactor (MBBR) loading rate on membrane fouling rate was studied in two parallel units combining MBBR and membrane reactor. Hollow fiber membranes with molecular weight cut-off of 30 kD were used. The HRTs of the MBBRs varied from 45 min to 4 h and the COD loading rates ranged from 4.1 to 26.6 g COD m(-2) d(-1). The trans-membrane pressure (TMP) was very sensitive to fluxes for the used membranes and the experiments were carried out at relatively low fluxes (3.3-5.6 l m(-2) h(-1)). Beside the test with the highest flux, there were no consistent differences in fouling rate between the low- and high-rate reactors. Also, the removal efficiencies were quite similar in both systems. The average COD removal efficiencies in the total process were 87% at 3-4 h HRT and 83% at 0.75-1 h HRT. At high loading rates, there was a shift in particle size distribution towards smaller particles in the MBBR effluents. However, 79-81% of the COD was in particles that were separated by membranes, explaining the relatively small differences in the removal efficiencies at different loading rates. The COD fractionation also indicated that the choice of membrane pore size within the range of 30 kD to 0.1 microm has very small effect on the COD removal in the MBBR/membrane process, especially with low-rate MBBRs.     

Technical and sanitary aspects of wastewater disinfection by UV irradiation for landscape irrigation.

Water reuse for landscape irrigation requires the production of high quality virus-free effluents to minimize risk for human health. In order to establish the relevance of MS2 phages as an appropriate biodosimeter for UV design, a pilot plant study has been carried out with different types of wastewater effluents. The two pilot systems tested (low-pressure high output and medium-pressure UV units) were able to achieve 4 and 5 log MS2 reduction in tertiary filtered effluent at high calculated UV doses of 170 _ 10 and 300 mJ/cm2, respectively. UV disinfection was extremely efficient for MS2 inactivation in high quality effluents after reverse osmosis: detention times as low as one second and UV dose of 40 mJ/cm2 were sufficient to reach 5 log inactivation of MS2. UV irradiation also produced rapid inactivation of human pathogens such as poliovirus type 1 and indigenous enteroviruses at UV doses up to 3 times lower that those for MS2 disinfection. It was concluded that accurate UV unit design for a given type of wastewater could be ensured by pilot tests using laboratory-propagated MS2 as biodosimeter and collimated-beam tests as the calibration-check.     

Effect of membrane characteristics on the performance of membrane bioreactors for oily wastewater treatment

Influence of membrane material and pore size on the performance of a submerged membrane bioreactor (sMBR) for oily wastewater treatment was investigated. The sMBR had a working volume of about 19 L with flat sheet modules at the same hydrodynamic conditions. Five types of micro- and ultra-polymeric membranes containing cellulose acetate (CA), cellulose nitrate (CN), polyamide (PA), polyvinylidene difluoride (PVDF) and polyethersulfone (PES) were used and their filtration performance in terms of permeability, permeate quality and fouling intensity were evaluated. Characterization of the membranes was done by performing some analysis such as pore size distribution; contact angle and scanning electronic microscopy (SEM) microphotograph on all membranes. The quality of permeates from each membrane was identified by measuring chemical oxygen demand (COD). The results showed more irreversible fouling intensity for membranes with larger pore size which can be due to more permeation of bioparticles and colloids inside the pores. Membrane characteristics have a major role in the preliminary time of the filtration before cake layer formation so that the PA with the highest hydrophilicity had the lowest permeability decline by fouling in this period. Also, the PVDF and PES membranes had better performance according to better permeate quality in the preliminary time of the filtration related to smaller pore size and also their better fouling resistance and chemical stability properties. However, all membranes resulted in the same permeability and permeate quality after cake layer formation. An overall efficiency of about 95% in COD removal was obtained for oily wastewater treatment by the membranes used in this study.     

Virus removal in a membrane separation process

Recently, membrane technology has been considered an alternative to conventional water purification. To study the fate of viruses in membrane processes, indigenous coliphages in pilot scale membrane processes located in the eastern part of Tokyo Metropolitan area have been surveyed for 6 months. This plant used river water as its resource and had two microfiltration membrane processes which had different pore sizes (0.2 µm and 0.1 µm) and one ultrafiltration process which had 13,000 nominal molecular weight cut off. To detect indigenous coliphages, E. coli K12 F⁺(A/λ) and E. coli C were used as host bacteria. E. coli K12 F+(A/λ) can detect both DNA and RNA phages and E. coli C can only DNA phage. The resource water contained E. coli K12 phages at 200–1500 PFU/100 mL and the removal ratio of these DNA and RNA phages was lower than that of DNA phage by E. coli C in both MF membrane processes through 6 months. It is thought to be caused by difference of phage size, because DNA phage is bigger than RNA phage in general. The removal ratio of E. coli K12 and E. coli C phages reached 100% in the UF membrane process. According to the comparison of the concentration of phages in solution and eluted from suspended solid in resource and drain, it is thought that most phages concentrated in the drain were absorbed in suspended solids. To make certain of the removal ratio in UF and NF (nanofiltration) processes, high concentrations of coliphage Qβ and poliomyelitis virus vaccine were fed into these processes. The removal ratio of coliphage Qβ in UF and NF processes are 10⁻⁸³ and 10^−6.3 respectively, and the ratio of poliomyelitis virus vaccine in UF and NF are µ10^−6.7 and µ10^−7.3 respectively.     

Diffusion analysis and modeling of kinetic behavior for treatment of brine water using electrodialysis process

In this study, the removal of monovalent and divalent cations, Na⁺, K⁺, Mg²⁺, and Ca²⁺, in a diluted solution from Chott-El Jerid Lake, Tunisia, was investigated with the electrodialysis technique. The process was tested using two cation-exchange membranes: sulfonated polyether sulfone cross-linked with 10% hexamethylenediamine (HEXCl) and sulfonated polyether sulfone grafted with octylamine (S-PESOS). The commercially available membrane Nafion® was used for comparison. The results showed that Nafion® and S-PESOS membranes had similar removal behaviors, and the investigated cations were ranked in the following descending order in terms of their demineralization rates: Na⁺ > Ca²⁺ > Mg²⁺ > K⁺. Divalent cations were more effectively removed by HEXCl than by monovalent cations. The plots based on the Weber–Morris model showed a strong linearity. This reveals that intra-particle diffusion was not the removal rate-determining step, and the removal process was controlled by two or more concurrent mechanisms. The Boyd plots did not pass through their origin, and the sole controlling step was determined by film-diffusion resistance, especially after a long period of electrodialysis. Additionally, a semi-empirical model was established to simulate the temporal variation of the treatment process, and the physical significance and values of model parameters were compared for the three membranes. The findings of this study indicate that HEXCl and S-PESOS membranes can be efficiently utilized for water softening, especially when effluents are highly loaded with calcium and magnesium ions.     

Removal of manganese from water using combined chelation/membrane separation systems.

The addition of the chelating polymer polyacrylic acid (PAA) to assist in the removal of manganese from groundwater by membranes was investigated using membranes with different pore sizes under various operating conditions. Negligible manganese removal was achieved with the UF and NF membranes at acidic pH values, but removals exceeding 90% could be achieved at elevated pH (pH 9), presumably due to the formation of manganese hydroxides. Mn removal increased substantially when PAA was added to the feed solution, due to chelation of Mn by the PAA and rejection of the chelates by the membranes. The chelate could be broken at acidic pH, releasing free PAA that could then be separated from the Mn ions and reused. Smaller PAA molecules were lost in the first regeneration cycle, but negligible PAA was lost in subsequent cycles. In the systems with PAA, nitrate ions were rejected more efficiently than in the PAA-free systems, presumably because of electrical repulsion between nitrate ions and PAA sorbed on the membrane surface. With increasing PAA dose, the volumetric flux first decreased and then increased; the latter result was accompanied by a change in the physical-chemical form of the polymers, as indicated by an increase in turbidity.     

Differences in major ions as well as hydrogen and oxygen isotopes of sediment pore water and lake water

Isotopic and chemical compositions of pore water(PW) are highly relevant to environmental and forensic study. Five lake water(LW)samples and five sediment samples were collected to investigate the effects of pore sizes of sediments on PW chemistry and stable isotopes and determine mechanisms controlling their variations. Six pore water fractions were extracted from different-sized pores in each sediment sample at six sequential centrifugal speeds for chemical and isotopic analysis. The sediments consisted mainly of quartz, feldspar, and clay minerals. The hydrogen and oxygen isotopic compositions of PW are mainly controlled by the overlying LW, although the lag effect of exchange between overlying LW and PW results in isotopic differences when recharge of LW is quicker than isotopic exchange in PW. Identical isotopic compositions of PW from sediments with different pore sizes indicate that isotopic exchange of water molecules with different pore sizes is a quick process. The ratio of average total dissolved solid(TDS) concentration of PW to TDS concentration of LW shows a strong relationship with adsorption capacity of sediments, demonstrating that remobilization of ions bound to sediments mainly causes a chemical shift from LW to PW.Concentrations of Ca~(2+), Mg~(2+),and Cl~-in PW remain unchanged,while concentrations of Na~+,K~+,and SO_4~(2-) slightly increase with decreasing pore size. Chemical differences of PW from sediments with different pore sizes are governed by ion adsorption properties and surface characteristics of different-si zed particles.     

Virus removal in a pilot-scale 'advanced' pond system as indicated by somatic and F-RNA bacteriophages.

Advanced pond systems (APS), incorporating high-rate ponds, algal settling ponds, and maturation ponds, typically achieve better and more consistent disinfection as indicated by Escherichia coli than conventional waste stabilisation ponds. To see whether this superior disinfection extends also to enteric viruses, we studied the removal of somatic phages ('model' viruses) in a pilot-scale APS treating sewage. Measurements through the three aerobic stages of the APS showed fairly good removal of somatic phage in the summer months (2.2 log reduction), but much less effective removal in winter (0.45 log reduction), whereas E. coli was removed efficiently (> 4 logs) in both seasons. A very steep depth-gradient of sunlight inactivation of somatic phage in APS pond waters (confined in silica test tubes) is consistent with inactivation mainly by solar UVB wavelengths. Data for F-RNA phage suggests involvement of longer UV wavelengths. These findings imply that efficiency of virus removal in APS will vary seasonally with variation in solar UV radiation.     

Respirometric assays of two different MBR (microfiltration and ultrafiltration) to obtain kinetic and stoichiometric parameters

A comparison of two different medium scale MBRs (ultrafiltration and microfiltration) using respirometric methods has been achieved. The ultrafiltration membrane plant (0.034 microm pore size) maintained recirculation sludge flow at seven times the influent flow, and membranes were backwashed every 5 min and chemically cleaned weekly. The microfiltration membrane plant (0.4 microm pore size) maintained recirculation sludge flow at four times the influent flow, membrane-relax was applied after the production phase and membranes were chemically cleaned in the event of high trans-membrane pressure. Both technologies showed a similar performance with regard to heterotrophic kinetic and stoichiometric parameters and organic matter effluent concentrations. The influent was characterized by means of its COD fractions and the average removal percentages for COD concentrations were around 97% for both plants in spite of influent COD fluctuation, temperature variations and sludge retention time (SRT) evolution. Both SRT evolution and temperature affect the heterotrophic yield (Y(H)) and the decay coefficient (bH) in the same range for both plants. Y(H) values of over 0.8 mg COD/mg COD were obtained during the unsteady periods, while under steady state conditions these values fell to less than 0.4 mg COD/mg COD. bH by contrast reached values of less than 0.05 d(-1).     

Evaluation of the purification capacity of nine portable, small-scale water purification devices.

A test was performed to evaluate the microbial and chemical purification capacity of nine portable, small-scale water purification filter devices with production capacity less than 100 L/h. The devices were tested for simultaneous removal capacity of bacteria (cultured Escherichia coli, Clostridium perfringens, Klebsiella pneumoniae and Enterobacter cloacae), enteric protozoans (formalin-stored Cryptosporidium parvum oocysts), viral markers (F-RNA bacteriophages) and microcystins produced by toxic cyanobacterial cultures. In general, the devices tested were able to remove bacterial contaminants by 3.6-6.9 log10 units from raw water. Those devices based only on filtration through pores 0.2-0.4 microm or larger failed in viral and chemical purification. Only one device, based on reverse osmosis, was capable of removing F-RNA phages at concentrations under the detection limit and microcystins by 2.5 log10. The present study emphasised the need for evaluation tests of water purification devices from the public safety and HACCP (Hazard Analysis and Critical Control Point) points of view. Simultaneous testing for various pathogenic/indicator microbes and microcystins was shown to be a useful and practical way to obtain essential data on actual purification capacity of commercial small-scale drinking-water filters.