Novel insights into the competitive adsorption behavior and mechanism of per- and polyfluoroalkyl substances on the anion-exchange resin

Journal of Colloid and Interface Science 557 (2019) 655–663

Contents lists available at ScienceDirect

Journal of Colloid and Interface Science journal homepage: www.elsevier.com/locate/jcis

Novel insights into the competitive adsorption behavior and mechanism of per- and polyfluoroalkyl substances on the anion-exchange resin Wei Wang a,b,1, Xin Mi a,1, Ziming Zhou a, Shuangxi Zhou a, Chunli Li a, Xue Hu a, Delin Qi a, Shubo Deng b,⇑ a

State Key Laboratory of Plateau Ecology and Agriculture, Qinghai University, Xi’ning, Qinghai Province 810016, China State Key Joint Laboratory of Environment Simulation and Pollution Control (SKLESPC), Beijing Key Laboratory for Emerging Organic Contaminants Control, School of Environment, Tsinghua University, Beijing 100084, China b

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

a r t i c l e

i n f o

Article history: Received 7 August 2019 Revised 15 September 2019 Accepted 18 September 2019 Available online 19 September 2019 Keywords: PFASs Anion exchange resin Competitive adsorption Adsorption behavior Adsorption mechanism

a b s t r a c t Per- and polyfluoroalkyl substances (PFASs) are widely used and co-exist in various aquatic environments, but their co-removal is not clear. In this study, the competitive adsorption behavior and mechanism of six traditional and emerging PFASs on anion-exchange resin IRA67 in the bisolute and mixed systems were studied. The adsorption equilibrium of the long-chain PFASs was at least 96 h whereas 48 h was required for the short-chain PFASs. When the PFASs were co-removed in the bisolute system, their competition was not obvious at low PFAS concentration of 0.01597 mmol/L due to the relatively adequate adsorption sites. When the concentrations of PFASs were increased to 0.07666 mmol/L, the removal of perfluorobutanoic acid (PFBA) and perfluorobutane sulfonate (PFBS) decreased by 77.78% and 72.09%, respectively. The competitive experiments showed that the adsorbed short-chain PFASs could be replaced by the long-chain ones, which was closely related to their hydrophobicity, backbone and functional groups. With the increase of solution pH from 3 to 7, the polyamine groups on the resin IRA67 were transferred to the base forms and the effective adsorption sites decreased, resulting in a more obvious competitive replacement behavior. This study suggested that the PFASs with long chain could be more effectively removed from the coexisting PFASs solution by the anion-exchange resins, and the short-chain PFASs in water may be removed when high dosage of anion-exchange resins is applied or the solution pH is decreased. Ó 2019 Elsevier Inc. All rights reserved.

1. Introduction ⇑ Corresponding author. 1

E-mail address: [email protected] (S. Deng). W. Wang and X. Mi contributed equally to the work.

https://doi.org/10.1016/j.jcis.2019.09.066 0021-9797/Ó 2019 Elsevier Inc. All rights reserved.

Per- and polyfluoroalkyl substances (PFASs) are widely used in the manufacturing of textiles, carpets, paper, leather shoes,

656

W. Wang et al. / Journal of Colloid and Interface Science 557 (2019) 655–663

packaging, printing, fire-fighting agents and hydraulic oils due to their hydrophobic transfer characteristics [1]. Once PFASs enter human or animal bodies, they are toxic to blood, liver, kidney, heart, muscle and other tissues [2,3]. The international agency for research on cancer research has ranked PFASs as ‘‘a possible human carcinogen” [4]. Perfluorooctane sulfonate (PFOS) and perfluorooctanoic acid (PFOA) are typical PFASs, and have been found worldwide in the aquatic environment, even in human blood [5]. Because of their high toxicity, bioaccumulation and stability, PFOS and PFOA have been listed in the Stockholm Convention as persistent organic pollutants [6,7]. A previous study has detected seven common PFASs in urban sewage treatment plants and industrial wastewater [8]. Different PFASs were detected in most samples. PFOA and PFOS were detected in all samples and the average concentration of PFOS was up to 340 ng/L, while that of PFOA was 220 ng/L [8]. Moreover, more than 70% of drinking water and human blood samples in 13 cities of China contained four PFASs including perfluorobutanoic acid (PFBA), perfluorobutane sulfonate (PFBS), PFOA and PFOS [9], while the actual coexistence of PFASs may be more complex and common. It is reported that some emerging PFOS and PFOA substitutes with short chain were detected in the PFOA and PFOS polluted water environments, and among them sodium pperfluorous nonenoxybenzene sulfonate (OBS) and perfluoro-2propoxypropanoic acid (GenX) were two typical alternatives with high concentrations at lg/L level in surface water [10,11]. The restrictions and prohibitions on the use of traditional PFASs have led to the emergence of these new PFAS substitutes [12,13]. These human-made chemicals have widespread application and are essential for modern society [14]. However, it isn’t a good phenomenon because some studies have found that these emerging substitutes still have the characteristics of durability and biological toxicity [5,15]. Therefore, it is particularly urgent and necessary to co-remove them from the PFASs coexistence system. Adsorption is one of the effective methods for PFAS removal [16,17]. Activated carbon [18,19], anion-exchange resins [20,21], fluorinated clay [22], modified biomass [23,24] and bcyclodextrin polymer [25] are effective adsorbents for PFASs removal. The resin IRA67 is an effective adsorbent to remove PFASs from water and is widely used in industry due to its low price and high adsorption capacity [20,26]. We ever evaluated the reusability and stability of the spent resin [20,21], and found that the PFASadsorbed IRA67 could be successfully regenerated by organic solvents with salt solution and ethanol solution, respectively, and the resin exhibited stable adsorption capacity. Moreover, we also conducted research on the treatment of PFASs in the actual environment using IRA67 resin, and found that inorganic ions and organic matter in real water would compete with PFASs for adsorption, and the organic matters had a relatively small effect on the adsorption of long-chain PFOA on the resin [20,21]. Due to the coexisting substances in the actual environment are complicated, the competitive adsorption behavior of the coexisting PFASs in the actual water has not been thoroughly explored, and the potential competition mechanism is unclear. Therefore, IRA67 was chosen as the adsorbent in this study and the competitive adsorption behavior and mechanism of PFAS were further explored. Most of the previous researches focused on the removal of single PFOS, PFOA, PFBS or PFBA from water [27,16], while few studies cared about the adsorptive removal of PFASs from the coexisting PFASs system, and the competitive adsorption behavior and mechanism of various PFASs on adsorbents are still not clear. Our previous study found that the adsorbed PFASs with short CAF chain could be replaced by the long-chain ones [7,12], but the condition of competitive adsorption experiments was simple and only a rough competitive adsorption behavior was obtained. Due to the diversity and differences of the structures and properties of emerging

PFAS, the removal relationship is not clear when they coexist with traditional PFASs. Therefore, it is very necessary to explore the underlying competitive adsorption mechanism among PFASs including the traditional PFASs and the emerging ones. In this study, the anion exchange resin IRA67 was chosen to remove six kinds of PFASs including PFOS, PFOA, PFBS, PFBA, OBS and GenX, and the competitive adsorption behaviors of different PFASs in the bisolute and the multisolute system were investigated. The influences of pH and PFASs concentration on their competitive adsorption were explored, and the corresponding adsorption mechanism and the exchange relationship were also discussed. 2. Materials and methods 2.1. Chemicals and materials PFOS, PFBS (all potassium salts, purity  97%), PFOA (purity  97%) and PFBA (purity  98%) were purchased from Macklin biochemical technology Co., Ltd. (Shanghai, China). GenX (purity  97%) was purchased from J & K Scientific Ltd. (Beijing, China). OBS was purchased from Wengjiang reagent Co. Ltd. (Guangzhou, China). Their chemical formulas are showed in Table S1. HPLC-grade methanol was purchased from Fisher Chemical (USA). The anion-exchange resin IRA67 was purchased from Sigma-Aldrich (St. Louis, MO). It is a weakly basic gel-type polyacrylic resin with tertiary amine functional groups [20] and its main properties are shown in Table S2. The ultrapure water was produced by a Pall cascada-III water purification system (Beijing, China). The other chemicals were all analytical reagent grade. 2.2. Adsorption experiments Adsorption experiments were conducted at 25 °C with a shaking speed of 165 rpm for 96 h in an orbital shaker, and the solid/liquid ratio performed in all adsorption experiments was 0.0194 g/L. All solutions were adjusted to pH 5 (±0.2) with NaOH and HCl except for the pH effect experiment. For the adsorption kinetics in single-solute systems, 180 mL of 0.01597 mmol/L PFASs solution was added into the conical flasks for adsorption. Co-adsorption experiments of two contaminants can be used to compare their adsorption behavior in the bisolute systems with an equimolar concentration [14,28]. The initial concentration of each PFAS was 0.01597 mmol/L in the bisolute systems in this study. The effect of pH experiment was conducted in the pH range of 3–7 without pH adjustment in the adsorption process, and the initial concentration of each PFAS was 0.07666 mmol/L, and other conditions were the same as bisolute adsorption system. The effect of PFASs concentration experiments were conducted using 0.01597, 0.07666 and 0.1597 mmol/L of PFASs and other conditions were the same as the bisolute adsorption system. All experiments were conducted twice, and the average value was adopted. 2.3. Analytical methods After the adsorption experiments, the mixture was filtered with a 0.22 lm nylon membrane, and about 1.0 mL of the sample was collected. The control experiment indicated that the adsorption of PFASs on the membrane was negligible. The PFASs concentrations in solution were determined by an ultra-high performance liquid chromatography-tandem mass spectrometry (UPLC-MS/ MS) equipped with a Thermo Fisher Scientific column (2.1 mm  50 mm i.d., 1.9 lm) and a Q Exactive Focus combined quadrupole Orbitrap mass spectrometer with an electrospray ionization source. The total analysis time of a sample was 9 min. The

W. Wang et al. / Journal of Colloid and Interface Science 557 (2019) 655–663

mobile phase was a binary mixture of 2 mmol/L ammonium acetate (solvent A) and methanol (solvent B) at a flow rate of 0.3 mL/ min. The gradient started with 90% A and 10% B, and linearly ramped to 20% A and 80% B from 2 to 6 min and kept constant of 20% A and 80% B from 6 to 8 min, finally turned to 90% A and 10% B from 8 to 9 min. The sample injection volume was 1.5 lL. The adsorbed amount was calculated according to the difference of PFAS concentrations before and after adsorption.

3. Results and discussion 3.1. Adsorption of individual PFASs

Adsorbed amount (mmol/g)

Fig. 1 shows the different adsorption kinetic curves of six individual PFASs at pH 5. To further investigate the adsorption kinetics, the pseudo-second order model (Table S3) was used to fit the kinetic curves. Although this model is not sensitive and almost fits all kinetic data well, the initial adsorption rate can be obtained from this model. The good fitting indicates possible chemical adsorption, but it is unable to identify the adsorption forces. However, the ion exchange involved in the PFASs adsorption on anionexchange resin has been clarified in our previous studies [20,26]. The fitting coefficients (R2) of all the adsorption data (Table S3) were higher than 0.96, indicating that the chemical adsorption between resin IRA67 and PFASs was possibly produced in the adsorption process. It took 48 h to reach the adsorption equilibrium of PFBA and GenX, while more than 96 h was needed for PFOS and PFOA to reach the adsorption equilibrium. The adsorption mechanism of anion pollutants on IRA67 is generally considered to be ion exchange [20]. If there is only anion-exchange interaction and all exchange sites are available, the adsorption capacity for these six PFASs should be the same [29]. The equilibrium adsorption amount decreased in the order of PFOS > PFBS > PFOA > OBS > GenX > PFBA, indicating that other interactions affected the adsorption kinetics in the adsorption process such as the hydrophobic interaction and backbone effect. Their CAF chain length and functional groups both had influence on the adsorption equilibrium. The PFASs with long CAF chain would block the pores on resin, making the diffusion of coexisting PFASs more difficult. Although the molecular lengths of PFBS and PFBA are similar (Table S1), the steric hindrance of the sulfonated group of PFBS is larger than that of the carboxyl group of PFBA (the molecular diameter of the sulfonate is 3.748 Å, while the carboxyl is 2.165 Å) [12], leading to a slower adsorption of PFBS. Because the chain length of PFOS is the longest among the six PFASs and the hydrophobic critical micelle concentration (CMC) is the lowest [16,30], PFOS is easy to form semi-micelles or micelles on the surface of the resin, block-

1.0

GenX PFBA PFOA PFOS

0.8

OBS PFBS

0.6 0.4

657

ing the pores and preventing other PFOS molecules from diffusing into the pores. For this reason, PFOS may require a longer adsorption equilibrium time. The adsorption rate (V0) of GenX was faster than OBS (Table S3), because OBS contained benzene rings and branched chain (Table S1), and it was not easy to diffuse into the pores. 3.2. Competitive adsorption between PFASs in bisolute systems The adsorption kinetics of bisolute PFASs are shown in Fig. 2. The adsorption amounts of all the six PFASs in the bisolute system were little lower than those in the single-solute system. In the bisolute system of perfluoroalkyl sulfonate (PFSAs), the adsorbed amounts of PFASs on IRA67 increased more slowly than those of PFSAs in the single-solute system with increasing time (Fig. 2a– c). In our previous study, the PFASs with similar molecular structure were found to have obvious competitive adsorption in the bisolute system [12], while this competitive behavior was not obvious in this study, and the possible reason was that the adsorption sites on IRA67 were sufficient for PFAS molecules. In the bisolute system of perfluoroalkyl carboxylic acid (PFCAs), the PFBAPFOA (Fig. 2d) and GenX-PFBA (Fig. 2f) bisolute systems were similar to the above three PFSAs experiments without obvious competitive adsorption, but in the GenX-PFOA bisolute system, there was slight competitive adsorption. In the GenX-PFOA bisolute system (Fig. 2e), the adsorption kinetic of GenX was different from that in the single-solute system, and the adsorbed amounts of GenX in the bisolute solution decreased significantly, whereas the adsorbed amounts of GenX on IRA67 increased first and then decreased from 0.30 mmol/g at 48 h to 0.24 mmol/g at 96 h, indicating that some GenX molecules adsorbed on IRA67 were replaced by PFOA. The PFASs can adsorb on the surface of resin by ion exchange through the negatively charged head [7]. The GenX molecule contains ether bond which is close to the negative head, making the adsorbed GenX can be replaced by PFOA. In general, the adsorbed amounts of PFASs decreased slightly in the presence of PFOS and PFOA in the bisolute system. Additionally, there was no exchange between PFOA and the adsorbed PFBA (Fig. 2d), while the adsorbed GenX could be replaced by PFOA. Since PFBA was shorter and more hydrophilic than GenX (Table S1), it should be more easily replaced by PFOA than GenX. However, the ether bond in the GenX molecule may cause the instability of the negatively charged head adsorbed on adsorbent surface, leading to the replacement of the adsorbed GenX by PFOA. There are three reasons for the lack of adsorption replacement of the three group experiments in PFSAs bisolute systems. Firstly, the sulfonate head and excessively long chain structure of the PFOS molecules lead to a large resistance for the spatial rotation in the replacement process, causing the adsorbed OBS in the OBS-PFOS bisolute system and the adsorbed PFBS in the PFBS-PFOS cannot be replaced. Secondly, the OBS molecules contain benzene ring structure and have high hydrophobicity, thus OBS was not easy to be replaced in the OBS-PFOS bisolute system after adsorption on IRA67. Thirdly, the previous studies indicated that PFASs were mainly adsorbed by ion exchange, and the adsorption sites should be the same [12,26]. When the adsorption sites on the resin were sufficient for all the PFSAs molecules, all the co-existed PFASs could be fully adsorbed on resin without competitive phenomenon.

0.2 —— Pseudo-second-order

0.0 0

20

40

60

80

100

Sorption time (h) Fig. 1. Adsorption kinetics of PFASs on IRA67 in single-solute solution as well as modeling using the pseudo-second-order model.

3.3. Effect of PFASs concentrations As shown in Fig. 3, the PFAS concentrations had a significant effect on the competitive adsorption of PFASs in the bisolute system. At the low PFAS concentration of 0.01597 mmol/L, the PFOS-PFBS bisolute system was completely free of competitive replacement and the adsorbed amounts of two PFASs were very

658

W. Wang et al. / Journal of Colloid and Interface Science 557 (2019) 655–663

1.0 0.8

PFBS-single PFOS-single PFBS-bisolute PFOS-bisolute

(a)

PFBA-single PFOA-single PFOA-bisolute PFBA-bisolute

(d)

OBS-single PFOS-single OBS-bisolute PFOS-bisolute

(b)

GenX-single PFOA-single GenX-bisolute PFOA-bisolute

(e)

(c)

GenX-single PFBA-single GenX-bisolute PFBA-bisolute

(f)

0.6 0.4 0.2

Adsorbed amount (mmol/g)

0.0

0.8 0.6 0.4 0.2 0.0

OBS-single PFBS-single OBS-bisolute PFBS-bisolute

0.8 0.6 0.4 0.2 0.0 0

20

40

60

80

100

t (h)

0

20

40

60 t (h)

80

100

Fig. 2. Competitive adsorption kinetics between PFBS and PFOS (a), OBS and PFOS (b), OBS and PFBS (c), PFBA and PFOA (d), GenX and PFOA (e) and GenX and PFBA (f).

similar. However, after increasing their concentrations to 0.07666 mmol/L (4.8 times), the adsorbed amounts of the two co-existing PFASs were completely different. After the concentrations of PFASs increased by 4.8 times, the removal percent of PFBA decreased by 77.78% and that of PFBS decreased by 72.09%, while the decrease of the PFOA and PFOS removal were relatively small (<22%). The phenomenon of competitive replacement between PFBS and PFOS was still not obvious. Until the concentrations of the co-existing PFASs were increased to 0.1597 mmol/L (10 times), the adsorbed amount of PFBS increased first and then decreased to about 0, indicating that the long-chain PFASs replaced the shortchain ones at high concentrations. The PFOA-PFBA bisolute system exhibited a partial replacement that the adsorbed PFBA was replaced by PFOA at low concentrations. When the concentrations of both PFASs in the bisolute system were increased by 4.8 times, a more obvious replacement phenomenon occurred. At 48 h, the adsorbed PFBA was basically replaced by PFOA, and all adsorbed PFBA was replaced by the longer-chain PFOA when the concentrations were increased by 10 times. The result of competitive adsorption at high concentrations is the same as the trend in our previous study [7,12], while the competitive adsorption at low concentrations is not obvious. The reason for the significant changes in replacement behavior at high

PFAS concentrations was due to the insufficient number of adsorption sites of IRA67 compared with PFAS molecules. A previous study has shown that the saturation of adsorption sites at higher pollutant concentrations could lead to lower removal percent [31], resulting in the enhanced competitive adsorption in this study. The previous studies showed that IRA67 had a high adsorption efficiency for PFASs [20,26], and no competition and replacement occurred when the adsorption sites were sufficient to simultaneously adsorb two PFASs. However, once increasing the concentrations of PFASs to fully occupy the entire adsorption sites of IRA67, the two PFASs competed for limited adsorption sites and the more hydrophobic PFASs can replace the adsorbed hydrophilic ones. By observing the change of the adsorbed amounts of PFASs in different concentration experiments, it was found that the two PFASs at the concentration of 0.01597 mmol/L did not completely occupy the adsorption sites of IRA67, resulting in high removal of both PFASs in the coexisting PFASs system. 3.4. Effect of solution pH Fig. 4 shows the effect of solution pH on the competitive adsorption kinetics of PFASs in the bisolute solution, and the total adsorbed amounts of the two co-existed PFASs decreased as the pH

659

W. Wang et al. / Journal of Colloid and Interface Science 557 (2019) 655–663

1.0

(a)

PFBS-1 time PFOS-1 time

(d)

PFBA-1 time PFOA-1 time

0.8 0.6 0.4 0.2

Adsorbed amount (mmol/g)

0.0

(b)

PFBS-4.8 times PFOS-4.8 times

(e)

PFBA-4.8 times PFOA-4.8 times

3

2

1

0

(c)

PFBS-10 times PFOS-10 times

6

(f)

PFBA-10 times PFOA-10 times

5 4 3 2 1 0 0

20

40

60

80

100

t (h)

0

20

40

60

80

100

t (h)

Fig. 3. Effect of PFAS concentrations on the competitive adsorption of PFBS and PFOS in bisolute solution with each PFAS concentration of 0.01597 mmol/L (a), 0.07666 mmol/ L (b) and 0.1597 mmol/L (c), as well as PFBA and PFOA in bisolute solution with each PFAS concentration of 0.01597 mmol/L (d), 0.07666 mmol/L (e) and 0.1597 mmol/L (f) on IRA67.

value increased from 3 to 7 (Fig. S1). At high solution pH, the polyamine groups of IRA67 would change to the base form, resulting in a decrease of PFASs adsorption due to the loss of anion exchange ability [32]. Since the anion exchange was the main adsorption interaction involved in the PFAS adsorption on the anionexchange resin, the adsorbed amounts declined with the increase of solution pH. The adsorption ratio analysis (Fig. S2) illustrated that the adsorbed amounts of PFBA and PFBS decreased with increasing solution pH and accounted for nearly 0% of the total PFASs adsorption amounts at pH 7, indicating a more obvious competitive adsorption at high solution pH. As a result of increasing solution pH, the conversion of polyamine functional groups on IRA67 to base forms reduced the quantity of effective adsorption sites, which was an important reason for the more obvious competitive adsorption between the long-chain PFASs and shortchain ones. Since IRA67 has the polyacrylic matrix with high hydrophilicity [20], the PFASs with short CAF chain have fast intra-particle diffusion rate to transport into the resin pores, causing their fast adsorption rates at the initial process (Fig. 4). How-

ever, the adsorbed short-chain PFASs were replaced by the longer ones with more hydrophobic skeleton, and the competition adsorption between PFASs was more obvious in the bisolute system with increasing solution pH. The effective adsorption sites were limited at high pH, and hydrophobic adsorption was the main interaction. Some studies have found that hydrophobic interaction is not sensitive to pH changes compared with anion-exchange interaction [26,33]. 3.5. Co-removal of six PFASs by IRA67 The kinetic curves of the six PFASs adsorbed on IRA67 at different concentrations were completely different (Fig. 5). The competitive adsorption of PFASs was not evident at low concentrations (0.01597 mmol/L), probably because the PFASs molecules did not completely occupy the adsorption sites on the resin at low concentration. The adsorbed amounts of PFASs at 96 h at the high concentration of 0.07666 mmol/L follow the order of PFOS > OBS > PFOA > PFBS > GenX > PFBA (Fig. 5b). According to the

660

W. Wang et al. / Journal of Colloid and Interface Science 557 (2019) 655–663

4.5

(a)

PFBS pH=3 PFOS pH=3

4.0

(d)

PFBA pH=3 PFOA pH=3

3.5 3.0 2.5 2.0 1.5 1.0 0.5

Adsorbed amount (mmol/g)

0.0

(b)

PFBS pH=5 PFOS pH=5

4.0

(e)

PFBA pH=5 PFOA pH=5

3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0

(c)

PFBS pH=7 PFOS pH=7

4.0

(f)

PFBA pH=7 PFOA pH=7

3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0

20

40

60

80

100

0

20

40

60

80

100

t (h)

t (h)

1.2

(a)

GenX OBS PFBA PFBS PFOA PFOS

1.0 0.8 0.6 0.4 0.2 0.0

0

20

40 60 80 Sorption time (h)

100

Adsorbed amount (mmol/g)

Adsorbed amount (mmol/g)

Fig. 4. Effect of pH on the competitive adsorption of PFBS and PFOS in bisolute solution at pH of 3 (a), 5 (b) and 7 (c), as well as PFBA and PFOA in bisolute solution at pH of 3 (d), 5 (e) and7 (f).

4.0

(b)

GenX OBS PFBA PFBS PFOA PFOS

3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0

0

20

40 60 80 Sorption time (h)

100

Fig. 5. Adsorption kinetics of the mixed six PFASs on IRA67 with each PFAS concentration of 0.01597 mmol/L (a) and 0.07666 mmol/L (b) at pH 5.

logKow of six PFASs (Table S1), their adsorption capacities was related to their hydrophobic sequences, suggesting that the hydrophobic interaction contributed to the competitive adsorption of different PFASs on IRA67. OBS was the only PFAS with an upward trend after 72 h (Fig. 5b), while PFOA, GenX, PFBS and PFBA

all had a downtrend, which proved that the functional group like benzene ring also played an important role in the competitive replacement. The adsorption on the resin did not have selectivity through anion exchange, and the difference in adsorption capacity was affected by hydrophobicity. The main mechanism of the

661

W. Wang et al. / Journal of Colloid and Interface Science 557 (2019) 655–663

3.6. Competitive adsorption mechanism To further verify the effect of solution pH and PFASs concentration on competitive adsorption, this study designed an experiment by adding PFOA and adjusting solution pH in the PFBA adsorption process containing the 1 time of PFASs concentration (0.01597 mmol/L) or the 4.8 times of PFASs concentration (0.07666 mmol/L). The volume change of the whole adsorption solution was less than 6% during the adsorption process (Fig. 6). After the 1-time concentration of PFBA adsorption for 48 h (Fig. 6a), the 1-time concentration of PFOA was added (the final PFOA concentration and PFBA concentration were both 0.01597 mmol/L). Although the adsorbed amounts of PFBA decreased, but still tended to increase. After adsorption for 120 h, the solution pH was adjusted to 7.5, the adsorption sites were reduced due to the deprotonation of polyamine groups on IRA67, and the adsorbed amounts of PFBA decreased. In this case, the hydrophobic PFOA replaced the adsorbed PFBA molecules. At the 4.8 times of PFASs concentration (Fig. 6b), the adsorbed amounts

Adsorbed amount (mmol/g)

0.8

(a)

PFBA PFOA 0.6

Add PFOA

0.4

Adjust pH to 7.5

only PFBA

0.2

0.0

of PFBA decreased significantly after adding the 4.8 times of PFOA at 48 h, which proved that the competitive adsorption was more obvious after increasing PFAS concentration due to the inadequate adsorption sites. After adjusting the solution pH to 3 at 120 h, the protonation of the polyamine groups on IRA67 resulted in an increase of the sorption sites, which alleviated the competitive adsorption, and the adsorbed amounts of PFBA increased. According to the above results and discussion, the underlying competitive adsorption mechanisms are proposed in Fig. 7. In the competitive adsorption experiments of various PFASs, it was possible to draw an important conclusion that there was no competition and replacement when the adsorption sites on resin were adequate. Competition occurred when the available adsorption sites on the resin were limited, and the replacement occurred when the adsorption sites were almost occupied. In order to ensure the good removal of all PFASs, it is necessary to increase the amount of adsorbent. When the targeted pollutant is the PFASs with strong hydrophobicity and high removal priority, the amount of adsorbent can be slightly reduced. The replacement phenomenon was different from the hydrophobic interaction found in previous studies [7,12], which was closely related to the backbone effect and functional groups [20]. The hydrophobic long-chain PFASs could replace the adsorbed short-chain PFASs and showed the high adsorbed amounts, but the hydrophobic OBS did not reflect the high adsorption capacity. At low concentrations of co-existing GenX and PFBA, the decrease in adsorption amounts of GenX reflected the effect of hydrophilic ether bond on competitive adsorption. During the replacement process, the negative head of

3.0

Adsorbed amount (mmol/g)

adsorption of PFASs on ion exchange resin was found to be ion exchange, but the backbone adsorption and the multilayer adsorption were also involved [34,35]. The matrix of IRA67 was linear polyacrylic acid, preferring to adsorb PFASs with linear structures according to the law of similarity and intermiscibility, but OBS didn’t have a linear straight-chain. It was the possible reason for OBS failing to show high adsorption capacity and being difficult to be replaced in the coexisting PFOS system.

(b)

PFBA PFOA

2.5

Adjust pH to 3.0

Add PFOA

2.0 1.5 only PFBA

1.0 0.5 0.0

0

20

40

60

80

100

120

Sorption time (h)

140

160

0

20

40

60

80

100

120

140

160

Sorption time (h)

Fig.6. The continuous adjustment of adsorbed solute and solution pH in the PFBA adsorption process with each PFAS concentration of 0.01597 mmol/L (a) and 0.07666 mmol/ L (b) at initial pH of 5.

Fig. 7. Schematic diagram for the PFASs competitive adsorption at different concentrations (a) and different pH (b).

662

W. Wang et al. / Journal of Colloid and Interface Science 557 (2019) 655–663

the long-chain PFASs competed with that of the short ones for the positive adsorption sites. For the PFCAs containing carboxylic groups, the competition of long-chain PFCAs replacing the shortchain ones was more obvious than that of the PFSAs containing sulfonic groups. Therefore, it was once again proved that functional groups had significant influence on PFASs competitive adsorption. 4. Conclusions The anion-exchange resin IRA67 exhibited good adsorption efficiency for six PFASs. The adsorption kinetic data of six PFASs on IRA67 were described well by the pseudo-second order model, but the competitive adsorption behavior between PFSAs and PFCAs in the bisolute system was completely different. The competitive adsorption of PFASs on IRA67 was closely related to their concentrations. The obvious competitive replacement between the longchain PFASs and short-chain ones occurred when the number of PFAS molecules was enough to completely occupy the adsorption sites on the resin. When the concentrations of PFASs increased from 0.01597 mmol/L to 0.07666 mmol/L, the removal of PFBA and PFBS decreased by 77.78% and 72.09%, respectively, while the decrease of PFOA and PFOS removal were both less than 22%. The effect of pH on the PFASs competitive adsorption on IRA67 was significant. The competitive replacement situation could be alleviated in low pH solution due to the protonation of polyamine groups on IRA67, while the hydrophobic PFASs significantly replaced hydrophilic ones previously adsorbed on IRA67 in high pH solution. The adsorbed amounts of PFBA and PFBS decreased with increasing solution pH from 3 to 7, and almost no adsorption was observed at pH 7. The PFASs with longer CAF chain could replace the adsorbed PFASs with shorter CAF chain. When ionexchange resins are used in the real wastewater treatment, the priority of targeted PFASs removal should be considered.

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

Acknowledgements We thank the independent subject of State Key Laboratory of Plateau Ecology and Agriculture (no. 2018-ZZ-1) and the Project of Qinghai Science & Technology Department (2018-ZJ-Y01 and 2019-ZJ-935Q) for financial support. Appendix A. Supplementary material Supplementary data to this article can be found online at https://doi.org/10.1016/j.jcis.2019.09.066. References [1] D. Herzke, E. Olsson, S. Posner, Perfluoroalkyl and polyfluoroalkyl substances (PFASs) in consumer products in Norway–a pilot study, Chemosphere 88 (2012) 980–987. [2] M. Houde, T.A. Bujas, J. Small, R.S. Wells, P.A. Fair, G.D. Bossart, K.R. Solomon, D. C. Muir, Biomagnification of perfluoroalkyl compounds in the bottlenose dolphin (Tursiops truncatus) food web, Environ. Sci. Technol. 40 (2006) 4138– 4144. [3] F. Fàbrega, V. Kumar, E. Benfenati, M. Schuhmacher, J.L. Domingo, M. Nadal, Physiologically based pharmacokinetic modeling of perfluoroalkyl substances in the human body, Toxicol. Environ. Chem. 97 (2015) 814–827. [4] X.C. Hu, D.Q. Andrews, A.B. Lindstrom, T.A. Bruton, L.A. Schaider, P. Grandjean, R. Lohmann, C.C. Carignan, A. Blum, S.A. Balan, Detection of poly-and perfluoroalkyl substances (PFASs) in US drinking water linked to industrial sites, military fire training areas, and wastewater treatment plants, Environ. Sci. Technol. Lett. 3 (2016) 344–350. [5] L. Xu, Y. Shi, C. Li, X. Song, Z. Qin, D. Cao, Y. Cai, Discovery of a novel polyfluoroalkyl benzenesulfonic acid around oilfields in Northern China, Environ. Sci. Technol. 51 (2017) 14173–14181. [6] N. Yu, H. Guo, J. Yang, L. Jin, X. Wang, W. Shi, X. Zhang, H. Yu, S. Wei, Non-target and suspect screening of per-and polyfluoroalkyl substances in airborne particulate matter in China, Environ. Sci. Technol. 52 (2018) 8205–8214. [7] W. Wang, A. Maimaiti, H. Shi, R. Wu, R. Wang, Z. Li, D. Qi, G. Yu, S.B. Deng, Adsorption behavior and mechanism of emerging perfluoro-2-

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29] [30]

[31]

propoxypropanoic acid (GenX) on activated carbons and resins, Chem. Eng. J. 364 (2019) 132–138. M. Clara, S. Scharf, S. Weiss, O. Gans, C. Scheffknecht, Emissions of perfluorinated alkylated substances (PFAS) from point sources—identification of relevant branches, Water Sci. Technol. 58 (2008) 59–66. S. Zhang, Q. Kang, H. Peng, M. Ding, F. Zhao, Y. Zhou, Z. Dong, H. Zhang, M. Yang, S. Tao, Relationship between perfluorooctanoate and perfluorooctane sulfonate blood concentrations in the general population and routine drinking water exposure, Environ. Int. 126 (2019) 54–60. S. Liu, Y. Lu, S. Xie, T. Wang, K.C. Jones, A.J. Sweetman, Exploring the fate, transport and risk of Perfluorooctane Sulfonate (PFOS) in a coastal region of China using a multimedia model, Environ. Int. 85 (2015) 15–26. X. Chen, X. Xia, X. Wang, J. Qiao, H. Chen, A comparative study on sorption of perfluorooctane sulfonate (PFOS) by chars, ash and carbon nanotubes, Chemosphere 83 (2011) 1313–1319. A. Maimaiti, S.B. Deng, P.P. Meng, W. Wang, B. Wang, J. Huang, Y.J. Wang, G. Yu, Competitive adsorption of perfluoroalkyl substances on anion exchange resins in simulated AFFF-impacted groundwater, Chem. Eng. J. 348 (2018) 494–502. Z. Wang, J.C. DeWitt, C.P. Higgins, I.T. Cousins, A never-ending story of per-and polyfluoroalkyl substances (PFASs)?, Environ Sci. Technol. 51 (2017) 2508– 2518. E. Sharifpour, E.A. Dil, A. Asfaram, M. Ghaedi, A. Goudarzi, Optimizing adsorptive removal of malachite green and methyl orange dyes from simulated wastewater by Mn-doped CuO-Nanoparticles loaded on activated carbon using CC-RSM: mechanism, regeneration, isotherm, kinetic, and thermodynamic studies, Appl. Organomet. Chem. 33 (2019) 4768. M. Sun, E. Arevalo, M. Strynar, A. Lindstrom, M. Richardson, B. Kearns, A. Pickett, C. Smith, D.R. Knappe, Legacy and emerging perfluoroalkyl substances are important drinking water contaminants in the Cape Fear River Watershed of North Carolina, Environ. Sci. Technol. Lett. 3 (2016) 415–419. Z.W. Du, S.B. Deng, Y. Bei, Q. Huang, B. Wang, J. Huang, G. Yu, Adsorption behavior and mechanism of perfluorinated compounds on various adsorbents—a review, J. Hazard. Mater. 274 (2014) 443–454. M.F. Rahman, S. Peldszus, W.B. Anderson, Behaviour and fate of perfluoroalkyl and polyfluoroalkyl substances (PFASs) in drinking water treatment: a review, Water Res. 50 (2014) 318–340. X. Xiao, B.A. Ulrich, B. Chen, C.P. Higgins, Sorption of poly-and perfluoroalkyl substances (PFASs) relevant to aqueous film-forming foam (AFFF)-impacted groundwater by biochars and activated carbon, Environ. Sci. Technol. 51 (2017) 6342–6351. Q. Yu, R. Zhang, S.B. Deng, J. Huang, G. Yu, Sorption of perfluorooctane sulfonate and perfluorooctanoate on activated carbons and resin: kinetic and isotherm study, Water Res. 43 (2009) 1150–1158. S.B. Deng, Q. Yu, J. Huang, G. Yu, Removal of perfluorooctane sulfonate from wastewater by anion exchange resins: effects of resin properties and solution chemistry, Water Res. 44 (2010) 5188–5195. Z.W. Du, S.B. Deng, Y. Chen, B. Wang, J. Huang, Y.J. Wang, G. Yu, Removal of perfluorinated carboxylates from washing wastewater of perfluorooctanesulfonyl fluoride using activated carbons and resins, J. Hazard. Mater. 286 (2015) 136–143. Z.W. Du, S.B. Deng, S. Zhang, B. Wang, J. Huang, Y.J. Wang, G. Yu, B.S. Xing, Selective and high sorption of perfluorooctanesulfonate and perfluorooctanoate by fluorinated alkyl chain modified montmorillonite, The J. Phys. Chem. C 120 (2016) 16782–16790. S.B. Deng, Y. Zheng, F. Xu, B. Wang, J. Huang, G. Yu, Highly efficient sorption of perfluorooctane sulfonate and perfluorooctanoate on a quaternized cotton prepared by atom transfer radical polymerization, Chem. Eng. J. 193 (2012) 154–160. Q. Zhang, S.B. Deng, G. Yu, J. Huang, Removal of perfluorooctane sulfonate from aqueous solution by crosslinked chitosan beads: sorption kinetics and uptake mechanism, Bioresource Technol. 102 (2011) 2265–2271. L. Xiao, Y. Ling, A. Alsbaiee, C. Li, D.E. Helbling, W.R. Dichtel, b-Cyclodextrin polymer network sequesters perfluorooctanoic acid at environmentally relevant concentrations, J. Am. Chem. Soc. 139 (2017) 7689–7692. Y.X. Gao, S.B. Deng, Z.W. Du, K. Liu, G. Yu, Adsorptive removal of emerging polyfluoroalky substances F-53B and PFOS by anion-exchange resin: a comparative study, J. Hazard. Mater. 323 (2016) 550–557. J. Yu, L. Lv, P. Lan, S. Zhang, B. Pan, W. Zhang, Effect of effluent organic matter on the adsorption of perfluorinated compounds onto activated carbon, J. Hazard. Mater. 225 (2012) 99–106. A. Asfaram, M. Ghaedi, K. Dashtian, G.R. Ghezelbash, Preparation and characterization of Mn0.4Zn0.6Fe2O4 nanoparticles supported on dead cells of Yarrowia lipolytica as a novel and efficient adsorbent/biosorbent composite for the removal of azo food dyes: central composite design optimization study, ACS Sustain, Chem. Eng. 6 (2018) 4549–4563. T.H. Boyer, P.C. Singer, Stoichiometry of removal of natural organic matter by ion exchange, Environ. Sci. Technol. 42 (2007) 608–613. S. Rayne, K. Forest, Perfluoroalkyl sulfonic and carboxylic acids: a critical review of physicochemical properties, levels and patterns in waters and wastewaters, and treatment methods, J. Environ. Sci. Heal. A 44 (2009) 1145– 1199. A. Asfaram, M. Ghaedi, S. Agarwal, I. Tyagi, V.K. Gupta, Removal of basic dye Auramine-O by ZnS: Cu nanoparticles loaded on activated carbon: optimization of parameters using response surface methodology with central composite design, RSC Adv. 5 (2015) 18438–18450.

W. Wang et al. / Journal of Colloid and Interface Science 557 (2019) 655–663 [32] A. Moldes, J. Alonso, J. Parajo, Recovery of lactic acid from simultaneous saccharification and fermentation media using anion exchange resins, Bioproc. Biosyst. Eng. 25 (2003) 357–363. [33] Q. Yu, S.B. Deng, G. Yu, Selective removal of perfluorooctane sulfonate from aqueous solution using chitosan-based molecularly imprinted polymer adsorbents, Water Res. 42 (2008) 3089–3097. [34] B. Pan, W. Zhang, B. Pan, H. Qiu, Q. Zhang, Q. Zhang, S. Zheng, Efficient removal of aromatic sulfonates from wastewater by a recyclable polymer: 2-

663

naphthalene sulfonate as a representative pollutant, Environ. Sci. Technol. 42 (2008) 7411–7416. [35] Y. Sumida, T. Oki, A. Masuyama, H. Maekawa, M. Nishiura, T. Kida, N. Yohji, I. Isao, M. Nojima, p A isotherms for triple-chain amphiphiles bearing two or three hydroxyl groups. Effect of the backbone structure on the adsorption behavior of the molecules on the surface, Langmuir 14 (1998) 7450–7455.