Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Mar 16.
Published in final edited form as: Environ Sci Technol. 2021 Mar 5;55(6):3706–3715. doi: 10.1021/acs.est.0c07355

CONTRIBUTION OF NON-AQUEOUS PHASE LIQUIDS TO PFAS RETENTION AND TRANSPORT

Sarah Van Glubt 1, Mark L Brusseau 1,*
PMCID: PMC8634874  NIHMSID: NIHMS1757285  PMID: 33666425

Abstract

Per and polyfluoroalkyl substances (PFAS) co-contamination with non-aqueous phase organic liquids (NAPLs) has been observed or suspected at various sites, particularly at fire-training areas at which aqueous film-forming foams (AFFF) were applied. The objectives of this study are to (1) delineate the relative significance of specific PFAS-NAPL processes on PFAS retention, including partitioning into the bulk NAPL phase and adsorption to the NAPL-water interface; (2) investigate the influence of NAPL properties, saturation, and mass-transfer constraints on PFAS retention; and (3) determine whether PFAS may impact NAPL distribution through mobilization or dissolution. Perfluorooctane sulfonic acid (PFOS) and perfluorooctanoic acid (PFOA) are used as representative PFAS and trichloroethene (TCE) and decane are used as representative NAPLs. NAPL-water interfacial adsorption was quantified with NAPL-water interfacial-tension measurements, partitioning into NAPL was quantified with batch experiments, and retardation factors (R) in the absence and presence of residual NAPL were determined with miscible-displacement transport experiments. R values increased in the presence of residual NAPL, with adsorption to the NAPL-water interface accounting for as much as ~77% of retention and solid-phase adsorption also significantly contributing to retention. Additionally, this study provides the first QSPR analysis focused on NAPL-water interfacial adsorption coefficients, with results consistent with those from previous air-water studies. Lastly, this initial investigation into PFAS impacts on NAPL behavior determined that PFOS/PFOA are unlikely to enhance solubilization or mobilization of NAPL under the conditions present at many AFFF legacy sites.

Keywords: perfluorooctanoic acid, perfluorooctane sulfonic acid, TCE, decane, interfacial adsorption, partitioning

Introduction

The use of per and polyfluoroalkyl substances (PFAS) for industrial, commercial, and military applications has resulted in widespread occurrence of PFAS in the environment (e.g., refs 1–4). High (up to mg/L) levels of PFAS have been observed in groundwater for sites at which aqueous film-forming foams (AFFF) were used, such as fire-training areas (e.g., refs 5–12). Furthermore, recent research has demonstrated that soils and vadose zones at these types of sites typically contain large quantities of PFAS and therefore are likely to serve as long-term sources.13,14 Soil concentrations of up to ~400 mg/kg, for example, were reported for PFOS in a recent meta-analysis.14 A complicating factor for some AFFF sites is that non-aqueous phase organic liquids (NAPLs) such as chlorinated solvents and hydrocarbon fuels have been observed to co-occur with PFAS (e.g., refs 7, 11). Additionally, PFAS have been observed at fire-training sites with known or suspected NAPL use.5,6,10 Understanding the transport and fate of PFAS in source zones is critical for conducting accurate risk assessments and developing effective mitigation efforts.

The majority of prior laboratory research on PFAS transport in porous media has focused on the contributions of solid-phase sorption (e.g., refs 15–20) and air-water interfacial adsorption.21–26 However, additional retention processes may occur in NAPL-contaminated source zones, including partitioning into bulk NAPL and adsorption at NAPL-water interfaces.22,27–29 McKenzie et al.28 conducted miscible-displacement column experiments to investigate the impacts of TCE NAPL on the transport of a suite of PFAS in a loamy sand. Increased PFAS retention was observed in the presence of residual NAPL. PFAS partitioning into bulk NAPL and adsorption at the NAPL-water interface were discussed as possible reasons for the observed enhanced retardation, but their specific contributions were not delineated or quantified. Brusseau29 developed a comprehensive retention model for PFAS and used it to conduct a theoretical evaluation of the relative contributions of different processes, including NAPL-water interfacial retention, to total retardation. Brusseau et al.22 conducted a preliminary investigation of the impact of NAPL on PFAS transport with a series of miscible-displacement column experiments. A single PFAS (PFOS) and NAPL (decane) were investigated. They explicitly distinguished between partitioning into the bulk NAPL and NAPL-water interfacial adsorption, finding that adsorption to the NAPL-water interface contributed more than 70% to the total retention. Further research is needed to determine to what extent PFAS partitioning into NAPL and adsorption to the NAPL-water interface contribute to retention, as well as delineating the impacts of PFAS and NAPL properties, magnitude of NAPL saturation, and potential mass-transfer constraints on retention.

As surfactants, PFAS may have a number of potential impacts on the behavior of the NAPL. For example, their presence may result in changes to NAPL distribution through mobilization as a result of reductions in NAPL-water interfacial tensions. PFAS may also affect magnitudes or rates of NAPL dissolution. They may in addition enhance the apparent aqueous solubilities of NAPL constituents. To the best of our knowledge, these phenomena have not been previously investigated.

The objectives of this study are to (1) delineate the relative significance of specific PFAS-NAPL processes on PFAS retention, including partitioning to the bulk NAPL phase and adsorption to the NAPL-water interface; (2) investigate the influence of NAPL properties, saturation, and mass-transfer constraints on PFAS retention; and (3) determine whether PFAS may impact NAPL distribution through mobilization or dissolution. Interfacial tensions measured in this study are combined with data collected from the literature to examine the impact of PFAS molecular structure on NAPL-water interfacial adsorption. The first quantitative-structure/property-relationship (QSPR) analysis focused on NAPL-water interfacial adsorption of PFAS is presented. Transport experiments are conducted using two different NAPLs, two different NAPL saturations, and two different pore-water velocities. Solubilization and mobilization experiments are conducted to investigate the potential impact of PFOS on NAPL distribution. The results of these experiments are compared to standard behavior observed for hydrocarbon-surfactant systems.

Materials and Methods

A brief overview of the methods used for this study are presented below. Additional details of the experimental methods are provided in section S1 of the supplemental information (SI). The experiments included in this study are presented in Table S1.

Materials

PFOS (CAS# 1763-23-1, Sigma-Aldrich, 98% purity) and PFOA (CAS# 335-67-1, Alfa Aesar, 95%) were used as representative PFAS in the experiments. Additionally, perfluoropentanoic acid (PFPeA) was used for interfacial-tension measurements (CAS# 2706-96-3, TCI America, 98%). Pentafluorobenzoic acid (Strem Chemicals, 99% purity), which is not a PFAS, was used as the nonreactive tracer (NRT). Trichloroethene (CAS# 79-01-6, Sigma-Aldrich, >99.5% purity) and decane (CAS # 124-15-5, Sigma-Aldrich, 99% purity) were used as representative dense (DNAPL) and light (LNAPL) NAPLs, respectively. Experiments were conducted with either 0.01 M NaCl or synthetic groundwater (SGW) solutions. A commercial quartz sand (Accusand, Unimin Corp., Ottawa, Mn) was used in the column experiments. SGW and sand characteristics are detailed in section S1 of the SI.

Experiments to Measure PFAS Partitioning, Retention, and Transport

Interfacial tensions between aqueous PFAS solutions (PFOS, PFOA, or PFPeA) and TCE or decane NAPL were measured to determine adsorption to the NAPL-water interface. Measurements were conducted with a Du Nouy ring tensiometer (Fischer Scientific Tensiomat 21 or Sigma 701 precision force tensiometer) following standard methods (ASTM D1331-14). Additional details of the interfacial tension measurements are provided in the SI.

Interfacial-tension data were also collected from published studies to generate a database of NAPL-water interfacial adsorption measurements for PFAS. The literature data were digitized using the open-source Engauge program30. A list of the sources of data sets collected from the literature is provided in Table S2.

Batch experiments were conducted to measure PFAS partitioning into the bulk NAPL phase. Initial PFOS or PFOA concentrations of 2, 20, and 200 mg/L were used. High aqueous concentrations were used to reduce the impact of interfacial adsorption on measurements, given the nonlinearity of the NAPL-water interfacial adsorption coefficient in this concentration range. Additional details of the batch experiment methods are provided in the SI.

Miscible-displacement column experiments were conducted to examine PFAS retention and transport in the absence and presence of residual NAPL. Experiments were first conducted with the NRT to determine the hydrodynamic properties of the packed columns and ensure that they were well packed. Experiments were then conducted with PFOS or PFOA to measure sorption and retardation during transport. Additionally, one PFOS column experiment was conducted with a background solution of dissolved TCE at approximately 1200 mg/L to investigate the effects of TCE presence in solution on PFOS sorption. The NRT and PFOS/PFOA experiments were repeated after residual NAPL saturation was established within the columns. PFOS or PFOA input concentrations (C0) of 10 mg/L, and one experiment with C0 of 25 mg/L, were used for the column experiments to represent AFFF source-zone scenarios where PFAS concentrations are anticipated to be relatively high. Additional details of the miscible-displacement experiment methods are provided in the SI.

Experiments to Characterize NAPL Solubilization and Mobilization in the Presence of PFAS

Batch solubilization experiments were conducted to determine whether the presence of PFOS increased the aqueous concentration of TCE. PFOS concentrations of 0, 0.1, 1, 10, 100, and 500 mg/L in SGW were used.

Column mobilization experiments were conducted to determine whether the presence of PFOS could reduce NAPL interfacial tension sufficiently to cause mobilization. Columns with residual TCE were prepared using the same methods as described in the SI for the transport experiments. A 3 mL pulse containing 500 mg/L PFOS in 0.01 M NaCl solution was applied to the top of the column while the bottom of the column was open to allow drainage to occur. Effluent was collected and examined for free-phase NAPL.

Analytical and Data Analysis Methods

The pentafluorobenzoic acid and aqueous TCE samples were analyzed using ultraviolet-visible (UV-Vis) spectrophotometry (Shimadzu model 1601). PFOS and PFOA samples were analyzed by the methylene blue active substances (MBAS) assay, a standard method for quantifying single-component anionic surfactant concentrations. Prior work has demonstrated that this method produces results consistent with those from High Performance Liquid Chromatography tandem mass spectrometry (LCMS).19 Data analysis methods employed in this study are provided in the SI.

Results and Discussion

NAPL-Water Interfacial Adsorption

NAPL-water interfacial-tension measurements for PFOS, PFOA, and PFPeA solutions are shown in Figure 1. It is observed that interfacial activity is greatest for the longest-chain PFAS (PFOS) and weakest for the shortest-chain (PFPeA). This is consistent with prior studies. NAPL-water interfacial adsorption coefficient (Ki) values for PFOS, PFOA, and PFPeA determined from the NAPL-water interfacial-tension data for a specified concentration of 10 mg/L to match the column experiments are provided in Table 1. The methods used to calculate Ki are presented in the SI.

Figure 1.

Figure 1.

Open in a new tab

NAPL-water interfacial-tension measurements for PFOS solutions (0.01 M NaCl and SGW) PFOA solution (0.01 M NaCl), and PFPeA solution (SGW). The NAPL is TCE. SGW is synthetic groundwater.

Table 1.

Ki values determined from interfacial-tension measurements and column experiments and Kn values determined from batch NAPL-partitioning experiments for PFOS or PFOA

Analytea NAPL Interfacial Tension Ki (cm)b Column Ki (cm)b,c,d Kn
PFOS (NaCl) TCE 0.0050 0.0046 (0.0040-0.0052) [Sn 0.28]e
0.0048 (0.0045-0.0052) [Sn 0.26]f
0.0026 (0.0024-0.0029) [Sn 0.42]		 0.11 (0.027-0.18)
PFOS (SGW) TCE 0.0056 - -
PFOS (NaCl) Decane 0.0038

0.0015g 		0.0038 (0.0036-0.0041) [Sn 0.29]
0.0029 (0.0027-0.0032) [Sn 0.54]
0.0015 (0.0013-0.0018) [Sn 0.27]g 		0.022 (−0.003-0.046)
PFOS (SGW) Decane 0.0070 - -
PFOA (NaCl) TCE 0.0003 0.0003 (0.0002-0.0004) [Sn 0.31] 0.19 (0.034-0.35)
PFOA (NaCl) Decane - - 0.065 (−0.0085-0.14)
PFPeA (SGW) TCE 0.000022 - -
Open in a new tab
a

parentheses indicate background solution

b

Ki values correspond to C0=10 mg/L, except where noted otherwise

c

parentheses indicate the 95% confidence interval range

d

brackets indicate NAPL saturation for column experiments

e

values from two replicate experiments

f

Q = 0.1 mL/min

g

C0 = 25 mg/L

The PFOS TCE-water Ki values for the 0.01 M NaCl and SGW solutions are 0.0050 cm and 0.0056 cm, respectively. The PFOS decane-water Ki values are 0.0038 cm and 0.0070 cm for 0.01 M NaCl and SGW, respectively. The Ki values are greater for the SGW solutions compared to the 0.01 M NaCl solutions, although both solutions have the same ionic strength. This is consistent with the findings of Brusseau and Van Glubt31 and can be attributed to the presence of both monovalent and divalent cations in the SGW versus only monovalent cations in the NaCl solution. Divalent cations produce a greater reduction in electrostatic repulsion within the interface, thereby enhancing interfacial adsorption of the PFAS. The TCE-water Ki value for PFOA is 0.0003 cm, which is an order of magnitude smaller than the TCE-water Ki value for PFOS. This difference reflects the greater interfacial activity of PFOS and is consistent with observed differences in air-water interfacial adsorption of PFOS versus PFOA.22,31 The PFPeA TCE-water Ki value is one additional order of magnitude smaller, due to its shorter carbon chain compared to PFOS and PFOA. The preceding results are consistent with prior research showing that PFAS surface/interfacial activity and Ki values are influenced by surfactant properties, including carbon chain length and headgroup properties, and solution properties, including electrolyte composition and ionic strength (e.g., refs 29, 31–34).

Brusseau34 and Brusseau and Van Glubt31 conducted quantitative-structure/property-relationship (QSPR) analyses of PFAS air-water interfacial adsorption coefficients. QSPR models use empirical-based relationships to allow estimation of compound properties based on molecular structure. While a few NAPL-water interfacial adsorption coefficients were included in the Brusseau34 study, the present study to the best of our knowledge provides the first QSPR analysis focused specifically on NAPL-water interfacial adsorption. Molar volume (Vm, cm3/mol) is used as the single molecular descriptor, the same descriptor employed in the previous studies. Further description of the QSPR methods are included in Brusseau34 (2019a) and Brusseau and Van Glubt.31

The results of the QSPR analysis are shown in Figure 2, including Ki values from this study and those calculated from the literature data reported in Table S2. Separate data sets are shown for solutions with SGW or NaCl and deionized water. Data for two common hydrocarbon surfactants sodium dodecyl sulfate (SDS) and sodium dodecyl benzene sulfonate (SDBS) are included for comparison. The Vm values span approximately 200 cm3/mol and the Ki values span almost four log units. The regression determined for the QSPR analysis of PFAS air-water interfacial adsorption in electrolyte solutions reported by Brusseau and Van Glubt31 is presented in the figure. Also presented is the regression determined for the QSPR analysis of air-water interfacial adsorption of PFAS and hydrocarbon surfactants in deionized water reported by Brusseau.34

Figure 2.

Figure 2.

Open in a new tab

QSPR model for NAPL-water interfacial adsorption coefficient (Ki) versus molar volume (Vm) for PFAS and hydrocarbon (Hydro) surfactants. The Ki values are determined for a concentration of 0.1 mg/L to match the target concentrations used in the prior QSPR analyses conducted by Brusseau (BRU)34 and Brusseau and Van Glubt (B&VG).31 Electrolyte means the aqueous solution comprises either synthetic groundwater or NaCl. DIW is deionized water. For the following compounds, values are calculated as the mean from the number of data sets shown in parentheses: PFOS electrolyte (7), PFOA electrolyte (6), FOSA electrolyte (3) PFNA electrolyte (2), PFDA electrolyte (2), SDBS electrolyte (3), PFOA DIW (4). The error bars represent 95% confidence intervals. The PFAS appear in the following order from left to right for the electrolyte data set: PFPeA, PFOA, PFNA, PFOS, FOSA, and PFDA; the two hydrocarbons are SDS and SDBS from left to right.

The QSPR regressions developed for air-water interfacial adsorption provide reasonable representations of the measured data sets. The variability among the measurements for the same constituent is relatively moderate as shown by the error bars. The similarity of Ki values for the same constituent obtained from different data sources illustrates the consistency of the measured interfacial-tension data sets. It is observed that the Ki values for the PFAS and the two hydrocarbon surfactants are adequately represented by the same regression, for both the electrolyte and deionized-water data sets. This result is consistent with the QSPR analysis reported by Brusseau for air-water interfacial adsorption of PFAS and hydrocarbon surfactants.34 Notably, molar volume serves as a good descriptor for both air-water surface activity and NAPL-water interfacial activity. As discussed by Brusseau34, molar volume provides a representation of the influence of molecular size on solvation and hydrophobic interaction, the primary driving force for interfacial adsorption.

The Ki values for the deionized-water data are smaller than the values for the SGW and NaCl data. Similar results were observed for deionized water versus SGW data sets by Brusseau and Van Glubt31 for air-water interfacial adsorption. This is consistent with the greater surface activity of surfactants in solutions of greater ionic strength. These results indicate that ionic strength has a measurable but moderate impact on Ki.

Adsorption of surfactants at fluid-fluid interfaces is inherently nonlinear, which results in Ki being a function of the aqueous surfactant concentration. However, Ki attains a maximum value below some critical concentration as demonstrated in prior studies.34,37 This nonlinearity of Ki as a function of aqueous concentration needs to be considered in the application of a QSPR model for determining Ki values for a specific concentration range.34 The analysis requires selection of a target or representative concentration for which to calculate the Ki values. If the selected concentration is within the range of concentration at which Ki attains its maximum value, the model can be used for all concentrations lower than that value.34 However, a new QSPR model employing a different set of Ki values would be required if the selected concentration is within the range wherein adsorption is nonlinear and Ki is concentration dependent. A target concentration of 0.1 mg/L was selected for the present study, which is within the range at which the selected surfactants attain maximum values.

Partitioning into Bulk NAPL

The Kn values determined from the batch NAPL-water partitioning experiments are provided in Table 1. The Kn value determined from batch experiments for PFOS with TCE is 0.11, while the Kn value for PFOS with decane is 0.02, an order of magnitude lower, although the 95% confidence intervals overlap. A similar trend is observed for PFOA, where the Kn value for PFOA with TCE (0.19) is greater than the Kn value for PFOA with decane (0.07), although the 95% confidence intervals also overlap. The Kn values for PFOA are greater than those determined for PFOS for both TCE and decane, although for a given NAPL the Kn values are of the same order of magnitude and have overlapping 95% confidence intervals.

Guelfo and Higgins27 and McKenzie et al.28 both investigated the partitioning of a suite of PFAS into bulk NAPL with batch experiments. Guelfo and Higgins27 report Kn values of 0.98 (−0.85–2.8) for PFOS with TCE and 2.3 (−0.10–4.7) for PFOA with TCE. Note that these and all other values reported here have been converted to nondimensional units (values in parentheses are 95% confidence intervals). McKenzie et al.28 report similar Kn values of approximately 0.95 (−0.72–2.6) and 0.98 (−0.74–2.7) for PFOS and PFOA with TCE, respectively, when 3 mL of TCE was used and pH was 6, and 2.2 (1.0–3.5) and 1.5 (−0.29–3.4) for PFOS and PFOA with TCE, respectively, when 10 mL of TCE was used and pH was 6. Additionally, Guelfo and Higgins27 report Kn values of 0.18 (−0.06–0.42) for PFOS with dodecane and 1.1 (0.68–1.6) for PFOA with dodecane. While the mean values reported by Guelfo and Higgins27 and McKenzie et al.28 are greater than those reported in the present study, nearly all data sets have overlapping 95% confidence intervals. One exception is the greater Kn determined by Guelfo and Higgins27 for PFOA with dodecane compared to the Kn determined in this study for PFOA with decane. The Kn values reported by Guelfo and Higgins27 follow the same trends observed in the present study, where the Kn values for PFOS or PFOA with TCE are greater than with decane or dodecane, and the Kn for PFOA with a given NAPL is greater than for PFOS. Conversely, the values reported by McKenzie et al.28 do not exhibit a similar trend.

Several factors may influence the partitioning of PFAS into the bulk NAPL. First, the magnitude of partitioning may be influenced by properties of the NAPL such as polarity. For example, the dialectric constant, related to polarity, is approximately 3.4 for TCE compared to approximately 2 for decane. The mean Kn values for both PFOS and PFOA are greater for the NAPLs with greater polarity. Partitioning behavior is also of course influenced by properties of the PFAS, such as chain length. Additionally, it is important to note that partitioning behavior may be influenced by the protonation/deprotonation status of PFOS or PFOA. Based on literature reports (discussed in section S3 of the SI), it is likely that the pKa value for PFOA is greater than for PFOS, meaning PFOA may have a greater fraction of protonated species compared to PFOS, which may contribute to differences in partitioning under certain conditions. While a number of factors may impact PFAS partitioning into bulk NAPL, it is important to note the relatively large magnitudes of uncertainty for all Kn values determined from the batch experiments reported in this and the prior studies. The 95% confidence intervals for almost all of the measurements encompass 0. Thus, the batch NAPL-partitioning experiments should be considered to provide semi-quantitative measures of partitioning. Based on the aggregate results, it appears that Kn values for PFOS and PFOA are relatively small for typical NAPLs. Similar small or negligible NAPL partitioning has been reported for hydrocarbon surfactants in previous studies.38–40

PFOS and PFOA Transport in the Presence of NAPL

BTCs for transport of the nonreactive tracer in packed columns containing no residual NAPL are observed to be ideal, with sharp arrival and elution waves and retardation factors of 1 (see Figures 3 and S1). A dispersivity of approximately 0.1 cm was obtained by curve-fitting of a mathematical model incorporating ideal advective-dispersive transport to the measured BTCs. The BTCs for NRT transport in the presence of NAPL at the lower saturation exhibit no measurable differences compared to those for transport in the absence of the NAPL (see Figure S1). This indicates that the presence of NAPL had no measurable impact on flow and solute transport. Conversely, the NRT BTC exhibits additional spreading for the system with larger NAPL saturation (Figure S1). These disparate results suggest that flow and transport were impacted by the presence of the larger quantity of NAPL.

Figure 3.

Figure 3.

Open in a new tab

Breakthrough curves for the transport of the nonreactive tracer (NRT), PFOS under saturated conditions, and PFOS with residual TCE saturations of 0.28 and 0.42. PFOS input concentration is 10 mg/L. Lines are included for visualization purposes.

BTCs for PFOS transport in the absence and presence of TCE NAPL are shown in Figure 3. The arrival wave for PFOS in the absence of TCE NAPL is delayed very slightly compared to the NRT due to sorption of PFOS by the sand. The transport of PFOS in the presence of TCE NAPL exhibits additional retardation. Similarly, PFOS transport exhibits greater retardation in the presence of decane NAPL compared to no NAPL.22 Similar to PFOS, PFOA transport exhibits greater retardation in the presence of TCE NAPL. The greater retardation of PFOS and PFOA transport in the presence of NAPL can be attributed to interactions between PFOS/PFOA and the NAPL. Specifically, some combination of partitioning into the bulk NAPL and adsorption at the NAPL-water interface. Extended elution tailing is observed for PFOS transport without and with the presence of NAPL. This tailing is likely the result of a combination of extended elution tailing associated with sorption, as noted in a prior investigation of PFOS transport19, and analytical uncertainty at low concentrations.

Mass recoveries, retardation factor (R) values, and percent contributions to retention are presented in Table 2. Mass recoveries for the column experiments range from 98% to 118%. As PFOS and PFOA are recalcitrant, variability in the mass recovery can be attributed to analytical uncertainty. The R value measured for PFOS transport in the fully water-saturated system (no NAPL present) is 1.3 (Kd = 0.064 cm3/g). This Kd value is consistent with that measured by batch and column isotherm experiments.41 An R of 1.3 was also measured for the PFOS column experiment conducted with TCE dissolved in solution, but not present as residual NAPL. This indicates that the presence of TCE in solution at high (~1200 mg/L) concentration did not influence the sorption and retardation of PFOS.

Table 2.

Retardation factors (R), percent mass recovery, and percent contributions to retention by solid-phase sorption (fKd), NAPL-water interfacial adsorption (fKi), and NAPL-water partitioning (fKn) for PFOS or PFOA column experiments, including experimental and predicted results

Experiment Results Predicted Results
Analyte Systema Recovery (%) R fKd (%) fKn (%) fKi (%) Rb fKd (%) fKn (%) fKi (%)
PFOS Saturated 105 1.3 100 - - NA
Saturatedc 114 1.3 100 - - NA
TCE (0.28)d 102 2.5 23.8 2.7 73.5 2.7 (2.4-3.1) 31 2.3 66
TCE (0.26)e 105 2.7 25.2 2.2 72.6 3.0 (2.6-3.4) 32 1.9 66
TCE (0.42) 104 3.3 27.6 3.4 69 5.0 (4.4-5.6) 23 1.9 75
Decane (0.29) 113 2.6 28.5 0.5 71 2.8 (2.5-3.3) 37 0.5 63
Decane (0.54) 109 4.0 22 0.8 77.2 5.1 (4.4-5.7) 24 0.6 75
Decane (0.27)f 98 1.7 48 1 51 1.9 (1.7-2.8) 58 0.9 41
PFOA Saturated 101 1.2 100 - - NA
TCE (0.31) 118 1.5 57 19 24 1.7 (1.4-2.3) 76 12 13
Open in a new tab
a

parentheses indicate the column NAPL saturation when applicable

b

parentheses indicate the 95% confidence interval range, determined using the lower and upper 95% confidence interval values for Kd, Kn, and NAPL-water interfacial area

c

dissolved TCE present in background solution

d

values from two replicate experiments

e

Q = 0.1 mL/min

f

C0 = 25 mg/L

The R values for PFOS transport in the columns containing residual TCE are 2.5 and 3.3 for the experiments with Sn = 0.28 and 0.42, respectively. These values are greater than the R (1.3) measured for the corresponding experiment with no TCE NAPL present. The R value for PFOA transport in the presence of residual TCE is 1.5 for a Sn = 0.31, which is smaller than the R value for PFOS with residual TCE of similar saturation. This difference primarily reflects the smaller Ki value for PFOA versus PFOS, given that PFOS and PFOA have similar Kn values and that bulk-NAPL partitioning has minimal contribution (as will be discussed below). The R values for PFOS transport in the residual decane columns, 2.6 and 4.0 for Sn = 0.29 and 0.54, respectively, are greater than the R (1.3) for transport in the decane-free column. The R value for PFOS transport with residual decane (Sn = 0.27) when C0 = 25 mg/L is 1.7, which is smaller than the R value of 2.5 for the experiment with C0 = 10 mg/L and similar Sn (0.29). The smaller R measured for the higher input concentration results from two factors. First, the sorption of PFOS by the sand is nonlinear, as demonstrated by prior batch and column isotherm measurements.41 Second, NAPL-water interfacial adsorption is nonlinear in this concentration range (see respective Ki values in Table 1).

The preceding results indicate that the presence of NAPL increases PFOS and PFOA retention and produces resultantly larger retardation factors. Notably, the R values of PFOS for a given NAPL Sn are greater for larger Sn. Previous studies have established that the NAPL-water interfacial area increases as water saturation decreases.42–45 Thus, as NAPL saturation increases, and water saturation decreases, more NAPL-water interfacial area is available for adsorption, resulting in increased retention. Additionally, partitioning into the bulk NAPL increases with greater Sn. However, as discussed further below, the additional interfacial adsorption that occurs with greater Sn is more significant than the additional bulk-partitioning due to their relative contributions to total retention.

Measured Ki values were determined from the column experiments through the NAPL-water interfacial adsorption term in equation 8 in the SI, using measured parameters for all other variables. The Kd values were determined from the saturated-flow column experiments with no NAPL present. The Kn values were determined from the batch NAPL-partitioning experiments. The NAPL-water interfacial area was determined as described in the SI. The resultant Ki values are reported in Table 1. The column Ki values for the lower-Sn experiments are in excellent agreement with the Ki values determined from the interfacial-tension data. Consistency of transport-measured Ki values with those determined from surface and interfacial-tension data has also been observed in our prior studies.21,22,37

One implication of the observed consistency between the transport and interfacial-tension determined Ki values is that NAPL-water interfacial adsorption of PFOS/PFOA was not significantly rate limited for the transport experiments. This is consistent with the results of mathematical modeling of measured PFAS transport data reported by Brusseau.23 The potential for rate-limited NAPL-water interfacial adsorption was tested herein by conducting a transport experiment at a pore-water velocity 10-times lower than that used for all of the other experiments. The Ki value determined for the lower-velocity experiment is similar to the value determined for the higher-velocity experiments, and both are similar to the value determined from the interfacial-tension data (Table 1). This further supports the contention that NAPL-water interfacial adsorption was not measurably rate limited during transport.

In contrast to the lower-Sn experiments, the column-measured Ki values for the experiments conducted with larger Sn (TCE Sn = 0.42 and decane Sn = 0.54) are smaller than the Ki values determined from the interfacial-tension data. This disparity may be due to changes in hydraulic accessibility of the NAPL at higher Sn. At higher Sn, some portion of the NAPL-water interfaces in the columns may be hydraulically inaccessible to the PFOS solution, which would result in reduced retention and thus smaller than expected Ki values. This is supported by the results observed for the NRT transport experiments discussed above. Further investigation is needed to substantiate these results.

Predicted R values were determined using equation 8 in the SI and column-independent data. Kd values for the specific input concentrations used in the transport experiments were determined from measured Freundlich parameters reported in a recent study investigating PFOS and PFOA sorption by the sand.41 The Kn values are those measured herein with the batch NAPL-partitioning experiments, and the Ki values are from the interfacial-tension measurements. The predicted R values are compared to the measured values in Table 2. The predicted and measured values are in good agreement for all experiments conducted with lower NAPL saturations, considering the reported uncertainty ranges for the predicted values and the typical uncertainty associated with the transport experiments (see the SI). This indicates that the independent data-input sources provided measurements that are representative for transport conditions. In addition, this demonstrates that the observed retardation and transport behavior is consistent with the comprehensive retention model proposed by Brusseau and colleagues.22,29

The predicted R values are greater than the measured values for the systems with higher NAPL saturations. One reason for the disparity may be due to hydraulic inaccessibility of NAPL-water interfaces as discussed above. Additionally, the NAPL may obstruct access of the solution to some pore domains, thereby reducing the overall magnitude of solid-phase sorption. This would also result in reduced measured retardation.

The measured percent contributions to retention by solid-phase sorption (fKd) were determined using the Kd value determined from saturated experiments with no NAPL present and equation 9 in the SI. The measured percent contributions to retention by partitioning into NAPL (fKn) were determined with Kn values from batch NAPL-partitioning experiments and equation 10 in the SI. The measured contributions to retention by NAPL-water interfacial adsorption (fKi) were determined by solving for the NAPL-water interfacial adsorption term in equation 8 (by subtracting the solid-phase sorption and partitioning into NAPL terms) and using the resulting term in equation 11 in the SI. For all PFOS experiments with residual NAPL, adsorption to the NAPL-water interface has the highest contribution to retention (51-77%) followed by solid-phase sorption (22-48%). The opposite is true for PFOA with residual TCE, where solid-phase sorption has the highest contribution to retention (57%) followed by adsorption to the NAPL-water interface (24%). Predicted fKd, fKi, and fKn values were determined by solving equations 9, 10, and 11, respectively, using the independently determined parameters. The predicted results for contributions by the different mechanisms generally closely mirror those determined for the measured data.

For all experiments, partitioning into the bulk NAPL has the lowest contribution to retention, with contributions ranging from <1% to ~3% for PFOS experiments. Thus, partitioning into NAPL is observed to provide a minor contribution to overall retention and retardation, with the one exception of PFOA and residual TCE (19%). These results are interesting, considering that the Kn values are greater than the Ki values by one to three orders of magnitude. This can be explained by the significance of the magnitude of the NAPL-water interfacial area in the third term of equation 8 in the SI compared to the volumetric NAPL content in the fourth term of equation 8. Based on these results, it appears that the contribution of partitioning into NAPL can be considered to be relatively minor for many cases.

The above results indicate that both NAPL-water interfacial adsorption and solid-phase sorption contribute significantly to retention in the systems employed herein. The magnitudes of the respective contributions of the two retention processes will depend upon the properties of the porous medium and the quantity of non-wetting phase present.46 For a soil with greater sorption capacity, the relative contribution of solid-phase sorption would be expected to be greater, and that of NAPL-water interfacial adsorption correspondingly lesser, than for the sand used in the present study (assuming similar magnitudes of NAPL-water interfacial area).

As has been discussed, NAPL-water interfacial adsorption is nonlinear at higher concentrations. In addition, solid-phase sorption of PFAS is also often nonlinear. Hence, the relative contributions of the two processes to total retention may vary as a function of concentration. This concentration effect is investigated in two ways. First, measured PFOS interfacial-tension and sorption data are used to calculate Ki and Kd values for a representative lower concentration of 100 μg/L, using the TCE-water system as an example (Sn = 0.28). These are then used with equation 8 in the SI to calculate an R value for this lower concentration. The calculated R is 10.5, which is larger than the R values measured for the experiments (for which C0 was 10 mg/L). The fKi, fKd, and fKn values are 86.3, 13.2, and 0.4%, respectively. The contribution from NAPL-water interfacial adsorption is almost 20% greater for the lower concentration, while the contributions from the other two terms are smaller. This reflects the greater nonlinearity of NAPL-water interfacial adsorption compared to solid-phase sorption over the relevant concentration range for this system. However, the order and relative magnitudes of the contributions are consistent for both the lower and higher concentrations.

The concentration dependency of the retention processes is also examined by assessing the results of a previously reported transport study. McKenzie et al.28 conducted column experiments to investigate the impact of residual TCE on the transport of a suite of PFAS, including PFOS and PFOA. They observed that the presence of TCE NAPL generally resulted in increased retardation of the PFAS. They discussed that partitioning into bulk NAPL and adsorption at the NAPL-water interface could both be contributing to the increased retention, but they did not delineate or quantify the specific contributions of either process. The authors report PFOS R values of approximately 10 and 16, respectively, for experiments without and with a residual TCE saturation of approximately 0.29.28 These values were determined to be statistically different, whereas Rs for PFOA were not. The data-analysis methods employed in the present study were used to predict R values for the McKenzie et al.28 PFOS data. The Kd was determined from the R value reported for their column experiment conducted without NAPL present, using their reported system properties. A Ki value corresponding to their input concentrations (2 μg/L) was determined using the interfacial-tension data measured in the present study. The Kn value was also determined from the present study. The NAPL-water interfacial area measured for the sand used in the present study was used as a first approximation for the McKenzie et al. porous medium. This analysis produced a predicted R value of approximately 15 for PFOS transport in the presence of NAPL. This value is quite close to the measured value of 16 reported by McKenzie et al.28, indicating the reasonableness of the parameter values used for the prediction. Given that their soil contains fractions of silt and clay, it is likely that the interfacial area for the soil is somewhat larger than for the sand and thus the predicted R may be somewhat greater than 15. However, the overall impact of the magnitude of NAPL-water interfacial area is muted for this system due to its secondary importance compared to that of solid-phase sorption, as noted in the next paragraph.

Solid-phase sorption has the highest (85%) percent contribution to retention for the McKenzie et al.28 PFOS data, versus 15% for adsorption to the NAPL-water interface. As observed in the present study, partitioning into NAPL has the lowest contribution to retention (essentially 0%). The greater contribution by solid-phase sorption and smaller contribution by NAPL-water interfacial adsorption compared to the present study reflects the impact of the greater organic-carbon content of the loamy sand used by McKenzie et al.28 compared to the sand used in this study and the correspondingly greater magnitude of sorption. Notably, the results from both studies are consistent with the comprehensive retention model proposed by Brusseau and colleagues.22,29 This is despite the great difference in input concentrations employed for the studies, 2 μg/L for McKenzie et al. vs 10 mg/L for the present study. This consistency illustrates the robustness of the retention model and the data-analysis methods employed herein.

PFAS Impacts on NAPL Solubility and Mobility

For the TCE solubilization experiments, the aqueous concentrations of TCE for PFOS concentrations of 0.1, 1, and 10 mg/L were statistically identical to the concentration measured for the controls (no PFOS present), with a mean value of approximately 1150 mg/L and overlapping 95% confidence intervals. Thus, no enhanced solubilization was observed for lower concentrations of PFOS. The aqueous concentrations of TCE increased for PFOS concentrations of 100 mg/L and 500 mg/L, with mean values of approximately 1550 mg/L. The 95% confidence intervals between the smaller and larger PFOS concentration ranges do not overlap. Thus, PFOS increased the aqueous solubility of TCE by approximately 35% for concentrations of 100 mg/L and greater.

Many studies have observed increased NAPL solubilization in the presence of surfactants. In some cases, NAPL solubilization occurred only when surfactant concentrations were greater than the critical micelle concentration (CMC) (e.g., refs 47–49). Conversely, in other studies solubilization was also observed when surfactant concentrations were below the CMC (e.g., refs 50–54). Based on the measured interfacial-tension data, the CMC is >500 mg/L for PFOS and TCE. Thus, the observed increase in apparent TCE solubility caused by PFOS in the present study appears to be pre-micellar. The NAPL, surfactant, and solution properties impact the solubilization behavior. Two proposed mechanisms for solubilization observed below the surfactant CMC are the reduction in NAPL-water interfacial tension and the formation of pre-micellar aggregates (e.g., refs 51–54).

No free-phase TCE was observed in the effluent of the TCE mobilization column experiment. This indicates that the PFOS pulse, though at a high concentration of 500 mg/L, did not reduce the TCE-water interfacial tension sufficiently to result in TCE mobilization. An NT value of approximately 5.6x10−6 was determined for the 500 mg/L PFOS column mobilization experiment. This value is lower than that determined to cause mobilization, between 2x10−5 to 5x10−5 (e.g., ref 55), and is consistent with the observation of no mobilization for the experiment.

The results presented in this subsection indicate that the presence of PFAS in concentrations in the 10s of mg/L and lower are likely to have minimal impact on NAPL solubilization and mobilization. Thus, it may be anticipated that NAPL solubilization and mobilization impacts may be of concern for only the most contaminated legacy AFFF sites, i.e. sites with PFAS concentrations exceeding 100s of mg/L. However, NAPL solubilization and mobilization may be of greater significance for sites with new AFFF releases and for reconstruction analysis of historical AFFF releases wherein high initial PFAS concentrations are relevant and considering that the presence of some of the other AFFF constituents such as hydrocarbon surfactants and cosolvents can influence solubilization and mobilization. It is important to note that the mobilization experiments conducted herein comprised two-phase systems. Mobilization impacts in particular may be of greater concern for three-phase systems such as in the vicinity of the capillary fringe. In addition, this study is focused on TCE as a single-component NAPL. However, NAPLs are often comprised of multiple compounds with different properties. Thus, further research is needed to assess the impacts of PFAS to NAPL behavior in multiphase, complex-mixture systems.

Implications

Understanding the processes contributing to the retention and transport of PFAS at contaminated sites is critical in order to address risk assessment and site management needs. As NAPLs are often known or suspected co-contaminants, the presence of NAPL may impact the transport and distribution of PFAS at contaminated sites, or vice versa. This study demonstrates that NAPL can contribute significantly to PFAS retention, primarily through NAPL-water interfacial adsorption. This was the case for systems with differing NAPL type, NAPL saturation, PFOS input concentration, and pore-water velocity. In the systems studied herein, adsorption to the NAPL-water interface accounted for as much as ~77% of the retention observed. Conversely, partitioning into bulk NAPL contributed minimally to retention. The relative contributions of solid-phase sorption versus NAPL-water interfacial adsorption will depend on the specific PFAS and on system conditions, including soil type, NAPL saturation, and solution properties. The comprehensive retention model developed by Brusseau and colleagues was demonstrated to provide good predictions of the measured data sets.

The results of the study indicate that PFAS will generally have significant impacts on NAPL solubilization and mobilization only at high concentrations (100s of mg/L). Such high concentrations are rarely observed at legacy AFFF source-zone sites. However, such high concentrations are relevant for new AFFF releases as well as for historical-release scenarios. Reconstructing historical releases requires investigation into the source history, including the time and amount of release and impacts by and to co-contaminants such as NAPLs. In these cases, accounting for initial high PFAS concentrations is important, and, as discussed above, these applications may induce changes to NAPL behavior. Understanding these interactions is crucial to understanding transport conditions and conducting accurate exposure analyses. Additional investigation, including systems with PFAS and NAPL mixtures, is warranted to further evaluate these complex systems.

Supplementary Material

SI

Acknowledgements

This research was supported by the NIEHS Superfund Research Program (grant #P42 ES 4940). We thank the reviewers for their constructive and helpful comments.

References

  • 1.Rayne S; Forest K Congener-specific organic carbon-normalized soil and sediment-water partitioning coefficients for the C1 through C8 perfluoroalkyl carboxylic and sulfonic acids. J. Environ. Sci. Health A Tox. Hazard. Subst. Environ. Eng 2009, 44, 1374–1387. [DOI] [PubMed] [Google Scholar]
  • 2.Ahrens L Polyfluoroalkyl compounds in the aquatic environment: a review of their occurrence and fate. J. Environ. Monit 2011, 13, 20–31. [DOI] [PubMed] [Google Scholar]
  • 3.Krafft MP; Riess JG Per- and polyfluorinated substances (PFASs): Environmental challenges. Curr. Opin. Colloid Interface Sci 2015, 20, 192–212. [Google Scholar]
  • 4.Cousins IT; Vestergren R; Wang Z; Scheringer M; McLachlan MS The precautionary principle and chemicals management: The example of perfluoroalkyl acids in groundwater. Environ. Int 2016, 94, 331–340. [DOI] [PubMed] [Google Scholar]
  • 5.Moody CA; Field JA Determination of perfluorocarboxylates in groundwater impacted by fire-fighting activity. Environ. Sci. Technol 1999, 33, 2800–2806. [Google Scholar]
  • 6.Moody CA; Field JA Perfluorinated surfactants and the environmental implications of their use in fire-fighting foams. Environ. Sci. Technol 2000, 34, 3684–3870. [Google Scholar]
  • 7.Moody CA; Hebert GN; Strauss SH; Field JA 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, 341–345. [DOI] [PubMed] [Google Scholar]
  • 8.Schultz MM; Barofsky DF; Field JA, 2004. Quantitative determination of fluorotelomer sulfonates in groundwater by LC MS/MS. Environ. Sci. Technol 2004, 38, 1828e1835. 10.1021/es035031j. [DOI] [PubMed] [Google Scholar]
  • 9.Weiß O; Wiesmüller GA; Bunte A; Göen T; Schmidt CK; Wilhelm M; Hölzer J Perfluorinated compounds in the vicinity of a fire training area – Human biomonitoring among 10 persons drinking water from contaminated private wells in Cologne, Germany. Int. J. Hyg. Environ. Health 2012, 215, 212–215. [DOI] [PubMed] [Google Scholar]
  • 10.Backe WJ; Day TC; Field JA Zwitterionic, cationic, and anionic fluorinated chemicals in aqueous film forming foam formulations and groundwater from U.S. military bases by nonaqueous large-volume injection HPLC-MS/MS. Environ. Sci. Technol 2013, 47, 5226–5234. [DOI] [PubMed] [Google Scholar]
  • 11.McGuire ME; Schaefer C; Richards T; Backe WJ; Field JA; Houtz E; Sedlak DL; Guelfo JL; Wunsch A; Higgins CP Evidence of remediation-induced alteration of subsurface poly-and perfluoroalkyl substance distribution at a former firefighter training area. Environ. Sci. Technol 2014, 48, 6644–6652. [DOI] [PubMed] [Google Scholar]
  • 12.Filipovic M; Woldegiorgis A; Norström, Bibi M; Lindberg M; Österås AH Historical usage of aqueous film forming foam: A case sutyd of the widespread distribution of perfluoroalkyl acids from a military airport to groundwater, lakes, soils and fish. Chemosphere 2015, 129, 39–45. [DOI] [PubMed] [Google Scholar]
  • 13.Anderson RH; Adamson DT; Stroo HF Partitioning of poly- and perfluoroalkyl substances from soilto groundwater within aqueous film-forming foam source zones. J. Contam. Hydrol 2019, 220, 59–65. [DOI] [PubMed] [Google Scholar]
  • 14.Brusseau ML; Anderson RH, Guo B PFAS concentrations in soils: background levels versus contaminated sites. Sci. Total Environ 2020, 740, 140017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.McKenzie ER; Siegrist RL; McCray JE; Higgins CP Effects of chemical oxidants on perfluoroalkyl acid transport in one-dimensional porous media columns. Environ. Sci. Technol 2015, 49, 1681–1689. [DOI] [PubMed] [Google Scholar]
  • 16.Aly YH; Liu C; McInnis DP; Lyon BA; Hatton J; McCarty M; Arnold WA; Pennell KD; Simcik MF In situ remediation method for enhanced sorption of perfluoro-alkyl substances onto Ottawa sand. J. Environ. Eng 2018, 144, 040180. [Google Scholar]
  • 17.Aly YH; McInnis DP; Lombardo SM; Arnold WA; Pennell KD; Hatton J; Simcik M Enhanced adsorption of perfluoroalkyl substances in in situ remediation. Environ. Sci.: Water Res. Technol 2019, 5, 1867–1875. [Google Scholar]
  • 18.Lv X; Sun Y; Ji R; Gao B; Wu J; Lu Q; Jiang H Physiochemical factors controlling the retention and transport of perfluorooctanoic acid (PFOA) in saturated sand and limestone porous media. Wat. Res 2018, 141, 251–258. [DOI] [PubMed] [Google Scholar]
  • 19.Brusseau ML; Khan N; Wang Y; Yan N; Van Glubt S; Carroll KC Nonideal transport and extended elution tailing of PFOS in soil. Environ. Sci. Technol 2019, 53, 10654–10664. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Guelfo JL; Wunsch A; McCray J; Stults JF; Higgins CP Subsurface transport potential of perfluoroalkyl acids (PFAAs): Column experiments and modeling. J. Contam. Hydrol 2020, 233, 103661. [DOI] [PubMed] [Google Scholar]
  • 21.Lyu Y; Brusseau ML; Chen W; Yan N; Fu X; Lin X Adsorption of PFOA at the air-water interface during transport in unsaturated porous media. Environ. Sci. Technol 2018, 52, 7745–7753. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Brusseau ML; Yan N; Van Glubt S; Wang Y; Chen W; Lyu Y; Dungan B; Carroll KC; Holguin FO Comprehensive retention model for PFAS transport in subsurface systems. Wat. Res 2019, 148, 41–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Brusseau ML Simulating PFAS transport influenced by rate-limited multi-process retention. Wat. Res 2020, 168, 115179. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Lyu Y; Brusseau ML The influence of solution chemistry on air-water interfacial adsorption and transport of PFOA in unsaturated porous media. Sci. Total Environ 2020, 713, 136744. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Yan N; Ji Y; Zhang B; Zheng X; Brusseau ML Transport of GenX in saturated and unsaturated porous media. Environ. Sci. Technol 2020, 54, 11876–11885. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Guo B, Zeng J, and Brusseau ML. A Mathematical Model for the Release, Transport, and Retention of Per- and Polyfluoroalkyl substances (PFAS) in the Vadose Zone. Water Resour. Res, 2020, 57, article: e2019WR026667. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Guelfo JL; Higgins CP Subsurface transport potential of perfluoroalkyl acids at aqueous film-forming foam (AFFF)-impacted sites. Environ. Sci. Technol 2013, 47, 4164–4171. [DOI] [PubMed] [Google Scholar]
  • 28.McKenzie ER; Siegrist RL; McCray JE; Higgins CP The influence of non-aqueous phase liquids (NAPL) and chemical oxidant application on perfluoroalkyl acid (PFAA) fate and transport. Water Res. 2016, 199–207. [DOI] [PubMed] [Google Scholar]
  • 29.Brusseau ML Assessing the potential contributions of additional retention processes to PFAS retardation in the subsurface. Sci. Total Environ 2018, 613-614, 176–185. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Mitchell Mark, Muftakhidinov Baurzhan, Winchen Tobias. Engauge Digitizer Software 2017. Webpage http://markummitchell.github.io/engauge-digitizer/
  • 31.Brusseau ML; Van Glubt S The influence of surfactant and solution composition on PFAS adsorption at fluid-fluid interfaces. Wat. Res 2019, 161, 17–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Downes N; Ottewill GA; Ottewill RH An investigation of the behaviour of ammonium perfluoro-octanoate at the air/water interface in the absence and presence of salts. Colloids Surf. A Physiochem. Eng. Asp 1995, 102, 203–211. [Google Scholar]
  • 33.Lunkenheimer K; Geggel K; Prescher D Role of counterion in the adsorption behavior of 1:1 ionic surfactants at fluid interfaces–Adsorption properties of alkali perfluoro-n-octanoates at the air-water interface. Langmuir 2017, 33, 10216–10224. [DOI] [PubMed] [Google Scholar]
  • 34.Brusseau ML The influence of molecular structure on the adsorption of PFAS to fluid-fluid interfaces: Using QSPR to predict interfacial adsorption coefficients. Wat. Res 2019a, 152, 148–158. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Schaefer CE; Callaghan AV; King JD; McCray JE Dense nonaqueous phase liquid architecture and dissolution in discretely fractured sandstone blocks. Environ. Sci. Technol 2009, 43, 1877–1883. [DOI] [PubMed] [Google Scholar]
  • 36.Silva JAK; Martin WA; Johnson JL; McCray JE Evaluating air-water and NAPL-water interfacial adsorption and retention of perfluorocarboxylic acids within the vadose zone. J. Contam. Hydrol 2019, 223, 103472. [DOI] [PubMed] [Google Scholar]
  • 37.Brusseau ML Examining the robustness and concentration dependency of PFAS air-water and NAPL-water interfacial adsorption coefficients. Water Res. 2021, 190, 116778. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Thangamani S; Shreve GS Effect of anionic biosurfactant on hexadecane partitioning in multiphase systems. Environ. Sci. Technol 1994, 28, 1993–2000. [DOI] [PubMed] [Google Scholar]
  • 39.Zhao B; Zhu L, Yang K Solubilization of DNAPLs by mixed surfactant: Reduction in partitioning losses of nonionic surfactant. Chemosphere 2006, 772–779. [DOI] [PubMed] [Google Scholar]
  • 40.Damrongsiri S; Tongcumpou C; Sabatini DA Partition behavior of surfactants, butanol, and salt during application of density-modified displacement of dense non-aqueous phase liquids. J. Hazard. Mater 2013, 248-249, 261–267. [DOI] [PubMed] [Google Scholar]
  • 41.Van Glubt S; Brusseau ML; Yan N; Huang D; Khan N; Carroll KC Column versus batch methods for measuring PFOS and PFOA sorption to geomedia. Environ. Pollut 2021, 268, article 115917. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Schaefer CE; DiCarlo DA; Blunt MJ Determination of water-oil interfacial area during 3-phase gravity drainage in porous media. J. Colloid Interface Sci 2000, 221, 308–312. [DOI] [PubMed] [Google Scholar]
  • 43.Brusseau ML; Narter M; Schnaar G; Marble J Measurement and estimation of organic-liquid/water interfacial areas for several natural porous media. Environ. Sci. Technol 2009, 43, 3619–3625. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Zhong H; El Ouni A; Lin D; Wang B; Brusseau ML The two-phase flow IPTT method for measurement of nonwetting-wetting liquid interfacial areas at higher nonwetting saturations in natural porous media. Water Resour. Res 2016, 10.1002/2016WR018783. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Brusseau ML; Taghap H NAPL-water interfacial area as a function of fluid saturation measured with the interfacial partitioning tracer test method. Chemosphere 2020, 127562. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Brusseau ML Estimating the relative magnitudes of adsorption to solid-water and air/oil-water interfaces for per- and poly-fluoroalkyl substances. Environ. Pollut 2019, 254, 113102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Clifford JS; Ioannidis MA; Legge RL Enhanced aqueous solubilization of tetrachloroethylene by a rhamnolipid biosurfactant. J. Colloid Interface Sci 2007, 305, 361–365. [DOI] [PubMed] [Google Scholar]
  • 48.Mir MA; Chat OA; Najar MH; Younis M; Dar AA; Rather GM Solubilization of triphenylamine, triphenylphosphine, triphenylphosphineoxide and triphenylmethanol in single and binary surfactant systems. J. Colloid Interface Sci 2011, 364, 163–169. [DOI] [PubMed] [Google Scholar]
  • 49.Tehrani-Bagha AR; Holmberg K Solubilization of hydrophobic dyes in surfactant solutions. Materials 2013, 6, 580–608. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Kile DE; Chiou CT Water solubility enhancements of DDT and trichlorobenzene by some surfactants below and above the critical micelle concentration. Environ. Sci. Technol 1989, 23, 832–838. [Google Scholar]
  • 51.Zhang Y; Miller RM Enhanced octadecane dispersion and biodegradation by a Pseudomonas rhamnolipid surfactant (biosurfactant). Appl. Environ. Microbio 1992, 58, 3276–3282. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Zhong H; Liu Y; Liu Z; Jiang Y; Tan F; Zeng G; Yuan X; Yan M; Niu Q; Liang Y Degradation of pseudo-solubilized and mass hexadecane by a Pseudomonas aeruginosa with treatment rhamnolipid biosurfactant. Int. Biodeterior. Biodegradation 2014, 94, 152–159. [Google Scholar]
  • 53.Zhong H; Yang L; Zeng G; Brusseau ML; Wang Y; Li Y; Liu Z; Yuan X; Tan F Aggregate-based sub-CMC solubilization of hexadecane by surfactants. RSC Adv. 2015, 5, 78142–78149. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Zhong H; Yang X; Tan F; Brusseau ML; Yang L; Liu Z; Zeng G; Yuan X Aggregate-based sub-CMC solubilization of n-alkanes by monorhamnolipid biosurfactant. New J. Chem 2016, 40, 2028–2035. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Pennell KD; Pope GA; Abriola LM Influence of viscous and buoyancy forces on the mobilization of residual tetrachloroethylene during surfactant flushing. Environ. Sci. Technol 1996, 30, 1328–1335. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

SI
NIHMS1757285-supplement-SI.pdf (290.8KB, pdf)

RESOURCES