On the performance of real grey water treatment using a submerged membrane bioreactor syst

Brought to you by:

Journals Download PDF

Export

Books

Shopping cart

Search ScienceDirect

Advanced search

Journal of Membrane Science Volume 476, 15 February 2015, Pages 40–49

On the performance of real grey water treatment using a submerged membrane bioreactor system Khalid Bani-Melhema, Zakaria Al-Qodahb,

, 1,

,

, Mohammad Al-Shannagc, Ahmad Qasaimehd,

Mohammed Rasool Qtaishatc, Malek Alkasrawie Show m ore

doi:10.1016/j.memsci.2014.11.010

Get rights and content

Highlights • Effective treatment of real grey water (GW) was achieved using SMBR system. • Membrane fouling behaviour and hydraulic performance were investigated extensively. • The cake layer contributed significantly in reducing the permeation flux. • An empirical model was proposed to correlate the time-dependent permeation flux.

Abstract The performance of a submerged membrane bioreactor (SMBR) system for grey water (GW) treatment was evaluated in terms of effluent quality and membrane fouling. The SMBR was operated for 42 days at constant transmembrane pressure (13 kPa) in six consecutive stages. A hollow fibre ultrafiltration membrane module (ZW-1) was used to treat real GW with the aim of producing effluent that meets reuse guidelines for nonpotable standards. A complete retention for activated sludge was maintained in the bioreactor to minimize the amount of sludge disposed into the environment. The results demonstrated that the SMBR system was able to reduce effectively the COD, NH3 –N, turbidity, and colour to have values of 45 mg/L, 0.26 mg/L, 3 FTU and 18 PtCo in the effluent, respectively. Furthermore, a complete removal of total suspended solids (TSS) was achieved and faecal coliform concentration was below the determination threshold. In terms of membrane permeability, the results showed that the cake layer contributed significantly (86%) in reducing the permeation flux. The time-dependent permeation flux was modelled adequately according to an exponential expression. Ultimately, the treated GW by SMBR system can be considered as a good source for the most stringent non-potable reuse standards in arid areas.

List of abbreviations and symbols Am , Membrane area (m2 ); AS, Anionic surfactants (mg/L); C, Effluent concentration (mg/L); C0 , Influent concentration (mg/L); COD, Chemical oxygen demand (mg/L); DO, Dissolved oxygen (mg/L); EPS, Extracellular polymeric substances (mg/gVSS); EPSs, EPS in mixed liquor (mg/gVSS); EPSm , EPS on membrane surface (mg/gVSS); J, Permeation flux, (m3 /m2 s) or (L/ m2 h); J0 , Initial permeation flux (m3 /m2 s) or (L/ m2 h); k , Empirical constant used in Eq. (6) (day−1 ); MBR, Membrane bioreactor; ML, Mixed liquor solution; MLSS, Mixed liquor suspended solids (mg/L); MLVSS, Mixed liquor volatile suspended solids (mg/L); NH3 –N, Ammonia nitrogen concentration (mg/L); NO3 –N, Nitrate nitrogen concentration (mg/L); PRPF, Percentage reduction in permeation flux (–); Q, Effluent flow rate (L/h); Rc, Cake resistance (m−1 ); Rf, Total fouling resistance (m−1 ); Rg , Gel layer resistance (m−1 ); Rm , Membrane resistance (m−1 ); Rt, Total resistance (m−1 ); SD, Standard deviation; SMP, Soluble microbial products (mg/L); SMBR, Submerged membrane bioreactor; t, Time (s); TP, Total phosphorus (mg/L); TDS, Total dissolved solids (mg/L); TMP, Transmembrane pressure (kPa); TSS, Total suspended solids (mg/L); V, Total volume of the collected permeate (m3 ); VSS, Volatile suspended solids (mg/L); μ , Permeate viscosity (N s/m2 ); Δ P , Transmembrane pressure difference (kPa)

Keywords Submerged membrane bioreactor; Grey water; Wastewater reuse; Membrane fouling

1. Introduction Water resources are increasingly under pressure from the continuous depletion in fresh water resources accompanying with population growth, economic activity and intensify competition for the water among users. Therefore, water resources development and management should be a hot issue especially in regions like the Middle East and North Africa (MENA) due to severe water scarcity, rainfall fluctuation and the rise in

Sign in

Help

Biblioteka Politechniki Lodzkiej

water pollution. Apart from traditional methods in water resources planning and management, nonconventional methods for utilization of water should be considered, like wastewater treatment and reclamation. This trend can lead to a reduction of the demand for water from existing water sources [1]. In this domain, grey water (GW) treatment has received considerable attention as a valuable source for wastewater recycling and reuse in MENA during the last few years [1], [2] and [3]. Grey water treatment and reuse can contribute to promoting the preservation of high-quality fresh water as well as reducing pollutants in the environment. Generally, GW is defined as any source of wastewater generated from the kitchen, laundry and bathroom (sink, bath and shower) and any other non-toilet household wastewater [4]. Quantitatively, grey water constitutes from 50% to 80% of the total wastewater generated in households [5] and [6], making it a good source for water reuse. Qualitatively, GW might include some chemicals and several millions of pathogenic bacteria per 100 mL, which can cause a health hazard if this water is reused without proper treatment. Therefore, the GW should undergo certain treatments to be ready for reuse; the treated grey water should fulfil the water reuse guidelines which may vary from one country to another. Different approaches have been applied for GW treatment including physical, chemical and biological operational methods. Among the different treatment methods, the submerged membrane bioreactor (SMBR) technology has been investigated as an attractive method for grey water treatment by many researchers over the last few years [7], [8], [9], [10], [11] and [12]. In comparison with physical [13] and [14] and chemical treatments [15], [16] and [17], the SMBR technology was addressed as the only technology able to achieve satisfactory removal efficiencies of organic substances, surfactants and microbial contaminations without a post-filtration and disinfection step [5]. In other words, the SMBR technology has proved to be the most efficient method for grey water treatment and reuse since it combines physical separation of colloidal substances, including pathogenic bacteria, together with aerobic biological treatment of dissolved organic matter [5]. However, the treatment of wastewater by SMBR system is still suffering from the membrane fouling phenomena which is currently considered the main obstruct in SMBR applications [7]. Likewise consideration is given to apply many traditional methods for reducing the membrane fouling like cleaning the membrane unit by backwashing [18], optimizing process operating parameters [19] and modifying the characteristics of the activated sludge by adding chemical flocculants such as aluminium sulphate (alum), polyaluminum chloride and iron salts [20], [21] and [22] or the addition of adsorptive materials such as powdered activated carbon [23] and zeolite [22]. In addition, wide range of innovative approaches were adopted to reduce membrane fouling effectively like patenting a novel submerged membrane electrobioreactor [24] and [25] and developing a new configuration of membranes such as the helical module [26]. Although the literature contains few studies about treatment of grey water by SMBR system, but less attention was given on investigating the operational performance of the SMBR technology for real grey water treatment in arid and semi-arid areas. Unlike previous studies, the current work presents a comprehensive investigation of real GW treatment using a submerged membrane bioreactor (SMBR) system. The study includes analyzing effluent quality, membrane fouling behaviour and the hydraulic performance of the filtration process. In addition, an empirical model was introduced for the first time to assess the permeation flux of real GW treatment using a hollow fibre ultrafiltration membrane module.

2. Materials and methods 2.1. Experimental setup The experimental setup of the current study consisted of two parts. In the first part, a laboratory scale experimental setup (Fig. 1) was used to treat real grey water continuously for 42 days. The setup of the first part consisted of polypropylene container with a capacity of 20 L served as grey water feed tank. The grey water was pumped from the feed tank via feed pump to another tank (20 L) through fine screen served as a pretreatment stage to remove large particles from entering the SMBR system. The grey water flowed up to the SMBR system by gravity. The SMBR system consisted of a bioreactor with a working volume of 3.65 L and a hollow fibre, ultrafiltration (UF) membrane module; ZeeWeed-1 (GE/Zenon Membrane Solutions, Canada) was submerged in the bioreactor. The membrane module consisted of 80 fibre units with 0.2 m length and pore size of 0.04 µm and a total surface area of 0.047 m2 . The effluent from the membrane module was withdrawn via a peristaltic pump operated at a constant transmembrane pressure (TMP). A level sensor was connected with the feed pump via a level controller system to maintain a constant volume in the bioreactor. Furthermore, to maintain the required dissolved oxygen for the microorganisms in the bioreactors, a stream of compressed air, at a flow rate of 3.4 L/min, was supplied through a perforated tube located at the bottom of the membrane module. Also, the compressed air was used for creating a shear stress for effective scouring of the membrane surfaces and providing good mixing of the sludge suspension in the bioreactors.

Fig. 1. Part one of experimental setup used for grey w ater treatment by SMBR. Figure options

The second part of the experimental setup shown in Fig. 2 was used to analyse the hydraulic performance of the mixed liquor (ML) solution produced from the operation of the first part of the experimental work. A 2500 mL of the mixed liquor (ML) solution was taken from the bioreactor into a 3000 mL beaker which was served as a small bioreactor. As can be seen in Fig. 2 another clean hollow fibre ultrafiltration (UF) membrane module (ZeeWeed-1) was mounted in the small bioreactor and connected to a peristaltic pump operated at the same TMP that was applied in part one to simulate the same operating conditions. The small bioreactor was fed with grey water from the same tank used in part one. The level in the bioreactor was controlled as described in part one of the experimental work. The mixed liquor solution in the small bioreactor was agitated with magnetic stirrer bar at a speed of 1000 rpm.

Fig. 2. Part tw o of experimental setup used for studying fouling resistances. Figure options

2.2. Experimental procedure and operating conditions The SMBR system illustrated in Fig. 1 was operated at room temperature (23.6±0.7 °C) for 42 days. The experimental period was divided into six consecutive filtration-cleaning stages and each stage was operated for 7 days. Before starting these experimental stages, the bioreactor was seeded with activated sludge from a recycle sludge tank of a Municipal Waste Water Treatment Plant located in Al Zarqa city, Jordan. The sludge was acclimatized with real grey water for one month prior to the operation of the membrane filtration process. The fill-and-draw technique was used to cultivate the activated sludge [27]. With starting the experimental stage I, the biomass concentration (mixed liquor suspended solid-MLSS) in the bioreactor was 2200 with standard deviation (SD) of 100 mg/L. The SMBR was operated at constant transmembrane pressure (TMP) of 13 kPa created by withdrawing the effluent via a peristaltic pump operated at a constant suction pressure. Moreover, a complete sludge retention time (SRT) was maintained to decrease the amount of sludge wasting from the process. This is one advantage of membrane bioreactor technology over the conventional activated sludge process in which a large amount of sludge is produced and 50–60% of the operating cost goes to sludge treatment [28]. This procedure was conducted by many studies in treating grey water by SMBR systems [1] and [7]. During the operation of the experimental stages (part one of the experimental work), the fouling behaviour was monitored by measuring the decline of permeation flux with time. Subsequently, no backwashing of the membrane module was performed during the operation of each stage. However, after each experimental stage, the cake layer on the membrane surface was removed physically by washing with tap water followed by chemical cleaning procedure (0.5% (v/v) NaClO solution, 2 h duration). The washing step was performed prior to next experimental stage in order to restore most of the membrane permeability by removing all the particulates that deposited within the membrane pores. Finally the membrane module was rewashed extensively by tap water and distilled water to grantee the removal of any traces of chemical solution before reusing in the next stage. In part two of the experimental work shown in Fig. 2, the hydraulic performance of the filtration processes was

performed during the washing step. The objective of the second part of the experimental setup was to compare the fouling behaviour correctly. Therefore, the second part was based on short experiments in order to avoid the variations in the chemical and physical characteristics of the grey water during the experimental period [7]. At the end of each experimental stage, a 2500 mL of the ML solution was taken from the original bioreactor and filled into a 3000 mL beaker (small bioreactor) to apply the resistance-in-series model according to Darcy‫׳‬s law. In order to exclude the effect of MLSS concentration on the permeation flux at the end of each experimental stage, the MLSS concentration taken was adjusted to 2300 (SD=125 mg/L) to be identical to the MLSS concentration at the end of Stage I (the lowest concentration of MLSS during the whole experimental stages). At the end of each hydraulic performance test, the ML solution in the small bioreactor was returned to the original ML solution in the bioreactor and the next filtration stage (7 days) was run. Meanwhile, the physical and chemical cleaning were applied on the membrane module of part two of the experimental work as described previously for the same module of part one.

2.3. Grey water characteristics Actual grey water used in this study was collected from one building of the Faculty of Natural Resources at the Hashemite University, AL-Zarqa, Jordan. The composition of this grey water varies according to the activities of building employees. The primary contributions to grey water were from cleaning and sink activities. Table 1 shows a summary of statistical analysis of the grey water characteristics used in this study during the experimental stages. Table 1. Characteristics of grey w ater used as feed to SMBR process. a

N

pH

21

6.5

7.3

6.9

0.2

DO (mg/L)

21

0.19

1.60

0.72

0.60

TDS (mg/L)

21

280

350

319

27

Turbidity (FTU)

21

37

173

80

34

Colour (PtCo)

21

206

550

378

108

TSS (mg/L)

21

16

59

34

11

COD (mg/L)

14

92

668

356

181

NH3–N (mg/L)

11

0.60

4.75

2.47

1.40

NO3–N (mg/L)

11

0.50

2.5

1.10

0.58

Total phosphorus,TP (mg/L)

14

0.89

2.07

1.64

0.37

Anionic surfactants (mg/L)

6

29.80

75.50

45.84

17.23

Faecal coliform (CFU/100 mL)

6

Total coliform (CFU/100 mL) a

N: num ber of sam ples.

b

SD: standard dev iation.

6

Min.

Max.

Average

SDb

Water quality indexes

1.0×105 6.0×105 2.2×10

5

10×10

5

3.1×105

2.3×105

5

3.5×105

4.4×10

Table options

2.4. Analytical methods The performance of the SMBR system was monitored by analyzing both influent and effluent samples regularly for COD, ammonia nitrogen (NH3 –N), nitrate nitrogen (NO3 –N), total phosphorus (TP), anionic surfactants (AS), total dissolved solids (TDS), dissolved oxygen (DO), pH, temperature, conductivity, colour and turbidity. The SensoDirect 150 m (Lovibond, Germany) was used to measure DO, pH, temperature, TDS, and electrical conductivity. The SensoDirect 150 m was calibrated once a week prior the usage. The other parameters (colour, turbidity, COD, NH3 –N, NO3 –N, TP, AS) were analyzed by the Hatch methods (Hatch, DR 2000, USA). The activated sludge in the bioreactor was sampled regularly twice a week for MLSS and mixed liquor volatile suspended solids (MLVSS). Both MLSS and MLVSS were performed according to standard methods [29]. Total and faecal coliforms were determined by the membrane filtration procedure. The size distribution of activated sludge particles was determined by Horiba Laser scattering particle size distribution analyzer Model LA-300 (Horiba, USA). Furthermore, the concentration of extracellular polymeric substances (EPS) was measured during the operation of each experimental stage. The concentrations of EPS in mixed liquor (EPSs) and on membrane surface (EPSm ) were analyzed using thermal treatment method described by Chang and Lee [30]. The EPS was represented by the total organic carbon (TOC) value of the EPS solution, which was measured using a TOC analyzer (Teledyne Tekmar, USA). The concentration of EPS was expressed in mg/gVSS; where VSS are the volatile suspended solids in mg/L.

2.5. Calculations In order to investigate the qualitative performance of the SMBR for GW treatment, the reduction levels in influent pollutants were determined by calculating the percent removal defined as: (1) Turn

where C0 is the influent concentration (mg/L) of a given pollutant at a specific time and C is the corresponding effluent concentration (mg/L).

on

The quantitative determination of the permeation flux (J) in L/(h m2 ) was calculated from: (2) where Q is the effluent flow rate (L/h) evaluated by measuring the accumulated effluent volume with time and Am is the membrane surface area (m2 ). The hydraulic performance of the filtration processes was analyzed at the end of each experimental stage (part two of the experimental work). The MLSS concentrations were adjusted to be identical in all experimental stages as mentioned previously. Hence, the resistance-in-series model was applied according to Darcy‫׳‬s law as [7]: (3) where J is the membrane permeation flux (m3 /m2 s), V is the total volume of the collected permeate (m3 ), Am is the membrane area (m2 ), Δ P is the transmembrane pressure (N/m2 ), μ is the permeate viscosity (N s/m2 ), Rt is the total resistance in the system (m−1 ), Rm is the membrane resistance (m−1 ) and Rf is the fouling resistance (m−1 ). Generally, the fouling resistance (Rf) is due to two types of resistances: the resistance of the cake layer (Rc) and the gel layer resistance (Rg ). The Rc represents the reversible fouling formed by MLSS particles and Rg is due to irreversible adsorption and pore blocking. Accordingly, the fouling resistance (Rf) could be expressed as: Rf =Rc +Rg

(4)

The above fouling resistances were determined experimentally for each experimental stage by the following procedure: First, the membrane resistance (Rm ) was determined by using deionized water prior each experimental stage (i.e. Rf=0) according to: (5) Consequently, the filtration process was applied to the activated sludge solution. At the end of each experimental run (~120 min), the total fouling resistance (Rt) was calculated from the final values of the permeation flux and TMP according to Eq. (3). Second, the fouling resistance, Rf, could be calculated by subtracting membrane resistance (Rm ) from the total resistance Rt. The cake layer on the membrane module was removed by rinsing the membrane module with tap water, and the filtration procedure was followed with deionized water until the permeation flux reach stabilization. From permeation flux and TMP values, the summation of two resistances (Rm +Rg ) can be determined using Darcy‫׳‬s law. The resistance of the cake layer (Rc) was calculated by subtracting the summation of membrane resistance and the gel layer resistance (Rm +Rg ) from the total resistance Rt. Finally, the gel layer, Rg , resistance was calculated according to Eq. (4).

3. Results and discussion 3.1. Removal of colour, turbidity and total suspended solids Fig. 3 shows the removal performance of the SMBR system with respect to the changes in the colour and the turbidity. It is clear in the figure that the colour and turbidity concentrations of the influent fluctuated from one stage to another, specifically during the last three stages. However, the SMBR system achieved stable and excellent water permeate quality during the whole experimental period. During 42 days of continuous operation, the SMBR system reduced the colour of the grey water from an average value of 378 PtCo to 18 PtCo corresponding to an average removal of 95.2%. Meanwhile, the average concentration of grey water turbidity reduced from 80 FTU to 3 FTU corresponding to an average removal of 96.3%. Furthermore, The SMBR system achieved complete removal of total suspended solids (TSS) in the effluent.

Fig. 3. Performance of the SMBR for (a) colour removal and (b) turbidity removal. Figure options

3.2. Removal of COD, NH3–N, total phosphorus (TP) and anionic surfactants (AS) Fig. 4 shows the performance of SMBR system with respect to the removal of COD, NH3 –N, TP and AS during the six experimental stages. During the first three stages (from day 1 to day 21), the influent COD concentration varied between 92 and 348 mg/L with an average value of 223 and standard deviation SD=101 mg/L (Fig. 4(a)). The average value of the corresponding effluent COD was 34 (SD=26) which gives an average removal efficiency of 86.1% (SD=10.7%). Starting from the fourth stage, a considerable increase in the COD influent concentration was observed in the grey water from 304 mg/L on day 22 up to 668 mg/L on day 27 as shown in Fig. 4(a). The average concentration of COD during the last three stages was 488 (SD=142 mg/L). This dramatic increase in influent COD could be attributed to the variations of activities of building employees. The average removal efficiency of COD during the last three stages was 89.3% (SD=8.5%) which was greater than the removal efficiency during the first three stages when the average influent COD was considerably smaller than the last three stages. This was attributed to a significant increase in the population of microorganisms given by the increase in the MLSS concentration in the last three stages (average MLSS concentration was 5930 mg/L with standard deviation SD=990 mg/L) in comparison with the MLSS concentration during the first three stages (average MLSS concentration was 2940 mg/L with standard deviation SD=715 mg/L). Over the whole entire experimental period (42 days), the SMBR reduced the influent COD from an average value of 356 mg/L (SD=181 mg/L) to 45 mg/L (SD=39 mg/L) in the effluent for an average removal efficiency of 88% (SD=9.5%).

Fig. 4. Performance of the SMBR for (a) COD removal, (b) NH3–N removal, (c) total phosphorus (TP) removal and (d) anionic surfactants (AS) removal. Figure options

The variation in ammonia–nitrogen concentration in the grey water was notable during the six experimental stages (Fig. 4(b)); the influent NH3 –N varied between 0.6 and 4.75 mg/L with an average value of 2.47 mg/L (SD=1.4 mg/L). Except for the first stage, the performance of the SMBR system with respect to ammonia removal was excellent with an average percentage removal equal to 89.8% (SD=7.3%). The lower percentage removal efficiency of NH3 –N during the initial stage was attributed to the lower growth rate of nitrifying bacteria because these bacteria need more time to establish and reach sufficient concentrations to nitrify the ammonium [31] and [32]. Over the entire experimental period, the SMBR system achieved percentage removal efficiency of NH3 –N of 88% on an average (SD=7.3%). The removal of total phosphorus (TP) by SMBR system was unstable during the operation period; see Fig. 4(c). The removal efficiency was 72% at the beginning of operation and decreased to 30% on day 8. Then the removal efficiency increased again up to 75% on day 18 and started to fluctuate between 52% and 80% during the last period. This fluctuation in the removal efficiency could be attributed to the fact that biological phosphorous removal would be limited in membrane bioreactor (MBR) applications when SMBR systems operated with minimum sludge removal as was the case in this study [33]. However, over the whole period of operation, the SMBR reduced the TP from 1.64 mg/L (SD=0.37 mg/L) on average in the influent to an effluent average of 0.74 mg/L (SD=0.38 mg/L) for an average removal efficiency of 56% (SD=18%). The performance of SMBR system with respect to TP removal was closed to the same results found in the literature [10]. Fig. 4(d) shows the performance of SMBR system with respect to anionic surfactants (AS) removal. The figure shows that the removal of AS was better in the first stages with an average value around 90%, while the overall performance decreased in subsequent stages. This could be attributed in part to the significant increase in AS concentration in the effluent in the later stages.

3.3. Performance of SMBR system for pathogenic removal Faecal and total coliforms decreased significantly by SMBR system due to rejection by the UF membrane used in the study. As shown in Fig. 5, the effluent is almost free of pathogenic content, especially during the last three stages. On average, the SMBR system reduced the total and faecal coliforms from 4.4×105 CFU/100 mL and 3.1×105 CFU/100 mL in grey water to less than 29 CFU/100 mL and 26 CFU/100 mL in effluent, respectively.

Fig. 5. Performance of the SMBR for pathogenic removal: (a) faecal coliform and (b) total coliform. Figure options

In summary and in terms of effluent quality, this study shows that the SMBR system can be an attractive method for GW treatment. The most significant result of this study is that the SMBR treatment produces average effluent values of COD, TSS, and coliforms that satisfy the local guideline for reuse of treated grey water [34] and very close to the international guidelines [5] for reuse for non-potable applications; see Table 2. Table 2. Comparison of the characteristics of treated grey w ater of this study w ith w astew ater reuse standards from different countries, adapted from Ref. [5]. This Water quality indexes

USA

Japan

China

Germany

study

pH

6–9

5.8–8.6

6–9

–

7.8

DO (mg/L)

–

–

>1.5

>50%

6.15

TDS (mg/L)

–

–

–

–

298

Turbidity (NTU)

–

≤5

<5

–

3as

TSS (mg/L)

30

–

–

–

NDa

COD (mg/L)

–

–

–

–

45

BOD5(mg/L)

30

≤3

<6

5 (BOD7)

–

TP (mg/L)

–

–

< 0.5

–

0.74

Faecal coliform

< 200

–

< 10,000

< 10

26

–

≤50

–

< 100

29

Restricted

Environmental (limited public

Restricted impoundments

Toilet

–

reuses b

contact)

and lakes

flushing

FTU

(CFU/100 mL) Total coliform (CFU/100 mL) Reuse application

a

ND: Not detected.

b

Irrigation of areas where public access is inf requent and controlled golf courses, cem eteries, residential, greenbelt.

Table options

3.4. Performance of the membrane permeability The permeation flux of pure water a cross a clean membrane can be described by Darcy‫׳‬s law presented in Eq. (3) where the total resistance (Rt) is equal to the hydraulic resistance of the clean membrane (Rm ). For suspension filtration like activated sludge in SMBR, the permeation flux will decrease with time as a result of the increase in total resistance resulting from membrane fouling from particles deposition on the membrane surface ( reversible fouling) or in the membrane pores (irreversible fouling). Because the SMER process in this study was operated on the basis of constant TMP, a decline in permeation flux would result during the whole experimental stages due to the fouling phenomenon as shown in Fig. 6. Fig. 6 shows that the variations in permeation flux varied from one stage to another stage which could be attributed to many reasons discussed in the next sections. In partial, the variations in permeation flux can be attributed to the variation in the characteristics of the grey water in each experimental stage. Also, the missing of applying back-flushing event might lead to a quite strong flux decrease in each stage specifically during the first operation period of each stage as shown in Fig. 6. It was demonstrated not to apply the event of back-flushing in this work

because one of the main objectives of the present study was to introduce an empirical model to assess the permeation flux of real GW treatment using a hollow fibre ultrafiltration membrane module. Therefore, if a back-flushing event was applied during the operation of the six stages, the empirical model will be definitely affected. This procedure was applied successfully by other researchers [1] and [7].

Fig. 6. Permeation flux decline during the membrane filtration stages of the SMBR process. Figure options

However, the common feature of all the experimental stages is the exponential decline of permeation flux with time. A simple regression model which can describe the evolution of membrane flux in each experimental stage can be given by:

J = J0 e− k t

(6)

where J0 is the initial permeation flux measured during the first minute in each stage and J is the permeation flux at any time during the operational stage. The empirical constant k in Eq. (6) can be obtained by plotting the ln (J/J0 ) versus t in each experimental stage as shown in Fig. 7. A liner trend was obtained for all six stages considered with values of squared correlation coefficient (R2 ) greater than 0.9313. The k values for stages I, II, III, IV, V and VI are 0.0926, 0.0990, 0.1084, 0.1129, 0.1532 and 0.1218 day−1 , respectively. The discrepancy in the k values can be attributed to the variation in the operational parameters characteristics.

Fig. 7. Regression analysis for permeation flux decline during the membrane filtration: (a) stage I, (b) stage II, (c) stage III, (d) stage IV, (e) stage V and (f) stage VI. Figure options

However, because the SMBR system was operated at long sludge retention time, the development of the MLSS concentration may have a significant role on membrane fouling which might lead to a rapid decline in permeation flux. It can be seen from Fig. 8 that there is no significant increase in MLSS during the operation of the first stage which can be attributed to the acclimation period of the microorganisms. A notable increase started to appear in stages II and III, when the MLSS increased from 2300 mg/L at the beginning of stage II to 4100 at the end of stage III. The exponential growth phase of the microorganisms appeared in stages IV and V when a significant exponential increase in MLSS concentration was observed. During the operation of the last stage, the growth of microorganisms reached stability and the MLSS concentration was around 6680 (SD=80 mg/L).

Fig. 8. MLSS and MLVSS concentrations during experimental operational stages. Figure options

In order to investigate the impact of MLSS variation on the membrane fouling in each experimental stage, the concept of percentage reduction in permeation flux (PRPF) developed by Bani-Melhem and Elektorowicz [35]

was calculated in this study from: (7) where J0 is the initial permeation flux measured during the first minute in each stage and J is the permeation flux at the end operation of the corresponded experimental stage. It is worth mentioning that the chemical and physical cleaning that were applied between the experimental stages achieved more than 99% recovery of membrane permeability. The percentage reduction in permeation flux (PRPF) is plotted in Fig. 9. As can be seen from Fig. 8 and Fig. 9, while the MLSS concentration increases from 2300 mg/L to 4100 mg/L at the end of stages I and III, respectively, no significant increase in the PRPF was observed; the increase in PRPF was only 1.8% (PRPF increased from 81.3% to 83.1%). However, a notable increase was observed during the operation of the last three stages when the PRPF increased from 83.1% to 88.0% at the end of stages III and VI when the MLSS concentration increased from 4100 mg/L to 6750 mg/L, respectively. This increase in the MLSS concentration might have play a significant role in decreasing the permeation flux more rapidly in the last three stages in comparison of the first three stages of the experimental operation. Generally speaking, it can be said that the dependence of membrane fouling on MLSS concentration is not significant at low level of MLSS concentration which was presented at the first three stages while the impact of MLSS concentration started to appear when the MLSS increased in the last three stages. This is in agreement with the results reported by Le-Clech et al. [36] who found that there was no notable difference in membrane fouling at low levels of MLSS concentrations; while a significant decrease of permeation rate was observed when the MLSS concentration increased sharply. On the other hand, Fig. 8 demonstrates a decrease in MLVSS/MLSS ratio from 0.81 to 0.77 at the end of stages III and VI, respectively. This indicated to the increase in the accumulation of inorganic materials inside the bioreactor which might have contribution in reducing the permeation flux in the last stages.

Fig. 9. Percentage reduction in permeation flux (PRPF) in each experimental stage. Figure options

Other parameters like microbial floc size may be another causing for membrane fouling. A sample of mixed liquor (ML) solution was taken at the end of each experimental stage and the mean particle size was analyzed as shown in Fig. 10. Apparently, Fig. 10 shows that there is no significant difference in the mean particle size during the whole experimental period, the mean particle diameters in all the experimental stages were around 54.7 µm. This means that the microbial floc size did not play a significant parameter in membrane fouling phenomenon.

Fig. 10. Particles size distribution of sludge suspensions in SMBR at the end of each experimental stage. Figure options

According to the literature, there are many other factors, which can affect the permeation flux in SMBR system [37]. In addition to the increases in the MLSS concentrations during the last operation, the increase in inorganic foulants might have a significant contribution in membrane fouling. Fig. 8 confirms this fact in which the percentage of inorganic foulants increased in the last three stages which in turn decreased the permeation flux with time shown in Fig. 6.

In addition to MLSS concentration, the microbial floc size and the inorganic foulants, the composition of the wastewater has a direct impact on the membrane fouling in SMBR applications [37]. Specifically, the COD concentration plays an important role in membrane bioreactor technology. It is clearly shown in Fig. 4(a) that the influent COD concentration ranged between 92 and 348 mg/L during the operation of stages I, II and III in comparison with a range between 304 and 668 mg/L during the last three stages which may have a significant impact on the permeation flux reduction. These results are totally in agreement with the results reported in the literature which demonstrated that the membrane fouling in SMBR technology is caused not only by the microbial floc, but also by the supernatant containing colloids and solutes [38], [39], [40] and [41]. In order to identify the fouling behaviour in a better manner, the hydraulic performance of the filtration processes was analyzed at the end of each experimental stage by applying the resistance-in-series model according to Darcy‫׳‬s law. Because the MLSS concentration has a direct impact on cake layer formation on the membrane surface [42] and in order to exclude the effect of MLSS concentrations on the permeation flux, the MLSS concentration taken at the end of each experimental stage was adjusted to 2300 mg/L (SD=125 mg/L). This is in order to have identical MLSS concentrations at the end of each stage similar to that at stage I as a reference concentration. Table 3 shows the results of the different fouling resistances at the end of each experimental stage using Eqs. (3), (4) and (5). The results indicated that the resistance of the cake layer (Rc) contributed largely to the total resistance (82.5–89.6%) taking into consideration that the impact of the MLSS concentration was eliminated in analyzing the fouling resistances. On the other hand, the contribution of the gel layer resistance (Rg ) in membrane fouling varied between 10.4% and 12.2% during the first three stages and its contribution increased up to 17.5% during the operation of the last stage. The elimination of the impact of the MLSS concentration confirmed that the contribution of the gel layer resistance increased in the last stages. It is well known that the gel layer (Rg ) accounts for pore plugging which is considered as irreversible membrane fouling due to the adsorption of foulants onto the membrane surface by colloids and solutes in the supernatant. The gel layer (Rg ) might be formed by biopolymers, such as soluble microbial products (SMP) and EPS or by inorganic foulants. Therefore, the concentration of EPS in mixed liquor (EPSs) and on membrane surface (EPSm ) was analyzed at the end of each experimental stage and the results are presented in Table 3. Apparently, the results of analyzing EPSs and EPSm did not show a significant difference during the operation of the first experimental stages. A small difference was observed in the last stages which might be partially due to the production of SMP (not measured in this study) as results of the increase in microbial growth. In summary, the results of Table 3 suggested that the cake layer is the predominate resistance of the membrane fouling with an average contribution of 86%. The contribution of the gel layer increased slightly in the last three stages due to many parameters that can play key roles in irreversible membrane fouling. Specifically, the increases in influent COD, organic foulants and microbial growth can be addressed as significant parameters in increasing the contribution of the gel layer. In this context, the significant increases in anionic surfactants in the last three stages might also be another important parameter in reducing the permeation flux. Therefore, the impact of anionic surfactants concentration in grey water on membrane fouling is addressed as a future research. Table 3. Membrane fouling resistances and variation in extracellular polymeric substances (EPS) during the experimental stages. Rc

Rg

EPS

SMBR system 1012 (m−1)

(%)

1011 (m−1)

(%)

EPSs (mg/gVSS)

EPSm (mg/gVSS)

Stage I

2.44

87.8

3.39

12.2

18

16

Stage II

2.61

89.6

3.03

10.4

21

18

Stage III

2.28

88.4

2.99

11.6

19

17

Stage IV

2.34

83.8

4.52

16.2

23

20

Stage V

2.04

84.2

3.83

15.8

27

22

Stage VI

2.11

82.5

4.48

17.5

26

21 Table options

4. Conclusions The results of this investigation confirmed that the SMBR system is an effective method for GW treatment and reuse with respect to physical impurities, organic matter and pathogenic content. In terms of physical impurities, the SMBR system exhibited excellent removal efficiencies for colour and turbidity. The total suspended solids were completely removed as well as the total coliforms retained by the UF membrane. In terms of organic matter, COD was reduced by an average of 88%. Ammonia nitrogen reduction was excellent with an average value of 88%. The performance of SMBR with respect to COD and NH3 –N removal was not adversely affected by even large variations in influent values. However, the reduction of total phosphorus (TP) was just 56%, and the reduction of anionic surfactants (AS) was around 73%. Removal efficiencies of TP and AS appeared to be related to fluctuations in influent concentrations. Missing of the back-flushing event and the operation of SMBR system at constant transmembrane pressure led to a significant increase in the fouling rate during the operation of all experimental stages. Additionally, the results of this study showed that the cake layer contributes significantly in membrane fouling. In conclusion, the SMBR treatment produces average effluent values of COD, TSS, and coliforms that can satisfy international and local guidelines for reuse for non-potable applications. It is hoped that continuing efforts to optimize SMBR treatment will render the reuse of grey water as a cost-effective asset in the overall water budget of many arid communities in the MENA region like Jordan. Future investigations aim at

integrating SMBR system with other physical/chemical treatment processes in order to reduce the membrane fouling and increase the performance while reducing costs of treatment of full-scale applications.

References [1]

E. Smith, K. Bani-Melhem Grey water characterization and treatment for reuse in an arid environment Water Sci. Technol., 66 (2012), pp. 72–78

View Record in Scopus | Full Text via CrossRef | Citing articles (5) [2]

M. Lamine, D. Samaali, A. Ghrabi Greywater treatment in a submerged membrane bioreactor with gravitational filtration Desalin. Water Treat., 46 (2012), pp. 182–187

View Record in Scopus | Full Text via CrossRef | Citing articles (2) [3]

M. Halalsheh, S. Dalahmeh, M. Sayed, W. Suleiman, M. Shareef, M. Mansour, M. Safi Grey water characteristics and treatment options for rural areas in Jordan Bioresour. Technol., 99 (2008), pp. 6635–6641

Article | [4]

PDF (173 K) | View Record in Scopus | Citing articles (30)

A.I. Schäfer, D.L. Nghiem, N. Oschmann Bisphenol A retention in the direct ultrafiltration of greywater J. Membr. Sci., 283 (2006), pp. 233–243

Article | [5]

PDF (458 K) | View Record in Scopus | Citing articles (26)

F. Li, K. Wichmann, R. Otterpohl Review of the technological approaches for grey water treatment and reuses Sci. Total Environ., 407 (2009), pp. 3439–3449

Article | [6]

PDF (372 K) | View Record in Scopus | Citing articles (99)

E. Friedler, M. Hadari Economic feasibility of on-site greywater reuse in multi-storey buildings Desalination, 190 (2006), pp. 221–234

Article | [7]

PDF (507 K) | View Record in Scopus | Citing articles (69)

K. Bani-Melhem, E. Smith Grey water treatment by a continuous process of an electrocoagulation unit and a submerged membrane bioreactor system Chem. Eng. J., 198–199 (2012), pp. 201–210

Article | [8]

PDF (928 K) | View Record in Scopus | Citing articles (15)

A. Huelgas, N. Funamizu Flat-plate submerged membrane bioreactor for the treatment of higher-load graywater Desalination, 250 (2010), pp. 162–166

Article | [9]

PDF (495 K) | View Record in Scopus | Citing articles (12)

C. Merz, R. Scheumann, B.E. Hamouri, M. Kraume Membrane bioreactor technology for the treatment of greywater from a sports and leisure club Desalination, 215 (2007), pp. 37–43

Article |

PDF (387 K) | View Record in Scopus | Citing articles (48)

[10] B. Lesjean, R. Gnirss Grey water treatment with a membrane bioreactor operated at low SRT and low HRT Desalination, 199 (2006), pp. 432–434

Article |

PDF (73 K) | View Record in Scopus | Citing articles (25)

[11] R. Liu, X. Huang, L. Chen, X. Wen, Y. Qian Operational performance of a submerged membrane bioreactor for reclamation of bath wastewater Process Biochem., 40 (2005), pp. 125–130

Article |

PDF (304 K) | View Record in Scopus | Citing articles (40)

[12] N. Oschmann, L.D. Nghiem, A.I. Schäfer Fouling mechanisms of submerged ultrafiltration membranes in greywater recycling Desalination, 179 (2005), pp. 225–233

[13] J.G. March, M. Gual, F. Orozco Experiences on greywater re-use for toilet flushing in a hotel (Mallorca Island, Spain) Desalination, 164 (2004), pp. 241–247

Article |

PDF (519 K) | View Record in Scopus | Citing articles (58)

[14] T. Itayama, M. Kiji, A. Suetsugu, N. Tanaka, T. Saito, N. Iwami, M. Mizuochi, Y. Inamori On site experiments of the slanted soil treatment systems for domestic gray water

Water Sci. Technol., 53 (2004), pp. 193–201

[15] M. Pidou, L. Avery, T. Stephenson, P. Jeffrey, S.A. Parsons, S. Liu, F.A. Memon, B. Jefferson Chemical solutions for greywater recycling Chemosphere, 7 (2008), pp. 147–155

Article |

PDF (166 K) | View Record in Scopus | Citing articles (31)

[16] C.-J. Lin, S.-L. Lo, C.-Y. Kuo, C.-H. Wu Pilot-scale electrocoagulation with bipolar aluminium electrodes for on-site domestic greywater reuse J. Environ. Eng., 131 (2005), pp. 491–495

View Record in Scopus | Full Text via CrossRef | Citing articles (16) [17] Kuntal A. Vakil, Meena K. Sharma, Akansha Bhatia, Absar A. Kazmi, Sudipta Sarkar Characterization of greywater in an Indian middle-class household and investigation of physicochemical treatment using electrocoagulation Sep. Purif. Technol., 130 (2014), pp. 160–166

Article |

PDF (985 K) | View Record in Scopus | Citing articles (2)

[18] K.-J. Hwang, C.-S. Chan, K.-L. Tung Effect of backwash on the performance of submerged membrane filtration J. Membr. Sci., 330 (2009), pp. 349–356

Article |

PDF (1651 K) | View Record in Scopus | Citing articles (24)

[19] P. Schoeberl, M. Brik, M. Bertoni, R. Braun, W. Fuchs Optimization of operational parameters for a submerged membrane bioreactor treating dyehouse wastewater Sep. Sci. Technol., 44 (2005), pp. 61–68

Article |

PDF (318 K) | View Record in Scopus | Citing articles (48)

[20] N. Nouri, M.R. Mehrnia, M.H. Sarrafzadeh, R. Nabizadeh Performance of membrane bioreactor in presence of flocculants Desalin. Water Treat., 52 (2014), pp. 2933–2938

View Record in Scopus | Full Text via CrossRef | Citing articles (1) [21] J. Wu, X. Huang Effect of dosing polymeric ferric sulfate on fouling characteristics, mixed liquor properties and performance in a long-term running membrane bioreactor Sep. Purif. Technol., 63 (2008), pp. 45–52

Article |

PDF (751 K) | View Record in Scopus | Full Text via CrossRef | Citing articles (38)

[22] J.C. Lee, J.S. Kim, I.J. Kang, M.H. Cho, P.K. Park, C.H. Lee Potential and limitations of alum or zeolite addition to improve the performance of a submerged membrane bioreactor Water Sci. Technol., 43 (2001), pp. 59–66

View Record in Scopus | Citing articles (70) [23] A.Y. Hu, D.C. Stuckey Activated carbon addition to a submerged anaerobic membrane bioreactor: effect on performance, transmembrane pressure, and flux J. Environ. Eng., 133 (2007), pp. 73–80

View Record in Scopus | Full Text via CrossRef | Citing articles (37) [24] K. Bani-Melhem (Ph.D. Dissertation) Development of a Novel Submerged Membrane Electro-Bioreactor for Wastewater TreatmentConcordia University, Montreal, Quebec, Canada (2008)

[25] M. Elektorowicz, J. Oleszkiewicz, K. Bani-Melhem, Submerged Membrane Electro-Bioreactor – SMEBR, US Patent 12/553,680; 2009. [26] L. Jie, L. Liu, F. Yang, F. Liu, Z. Liu The configuration and application of helical membrane modules in MBR J. Membr. Sci., 392–393 (2012), pp. 112–121

Article |

PDF (2721 K) | View Record in Scopus | Citing articles (9)

[27] I.S. Chang, C.H. Lee, K.H. Ahn Membrane filtration characteristics in membrane-coupled activated sludge system: the effect of floc structure on membrane fouling Sep. Sci. Technol., 34 (1999), pp. 1743–1758

View Record in Scopus | Full Text via CrossRef | Citing articles (66) [28] M. Al-Shannag, K. Bani-Melhem, Z. Al-Anber, Z. Al-Qodah Enhancement of COD-nutrients removals and filterability of secondary clarifier municipal wastewater influent using electrocoagulation technique

Sep. Sci. Technol., 48 (2013), pp. 673–680

View Record in Scopus | Full Text via CrossRef | Citing articles (4) [29] American Public Health Association, American Water Works Association, Water Environment Federation Standard Methods for the Examination of Water and Wastewater (20th ed.)American Public Health Association, Washington, DC, USA (1998)

[30] I.S. Chang, C.H. Lee Membrane filtration characteristics in membrane-coupled activated sludge system – the effect of physiological states of activated sludge on membrane fouling Desalination, 120 (1998), pp. 221–233

Article |

PDF (815 K) | View Record in Scopus | Citing articles (343)

[31] S. Rosenberger, U. Kruger, R. Witzig, W. Manz, U. Szewzyk, M. Kraume Performance of a bioreactor with submerged membranes for aerobic treatment of municipal waste water Water Res., 36 (2002), pp. 413–420

Article |

PDF (191 K) | View Record in Scopus | Citing articles (219)

[32] K. Bani-Melhem, M. Elektorowicz Performance of the submerged membrane electro-bioreactor (SMEBR) with iron electrodes for wastewater treatment and fouling reduction J. Membr. Sci., 379 (2011), pp. 434–439

Article |

PDF (497 K) | View Record in Scopus | Citing articles (18)

[33] C. Adam, R. Gnirss, B. Lesjean, H. Buisson, M. Kraume Enhanced biological phosphorous removal in membrane bioreactors Water Sci. Technol., 46 (2002), pp. 281–286

View Record in Scopus | Citing articles (31) [34] Jordan Standards and Metrology Organization Guidelines for Reuse of Treated Grey Water (Report JS 1776) Jordan Standards and Metrology Organization, Amman, Jordan (2008)

[35] K. Bani-Melhem, M. Elektorowicz Development of a novel submerged membrane electro-bioreactor (SMEBR): performance for fouling reduction Environ. Sci. Technol., 44 (2010), pp. 3298–3304

View Record in Scopus | Full Text via CrossRef | Citing articles (38) [36] P. Le-Clech, B. Jefferson, S.J. Judd Impact of aeration, solids concentration and membrane characteristics on the hydraulic performance of a membrane bioreactor J. Membr. Sci., 218 (2003), pp. 117–129

Article |

PDF (276 K) | View Record in Scopus | Citing articles (175)

[37] P. Le-Clech, V. Chen, T.A.G. Fane Fouling in membrane bioreactors used in wastewater treatment J. Membr. Sci., 284 (2006), pp. 17–53

Article |

PDF (976 K) | View Record in Scopus | Citing articles (841)

[38] S. Judd Fouling control in submerged membrane bioreactors Water Sci. Technol., 51 (2005), pp. 27–34

View Record in Scopus | Citing articles (39) [39] W. Lee, S. Kang, H. Shin Sludge characteristics and their contribution to microfiltration in submerged membrane bioreactors J. Membr. Sci., 216 (2003), pp. 217–227

Article |

PDF (179 K) | View Record in Scopus | Citing articles (353)

[40] L. Defrance, M.Y. Jaffrin, B. Gupta, P. Paullier, V. Geaugey Contribution of various constituents of activated sludge to a membrane bioreactor fouling Bioresour. Technol., 73 (2000), pp. 105–112

Article |

PDF (713 K) | View Record in Scopus | Citing articles (225)

[41] N. Yamato, K. Kimura, T. Miyoshi, Y. Watanabe Difference in membrane fouling in membrane bioreactors (MBRs) caused by membrane polymer materials J. Membr. Sci., 280 (2006), pp. 911–919

Article |

PDF (490 K) | View Record in Scopus | Citing articles (101)

[42] I.-S. Chang, S.-N. Kim Wastewater treatment using membrane filtration-effect of biosolids concentration on cake resistance Process Biochem., 40 (2005), pp. 1307–1314

Article |

PDF (389 K) | View Record in Scopus | Citing articles (86)

Correspondence to: Chemical Engineering Department, Taibah University P.O. Box 344, Madinah, KSA. Fax: +966 48475837. 1

Present address: Chemical Engineering Department, Al-Balqa Applied University, P.O. Box 340558, Marka, Amman 11134, Jordan.

Copyright © 2014 Elsevier B.V. All rights reserved.

About ScienceDirect

Contact and support

Terms and conditions

Privacy policy

Copyright © 2015 Elsevier B.V. or its licensors or contributors. ScienceDirect® is a registered trademark of Elsevier B.V.

Cookies are used by this site. To decline or learn more, visit our Cookies page. Sw itch to Mobile Site

Recommended articles

Grey water treatment in urban slums by a filtratio… 2014, Journal of Environmental Management

m ore

Effects of hydrophilicity/hydrophobicity of membr… 2015, Bioresource Technology

m ore

Modeling biodegradation characteristics of grey … 2013, Journal of Membrane Science

View more articles »

Citing articles (0)

Related book content

m ore