The role of a combined coagulation and disk filtration process as a pre-treatment to microfiltration and reverse osmosis membranes in a municipal wastewater pilot plant

Chemosphere 117 (2014) 20–26

Contents lists available at ScienceDirect

Chemosphere journal homepage: www.elsevier.com/locate/chemosphere

Technical Note

The role of a combined coagulation and disk filtration process as a pre-treatment to microfiltration and reverse osmosis membranes in a municipal wastewater pilot plant Kangmin Chon a,b, Jaeweon Cho c, Seung Joon Kim d, Am Jang e,⇑ a

Jeju Global Research Center (JGRC), Korea Institute of Energy Research (KIER), 200 Haemajihaean-ro, Gujwa-eup, Jeju-si, Jeju-do 695-971, Republic of Korea School of Environmental Science and Engineering, Gwangju Institute of Science and Technology (GIST), 261 Cheomdan-gwagiro, Buk-gu, Gwangju 500-712, Republic of Korea Department of Civil and Environmental Engineering, College of Engineering, Yonsei University, 50 Yonsei-ro, Seodaemun-gu, Seoul 120-749, Republic of Korea d Environmental Technology Research Laboratory, Research and Business Development Center, Kolon Global, 199-5 Jeondae-ri, Pogok-eup, Cheoin-gu, Yongin, Gyeonggi-do 449-815, Republic of Korea e School of Civil and Architecture Engineering, Sungkyunkwan University, 2066 Seobu-ro, Jangan-Gu, Suwon, Gyeonggi-do 440-746, Republic of Korea b c

h i g h l i g h t s

g r a p h i c a l a b s t r a c t

 CC–DF was effective to reduce

turbidity associated with cake layer formation.  Residual coagulants after CC–DF could lead to formation of aluminium–EfOM complexes.  Aluminium–EfOM complexes had a higher MW than the pore size of MF membranes.  Formation of aluminium–EfOM complexes might increase the fouling potential of EfOM.  Control of residual coagulants was required to use CC–DF as a pretreatment of MF and RO membranes.

a r t i c l e

i n f o

Article history: Received 17 February 2014 Received in revised form 7 May 2014 Accepted 8 May 2014

Handling Editor: O. Hao Keywords: Combined coagulation–disk filtration Effluent organic matter Microfiltration Municipal wastewater reclamation Reverse osmosis

: Hydrophobic fractions Effluent organic matter (EfOM)

http://dx.doi.org/10.1016/j.chemosphere.2014.05.042 0045-6535/Ó 2014 Elsevier Ltd. All rights reserved.

CC-DF process

Coagulants: Al2(SO4)3 (dosage: 3 mg Al L-1)

Formation of Al–EfOM complexes (Molecular weight > 0.1 µm) –Al

–Al

–Al

–Al

–Al

–Al

–Al

–Al

–Al

a b s t r a c t A pilot study was conducted to assess the performance of a municipal wastewater reclamation plant consisting of a combined coagulation–disk filtration (CC–DF) process, microfiltration (MF) and reverse osmosis (RO) membranes, in terms of the removal of water contaminants and changes in characteristics of effluent organic matter (EfOM). The CC–DF and MF membranes were not effective for the removal of dissolved water contaminants. However, they could partially reduce the turbidity associated with the cake layer formation by particulate materials on the membrane surfaces. Furthermore, most of water contaminants were completely removed by the RO membranes. Although the CC–DF process could remove approximately 20% of turbidity, the aluminium concentrations considerably increased after the CC–DF process due to the residual coagulants complexed with both carboxylic acid and alcohol functional groups of EfOM. Those aluminium–EfOM complexes had a lower negative charge and higher molecular weight (>0.1 lm pore size of the MF membranes) compared to non-complexed EfOM. These results indicate that the control of the formation of the aluminium–EfOM complexes should be considered as a key step to use the CC–DF process as a pre-treatment of the MF and RO membranes for mitigation of membrane fouling in the tested pilot plant. Ó 2014 Elsevier Ltd. All rights reserved.

⇑ Corresponding author. Tel.: +82 31 290 7526; fax: +82 31 290 7549. E-mail address: [email protected] (A. Jang).

: Hydrophilic fractions

21

K. Chon et al. / Chemosphere 117 (2014) 20–26

1. Introduction Municipal wastewater reclamation and reuse have become an attractive option to overcome the problem of water scarcity caused by industrialisation, urbanisation, rapid growth of population, and climate change (Wintgens et al., 2005; Chon et al., 2012a). Nevertheless, the direct use of the reclaimed wastewater for drinking and/or irrigation may provide adverse impacts on the aquatic ecosystem and/or human health due to its high contents of heavy metals, metalloids, and micropollutants, including pharmaceuticals and personal care products, endocrine disrupting chemicals, disinfection by-products and other persistent organic compounds (Tanaka et al., 2001; Kolpin et al., 2002; Kim and Aga, 2007; Alturki et al., 2012; Chon et al., 2013a). Therefore, the application of a tertiary treatment process to remove micropollutants (i.e., pharmaceuticals and personal care products, endocrine disrupting chemicals, disinfection by-products and other persistent organic compounds) from secondary effluents after conventional biological wastewater treatment is considered as an essential step for municipal wastewater reclamation and reuse (Chon et al., 2012b). Among various wastewater treatment processes, membrane processes, including membrane bioreactor (MBR), microfiltration (MF), ultrafiltration (UF), nanofiltration (NF), reverse osmosis (RO), and membrane distillation, have received great attention as a promising technology for reclamation and reuse of municipal wastewater due to its high efficiency and cost effectiveness (Wang et al., 2011; Alkhudhiri et al., 2012; Chon et al., 2012c). In general, biological wastewater treatment processes (i.e., activated sludge and MBR) and/or low-pressure membrane processes (i.e., MF and UF) are not effective for the removal of dissolved organic, inorganic materials, nutrients and micropollutants (Qin et al., 2004; Bellona and Drewes, 2007; Kim et al., 2007). Thus, high-pressure membrane processes (e.g., NF and/or RO) are usually applied to municipal wastewater reclamation systems after MF/UF or MBR for the removal of dissolved water contaminants, mainly micropollutants from secondary effluents (Chon et al., 2011, 2012b). Based on these reasons, the dual membrane process composed of MF/UF and RO is regarded to be one of the best combinations to meet the water quality criteria for municipal wastewater reclamation and reuse (Del Pino and Durham, 1999; Comerton et al., 2005). In recent years, there has been a rapid growth and development of membranes technologies, in terms of membrane properties related to fouling resistance (e.g., hydrophobicity), operating and management protocols (e.g., pre-treatment, procedures for backwashing, physical and chemical cleaning, dosage of anti-fouling and anti-scaling agents) (Chon et al., 2013b). However, membrane fouling caused by effluent organic matter (EfOM) composed of natural organic matter derived from drinking water and soluble microbial products produced during biological wastewater treatment processes, such as activated sludge and MBR, remains a major obstacle for practical applications of membrane processes

Coagulants: Al 2(SO4)3 (dosage: 3 mg Al L -1)

Ansan wastewater treatment plant

in wastewater reclamation and reuse since it may lead to flux decline which can increase the operating cost (Choi et al., 2006; Choi and Ng, 2008; Her et al., 2008; Chon et al., 2012a). Previous studies reported that hydrophilic fractions of EfOM capable of passing through pre-treatments (i.e., coagulation, MBR, MF and UF) play a key role in the formation of membrane fouling on the surfaces of NF or RO membranes (Jarusutthirak et al., 2002; Li et al., 2007). Even though many studies have been performed to investigate the effects of pre-treatment methods on fouling mitigation of membranes processes for wastewater reclamation and reuse (Chon et al., 2012a), the knowledge on a combined coagulation and disk filtration (CC–DF) process as a pre-treatment of the integrated membrane system (i.e., MF/UF and RO membranes) is still lacking as coagulation has been commonly used as a pre-treatment of MF/UF and RO membranes during the past decades (Chon et al., 2012c). Recent studies have investigated the effects of a CC– DF process on fouling characteristics of UF and RO membranes (Chon et al., 2012a, 2013b). Nevertheless, they have primarily focused on characterisation of membrane flaunts to elucidate fouling mechanisms in a pilot plant. Therefore, there is a great need for investigating changes in physicochemical characteristics of EfOM by a CC–DF process (i.e., functional group compositions and relative hydrophobicity) which are closely related to the formation of EfOM complexes with coagulants. The main aim of this study is to evaluate the removal performance and mechanism of water contaminants, in terms of organic materials, metals, metalloids and nutrients (i.e., nitrogen and phosphorous species) through the municipal wastewater reclamation pilot plant consisting of a CC–DF process, MF and RO membranes. Furthermore, the physicochemical properties of EfOM in the raw and treated wastewaters were rigorously characterised using various analytical methods to provide valuable insights into the viability of the CC–DF process as a pre-treatment of the MF and RO membranes for membrane fouling mitigation. 2. Materials and methods 2.1. Configuration of the pilot plant During this study, the tested municipal wastewater reclamation pilot plant comprised of the CC–DF, MF and RO membranes was continuously operated at Ansan wastewater treatment plant (Ansan, Gyeonggi-do, Korea), as illustrated in Fig. 1. Secondary effluents of Ansan wastewater treatment plant (498.5 m3 d 1) equipped with a biological nitrification–denitrification process were treated by coagulation (dosage of aluminium sulphate (Al2(SO4)3): 3 mg Al L 1) and coagulated particles were directly removed by DF consists of two layers of mesh fabrics (effective surface area: 10.5 m2; recovery efficiency: 97%; PETEX, Sefar, Heiden, Switzerland) and then filtered with hollow fibre MF membranes (effective surface area: 126 m2; recovery efficiency: 92%; Kolon Industries, Gwacheon, Gyeonggi-do, Korea). The sodium

Backwashing

RO retentate

Secondary effluents

Combined coagulation and disk filtration process

Microfiltration membranes

Reverse osmosis membranes

Reservoir for RO permeate

Fig. 1. Configuration of the municipal wastewater reclamation pilot plant consisting of the CC–DF, MF and RO membranes.

22

K. Chon et al. / Chemosphere 117 (2014) 20–26

hypochlorite (13.2 mg L 1) was used for backwashing with an air stripping every half hours for 30 s to minimise the membrane pore blocking. The permeate from the MF membrane modules was additionally treated by spiral wound RO membranes (effective surface area: 148.8 m2; recovery efficiency: 73%; Woongjin Chemical, Seoul, Korea). Therefore, four different water samples were used to assess the removal of dissolved water contaminants (e.g., dissolved organic carbon (DOC), nitrogen, phosphorous, heavy metals and metalloids) and investigate the characteristics of EfOM in the municipal wastewater effluents: (i) feed water, (ii) CC–DF permeate, (iii) MF permeate and (iv) RO permeate. Detailed information of tested MF and RO membranes are summarised in Table 1 (Chon et al., 2012a).

2.2. Analytical methods All the samples were filtered with 0.45 lm MF filters (Advantec, Tokyo, Japan) before all analyses. The turbidity was monitored using an online turbidimeter (1720E, Hach, Loveland, USA). Concentrations of DOC and total nitrogen (TN) were determined by the catalytically aided combustion total organic carbon analyser which is equipped with a TN analyser and a non-dispersive Infrared detection (TOC-VCPH with TNM-1 unit, Shimadzu, Kyoto, Japan) (Chon et al., 2013c). Aromatic chromophores of the water samples were investigated using an UV–Vis spectrophotometer at 254 nm (UV-1601, Shimadzu, Kyoto, Japan) and the specific UV absorbance (SUVA) value was calculated from the ratio of UV absorbance (UVA) at 254 nm to DOC concentration. The concentration of most metals and metalloids in feed water and treated wastewater were analyzed using inductively coupled plasma-mass spectrometry (7500ce, Agilent, Santa Clara, CA, USA). Anions (e.g., nitrite, nitrate, and phosphate) and cations (e.g., ammonium, calcium and magnesium) were measured using ion chromatography (DX120, Dionex, Sunnyvale, CA, USA) with equipped with IonPac AS 14 and CS12A Columns (Dionex, Sunnyvale, CA, USA), and concentrations of total phosphorous (TP) were investigated by ascorbic acid method at 880 nm (4500-P E, American Public Health Association (APHA), 1998). Analyses for molecular weight distribution of aromatic and protein-like substances were performed by high performance size exclusion chromatography equipped with a Protein-Pak 125 column (Waters, Milford, MA, USA), UVA (SPD-10AVP, Shimadzu, Kyoto, Japan) at 254 nm and fluorescence detections (RF-10AXL, Shimadzu, Kyoto, Japan) at excitation and emission wavelength of 278 nm and 353 nm, respectively. A fluorescence spectrophotometer (F-2500, Hitachi, Tokyo, Japan) was used to confirm fluorescence characteristics of EfOM, and Raman scattering peaks of the samples were subtracted using a deionised water blank (Chon et al., 2013d). Functional group compositions were confirmed by Fourier transform infrared (FT/IR) spectrometer (FT/IR60 plus, Jasco, Tokyo, Japan) with a KBr pellet (Pike, Madison, WI, USA). Relative hydrophobicity of EfOM in water samples were

determined using the fractionation technique with Amberlite XAD-8/4 resins (Supelco, Bellefonte, PA, USA). 3. Results and discussion 3.1. Removal of bulk parameters Changes in water characteristics, such as pH, conductivity, turbidity, and SUVA and removal of water contaminants, including DOC, TN, metals, and metalloids through the CC–DF, MF and RO membranes were monitored during the operation of the pilot plant (Table 1). The pH of the MF permeated was reduced from 6.9 to 5.7 after the RO membranes as carbon dioxide could pass through the RO membranes while bicarbonate was rejected by the RO membranes (Farhat et al., 2013). The conductivity and DOC were not effectively removed by the CC–DF and MF membranes whereas the turbidity was substantially reduced after the CC–DF (>20%) and MF membranes (>97%), meaning that the CC–DF process can mitigate the flux decline of the MF membranes caused by the cake layer formation of particulate materials on the membrane surfaces (Howe and Clark, 2002). In the case of the RO membranes, they could remove approximately 97% of conductivity and 93% of DOC from the MF permeate. Through the RO membranes, SUVA value in the feed water was considerably decreased from 1.39 to 0.29 L mg 1 m 1 (removal efficiency: 79%). This observation indicates that the electrostatic interaction between humic substances with a negative charge and negative surface charges of the RO membranes strongly contributed to the preferential removal of hydrophobic fractions of EfOM (Her et al., 2007). However, the electrostatic interaction between humic substances and membrane surfaces may be reduced when multivalent metal ions complexes with humic substances (Yoon et al., 1998). 3.2. Removal of metals and metalloids Variations in the concentrations of inorganic materials in the feed and treated wastewater samples through the pilot plant are summarised in Table 2. Most of metals and metalloids were inefficiently removed by the CC–DF process but significant changes were observed for the concentrations of aluminium and iron. The concentration of aluminium was increased from 12 lg L 1 to 29 lg L 1 after the CC–DF process due to the residual coagulants (i.e., aluminium sulphate) whereas the concentrations of iron and zinc were decreased from 31 to 6 lg L 1 and 20 to 13 lg L 1, respectively. Based on these findings, it can be postulated that approximately 80% and 35% of iron zinc were in the form of the metal–EfOM complexes with high MW which were larger than the pore size of the DF (>10 lm). The MF membranes were also not effective for the removal of metals and metalloids, with the exception of aluminium. A considerable amount of aluminium (>38%) in the CC–DF permeate was removed by the MF mem-

Table 1 Characteristics of the DF, MF and RO membranes. Type

Materials

Pore size (lm)

MWCO (Da)

Contact angle (°)

Zeta potential at pH 7 (mV)

Roughness (nm)

DF

PEa

1st layer: 15 2nd layer: 10

N.A.

N.A.

N.A.

N.A.

MF

PVDFb

0.1

N.A.

N.A.

N.A.

N.A.

RO

PA TFCc

N.A.

100–400d

30.4d

35.5d

PE: polyethylene; PVDF: polyvinylidene fluoride; PA TFC: polyamide thin-film composite; MWCO: molecular weight cut off; N.A.: not available. a Sefar. b Kolon Membrane. c Woongin Chemical. d Chon et al. (2012a).

63.5d

23

K. Chon et al. / Chemosphere 117 (2014) 20–26

branes. This result implies that a large portion of the residual aluminium sulphates from the coagulants was in the form of the aluminium–EfOM complexes which were bigger than the pore size of the MF membranes (>0.1 lm). In contrary to the CC–DF and MF membranes, the RO membranes were effective for the removal of most metals and metalloids. Nevertheless, boron was poorly removed by the RO membranes due to its non-ionised form at neutral pH (Bick and Oron, 2005). Hence, an additional step for pH control should be considered to enhance the removal of boron by the RO membranes (pKa of boric acid: 9.24) (Chon et al., 2012b). 3.3. Removal of nutrients Concentrations of nitrogen and phosphorous species in the feed and treated wastewaters are listed in Table 2. Nitrate ions were found to be the dominant nitrogen species for the feed water (>97%), CC–DF permeate (>96%), and the MF permeate (>97%) since a biological nitrification–denitrification process was used as a second stage of the municipal wastewater treatment. The nitrate ions were not sufficiently removed by the CC–DF and MF membranes while the RO membranes could substantially remove them from the MF permeate (removal efficiency >86%) due to the combined effects of both size exclusion and charge repulsion of the RO membranes (Lee and Lueptow, 2001). The removal trends of phosphorous species through the pilot plant were similar to those of nitrogen species. However, the removal efficiency of phosphate ions was higher than that of nitrate ions since the RO membranes could remove multivalent ions more effectively compared to monovalent ions (Miller et al., 2013). In the feed and treated wastewaters, TP concentrations were primarily comprised of phosphate ions (feed: 81%; CC–DF permeate: 76%; MF permeate: 92%). Although phosphorous species were inefficiently removed by the CC–DF process (removal efficiency of TP: 10%; removal efficiency of phosphate ions: 15%), both the MF (removal efficiency of TP: 47%; removal efficiency of phosphate ions: 36%) and RO membranes (removal efficiency of TP: 85%; removal efficiency of phosphate ions: 98%) were effective for the removal of TP and phosphate ions. Therefore, the amount of DOC and TN in the

reclaimed wastewater through the pilot plant could meet the draft regulations for wastewater reclamation and reuse proposed by California Department of Public Health (DOC < 0.5 mg L 1; TN < 5 mg L 1) (California Department of Public Health, 2013). 3.4. Removal of EfOM 3.4.1. Fluorescence spectroscopy 3-Dimensional fluorescence excitation–emission matrix of the feed and treated wastewaters was investigated using a fluorescence spectrophotometer (Fig. 2). It has been known that a fluorescence peak at excitation (Ex) = 280 nm/emission (Em) = 340 nm is indicative of the protein-like fluorescence of bovine serum albumin and the humic-like fluorescence peaks of Suwannee River humic acid and Suwannee River fulvic acid can be found at Ex = 270 nm/Em = 450 nm (Suwannee River humic acid I), Ex = 315 nm/Em = 440 nm (Suwannee River humic acid II) and Ex = 260 nm/Em = 440 nm (Suwannee River fulvic acid I), Ex = 320 nm/Em = 440 nm (Suwannee River fulvic acid II), respectively (Chon et al., 2011). A strong humic-like fluorescence peak was observed for the feed water at Ex = 340 nm and Em = 420 nm (maximum intensity: 3,192 mV) and its intensity was slightly reduced after the CC–DF process (maximum intensity: 3065 mV). However, its decreasing rate (4%) was substantially lower than that of the DOC concentrations (removal efficiency of DOC: 12%), which was probably due to the aluminium–EfOM complexes. The humiclike chromophores in the CC–DF permeate were poorly removed by the MF membranes (maximum intensity: 3016 mV) and then completely eliminated through the RO membranes. Consequently, a weak protein-like fluorescence peak was found for the RO permeate at Ex = 280 nm and Em = 360 nm (maximum intensity: 105 mV). These removal patterns were consistent with the removals of DOC, SUVA, and aluminium in the feed and treated wastewaters. 3.4.2. High performance size exclusion chromatography analysis The UVA and fluorescence chromatograms of aromatic and protein-like substances in the feed and treated wastewaters as func-

Table 2 Removal of water contaminants through the municipal wastewater reclamation pilot plant consisting of the CC–DF, MF and RO membranes. NOD

Feed

CC–DF permeate

MF permeate

RO permeate

1 1 3 3 3 3

6.9 362 1.34 (±0.16) 5.6 (±0.6) 0.091 (±0.001) 1.64 (±0.17)

7.0 376 1.07 (±0.49) 4.9 (±0.3) 0.068 (±0.001) 1.39 (±0.08)

6.9 341 0.03 (±0.0) 4.82 (±0.1) 0.069 (±0.000) 1.39 (±0.04)

5.7 11 N.D. 0.4 (±0.1) 0.001 (±0.000) 0.29 (±0.09)

Al (lg L 1) Ca (mg L 1) Co (lg L 1) Cr (lg L 1) Fe (lg L 1) Mg (mg L 1) Mn (lg L 1) Ni (lg L 1) Zn (lg L 1) As (lg L 1) B (lg L 1)

2 2 2 2 2 2 2 2 2 2 2

11.8 (±0.4) 33 (±5) 0.8 (±0.1) 2.0 (±0.3) 31.4 (±3.0) 10.7 (±4.3) 16.1 (±8.0) 8.6 (±1.4) 20 (±1) 1.7 (±0.1) 45.6 (±9.5)

28.7 (±0.1) 32 (±4) 0.6 (±0.2) 2.0 (±0.4) 6.3 (±0.8) 10.4 (±4.0) 15.2 (±8.1) 8.3 (±1.1) 13 (±1) 1.6 (±0.0) 48.6 (±7.1)

17.6 (±0.4) 31 (±7) 0.6 (±0.2) 2.0 (±0.5) 6.8 (±1.6) 9.8 (±4.2) 15.3 (±8.2) 7.4 (±1.1) 13 (±0) 1.5 (±0.2) 48.9 (±7.4)

2.8 (±1.0) 0. (±0) 0.2 (±0.0) 0.1 (±0.1) 0.6 (±0.7) 0.0 (±0.0) 0.2 (±0.1) 0.1 (±0.0) 3 (±1) 0.2 (±0.2) 41.0 (±4.9)

TN (mg L 1) NO2 (mg N L 1) NO3 (mg N L 1) NH+4 (mg N L 1) TP (mg L 1) PO34 (mg P L 1)

2 2 2 2 2 2

7.2 (±0.2) N.D. 7.1 (±0.3) N.D. 1.09 (±0.26) 0.88 (±0.11)

7.3 (±0.1) N.D. 7.1 (±0.1) N.D. 0.99 (±0.14) 0.75 (±0.14)

7.0 (±0.2) N.D. 6.8 (±0.3) N.D. 0.52 (±0.12) 0.48 (±0.04)

0.9 (±0.2) N.D. 0.9 (±0.2) N.D. 0.08 (±0.00) 0.01 (±0.01)

pH Conductivity (lS cm 1) Turbidity (NTU) DOC (mg L 1) UVA at 254 nm (cm 1) SUVA (L mg 1 m 1) Metals

Metalloids Nutrients

NOD: numbers of measurements; N.D.: not detected.

24

K. Chon et al. / Chemosphere 117 (2014) 20–26

500

(a)

(Ex= 330 nm and Em= 410 nm)

450

Excitation (nm)

(b)

Humic-like fluorescence

Humic-like fluorescence (Ex= 330 nm and Em= 410 nm)

400

350

300

250 500

(d)

(c)

Humic-like fluorescence (Ex= 330 nm and Em= 410 nm)

Excitation (nm)

450

400

350 Protein-like substances (Ex = 280 nm and Em = 360 nm)

300

250 250

300

350

400

450

Emission (nm)

500 250

300

350

400

450

500

Emission (nm)

Fig. 2. Variations in fluorescence characteristics of EfOM through the municipal wastewater reclamation pilot plant consisting of the CC–DF, MF and RO membranes: (a) feed, (b) CC–DF permeate, (c) MF permeate, and (d) RO permeate.

tion of MW are provided in Fig. 3. The aromatic substances of EfOM in the feed water ranged from 580 to 1300 Da, with the highest peak of UVA detection at 580 Da. The relatively high MW fractions of the aromatic substances (MW: 960–1300 Da) were more effectively removed by the CC–DF than the relative low MW fractions of them (MW: 580–680 Da). The MF membranes were ineffective for the removal of both the low and high MW fractions due to its pore size (0.1 lm) while they were sufficiently removed by the RO membranes. Similar removal trends were observed in the removal of the protein-like substances through the pilot plant. The protein-like substances of the feed water were composed of small (90–310 Da) and large MWs (52 900 Da), with the highest peak of fluorescence detection at 210 Da. The CC–DF process effective for the removal of the high MW fractions (MW: 52 900 Da) whereas it was not effective form the removal of the low MW fractions (MW: 90–560 Da). The protein-like substances were poorly removed by the MF membranes and completely removed after the RO membranes also. From the changes in MW distribution of aromatic and protein-like substances through the pilot plant, it was confirmed that the CC–DF process could preferentially remove the high MW fractions of the aromatic and protein-like substances.

3.4.3. FT/IR analysis The infrared spectra of the feed and treated wastewaters were confirmed using Fourier transform infrared spectroscopy (Fig. 4). The strong peak in the region of 3500–3300 cm 1 is an indicative of N–H stretching of amides and the relatively strong peak of primary amides occurred in the region of 1680–1630 cm 1 that was

derived from peptidoglycans produced by lysis of microbial cells (Chon et al., 2013d). The O–H stretching of carboxylic acids from the hydrophobic components of EfOM were found in the region of 3100–2900, 1440–1395 and 960–875 cm 1. The peak in the region of 2900–2800 was associated with the CH band of aldehydes and the C–O stretching of alcohols from polysaccharides which were major components of bacterial cell was observed in the region of 1210–1100 and 1125–1090 cm 1 (Chon et al., 2010). Although EfOM in the feed and treated waters exhibited identical infrared peaks, the intensities of the strong infrared peaks in the feed water and the CC–DF permeate from the O–H stretching of carboxylic acids in the region of 1440–1395 cm 1 and 960– 875 cm 1 and the C–O stretching of alcohols in the region of 1210–1100 and 1125–1090 cm 1 were substantially reduced by the MF membranes. These results are evident that a decrease of the infrared peaks for carboxylic acids and alcohols were closely associated with the removal of the aluminium–EfOM complexes.

3.4.4. Structural analysis The structural analysis using Amberlite XAD-8/4 resins was conducted to investigate the relative hydrophobicity of EfOM in the feed and treated wastewaters (Fig. 5). Hydrophobic fractions of the feed water was approximately two times higher than its hydrophilic fractions (hydrophobic fractions: 45%, hydrophilic fractions: 26%, and transphilic fractions: 29%) and the CC–DF permeate was comprised of 42% of hydrophobic fractions, 35% of hydrophilic fractions, and 23% of transphilic fractions. This means that the increase of hydrophilic fractions after the CC–DF process

25

K. Chon et al. / Chemosphere 117 (2014) 20–26

30000

60

(a) 25000

580

20000

960

15000

1,300

10000

Hydrophobic Hydrophilic Transphilic

50

45

Fractions (%)

UV response (mV)

Feed CC-DF permeate MF permeate RO permeate

680

42

42

40

35 30

26

31

29

27 23

20

5000

10 0

Fluorescence response (mV)

(b)

Feed CC-DF permeate MF permeate RO permeate

210

75000

310 60000

560

90 45000

Feed

CC-DF permeate MF permeate

Fig. 5. Variations in the relative hydrophobicity of EfOM through the municipal wastewater reclamation pilot plant consisting of the CC–DF, MF and RO membranes.

52,900

30000

ute to the reduction in the permeate flux of the MF and RO membranes in the tested pilot plant.

15000

4. Conclusions

0 1e+1

1e+2

1e+3

1e+4

1e+5

Molecular weight (Da) Fig. 3. Changes in MW distribution of EfOM through the municipal wastewater reclamation pilot plant consisting of the CC–DF, MF and RO membranes: (a) aromatic substances and (b) protein-like substances.

1.2

3500-3300 N-H Amides

3100-2900 O-H Carboxylic acids

Feed CC-DF permeate UF permeate

2900-2800 CH Aldehydes

1440-1395 O-H 1210-1100 Carboxylic acids C-O Alcohols 1680-1630 1125-1090 C=O Amides C-O Alcohols

1.0

Absorbance

0

960-875 O-H Carboxylic acids

0.8

0.6

0.4 4000

3500

3000

2500

2000

1500

1000

Wavenumbers (cm -1) Fig. 4. Changes in the functional group compositions of EfOM through the municipal wastewater reclamation pilot plant consisting of the CC–DF, MF and RO membranes.

was probably due to the formation of EfOM complexes with the residual aluminium sulphates from the coagulants which were larger than the pore size of the DF (>10 lm). After the MF membranes, the hydrophobic and hydrophilic fractions were decreased to 42% and 31%, respectively, and the transphilic fraction was increased by 27%. These observations suggest that both hydrophobic and hydrophilic fractions of EfOM were intimately related to the formation of the aluminium–EfOM complexes which might contrib-

In this study, the municipal wastewater reclamation pilot plant comprised of the CC–DF, MF and RO membranes was tested to investigate its performance, with respect to the removal of organic materials, metals, metalloids and nutrients (i.e., nitrogen and phosphorous species), and the effects of the CC–DF process on characteristics of EfOM. The CC–DF and MF membranes were effective for the removal of turbidity related to the cake layer formation by particulate materials on the membrane surfaces but ineffective for the dissolved form of water contaminants, including conductivity, DOC, metals, metalloids, nitrogen, and phosphorous species. In opposition to the CC–DF and MF membranes, most of water contaminants were efficiently removed by the RO membranes, with the exception of boron due to its non-ionised form at the neutral pH (i.e., boric acids). Thus, the reclaimed wastewater might satisfy the draft regulations on DOC (<0.5 mg L 1) and TN (<5 mg L 1) for wastewater reclamation and reuse suggested by California Department of Public Health. An increase of aluminium concentrations after the CC–DF process was attributed to the residual coagulants in the form of the aluminium–EfOM complexes and a large portion of the aluminium–EfOM complexes had a higher MW (38%) than the pore size of the MF membranes (>0.1 lm) which were effectively removed by the MF membranes. Furthermore, the rigorous characterisation of variations in characteristics of EfOM through the pilot plant revealed that the use of the CC–DF process as a pre-treatment of the MF and RO membranes may stimulate the formation of the aluminium–EfOM complexes via complexation of carboxylic acids and alcohols with the residual coagulants which could play an important role in the fouling formation on the MF and RO membrane surfaces in the tested pilot plant.

Acknowledgements This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (No. 2013R1A1A2057920) and partially supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MEST) (No. 2012047029 (TOC)).

26

K. Chon et al. / Chemosphere 117 (2014) 20–26

References Alkhudhiri, A., Darwish, N., Hilal, N., 2012. Membrane distillation: A comprehensive review. Desalination 287, 2–18. Alturki, A., McDonald, J., Khan, S.J., Hai, F.I., Price, W.E., Nghiem, L.D., 2012. Performance of a novel osmotic membrane bioreactor (OMBR) system: flux stability and removal of trace organics. Bioresour. Technol. 113, 201–206. APHA, 1998. Standard Methods for the Examination of Water and Wastewater. APHA, Washington DC, USA. Bellona, C., Drewes, J.E., 2007. Viability of a low-pressure nanofilter in treating recycled water for water reuse applications: a pilot-scale study. Water Res. 41, 3948–3958. Bick, A., Oron, G., 2005. Post-treatment design of seawater reverse osmosis plants: boron removal technology selection for potable water production and environmental control. Desalination 178, 233–246. California Department of Public Health, 2013. Draft Regulations: Ground Water Replenishment with Recycled Water. California Department of Public Health, Sacramento, CA, USA. Choi, J.H., Ng, H.Y., 2008. Effect of membrane type and material on performance of a submerged membrane bioreactor. Chemosphere 71, 853–859. Choi, H., Zhang, K., Dionysiou, D.D., Oerther, D.B., Sorial, G.A., 2006. Effect of activated sludge properties and membrane operation conditions on fouling characteristics in membrane bioreactor. Chemosphere 63, 1699–1708. Chon, K., Lee, S., Chon, K., Hussain, A.A., Cho, J., 2010. Developing organic fouling indices of microfiltration and nanofiltration membranes for wastewater reclamation. Desalin. Water Treat. 18, 61–70. Chon, K., Sarp, S., Lee, S., Lee, J.H., Lopez-Ramirez, J.A., Cho, J., 2011. Evaluation of a membrane bioreactor and nanofiltration for municipal wastewater reclamation: trace contaminants control and fouling mitigation. Desalination 272, 128–134. Chon, K., Kim, S.J., Moon, J., Cho, J., 2012a. Combined coagulation–disk filtration process as a pretreatment of ultrafiltration and reverse osmosis membrane for wastewater reclamation: an autopsy study of a pilot plant. Water Res. 46, 1803– 1806. Chon, K., Shon, H.K., Cho, J., 2012b. Membrane bioreactor and nanofiltration hybrid system for reclamation of municipal wastewater: removal of nutrients, organic matter, micropollutants. Bioresour. Technol. 233, 181–188. Chon, K., Cho, J., Shon, H.K., Chon, K., 2012c. Advanced characterization of organic foulants of ultrafltration and reverse osmosis from water reclamation. Desalination 301, 59–66. Chon, K., Cho, J., Shon, H.K., 2013a. A pilot-scale membrane hybrid system for municipal wastewater reclamation: removal of nutrients and micropollutants, characterization of membrane foulants. Bioresour. Technol. 141, 109–116. Chon, K., Cho, J., Shon, H.K., 2013b. Fouling characteristics of a membrane bioreactor and nanofiltration hybrid system for municipal wastewater reclamation. Bioresour. Technol. 130, 239–247. Chon, K., Lee, Y., Jacqueline, T., von Gunten, U., 2013c. Quantification and characterization of dissolved organic nitrogen in wastewater effluents by electrodialysis treatment followed by size-exclusion chromatography with nitrogen detection. Water Res. 47, 5381–5391. Chon, K., Cho, J., Shon, H.K., 2013d. Advanced characterization of algogenic organic matter, bacterial organic matter, humic acids and fulvic acids. Water Sci. Technol. 67, 2228–2235.

Comerton, A.M., Andrews, R.C., Bagley, D.M., 2005. Evaluation of an MBR–RO system to produce high quality reuse water: microbial control, DBP formation and removal. Water Res. 39, 3982–3990. Del Pino, M.P., Durham, B., 1999. Wastewater reuse through dual-membrane processes: opportunities for sustainable water resources. Desalination 124, 271–277. Farhat, A., Ahmad, F., Hilal, N., Arafat, H.A., 2013. Boron removal in new generation reverse osmosis (RO) membranes using two-pass RO without pH adjustment. Desalination 310, 50–59. Her, N., Amy, G., Plottu-Pecheux, A., Yoon, Y., 2007. Identification of nanofiltration membrane foulants. Water Res. 41, 3936–3947. Her, N., Amy, G., Chung, J., Yoon, J., Yoon, Y., 2008. Characterizing dissolved organic matter and evaluating associated nanofiltration membrane fouling. Chemosphere 70, 495–502. Howe, K.J., Clark, M.M., 2002. Fouling of microfiltration and ultrafiltration membranes by natural waters. Environ. Sci. Technol. 36, 3571–3576. Jarusutthirak, C., Amy, G., Croué, J.P., 2002. Fouling characteristics of wastewater effluent organic matter (EfOM) isolates on NF and UF membranes. Desalination 145, 247–255. Kim, S., Aga, D.S., 2007. Potential ecological and human health impacts of antibiotics and antibiotic-resistant bacteria from wastewater treatment plants. J. Toxicol. Environ. Health, Part B 10, 559–573. Kim, S.D., Cho, J., Kim, I.S., Vanderford, B.J., Syder, S.A., 2007. Occurrence and removal of pharmaceuticals and endocrine disruptors in South Korea surface, drinking and waste waters. Water Res. 41, 1013–1021. Kolpin, D.W., Furlong, E.T., Meyer, M.T., Thurman, E.M., Zuggg, S.D., Barber, L.B., Buxton, H.T., 2002. Pharmaceuticals, hormones, and other organic wastewater contaminants in US streams, 1999–2000: a national reconnaissance. Environ. Sci. Technol. 36, 1202–1211. Lee, S., Lueptow, R.M., 2001. Membrane rejection of nitrogen compounds. Environ. Sci. Technol. 35, 3008–3018. Li, G., Xu, Z., Pinnau, I., 2007. Fouling of reverse osmosis membranes by biopolymers in wastewater secondary effluent: role of membrane surface properties and initial permeate flux. J. Membr. Sci. 290, 173–181. Miller, D.J., Huang, X., Li, H., Kasemset, S., Lee, A., Agnihotri, D., 2013. Foulingresistant membranes for the treatment of flowback water from hydraulic shale fracturing: a pilot study. J. Membr. Sci. 437, 265–275. Qin, J.J., Oo, M.H., Wai, M.N., Lee, H., Hong, S.P., 2004. Pilot study for reclamation of secondary treated sewage effluent. Desalination 171, 299–305. Tanaka, H., Yakou, Y., Takahashi, A., Higashitani, T., Komori, K., 2001. Comparison between estrogenicities estimated from DNA recombinant yeast assay and from chemical analyses of endocrine disruptors during sewage treatment. Water Sci. Technol. 43, 125–132. Wang, X.M., Sun, F.Y., Li, X.Y., 2011. Investigation of the role of biopolymer clusters in MBR membrane fouling using flash freezing and environmental scanning electron microscopy. Chemosphere 85, 1154–1159. Wintgens, T., Melin, T., Schiller, A., Khan, S., Muston, M., Bixio, D., Thoeye, C., 2005. The role of membrane processes in municipal wastewater reclamation and reuse. Desalination 178, 1–11. Yoon, S.H., Lee, C.H., Kim, K.J., Fane, A.G., 1998. Effect of calcium ion on the fouling of nanofilter by humic acidin drinking water production. Water Res. 32, 2180– 2186.