literature review BRILLIANT ANSWERS

A comparative study of adsorption of perﬂuorooctane sulfonate (PFOS) onto granular activated carbon, ion-exchange polymers and non-ion-exchange polymers S.T.M.L.D. Senevirathna a, * , S. Tanaka b, S. Fujii b, C. Kunacheva b, H. Harada b, B.R. Shivakoti b, R. Okamoto b aResearch Center for Environmental Quality Management, Graduate School of Engineering, Kyoto University, 1-2 Yumihama, Otsu, Shiga 520-0811, JapanbGraduate School of Global Environmental Studies, Kyoto University, Yoshida-Honmachi, Sakyo-ku, Kyoto 606-8501, Japan article info Article history:

Received 26 November 2009 Received in revised form 7 April 2010 Accepted 22 April 2010 Available online 23 May 2010 Keywords:

Adsorption PFOS Polymers Sorption isotherm Sorption kinetics abstract Perﬂuorooctane sulfonate (PFOS) is the latest chemical categorized as persistent organic pollutants (POPs). PFOS appears in the environmental water and tap water in ng L 1level. The process of adsorption has been identiﬁed as an effective technique to eliminate PFOS in water. Three non-ion-exchange poly- mers (DowV493, DowL493 and AmbXAD4), two ion-exchange polymers (DowMarathonA and AmbI- RA400) and one granular activated carbon (GAC) (Filtersorb400) were tested with regard to their sorption kinetics and isotherms at low PFOS concentrations (100–1000 ng L 1 equilibrium concentra- tions). The sorption capacities at 1 lgL 1 equilibrium concentration decreased in the following order:

Ion-exchange polymers > non-ion-exchange polymers > GAC, but at further low equilibrium concentra- tion (100 ng L 1) non-ion-exchange polymers showed higher adsorption capacity than other adsorbents.

In the case of sorption kinetics, GAC and ion-exchange polymers reached the equilibrium concentration within 4 h and AmbXAD4 within 10 h. DowV493 and DowL493 took more than 80 h to reach equilibrium concentration. AmbIRA400 was identiﬁed as the best ﬁlter material to eliminate PFOS at equilibrium con- centration >1000 ng L 1. Considering both adsorption isotherms and adsorption kinetics, AmbXAD4 and DowMarathonA were recommended to eliminate PFOS at ng L 1equilibrium concentration.

2010 Elsevier Ltd. All rights reserved. 1. Introduction Perﬂuorooctane sulfonate (PFOS) is an anthropogenic organic pollutant, which is recognized as an emerging problem in water environment due to its persistence, bio-accumulation, long range transportation and toxic effects. PFOS was categorized as persistent organic pollutants (POPs) in the 4th meeting of the conference of the parties to the Stockholm Convention in May 2009 (Earth Nego- tiations Bulletin, 2009). From the available literature, tap water and surface water samples in several countries were found to be contaminated with PFOS (Fujii et al., 2007; Lien et al., 2008; Jin et al., 2009;Quinete et al., 2009). For example in Osaka (Japan), PFOS concentration in portable tap water varies from 0.16 ng L 1 to 22.00 ng L 1 (Takagi et al., 2008). Also there is a positive corre- lation between the PFOS concentration in raw water and tap water samples suggesting minimum removal efﬁciency at conventional water puriﬁcation systems (Fujii et al., 2007; Takagi et al., 2008).Since the conventional techniques are not sufﬁcient to treat PFOS, different alternative treatment techniques have been sug- gested. The methods that have been already tested are UV–visible light irradiation (Hori et al., 2004), photochemical decomposition with persulfate ions (S 2O2 8)(Hori et al., 2005), membrane process (Tang et al., 2006) and sonochemical degradations (Moriwaki et al., 2005).

It has been demonstrated in many cases that, adsorption is an effective and economical method to remove many polar organic pollutants from water. However most of these researchers have fo- cused on industrial wastewater where PFOS concentration is in lgL 1 level. Few researchers have reported the effectiveness of some anion-exchange synthetic polymers (Yu et al., 2009) and acti- vated carbon (Schaefer, 2006;Yong et al., 2007;Ochoa-Herrera and Sierra-Alvarez, 2008;Yu et al., 2009) to eliminate PFOS in environ- mental water. The published data on adsorption of PFOS onto non- ion-exchange synthetic polymer materials is still not available.

The objectives of this study were to investigate and to compare the sorption behaviors of PFOS on ion-exchange and non-ion-ex- change commercial adsorbents including activated carbon at low equilibrium concentration (<1 lgL 1). In this study, the sorption kinetics and sorption isotherms of PFOS for Dowex polymers (DowV493, DowL493 and DowMarathonA), Amberlite polymers (AmbXAD4 and AmbIRA400) and Filtrasorb 400 (GAC) were stud- ied in detail. 0045-6535/$ - see front matter 2010 Elsevier Ltd. All rights reserved.

doi:10.1016/j.chemosphere.2010.04.053 *Corresponding author. Tel.: +81 8037850520.

E-mail addresses:[email protected](S.T.M.L.D. Senevirathna), [email protected](S. Tanaka),[email protected](S. Fujii), [email protected](C. Kunacheva),[email protected] (H. Harada),[email protected](B.R. Shivakoti),[email protected].

kyoto-u.ac.jp(R. Okamoto). Chemosphere 80 (2010) 647–651 Contents lists available atScienceDirect Chemosphere journal homepage: www.elsevier.com/locate/chemosphere 2. Materials and methods 2.1. PFOS and adsorbents PFOS (98%) was purchased from Wako Chemicals (Japan). Ion- exchange polymers, non-ion-exchange polymers and Filtrasorb 400 (GAC) were purchased from Dow Chemicals (USA) and Sigma Aldrich (USA).

2.2. Adsorbent pretreatment Prior to the use in the sorption experiment, synthetic polymers were ﬁrst washed in deionized water to remove dirt and then dried at 50 C until reached constant weight. Similarly, the coal-based activated carbon of Filtrasorb 400 was ﬁrst rinsed with deionized water for several times and then washed in 80 C deionized water for 2 h to remove the impurities. After being dried in an oven at 105 C for 48 h, they were crushed by a mortar and passed through 0.25–0.5 mm sieve.

2.3. Analytical equipments and methods for PFOS determination PFOS was directly analyzed and measured by using HPLC/MS/ MS. In 1200SL HPLC, 5 mM ammonium acetate and acetonitrile were used as mobile phases. A triple quadrapole MS/MS was used in multiple reaction monitoring (MRM) at negative ionization mode for the detection ofm/zof parent ion and daughter ion of PFOS.

2.4. Isotherm and kinetic experiment Adsorption isotherm experiments were performed with the six polymers using a bottle-point technique. For single-solute adsorp- tion isotherm experiments, weighed adsorbent (0.1 g) were placed into 125 mL PET bottles that have been ﬁlled with 75 mL deionized water. PFOS stock solutions were added to yield the initial concen- trations of 10, 30, 60, 120, 250, 500, 1000, 2000 and 5000 lgL 1 ones each bottle was ﬁlled to 100 mL level. After preparations, bot- tles were immediately closed, placed horizontally in a thermo sha- ker (EYELA-NTS4000) and shaken for 100 h at the speed of 150 RPM and 25 C. After shaking, two samples were taken from each bottle, centrifuged (4000 RPM, 10 min), diluted with acetoni- trile (40%) and analyzed with LC/MS/MS.

Similar type of shaking experiment was conducted for the ki- netic study with the initial concentration of 5000 lgL 1. pH was measured before the experiment and it was recorded as 6.4. Sam- pling was done at different time intervals of 0, 2, 4, 19, 27, 50, 74 and 100 h.

3. Results and discussion 3.1. Sorption isotherms Most of the polymers tested in this study are widely used in water and wastewater treatment, particularly for organics. The adsorptive capacities of six polymers at a low concentration (100 ng L 1) were calculated by ﬁtting the experimental data to Freundlich models. Freundlich equation is an empirical relation- ship describing the adsorption of solutes from a liquid to a solid surface. It is widely applied to describe adsorption process for many compounds onto heterogeneous surfaces, including acti- vated carbon, metals and polymers in dilute solutions (Robert et al., 2001; Siriniwasan et al., 2008).

The Freundlich equation is deﬁned by qc¼K fCn e ð1Þ whereq c(lgg 1) is the concentration in the solid phase,C e(lgL 1) is the equilibrium concentration of solute in solution,K f (lgg 1)(lgL 1) n is the Freundlich adsorption constant or capacity factor andnis the Freundlich exponent. According to the Eq.(1), adsorption capacity of a given sorbent is proportional to the equilib- rium concentration of the given sorbate. The Freundlich isotherm is known to be operative only within certain concentration limits. This is because, given an exponential distribution of binding sites, the number of sites increases indeﬁ- nitely with a decreasing association constant, implying that there are inﬁnite number of sites. But the Freundlich isotherm will be a more accurate approximation at lower concentrations (Robert et al., 2001).

Absorption isotherms of PFOS onto various adsorbents are shown inFig. 1and the determined Freundlich constants are listed inTable 1. The shaking duration of isotherm experiment is 100 h and which ensures that all ﬁlter materials have reached their equi- librium concentrations (Fig. 2). It was found that isotherms of all granular medias were nicely ﬁxed with Freundlich isotherm (R 2= 0.92–0.99).

Freundlich equation was used to calculate PFOS adsorption capacities onto various ﬁlter materials at further low equilibrium concentration of 100 ng L 1. Some possible errors can be expected with this calculation. Even though it is assumed in the calculation that all the materials are agreed with Freundlich equation, practi- cally all data points are not coincide with Freundlich curve (R 2–1). Also it should be emphasized that the minimum equilib- rium concentration received in this experiment is 150 ng L 1.

3.1.1. Adsorption of PFOS onto granular activated carbon Granular activated carbons are commonly used adsorbent materials to eliminate organic compounds in water. It has been identiﬁed that F400 (coal based) is the best type of GAC to elimi- nate PFOS (Ochoa-Herrera and Sierra-Alvarez, 2008). Same type of GAC was used in this experiment with the physical properties as tabulated inTable 2.

In this study GAC gave comparatively higher Freundlich expo- nent (n) than other polymer materials.nis an indicator of nonlin- earity of Freundlich curve. Nonlinearity can occur due to many reasons including the sorption site heterogeneity and sorbate–sor- bate interactions (Cheung et al., 2001). The reason for highernfor GAC may be the sorption site heterogeneity of GAC, which is higher than that of synthetic polymers. More than 78% total pore volume in GAC is occupied by micro pores, 14% meso pores and the rest is occupied by macro pores (Table 2).

With the isotherm experiment, it was determined that the PFOS adsorption of GAC (F400) at unit equilibrium concentration (1 lgL 1) is 28.4 lgg 1. Ochoa-Herrera and Sierra-Alvarez re- ported that the amount of PFOS adsorbed onto GAC (F400) at unit equilibrium concentration (mg L 1) is 25.9 mg g 1. According to the available literature data, 2.6 lgg 1 of methyl tertiary butyl ether (MTBE) detected in GRC-22 (coconut based GAC) at 1 lgL 1 of equilibrium concentration (Melin et al., 1999). Hydrophobicity of PFOS than MTBE may be the reason to have higherK fthan MTBE.

Among the six materials tested in this study, GAC gave lowest sorption capacities at 1 lgL 1 (Kf) and 0.1 lgL 1 equilibrium con- centrations suggesting its incapability at low equilibrium concen- trations (<1 lgL 1). It has been reported that the application of GAC to eliminate PFOS in German drinking water treatment plants has been failed (Schaefer, 2006).

3.1.2. Adsorption of PFOS onto ion-exchange polymers PFOS is a fully ﬂuorinated compound, which is commonly ap- peared as a salt with a functional group of sulfonate attached at the end of the molecule. The carbon–ﬂuorine bond length is short- er than any other carbon–halogen bond, and shorter than carbon– 648S.T.M.L.D. Senevirathna et al. / Chemosphere 80 (2010) 647–651 nitrogen and carbon–oxygen bonds. Short bond length and the high electronegativity of ﬂuorine give the carbon–ﬂuorine bond a signiﬁcant polarity/dipole moment. The electron density is concen- trated around the ﬂuorine, leaving the carbon relatively electron poor. This introduces ionic character to the bond through partial charges (Cd+AFd )O’Hagan, 2008. Negatively charged molecular structure of PFOS suggests higher adsorption onto anion-exchange polymers.

Among the all polymers tested in this experiment, anion-ex- change polymers of AmbIRA400 and DowMarathonA gave the highestK fof 108.9 and 95.9 ( lgg 1)(lgL 1) n respectively. Some characteristics of these polymers are tabulated inTable 3. Yu et al.

investigated AmbIRA400 for PFOS adsorption and reported thatK f is 250 for high equilibrium concentration of 1 mg L 1 (Yu et al.

2009).

Among the two ion-exchange polymers tested, DowMarathonA showed smallern(1.68) giving ﬂatter frundich curve than AmbI- RA400 (n= 2.8), which indicates better performance of DowMa- rathonA at equilibrium concentrations in ng L 1 level. Calculated adsorption capacity for DowMarathonA is 2.01 lgg 1 whereas for AmbIRA400 is 0.9 lgg 1.

3.1.3. Adsorption of PFOS onto non-ion-exchange polymers A signiﬁcant aspect of the non-ion-exchange polymer adsorp- tion is that, the bonding forces between the adsorbent and the adsorbate are usually weaker than those encountered in activated carbon adsorption. Regeneration of the resin can be accomplished by simple, nondestructive means, such as solvent washing, thus providing the potential for solute recovery (Busca et al., 2008; Lin and Juang, 2009).

DowL493, DowV493 and AmbXAD4 were identiﬁed as possible candidate materials to adsorb PFOS with our previous research 0 0.51 1.52 2.53 3.54 -2.5 -1.5 -0.5 0.5 1.5 2.5 Log (Sorption capacity) (µg PFOS/g sorbent) Log (Equilibrium PFOS concentration) (µg/L) 0 0.51 1.52 2.53 3.54 -2.5 -1.5 -0.5 0.5 1.5 2.5 B Log (Equilibrium PFOS concentration) (µg/L) Log (Sorption capacity) (µg PFOS/g sorbent) A 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 -2.5 -1.5 -0.5 0.5 1.5 2.5 C Log (Sorption capacity) (µg PFOS/g sorbent) Log (Equilibrium PFOS concentration) (µg/L) Fig. 1.(A) Adsorption isotherm of PFOS onto GAC: Filtrasorb 400 (4). (B) Adsorption isotherm of PFOS onto ion-exchange polymers: DowMarathonA (}), Amb IRA-400 (4). (C) Adsorption isotherm of PFOS onto non-ion-exchange polymers: Amb XAD 4 (j), DowV493 (4), DowL493 (}). Table 1 Freundlich isotherm constantsK f((lgg 1)(lgL 1) n) andnfor the adsorption of PFOS onto various sorbent materials at 25 C and calculated solid phase concentration q c(lgg 1) at 100 ng L 1equilibrium concentration. Adsorbent Freundlich Isotherm parameters K f nR 2 qc DowL493 54.60 0.84 0.99 7.93 Amb XAD 4 79.10 1.61 0.94 1.95 DowV493 81.30 0.94 0.92 9.29 Amb IRA-400 108.89 2.08 0.97 0.90 DowMarathonA 95.90 1.68 0.92 2.01 Filtrasorb 400 28.40 2.20 0.93 0.18 Time (hrs) 15 20 25 30 35 40 45 020406080100 Sorbent Phase Concentration (mg PFOS/g sorbent) Fig. 2.Adsorption kinetic of PFOS onto GAC and synthetic polymers. Legend:

DowL493 (}), Amb XAD 4 (h), V493 (4), Amb IRA-400 (j), DowMarathon A ( ), Filtrasorb 400 (N).

Table 2 Physical properties of GAC (Filtersorb 400, coal based).

Total pore volume 0.61 cm 3g 1 Macropores (>500 A) 0.04 cm 3g 1 Mesopores (20–500 A) 0.09 cm 3g 1 Micropores (<20 A) 0.48 cm 3g 1 Surface area 900–1100 m 2g 1 Matrix Stacked layers of fused hexagonal ring of C atoms Diameter 0.25–0.5 mm S.T.M.L.D. Senevirathna et al. / Chemosphere 80 (2010) 647–651649 works. Some important physical characteristics of these polymers are shown inTable 4. AmbXAD4 and DowV493 showed sorption capacities of 79.1 lgg 1 and 81.3 lgg 1 respectively at unit equi- librium concentration (1 lgL 1). These values are approximately in the range of 80% of sorption capacity determined for ion-ex- change polymers and 250% of sorption capacity determined for GAC at 1 lgL 1 equilibrium concentration. Distribution of homo- geneous pore size in the surface of synthetic non-ion-exchange polymers may be the reason for higherK fthan GAC. Also it was no- ticed that the adsorption capacities of DowV493 and DowL493 at 100 ng L 1 (calculated vales) are better than ion-exchange poly- mers. Comparing with ion-exchange polymers and GAC, adsorp- tion capacity of non-ion-exchange polymers showed a linear (comparatively) correlation to equilibrium concentration (n= 0.84, 1.61 and 0.94) (Fig. 1).

3.2. Sorption kinetics Fig. 2shows the amount of PFOS adsorption onto each adsor- bent with time (with 5000 lgL 1 initial concentration). Two ion- exchange polymers and GAC reached the equilibrium concentra- tion within 4 h. non-ion-exchange polymer materials of AmbXAD4 reached it within 10 h and DowL493 and DowV493 took 90 h to reach the equilibrium concentration.

To further understand the sorption kinetics, the pseudo second- order model (Eq.(2)) was selected to ﬁt the kinetic data. The model assumes that the sorption rate is controlled by chemical sorption and the sorption capacity is proportional to the number of active sites on the sorbent (Gomez et al., 2007);McKay and Ho, 1999). t q t¼ 1 kq2 eþ t q e ð2Þ whereq eandq tare the amount of PFOS adsorbed onto the adsor- bents at equilibrium at timet(mg g 1),kis the sorption rate con- stant (g mg 1 h 1). It was observed that our kinetic data of non-ion-exchange poly- mers ﬁxed well with the Eq.(2)(R 2= 0.99–1) suggesting the dom- inancy of the chemisorption at adsorption process (Table 5).

Since ion-exchange polymers and GAC reached the equilibrium concentration within a very short time (<4 h), the number of ki- netic data was not sufﬁcient to check with pseudo second-order ki- netic model.

4. Conclusions The adsorption of PFOS onto ion-exchange polymers, non-ion- exchange polymers and GAC was demonstrated in this study. Syn- thetic polymer materials were identiﬁed as better ﬁlter materials (in terms of adsorption capacity) to eliminate PFOS in water at low concentration (1 lgL 1). The magnitude ofK fdecreases in the following order: Ion exchange polymers > non-ion-exchange polymers > GAC, but at further low equilibrium concentration (100 ng L 1) non-ion-exchange polymers showed higher adsorp- tion capacity than other adsorbents. With regard to sorption kinet- ics, GAC and ion-exchange polymers reached the equilibrium concentration within 4 h, AmbXAD4 reached within 10 h and oth- ers took more than 80 h. AmbIRA400 was identiﬁed as the best ﬁl- ter material to eliminate PFOS at equilibrium concentration >1 lgL 1. considering both adsorption isotherm and adsorption kinetics, AmbXAD4 and DowMarathonA can be rec- ommended for eliminating PFOS at ng L 1 equilibrium concentra- tion. Further studies are to be done to understand the sorption mechanism of PFOS, but it can be suspected that the chemisorption is the dominant process.

References Busca, G., Berardinelli, S., Resini, C., Arrighi, L., 2008. Technologies for the removal of phenol from ﬂuid streams: a short review of recent developments. J. Hazard.

Mater. 160, 265–288.

Cheung, C.W., Porter, J.F., McKay, G., 2001. Sorption kinetic analysis for the removal of cadmium ions from efﬂuents using bone char. Water Res. 35, 605–612.

Earth Negotiations Bulletin, 2009. Published by the International Institute for Sustainable Development (IISD), vol. 15, No. 174.

Fujii, S., Polprasert, C., Tanaka, S., Lien, P.H., Qiu, Y., 2007. New POPs in the water environment: distribution, bioaccumulation and treatment of perﬂuorinated compounds – a review paper. Water Supply Res. Technol. – AQUA 56 (5), 313– 326.

Gomez, V., Larrechi, M.S., Callao, M.P., 2007. Kinetic and adsorption study of acid dye removal using activated carbon. Chemosphere 69, 1151–1158.

Hori, H., Yamamoto, A., Hayakawa, E., 2005. Efﬁcient decomposition of environmentally persistent perﬂuorocarboxylic acids by use of persulfate as a photochemical oxidant. Environ. Sci. Technol. 39 (7), 2383–2388.

Hori, H., Hayakawa, E., Einaga, H., 2004. Decomposition of environmentally persistent perﬂuorooctanoic acid in water by photochemical approaches.

Environ. Sci. Technol. 38, 6118–6124.

Jin, Y.E., Liu, W., Sato, I., Nakayama, S.F., Sasaki, K., Saito, N., Tsuda, S., 2009. PFOS and PFOA in environmental and tap water in China. Chemosphere 77, 605–611.

Lien, N.P.H., Fujii, S., Tanaka, S., Nozoe, M., Tanaka, H., 2008. Contamination of perﬂuorooctane sulfonate (PFOS) and perﬂuorooctanoate (PFOA) in surface water of the Yodo River basin (Japan). Desalination 226, 338–347.

Lin, S., Juang, R., 2009. Adsorption of phenol and its derivatives from water using synthetic resins and low-cost natural adsorbents: a review. J. Environ. Manage.

90, 1336–1349.

McKay, G., Ho, Y.S., 1999. Pseudo-second order model for sorption processes.

Process Biochem. 34, 451–465.

Melin, G., 1999. Evaluation of the Applicability of Synthetic Resin Sorbents for MTBE Removal from Water, Center for Groundwater Restoration and Protection, National Water Research Institute, California.

Moriwaki, H., Takagi, Y., Tanaka, M., Tsuruho, K., Okitsu, K., Maeda, Y., 2005.

Sonochemical Decomposition of Perﬂuorooctane Sulfonate and Perﬂuorooctanoic Acid. Environ. Sci. Technol. 39 (9), 3388–3392.

Ochoa-Herrera, V., Sierra-Alvarez, R., 2008. Removal of perﬂuorinated surfactants by sorption onto granular activated carbon, zeolite and sludge. Chemosphere 72, 1588–1593.

O’Hagan, D., 2008. Understanding organoﬂuorine chemistry. An introduction to the C–F bond. Chem. Soc. Rev. 37 (2), 308–319.

Quinete, N., Wub, Q., Zhang, T., Yun, H.S., Moreira, I., Kannan, K., 2009. Speciﬁc proﬁles of perﬂuorinated compounds in surface and drinking waters and accumulation in mussels, ﬁsh, and dolphins from southeastern Brazil.

Chemosphere 77, 863–869. Table 3 Physical properties of ion-exchange polymers.

Adsorbent Matrix Exchange capacityFunctional groupDiameter (mm) Amb IRA-400 Styrene- DVB3.0–3.5 (eq kg 1)–N +R3(Cl ) 0.8 DowMarathonA Styrene- DVB1.3 (eq L 1) Quaternary Amine0.57 Table 4 Physical properties of non-ion-exchange polymers.

Adsorbent Surface area m 2g 1 Matrix Diameter (mm) DowL493 1100 Styrene-DVB 0.8 Amb XAD 4>750 Macroreticular crosslinked aromatic polymer0.35–1.18 DowV493 1025 Styrene-DVB 0.8 Table 5 Pseudo-second-order kinetic parameters of PFOS onto non-ion-exchange polymers.

Adsorbent Pseudo-second-order kinetic parameter q eb ka R2 DowL493 40.00 0.07 0.99 Amb XAD 4 41.66 0.48 1.00 DowV493 38.46 0.08 0.99 aSorption rate constant [g sorbent (mg PFCs) 1(h) 1]. bPFCs adsorbed on the adsorbent at equilibrium (mg g 1). 650S.T.M.L.D. Senevirathna et al. / Chemosphere 80 (2010) 647–651 Robert, J., Sarah, C.B., Miguel, B., John, K.B.J., Ripal, N.S., Ken, D.S., 2001. Application of the Freundlich adsorption isotherm in the characterization of molecularly imprinted polymers. Anal. Chim. Acta 435, 35–42.

Schaefer, A., 2006. Perﬂuorinated surfactants contaminate German waters. Environ.

Sci. Technol. 40, 7108–7109.

Siriniwasan, R., Sorial, G.A., Ononye, G., Husting, C., Jackson, E., 2008. Elimination of persistent odorous compounds from drinking water. Water Sci. Technol. – Water Supply 8 (2), 121–127.

Takagi, S., Adachi, F., Miyano, K., Koizumi, Y., Tanaka, H., Mimura, M., Watanabe, Y., Tanabe, S., Kannan, K., 2008. Perﬂuorooctanesulfonate and perﬂuorooctanoatein raw and treated tap water from Osaka, Japan. Chemosphere 72 (10), 1409– 1412.

Tang, C.Y., Fu, Q.S., Robertson, A.P., Criddle, C.S., Leckie, J.O., 2006. Use of reverse osmosis membranes to remove perﬂuorooctane sulfonate (PFOS) from semiconductor wastewater. Environ. Sci. Technol. 40, 7343–7349.

Yong, Q., Fujii, S., Tanaka, S., 2007. Removal of perﬂuorochemicals from wastewater by granular activated carbon adsorption. Environ. Eng. Res. 44, 185–193.

Yu, Q., Zhang, R., Deng, S., Huang, J., Yu, G., 2009. Sorption of perﬂuorooctane sulfonate and perﬂuorooctanoate on activated carbons and resin: kinetic and isotherm study. Water Res. 43, 1150–1158. S.T.M.L.D. Senevirathna et al. / Chemosphere 80 (2010) 647–651651