ARTICLE pubs.acs.org/est
Effects of Monovalent Cations on the Competitive Adsorption of Perfluoroalkyl Acids by Kaolinite: Experimental Studies and Modeling Feng Xiao,†,‡ Xiangru Zhang,§ Lee Penn,z John S. Gulliver,† and Matt F. Simcik*,‡ †
Department of Civil Engineering, University of Minnesota, Minneapolis, Minnesota 55455, United States Division of Environmental Health Sciences, School of Public Health, University of Minnesota, Minneapolis, Minnesota 55455, United States § Department of Civil and Environmental Engineering, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong SAR, China z Department of Chemistry, University of Minnesota, 207 Pleasant Street SE, Minneapolis, Minnesota 55455, United States ‡
bS Supporting Information ABSTRACT: Our hypothesis that longer-chained perfluoroalkyl acids (PFAAs) outcompete shorter-chained PFAAs during adsorption was tested in this study, wherein the adsorption interactions of six frequently detected PFAAs with kaolinite clay were modeled and examined experimentally using various suspension compositions. Competitive adsorption of PFAAs on the kaolinite surface was observed for the first time, and longer-chained PFAAs outcompeted those with a shorter chain. The electrostatic repulsion between adsorbed PFAA molecules is a primary inhibitory factor in PFAA adsorption. An increase in aqueous sodium or hydrogen ion concentration weakened electrostatic repulsions and changed the adsorption free energy. Therefore, the adsorption of a shorter-chained PFAA with weaker hydrophobicity could occur at high sodium or hydrogen ion concentrations. The experimental and modeling data suggest that the adsorption of shorter-chained PFAAs (e4 perfluorinated carbons) in freshwater with a typical ionic strength of 102.5 is not thermodynamically favorable. Furthermore, by measuring the electrokinetic potential of kaolinite suspension in the presence of PFAAs, we found that the kaolinite surface became more negatively charged because of the adsorption of PFAAs. This observation indicates that the adsorbed PFAA molecules were within the electrical double layer of the kaolinite surface and that they contributed to the potential at the slipping plane. The possible alignments of adsorbed PFAA molecules on the kaolinite surface were then proposed.
’ INTRODUCTION The scientific community and the public have become increasingly concerned about the contamination of worldwide freshwater and marine sites by perfluoroalkyl acids (PFAAs). PFAAs are synthetic organic compounds in which all the CH bonds are replaced by CF bonds. PFAAs and their precursors have been produced and used widely for decades in cookware, fast food containers, fire-fighting foams, and painting materials.1,2 Longchain PFAAs (g7 perfluorinated carbons) are persistent organic pollutants that cannot be hydrolyzed, directly photolyzed, or biodegraded under environmental conditions.3,4 In surface water, the levels of PFAAs range from tens to hundreds of ng/L. In some seriously contaminated groundwater, the concentrations can be thousands of ng/L. Table S1 in the Supporting Information summarizes the previously reported concentrations of PFAAs in rivers and groundwater at a variety of geographical locations. Commonly detected PFAAs include perfluorooctanesulfonic acid (PFOS) and perfluorinated carboxylates (perfluoroheptanoic acid (PFHpA), perfluorooctanoic acid (PFOA), perfluorononanoic acid (PFNA), perfluorodecanoic acid (PFDA), and perfluoroundecanoic acid (PFUnDA)). Long-chain PFAAs can accumulate in the blood and disturb human hormone activity.57 r 2011 American Chemical Society
Lau et al. have reported on their developmental toxicity for mammals.8 More recently, serum PFAA levels have been linked to the attention deficit/hyperactivity disorders in children,9 hyperuricemia,10 and thyroid disease.11 In 2009, PFOS was listed as a persistent organic pollutant in Annex B of the Stockholm Convention.12 A fundamental characteristic of surfactants is their tendency to adsorb at solidwater interfaces in an oriented fashion.13 PFAA adsorption by natural adsorbents (soils, suspended solids, sediments, and the rocks of aquifers) is an important determinant of their transport and fate in the environment. Previous studies have shown that the adsorption of PFAA is controlled by both hydrophobic and electrostatic effects.1420 Greater organic carbon content of the adsorbent, higher aqueous Ca2+ or H+ concentration, or longer perfluorocarbon chains can increase the adsorption of PFAAs.1420 Nevertheless, the published reports do not provide a complete and consistent picture of PFAA adsorption. First, the effect of Received: July 21, 2011 Accepted: October 10, 2011 Revised: September 28, 2011 Published: October 10, 2011 10028
dx.doi.org/10.1021/es202524y | Environ. Sci. Technol. 2011, 45, 10028–10035
Environmental Science & Technology Na+ and the role of electrostatic interactions need to be better understood. NaCl has been used as a road deicing agent for decades in many northern regions, where snowmelt during the spring carrying a significant load of NaCl runs into storm sewers, ditches, and small streams and then empties into wetlands, lakes, and rivers.21 In seawater, the concentration of Na+ can reach 0.5 M.22 Knowledge of the effect of Na+ on PFAA adsorption may provide key insights into the environmental fate of PFAAs and the associated ecosystem risks.1 A recent study from Tang et al.17 described the effect of sodium on the electrostatic interaction between PFAA molecules adsorbed on mineral surfaces. However, there is a lack of quantitative estimations about this interaction. In the present study, the effects of sodium on the electrokinetic potential (ζ) of kaolinite suspensions with and without the presence of PFAAs are investigated and modeled to provide both the qualitative and the quantitative descriptions of the electrostatic interaction. Second, it is not clear whether PFAA molecules compete for suitable sites during adsorption. No competition was found during the sorption of PFAAs to sediment.15 However, the hydrophobic effect increases with the hydrophobic length of the PFAA,15 and thus, a longer-chained PFAA may outcompete a shorter-chained PFAA during adsorption. One study has observed the competitive adsorption of PFAAs to aerobic active sludge.20 The purpose of this paper is to experimentally examine and model important chemical and mineralogical factors affecting PFAA adsorption. With this goal, PFAA adsorption by kaolinite clay is investigated with respect to the influences of sodium concentration and PFAA perfluorocarbon chain length. The effect of PFAA adsorption on kaolinite surface charges is also studied. Kaolinite was selected as the adsorbent because it is a major type of clay mineral in the soils and sediments in warm climates.23 Understanding the interactions between kaolinite and PFAAs can provide mechanistic insights into the adsorption behavior of PFAAs and can facilitate the assessment of the relative contribution from mineral components to PFAA adsorption by soils and sediments.
’ MATERIALS AND METHODS Chemicals. PFHpA, PFOA, PFNA, PFOS, PFDA, and PFUnDA were purchased from Sigma(Aldrich) (Milwaukee, WI, USA, and Steinheim, Switzerland; see Table S2 in the Supporting Information for detailed information). Methanol (optima grade from Fisher Scientific) and water (HPLC grade from Fisher Scientific) were used to clean the containers and tubes used in the experiments. No PFAAs were found in the methanol or water when this was checked using a mass spectrometer (MS) equipped with an electrospray ionization (ESI) source and interfaced to a high-pressure liquid chromatograph (HPLC). Two kinds of stock solutions were prepared. The individual PFAA stock solutions were prepared by dissolving a single PFAA in methanol (1 mM). A multiple PFAA stock solution was made by dissolving all the PFAAs except for PFHpA (not available at the time) in methanol (1 mM for each PFAA). Each stock solution was split into three aliquots. One was used in the experiment, and the other two were used to periodically check for any changes in the concentration. All the stock solutions were stored at 20 °C. Adsorption Experiments. The batch adsorption experiments were carried out in triplicate in 50 mL polystyrene tubes containing 40 mL of test suspension, which were continuously shaken by a wrist action shaker (Burrell) for 48 h at 22.2 ( 0.5 °C.
ARTICLE
Preliminary tests had shown that the time required to reach adsorption equilibrium was 48 h. Unless otherwise specified, the test suspension was composed of 1.0 mM NaHCO3 (alkalinity), 1.0 mM NaCl, and 20 mg kaolinite clay (Fluka) that had been ultrasonically dispersed in the water. The size distribution of the kaolinite suspension was measured by a Coulter laser diffraction particle size analyzer (LS 320, Coulter Electronics, USA). The kaolinite had a mean diameter of 1.1 μm with a narrow size distribution (Figure S1 in the Supporting Information). The surface area of the kaolinite particles was 10 m2/g, and the cation exchange capacity of kaolinite was 3.3 meq/100 g measured by a summation method (Research Analytical Laboratory, University of Minnesota). Before being used, the kaolinite was stored in a desiccator. Except for the isotherm tests, the initial concentration of PFAA in the tests was 1.0 μM (413 μg/L for PFOA, 463 μg/L for PFNA, 499 μg/L for PFOS, 513 μg/L for PFDA, and 563 μg/L for PFUnDA; these concentrations are higher than most reports for environmental samples, but were in the linear adsorption range as determined by isotherm tests). This was ensured by spiking 40 μL of PFAA stock solution into 40 mL of the test suspension. The tiny amount of methanol (40 μL) from the stock solution did not have a detectable effect on the adsorption results (Figure S2 in the Supporting Information). The adsorption of PFAAs by kaolinite was conducted at different aqueous Na+ concentrations to study the effect of Na+ concentration. The solution pH was adjusted to 7.5 with 0.1 M HCl or NaOH. Because the effect of H+ on the adsorption of PFAAs is wellknown,1420 there is no need to study the pH effect further. Adsorption isotherms were obtained to assess the PFAA distribution between the solid and aqueous phases as a function of the aqueous PFAA concentration at equilibrium (Cw). Both single-compound and multicompound adsorption experiments were studied. In the single-compound system, only one PFAA was used as the adsorbate, while all the PFAAs were dosed in the multicompound system. Determination of SolidWater Distribution Coefficient (Kd). After 48 h of equilibration, the samples were pretreated using an approach similar to that reported by Johnson et al.16 The detailed steps for determining the values of Kd can be found in the Supporting Information. Quantification of PFAAs. Each sample for HPLC/ESI-MS analysis was placed in a 300 μL insert (Chrom Tech) in a Wheaton vial and was crimp-sealed with a natural rubber septum (Chrom Tech). The analytical column was a Luna C18 column (50 1.0 mm, 5 μm) (Phenomenex). The guard column was a KJO-4282 one (Phenomenex). The parameters for operating the HPLC/ESI-MS system were set according to the methods reported in previous studies.16,24,25 The detailed information about the HPLC/ESI-MS analysis of samples, instrumental detection/quantification limits, quality assurance and quality control results (blanks and recovery rates), and typical HPLC/ ESI-MS chromatograms for the calibration standard and blank (Figure S3) can be found in the Supporting Information. ζ-Potential Measurement. Kaolinite is a 1:1 layered clay composed of repeating tetrahedral sheets of SiO44 and octahedral sheets of Al(OH)63. It is negatively charged in water in the circumneutral pH range because of the deprotonation of the hydroxyl groups on the octahedral sheets and the permanent charge deficiency on the tetrahedral layers.26,27 Cationic counterions (e.g., Na+) are attracted to the negatively charged kaolinite surface resulting in the so-called electric double layer (EDL). A measure of ζ-potential of kaolinite suspension in the presence 10029
dx.doi.org/10.1021/es202524y |Environ. Sci. Technol. 2011, 45, 10028–10035
Environmental Science & Technology
ARTICLE
Table 1. SolidWater Distribution Coefficients in Logarithmic Form (log Kd, L/kg)a single-compound system
multicompound system
Δ (%)
PFHpA
ND
NA
PFOA PFNA
0.36 ( 0.08 0.74 ( 0.06
ND 0.30 ( 0.11
63.6b
PFOS
1.16 ( 0.05
0.88 ( 0.03
48.6b
PFDA
1.30 ( 0.04
1.05 ( 0.04
43.4b
PFUnDA
1.70 ( 0.05
1.72 ( 0.04
5.7
a
Measured at pH 7.5 with 1 mM NaCl and NaHCO3. ND: not detected. NA: not applicable (PFHpA was not involved in the multicompound system). b Bold values are significant at the alpha level of 0.05 (see Table S0 in the Supporting Information).
of PFAAs can provide useful information about the interactions of kaolinite particles and PFAA molecules. The ζ-potential was measured by means of a zeta potential analyzer (ZetaPlus, Brookhaven) in PFOS/kaolinite suspensions at different Na+ concentrations and in PFAA/kaolinite suspensions at a single Na+ concentration. The ζ-potential analyzer was calibrated at 22 °C. The measurements were repeated to provide 10 ζpotential data points for each sample.
’ RESULTS AND DISCUSSION Linear Adsorption Isotherms and the Effect of PFAA Structure. The observed adsorption isotherms were linear over
the concentration range (1003000 μg/L) examined in this study (Figure S4 of the Supporting Information). The linearity indicates that the value of Kd was insensitive to the concentration within the examined range. The results also indicate that the Kd values calculated from a single-compound concentration could be used for comparing PFAA adsorption. This finding about the linear adsorption isotherms is consistent with previous studies conducted by Ahrens et al.28 and by Liu and Lee.29 The single-point Kd values are presented in Table 1. The data show that each CF2 moiety increased the distribution coefficient by 0.46 log units in the single-compound system. The difference in log Kd per CF2 moiety can vary when the adsorbent is different, for example, mineral surfaces as compared to organic matter in sediments. Higgins and Luthy15 reported a greater contribution from each CF2 to log Kd during the sorption of PFAAs to sediments. Furthermore, a comparison of the partitioning characteristics of PFOS and PFNA reveals that the sulfonate moiety contributed 0.42 log units to the distribution coefficient (Table 1). The slightly larger size of the sulfonate moiety as compared to the carboxylate moiety (leading to slightly more hydrophobicity)15,26 is not sufficient to explain the observed difference in adsorption potentials between PFOS and PFNA. The difference is more likely due to specific electrostatic interactions. According to Pearson’s concept of hard and soft acids and bases, the carboxylate group is a soft base while the sulfonate group is a relatively hard one.22 A hard base is more readily adsorbed on oxide surfaces, which are hard acids.22,30 In short, the results indicate that both the hydrophobic chain length (m) of a PFAA and the functionality of the headgroup affect its adsorption to kaolinite. Amphiphilic compounds have the potential to adsorb onto minerals in hemimicelles when the organic ions are present at 0.0010.01 of the critical micelle concentration.26 Hemimicelle adsorption is usually characterized by a normal adsorption
Figure 1. Effects of sodium concentration on the adsorption of PFOS and PFOA by kaolinite in a single-compound system and in a multicompound system (pH: 7.5; initial PFAA concentration: 1 106 mol/L). Kd,salting‑out is the calculated distribution coefficient considering the salting-out effect only. In the multicompound system, the mass of adsorbed PFOA is lower than the instrumental quantification limit.
isotherm at low concentrations and then by a sharp increase in adsorption at hemimicelle concentrations.16 In the current study, the adsorption isotherms indicate that hemimicelles were unlikely to have formed. Effects of Monovalent Cations: Modeling. Higgins and Luthy31 developed a model including both hydrophobic and electrostatic components to predict the organic carbon normalized distribution coefficients of PFAAs. While their work offers a knowledgeable discussion of an important topic, their model should be more applicable to PFAA sorption by sediments or sorbents rich in organic matter. This is because their model was built on the basis that sediment sorption of PFAA and the electrostatic potential of the sediment organic matter were both insensitive to aqueous [Na+].31 In the present study, we observed a change of kaolinite surface charge with the addition of Na+ and the effect of Na+ addition on kaolinite adsorption of PFAAs (Figures 14). Therefore, a new model of the adsorption of PFAAs on mineral surfaces needs to be established. In addition, Tang et al.17 suggested the electrostatic repulsion between adsorbed PFAA molecules. This effect was considered in our model. The model developed here was mainly based on the conceptual model by Tang et al.17 and on the work of Schwarzenbach et al.26 Tang et al.17 documented that PFAA adsorption to a mineral surface is controlled by several effects or forces: (1) the hydrophobic effect that is the tendency of a PFAA molecule to exclude water molecules and to adsorb on an interface. It also includes the interaction between the perfluorocarbon chain of the PFAA and a hydrophobic moiety on the adsorbent surface; (2) the electrostatic interactions between the negatively charged groups of PFAAs and the adsorbent surface; and (3) the electrostatic repulsion between adsorbed PFAA molecules. Correspondingly, the change in total free energy ΔGadsorption associated with PFAA adsorption can be broken into hydrophobic and electrostatic parts ΔGadsorption ¼ ΔGhydrophobic þ ΔGelectrostatic;adsorbateadsorbent þ ΔGelectrostatic;adsorbateadsorbate
ð1Þ
The three terms on the right-hand side of eq 1 correspond to the hydrophobic effect, the PFAAsurface electrostatic interaction, and the electrostatic repulsion between the adsorbed PFAA 10030
dx.doi.org/10.1021/es202524y |Environ. Sci. Technol. 2011, 45, 10028–10035
Environmental Science & Technology
ARTICLE
Figure 2. The adsorption free energy (ΔGadsorption) as a function of m at different sodium concentrations (pH 7.5; initial PFAA concentration: 1 106 mol/L; ΔGadsorption = ΔGhydrophobic + ΔGelectrostatic = m ΔGCF2 + b). ΔGadsorption = 2.58m + 16.33 at [Na+] of 103.00 mol/L; ΔGadsorption = 2.69m + 13.27 at [Na+] of 101.18 mol/L; and ΔGadsorption = 2.47m + 9.83 at [Na+] = 100.18 mol/L. R2 > 0.97 for the linearity of all the fitted lines. mc is m corresponding to spontaneous adsorption (ΔGadsorption = 0). PFHxA: perfluorohexanoic acid; PFPA: perfluoropentanoic acid; PFBA: perfluorobutanoic acid; PFDoA: perfluorodecanoic acid.
molecules. The contribution of the CF2 moiety to ΔGhydrophobic can be expressed as26 ΔGhydrophobic ¼ m ΔGCF2
ð2Þ
where ΔGCF2 is the hydrophobic contribution made by each CF2 moiety driving these sorbates into the diffuse double layervicinal water layer, and m is the number of perfluorinated carbon of PFAA. This approach has also been applied to the adsorption of other surfactants.26,32,33 When the adsorption reaches equilibrium, we have the following relationship: ΔGadsorption ¼ RT ln Kd ¼ ΔGhydrophobic þ ΔGelectrostatic;adsorbateadsorbent þ ΔGelectrostatic;adsorbateadsorbate ¼ m ΔGCF2 þ b
ð3Þ
where b = ΔGelectrostatic,adsorbateadsorbent + ΔGelectrostatic,adsorbate adsorbate = ΔGelectrostatic. The slope of ΔGadsorption versus m is the value of ΔGCF2. Effects of Monovalent Cations: Experimental Results. The adsorption of PFAAs on kaolinite increased with increasing Na+ concentration (Figure 1). This result is consistent with the finding of Pavan et al.,34 who found that increasing Na+ concentration enhanced the adsorption of sodium dodecylsulfate (a C12 anionic surfactant) onto negatively charged clay. To gain more insight into PFAA adsorption, the Kd values of PFAA obtained at different Na+ concentrations were fitted to eq 3. The results are presented in Figure 2. As the figure shows, each CF2 moiety contributed 2.52.7 kJ/mol to ΔGhydrophobic. The values of ΔGhydrophobic and ΔGCF2 varied little at different Na+ concentrations, whereas ΔGelectrostatic was overwhelmingly affected by sodium. As illustrated in Figure 2, increasing Na+ concentration shifted ΔGelectrostatic to a less positive value and
Figure 3. (a) Effects of sodium on the surface charge of kaolinite at different ionic strengths with and without the presence of PFOS (pH: 7.5; initial PFAA concentration: 1 106 mol/L); (b) changes in the surface charge of kaolinite because of the adsorption of a PFAA ([NaCl]: 1 103 mol/L; pH: 7.5; initial PFAA concentration: 1 106 mol/L).
thus reduced the minimum m required for spontaneous adsorption (ΔGadsorption e 0). For example, the value of mc declined from 7 to 5 with the increasing Na+ concentration from 103.0 to 101.2 (see Figure 2) (the term mc refers to the value of m corresponding to ΔGadsorption = 0). As a result, the adsorption of PFHpA (m = 6) by kaolinite became possible at the higher Na+ concentration (Figure 2). The variation of ΔGelectrostatic with solution sodium concentration should be related with changes in the kaolinite’s surface charge. The pH of the zero point of charge of the kaolinite used in this study was 5.1; therefore, there is essentially no positive charge on the kaolinite surface at pH 7.5. As Schwarzenbach et al.26 have pointed out, because the kaolinite surface charge is of the same sign as the functional groups of PFAAs, there will be electrostatic repulsion between them. As apparent in Figure 3a, the kaolinite surface became less negatively charged at a high Na+ concentration because the EDL around the kaolinite surface was compressed by sodium ions. Consequently, the PFAAsurface electrostatic repulsion was reduced and the PFAA adsorption was increased. Figure 4 presents the values of mc as functions of Na+ concentration and ionic strength I. A strong linear relationship (R2 = 0.99) was found between mc and log[Na+]. The t test of the slope (see the Supporting Information) shows that the linear relationship is significant (p = 0.037) at a significance level of 0.05. The results imply that the adsorption of PFAAs with m smaller than six is not thermodynamically favorable in freshwater with a typical log I = 2.5.22 This is consistent with reports in the literature that perfluorohexanoic acid (PFHxA), perfluoropentanoic acid (PFPA), and perfluorobutanoic acid (PFBA) are rarely found to be adsorbed on particles in freshwater (Table S3 in the Supporting Information, which summarizes the reported PFAAs on sediments). According to Figure 4, for these shorter-chained PFAAs, higher Na+ concentrations are required to induce spontaneous adsorption. The adsorption of the shorter-chained PFAAs on particles can be thermodynamically possible in seawater, where the Na+ concentration can reach 0.5 M (corresponding to mc = 3.5),22 or under other high-ionic-strength environmental conditions. 10031
dx.doi.org/10.1021/es202524y |Environ. Sci. Technol. 2011, 45, 10028–10035
Environmental Science & Technology
ARTICLE
effect. These findings agree with the recent observations by Jeon et al.36 who found that the salting-out effect was not evident for PFAA adsorption on clay. Electrostatic Repulsion between Adsorbed PFAA Molecules. The value of ΔGelectrostatic,adsorbateadsorbent was evaluated by eq 526 ΔGelectrostatic;adsorbateadsorbent ¼ zFΨd ¼ FΨd ≈ Fζ
Figure 4. The change in mc as a function of ionic strength (I) or sodium concentration (pH 7.5; initial PFAA concentration: 1 106 mol/L.). p[Na+] = log [Na+]. mc = 0.82 log[Na+] + 3.89 or mc = 0.82p[Na+] + 3.89. mc is the number of perfluorinated carbons of PFAA corresponding to ΔGadsorption = 0. The dashed lines are the 95% confidence bands.
Previous studies have documented that solution pH has a remarkable influence on the adsorption of PFAAs onto charged solid surfaces.1518 It is illustrative to revisit the data of Higgins and Luthy.15 A strong linear relationship (R2 > 0.97) could be obtained between ΔGadsorption and m by processing their PFAAsediment interaction data at different pH values using eq 3 (see Figure S5 in the Supporting Information). As [H+] increased, the sediment surface charge became less negative,31 and thus, ΔGelectrostatic shifted to a less negative value, which in turn increased PFAA sorption to sediments. Similar to Figure 4, a moderately strong linear relationship (R2 = 0.60) could be obtained between pH/[H+] and mc from Figure S5 (see Figure S6 in the Supporting Information); the spontaneous sorption of a shorter-chained PFAA to sediments is more thermodynamically favorable at a higher hydrogen ion concentration or at a lower pH. For example, the spontaneous adsorption of PFHpA on sediment in 0.50 mM CaCl2 could occur only at pH less than 5.6 (see Figure S6 in the Supporting Information). In addition to the electrostatic effect, a salting-out effect has been suggested as the cause of the positive effects of cations on PFAA adsorption.19 The salting-out effect during PFAA adsorption can be estimated using Kd, salting-out ¼ Kd 10a 3 K
s
3 ½NaCl
ð4Þ
where Ks is the salting-out (Setschenow) constant, which is usually about 0.10.4.35 It can be calculated using an empirical relationship based on 101 organic compounds including halogenated acetic acids35 as 0.23, 0.27, 0.30, and 0.34 for PFOA, PFNA, PFDA, and PFUnDA, respectively. The value of a was 0.35, which was obtained by the linear free energy relationship between log Kd and log Cwsat (Cwsat is the solubility of PFAAs; see Figure S7 and Table S5 in the Supporting Information). As evident in Figure 1, the increase in PFOA’s Kd modeled by the empirical relationship for the salting-out effect (eq 4) was negligible compared to the observed increase in Kd, implying that the salting-out effect was unimportant at least for the range of Na+ concentrations examined in this work. In addition, the Kd values of PFOS and PFOA increased linearly instead of the exponential increment as predicated by eq 4 hinting that the effect of sodium on the sorption was not caused by the salting-out
ð5Þ
where z is the valence of PFAA (1), F is the Faraday constant (96 485 C/mol), and Ψd is the diffuse layer potential (V). Ψd can be approximated by the ζ-potential because the shear plane is often located close to the inner boundary of the diffuse layer.37,38 Therefore, on the basis of eqs 15, ΔGelectrostatic,adsorbateadsorbate can be calculated by ΔGelectrostatic;adsorbateadsorbate ¼ ΔGadsorption ðΔGhydrophobic þ ΔGelectrostatic;adsorbateadsorbent Þ ¼ ΔGadsorption m ΔGCF2 FΨd
ð6Þ The values of ΔGadsorption, ΔGhydrophobic, ΔGelectrostatic,adsorbate adsorbent, and ΔGelectrostatic,adsorbateadsorbate are tabulated in Table S4 in the Supporting Information. As shown, the value of ΔGelectrostatic,adsorbateadsorbate differed insignificantly among PFAAs and was ∼11.5 kJ/mol, which was about double that of the ΔGelectrostatic,adsorbateadsorbent. The result indicates that the electrostatic repulsion between adsorbed PFAA molecules is an important thermodynamic inhibitor for PFAA adsorption. The results agree with the finding from Tang et al.17 who observed that increasing the Na+ concentration increased the adsorption of PFOS on a weakly positively charged surface, and they attributed this counterintuitive phenomenon to the reduced electrostatic repulsions between adsorbed PFOS molecules. Remarkably, the electrostatic intermolecular repulsion between adsorbed PFAA molecules may be a unique characteristic for these chemicals. Johnson et al.16 have pointed out that the negative charge of PFAAs in water comes not only from their functional groups but also from their unique molecular structures. Figure S8 in the Supporting Information shows the partial atomic charges in PFOS and its nonfluorinated counterpart, octanesulfonate. The partial atomic charges were calculated by a method developed by Miller and Savchik39 (see the Supporting Information). The electrostatic partial maps of PFOS and octanesulfonate were then built using MarvinSketch software (ChemAxon, Hungary) on the basis of the calculation results and presented in Figure S9 of the Supporting Information. As shown, PFOS and octanesulfonate have distinctively different charge characteristics (Figure S9). In PFOS, the partially negatively charged fluorine, oxygen, and sulfur form an electron shell around the partially positively charged carbon framework; the carbon atoms are screened by fluorine, oxygen, and sulfur. A PFOS molecule is negatively charged to the surrounding water because of both its ionized functional group and its partially negatively charged hydrophobic chain. Competitive Adsorption. Previous studies have observed competitive adsorption between ionizable surfactants as summarized by Parida et al.40 For PFAAs, there is limited information available on the possibility of competitive adsorption. In the present study, the competition among PFAAs during their adsorption on the kaolinite surface was observed. The Kd values 10032
dx.doi.org/10.1021/es202524y |Environ. Sci. Technol. 2011, 45, 10028–10035
Environmental Science & Technology of PFAAs obtained from the single-compound system and the multicompound system are tabulated in Table 1. The adsorption of the shorter-chained PFAAs (PFOA, PFNA, and PFOS) was significantly influenced by the presence of other PFAAs. The Kd values of PFNA and PFOS in the multicompound system were around half of the values in the single-compound system (Table 1). Table S0 in the Supporting Information includes the 95% confidence intervals for the difference in the Kd values obtained from the two systems. The 95% confidence intervals do not contain zero for PFNA, PFOS, and PFDA; therefore, the differences in the Kd values obtained between the two systems for these relatively shorter-chained PFAAs are significant at the alpha level of 0.05. On the other hand, the 95% confidence interval contains zero for PFUnDA, the longest-chained PFAA in this study, indicating that its Kd values in the two systems differ insignificantly at the alpha level of 0.05 (Table 1). The statistical analysis results show that the adsorption of PFUnDA was not significantly influenced by the presence of shorter-chained PFAAs but that the adsorption of shorter-chained PFAAs was greatly influenced by the presence of other PFAAs. The competitive adsorption among PFAAs is also evident after comparing the effects of Na+ in the two adsorption systems (Figure 1). In the single-compound system, the addition of the counterion significantly increased the adsorption of PFAAs (Figure 1) as has been discussed. In the multicompound system, adding Na+ also increased the adsorption of PFOS, but the degree of increase was much smaller than that in the singlecompound system. For PFOA, the Na+ effect on its adsorption by kaolinite was so weak in the multicompound system that the mass of adsorbed PFOA was below the instrumental quantification limit in contrast with the substantially positive effect of Na+ on PFOA adsorption in the single-compound system (Figure 1). The competitive adsorption should be caused by site competition and electrostatic effects. A longer-chained PFAA can outcompete a shorter-chained PFAA because of the stronger hydrophobic effect. Furthermore, as mentioned earlier, PFAAs have a unique negatively charged shell (Figure S9 of the Supporting Information). Within the electrical field of a PFAA molecule, other molecules may not be allowed to adsorb. Kaolinite’s ζ-Potential in the Presence of PFAAs and Possible Orientation of PFAA Molecules on the Kaolinite Surface. Figure 3 shows the ζ-potentials of the kaolinite suspensions in the presence of PFOS at different Na+ concentrations and in the presence of different PFAAs at a single Na+ concentration. After adsorbing PFAAs, the kaolinite surface’s ζ-potential was shifted to a more negative value (Figure 3). Previous studies have observed that the surface ζ-potential can be shifted to a more negative value because of the adsorption of an anionic surfactant.33,34 As evident in Figure S10 (see Supporting Information) which links the kaolinite surface’s ζ-potentials in the presence of PFAAs and the Kd values of PFAAs, a PFAA with a higher Kd value can change the kaolinite surface’s ζ-potential to a larger degree. The results (illustrated in Figures 3 and S10) indicate that the adsorbed PFAA molecules were within the EDL of the kaolinite surface and contributed to the potential at the slipping plane (ζ-potential). Furthermore, the ζ-potential of PFOS-adsorbed kaolinite was much less negative than that of PFNA-adsorbed kaolinite, and it was close to that of PFOAadsorbed kaolinite (Figure 3b). This was observed despite the fact that PFOS had a higher Kd than PFOA and PFNA. The observation indicates that different functional groups of PFAAs have different effects in changing the ζ-potential of the kaolinite
ARTICLE
surface. This is consistent with the above discussion of the possible specific interactions between the sulfonate group and the oxide surface; the sulfonate group of PFOS was further away from the shear plane or closer to the mineral surface, which resulted in greater EDL screening than PFNA. An interesting question that has yet to be answered is the orientation and packing of PFAA molecules at the kaolinite water interface. The thickness of the EDL or the Debye length can be estimated as 0.28 I0.5 (nm).26 For the ionic compositions of the test suspensions in this study, this means that the adsorbed PFAA molecules were packed into a layer of water 0.36.3 nm thick. This leads to a basic question as to how the head or tail of each PFAA was aligned on the kaolinite surface in this thin water layer. In a low ionic strength environment, the hydrophilic head of an adsorbed PFAA molecule should be oriented predominately toward the aqueous phase (Figure S11a in the Supporting Information) so as to minimize both the contact between the charged head and the charged kaolinite surface and the contact between the hydrophobic tail and the water molecules. Interestingly, at high ionic strength (e.g., log[Na+] = 0.18) when the EDL is only ∼0.3 nm thick, PFAA molecules cannot be aligned as they are at low ionic strength because of their relatively long perfluorocarbon chains (∼1 nm, see Supporting Information). We propose that the adsorbed PFAA molecules should thus be aligned as illustrated in Figure S11b when the repulsion between the mineral surface and the PFAA is low in high ionic strength conditions. Environmental Significance. PFAAs have been found to be ubiquitous in the environment.14,4155 The fate and transport of PFAAs in aquatic environments strongly depend on their interactions with solid surfaces. To adequately describe their transport and fate, the adsorption characteristics of PFAAs need to be better understood. This study systematically examined and modeled the effects of sodium on PFAA adsorption characteristics in an attempt to elucidate the relative importance of hydrophobic and electrostatic interactions and the possible competitive adsorption. The current work advances the understanding of PFAA adsorption to kaolinite and facilitates estimating the fate and transport of PFAAs in the environment. The results reveal that PFAA transport can be seriously retarded in high ionic strength conditions such as brackish groundwater, river estuaries, seawater, or snowmelt containing road salts. This supports the observation made by Pan and You56 who found that PFOS was largely adsorbed to the sediment in estuaries. This study provided the first observation of the competitive adsorption of PFAAs on a natural adsorbent, which contributes to the understanding of the behavior of ionizable surfactants at a solidwater interface. Furthermore, industry has been using a short-chain PFAA, perfluorobutane sulfonate (PFBS), which has four carbons, in place of PFOS.57 Very little is known about the fate and activity of short-chain PFAAs in the environment. The results of this study imply that the adsorption of C4 PFAAs in freshwater systems is unlikely.
’ ASSOCIATED CONTENT
bS Supporting Information. Figures S1S11, Tables S0S5, QA/QCs, electrostatic partial maps of PFOS and octanesulfonate, lengths of PFOA and PFOS, and t test of the association between mc and log[Na+]. This material is available free of charge via the Internet at http://pubs.acs.org. 10033
dx.doi.org/10.1021/es202524y |Environ. Sci. Technol. 2011, 45, 10028–10035
Environmental Science & Technology
’ AUTHOR INFORMATION Corresponding Author
*Phone: +1-612-626-6269; fax: +1-612-626-0650; e-mail: msimcik@ umn.edu.
’ ACKNOWLEDGMENT This work was supported by the Minnesota Water Research Center (2009MN 253B). We thank Melissa Jones and Dr. Christy Haynes of The University of Minnesota for help of parts of the ζ-potential analysis. We are grateful for the constructive comments of the anonymous referees. ’ REFERENCES (1) Jeon, J.; Kannan, K.; Lim, H. K.; Moon, H. B.; Kim, S. D. Bioconcentration of perfluorinated compounds in blackrock fish, Sebastes schlegeli, at different salinity levels. Environ. Toxicol. Chem. 2010, 29 (11), 2529–2535. (2) Giesy, J. P.; Kannan, K. Global distribution of perfluorooctane sulfonate in wildlife. Environ. Sci. Technol. 2001, 35 (7), 1339–1342. (3) Co-operation on Existing Chemicals: Hazard assessment of perfluorooctane sulfonate and its salts; Environment Directorate Joint Meeting of the Chemicals Committee and the Working Party on Chemicals, Pesticides and Biotechnology; Organization for Economic Co-operation and Development (OECD): Paris, 2002. (4) Prevedouros, K.; Cousins, I. T.; Buck, R. C.; Korzeniowski, S. H. Sources, fate and transport of perfluorocarboxylates. Environ. Sci. Technol. 2006, 40 (1), 32–44. (5) Calafat, A. M.; Wong, L. Y.; Kuklenyik, Z.; Reidy, J. A.; Needham, L. L. Polyfluoroalkyl chemicals in the U.S. population: Data from the National Health and Nutrition Examination Survey (NHANES) 20032004 and comparisons with NHANES 19992000. Environ. Health Perspect. 2007, 115 (11), 1596–1602. (6) Olsen, G. W.; Burris, J. M.; Ehresman, D. J.; Froehlich, J. W.; Seacat, A. M.; Butenhoff, J. L.; Zobel, L. R. Half-life of serum elimination of perfluorooctanesulfonate, perfluorohexanesulfonate, and perfluorooctanoate in retired fluorochemical production workers. Environ. Health Perspect. 2007, 115 (9), 1298–1305. (7) Weiss, J. M.; Andersson, P. L.; Lamoree, M. H.; Leonards, P. E. G.; van Leeuwen, S. P. J.; Hamers, T. Competitive binding of poly- and perfluorinated compounds to the thyroid hormone transport protein transthyretin. Toxicol. Sci. 2009, 109 (2), 206–216. (8) Lau, C.; Butenhoff, J. L.; Rogers, J. M. The developmental toxicity of perfluoroalkyl acids and their derivatives. Toxicol. Appl. Pharmacol. 2004, 198 (2), 231–241. (9) Hoffman, K.; Webster, T. F.; Weisskopf, M. G.; Weinberg, J.; Vieira, V. M. Exposure to polyfluoroalkyl chemicals and attention deficit/hyperactivity disorder in U.S. children 1215 years of age. Environ. Health Perspect. 2010, 118 (12), 1762–1767. (10) Steenland, K.; Tinker, S.; Shankar, A.; Ducatman, A. Association of perfluorooctanoic acid (PFOA) and perfluorooctane sulfonate (PFOS) with uric acid among adults with elevated community exposure to PFOA. Environ. Health Perspect. 2010, 118 (2), 229–233. (11) Melzer, D.; Rice, N.; Depledge, M. H.; Henley, W. E.; Galloway, T. S. Association between serum perfluorooctanoic acid (PFOA) and thyroid disease in the U.S. National Health and Nutrition Examination survey. Environ. Health Perspect. 2010, 118 (5), 686–692. (12) Wang, T.; Wang, Y. W.; Liao, C. Y.; Cai, Y. Q.; Jiang, G. B. Perspectives on the inclusion of perfluorooctane sulfonate into the Stockholm Convention on persistent organic pollutants. Environ. Sci. Technol. 2009, 43 (14), 5171–5175. (13) Rosen, M. J. Surfactants and Interfacial Phenomena, 3rd ed.; John Wiley & Sons: Hoboken, NJ, 2004. (14) Deng, S. B.; Zhou, Q.; Yu, G.; Huang, J.; Fan, Q. Removal of perfluorooctanoate from surface water by polyaluminium chloride coagulation. Water Res. 2011, 45 (4), 1774–1780.
ARTICLE
(15) Higgins, C. P.; Luthy, R. G. Sorption of perfluorinated surfactants on sediments. Environ. Sci. Technol. 2006, 40 (23), 7251–7256. (16) Johnson, R. L.; Anschutz, A. J.; Smolen, J. M.; Simcik, M. F.; Penn, R. L. The adsorption of perfluorooctane sulfonate onto sand, clay, and iron oxide surfaces. J. Chem. Eng. Data 2007, 52 (4), 1165–1170. (17) Tang, C. Y.; Fu, Q. S.; Gao, D. W.; Criddle, C. S.; Leckie, J. O. Effect of solution chemistry on the adsorption of perfluorooctane sulfonate onto mineral surfaces. Water Res. 2010, 44 (8), 2654–2662. (18) Wang, F.; Shih, K. M. Adsorption of perfluorooctanesulfonate (PFOS) and perfluorooctanoate (PFOA) on alumina: Influence of solution pH and cations. Water Res. 2011, 45 (9), 2925–2930. (19) You, C.; Jia, C. X.; Pan, G. Effect of salinity and sediment characteristics on the sorption and desorption of perfluorooctane sulfonate at sedimentwater interface. Environ. Pollut. 2010, 158 (5), 1343–1347. (20) Zhou, Q.; Deng, S. B.; Yu, Q.; Zhang, Q. Y.; Yu, G.; Huang, J.; He, H. P. Sorption of perfluorooctane sulfonate on organo-montmorillonites. Chemosphere 2010, 78 (6), 688–694. (21) Novotny, E. V. Road deicing salt impacts on urban water quality. Ph.D. Dissertation, University of Minnesota, Minneapolis, MN, 2009. (22) Snoeyink, V.; Jenkins, D. Water Chemistry; John Wiley & Sons: New York, 1980. (23) Li, Z. H.; Schulz, L.; Ackley, C.; Fenske, N. Adsorption of tetracycline on kaolinite with pH-dependent surface charges. J. Colloid Interface Sci. 2010, 351 (1), 254–260. (24) Simcik, M. F.; Dorweiler, K. J. Ratio of perfluorochemical concentrations as a tracer of atmospheric deposition to surface waters. Environ. Sci. Technol. 2005, 39 (22), 8678–8683. (25) Dorweiler, K. J. Detection and quantitation of perfluorooctane sulfonate and perfluorooctanoic acid in environmental matrices by single quadropole mass spectrometry. Master’s Dissertation, University of Minnesota, Minneapolis, MN, 2004. (26) Schwarzenbach, R. P.; Gschwend, P. M.; Imboden, D. M. Environmental Organic Chemistry, 1st ed.; John Wiley & Sons: New York, 1993. (27) Ma, C.; Eggleton, R. A. Cation exchange capacity of kaolinite. Clays Clay Miner. 1999, 47 (2), 174–180. (28) Ahrens, L.; Yeung, L. W. Y.; Taniyasu, S.; Lam, P. K. S.; Yamashita, N. Partitioning of perfluorooctanoate (PFOA), perfluorooctane sulfonate (PFOS) and perfluorooctane sulfonamide (PFOSA) between water and sediment. Chemosphere 2011, DOI: 10.1016/j. chemosphere.2011.06.046. (29) Liu, J. X.; Lee, L. S. Solubility and sorption by soils of 8:2 fluorotelomer alcohol in water and cosolvent systems. Environ. Sci. Technol. 2005, 39 (19), 7535–7540. (30) Walters, F. H. Design of corrosion inhibitors: Use of the hard and soft acidbase (HSAB) theory. J. Chem. Educ. 1991, 68 (1), 29–31. (31) Higgins, C. P.; Luthy, R. G. Modeling sorption of anionic surfactants onto sediment materials: An a priori approach for perfluoroalkyl surfactants and linear alkylbenzene sulfonates. Environ. Sci. Technol. 2007, 41 (9), 3254–3261. (32) Somasundaran, P.; Healy, T. W.; Fuerstenau, D. W. Surfactant adsorption at the solidliquid interface: Dependence of mechanism on chain length. J. Phys. Chem. 1964, 68 (12), 3562–3566. (33) Wang, J. B.; Han, B. X.; Dai, M.; Yan, H. K.; Li, Z. X.; Thomas, R. K. Effects of chain length and structure of cationic surfactants on the adsorption onto Na-kaolinite. J. Colloid Interface Sci. 1999, 213 (2), 596–601. (34) Pavan, P. C.; Crepaldi, E. L.; Gomes, G. D.; Valim, J. B. Adsorption of sodium dodecylsulfate on a hydrotalcite-like compound: Effect of temperature, pH and ionic strength. Colloids Surf., A: Physicochem. Eng. Aspects 1999, 154 (3), 399–410. (35) Ni, N.; Yalkowsky, S. H. Prediction of Setschenow constants. Int. J. Pharm. 2003, 254 (2), 167–172. (36) Jeon, J.; Kannan, K.; Lim, B. J.; An, K. G.; Kim, S. D. Effects of salinity and organic matter on the partitioning of perfluoroalkyl acid (PFAs) to clay particles. J. Environ. Monit. 2011, 13 (6), 1803–1810. 10034
dx.doi.org/10.1021/es202524y |Environ. Sci. Technol. 2011, 45, 10028–10035
Environmental Science & Technology (37) Carnali, J. O.; Shah, P. Correlation of surfactant/polymer phase behavior with adsorption on target surfaces. J. Phys. Chem. B 2008, 112 (24), 7171–7182. (38) Kretzschmar, R.; Hesterberg, D.; Sticher, H. Effects of adsorbed humic acid on surface charge and flocculation of kaolinite. Soil Sci. Soc. Am. J. 1997, 61 (1), 101–108. (39) Miller, K. J.; Savchik, J. A new empirical method to calculate average molecular polarizabilities. J. Am. Chem. Soc. 1979, 101 (24), 7206–7213. (40) Parida, S. K.; Dash, S.; Patel, S.; Mishra, B. K. Adsorption of organic molecules on silica surface. Adv. Colloid Interface Sci. 2006, 121 (13), 77–110. (41) Boulanger, B.; Vargo, J.; Schnoor, J. L.; Hornbuckle, K. C. Detection of perfluorooctane surfactants in Great Lakes water. Environ. Sci. Technol. 2004, 38 (15), 4064–4070. (42) Furdui, V. I.; Stock, N. L.; Ellis, D. A.; Butt, C. M.; Whittle, D. M.; Crozier, P. W.; Reiner, E. J.; Muir, D. C. G.; Mabury, S. A. Spatial distribution of perfluoroalkyl contaminants in lake trout from the Great Lakes. Environ. Sci. Technol. 2007, 41 (5), 1554–1559. (43) Jin, Y. H.; Liu, W.; Sato, I.; Nakayama, S. F.; Sasaki, K.; Saito, N.; Tsuda, S. PFOS and PFOA in environmental and tap water in China. Chemosphere 2009, 77 (5), 605–611. (44) K€arrman, A.; Mueller, J. F.; Van Bavel, B.; Harden, F.; Toms, L. M. L.; Lindstr€om, G. Levels of 12 perfluorinated chemicals in pooled Australian serum, collected 20022003, in relation to age, gender, and region. Environ. Sci. Technol. 2006, 40 (12), 3742–3748. (45) Loganathan, B. G.; Sajwan, K. S.; Sinclair, E.; Kumar, K. S.; Kannan, K. Perfluoroalkyl sulfonates and perfluorocarboxylates in two wastewater treatment facilities in Kentucky and Georgia. Water Res. 2007, 41 (20), 4611–4620. (46) Martin, J. W.; Mabury, S. A.; Solomon, K. R.; Muir, D. C. G. Bioconcentration and tissue distribution of perfluorinated acids in rainbow trout (Oncorhynchus Mykiss). Environ. Toxicol. Chem. 2003, 22 (1), 196–204. (47) McLachlan, M. S.; Holmstr€om, K. E.; Reth, M.; Berger, U. Riverine discharge of perfluorinated carboxylates from the European continent. Environ. Sci. Technol. 2007, 41 (21), 7260–7265. (48) Meyer, T.; De Silva, A. O.; Spencer, C.; Wania, F. Fate of perfluorinated carboxylates and sulfonates during snowmelt within an urban watershed. Environ. Sci. Technol. 2011, DOI: 10.1021/es200106q. (49) Moody, C. A.; Hebert, G. N.; Strauss, S. H.; Field, J. A. Occurrence and persistence of perfluorooctanesulfonate and other perfluorinated surfactants in groundwater at a fire-training area at Wurtsmith Air Force Base, Michigan, USA. J. Environ. Monit. 2003, 5 (2), 341–345. (50) Murakami, M.; Imamura, E.; Shinohara, H.; Kiri, K.; Muramatsu, Y.; Harada, A.; Takada, H. Occurrence and sources of perfluorinated surfactants in rivers in Japan. Environ. Sci. Technol. 2008, 42 (17), 6566–6572. (51) Nakayama, S. F.; Strynar, M. J.; Reiner, J. L.; Delinsky, A. D.; Lindstrom, A. B. Determination of perfluorinated compounds in the Upper Mississippi River Basin. Environ. Sci. Technol. 2010, 44 (11), 4103–4109. (52) Ruan, T.; Wang, Y. W.; Wang, T.; Zhang, Q. H.; Ding, L.; Liu, J. Y.; Wang, C.; Qu, G. B.; Jiang, G. B. Presence and partitioning behavior of polyfluorinated iodine alkanes in environmental matrices around a fluorochemical manufacturing plant: Another possible source for perfluorinated carboxylic acids? Environ. Sci. Technol. 2010, 44 (15), 5755–5761. (53) Scott, B. F.; Spencer, C.; Mabury, S. A.; Muir, D. C. G. Poly and perfluorinated carboxylates in North American precipitation. Environ. Sci. Technol. 2006, 40 (23), 7167–7174. (54) Sinclair, E.; Mayack, D. T.; Roblee, K.; Yamashita, N.; Kannan, K. Occurrence of perfluoroalkyl surfactants in water, fish, and birds from New York State. Arch. Environ. Contam. Toxicol. 2006, 50 (3), 398–410. (55) Washington, J. W.; Ellington, J. J.; Jenkins, T. M.; Evans, J. J.; Yoo, H.; Hafner, S. C. Degradability of an acrylate-linked, fluorotelomer polymer in soil. Environ. Sci. Technol. 2009, 43 (17), 6617–6623.
ARTICLE
(56) Pan, G.; You, C. Sediment-water distribution of perfluorooctane sulfonate (PFOS) in Yangtze River Estuary. Environ. Pollut. 2010, 158 (5), 1363–1367. (57) Betts, K. S. Perfluoroalkyl acids: What is the evidence telling us? Environ. Health Perspect. 2007, 115 (7), A344–A344.
10035
dx.doi.org/10.1021/es202524y |Environ. Sci. Technol. 2011, 45, 10028–10035
