An exponential increase in understanding of the provenance, migration, and fate of per- and polyfluoroalkyl substances (PFAS) is evident over the past decade. The environmental and health concerns from this group of contaminants are well recognized, but they are still evolving with advancing studies. A key property of PFAS compounds at a molecular level is a strong carbon-fluorine bond that makes them highly resistant to degradation, resulting in waste and containment facilities being long-term sources of PFAS concentration due to the day-to-day consumer items historically containing PFAS compounds that have been, and continue to be, disposed of at these sites (Bouazza 2021).
Similar to what has been observed globally within the waste and containment industry, an Australian national study of 27 landfills identified PFAS in all sites of various ages and waste types, and, most significantly, in construction and demolition waste sites (Gallen et al. 2017). Elevated PFAS concentrations were associated with increasing pH and total organic carbon (TOC), with several compounds showing a significant increase in concentration with decreasing landfill age. PFAS concentrations were higher in leachates collected from currently operating landfills than those collected from closed landfills. Australian environmental management plans (currently PFAS National Environmental Management Plan [NEMP] 2.0) have been, and continue to be, updated to prevent and limit further migration, accumulation and exposure of PFAS within the environment.
Modern landfill sites are highly engineered, with licensed facilities designed to protect the surrounding environment from contaminants leaching out from the waste stream. Geosynthetic liners are commonly used as liquid and gas barriers within composite lining systems, but studies to date are limited in understanding the interaction of PFAS compounds with the geosynthetic lining system. Geosynthetic clay liners (GCLs) are widely used for secondary containment in landfill applications, and detailed Australian-based research and development is focusing efforts in determining the attenuation potential of enhanced powdered GCLs when exposed to landfill leachates containing PFAS (Figure 1).
GCL performance
GCLs are typically assessed for their hydraulic performance in a triaxial permeability testing apparatus. Standardized test methodologies, such as American Society for Testing and Materials (ASTM) D6766, are used to hydrate and permeate GCLs with potentially incompatible aqueous solutions. Traditional assessment using these methods results in several pore volumes of permeating liquid passing through the GCL until hydraulic and chemical equilibrium is reached. With precise volumetric measurement changes of the influent and effluent levels taken from the connected pipettes over periods of documented time, plus measurements including sample area, hydraulic head and hydrated clay thickness, index flux (m3/m2/s), and hydraulic conductivity (m/s) of the GCL in contact with the test liquid can be calculated.
To assess PFAS concentrations in the permeating liquid, the leachate is sampled and analyzed for 27 PFAS analytes before it is introduced to the GCL, and then later sampled at repeated intervals throughout the permeation period by collecting the effluent liquid that has passed through the GCL. From this data the PFAS attenuation capacity of GCLs can then be determined.
Initial studies on a landfill leachate (Leachate A) containing ~35 ppb (~35 µg/L) total PFAS were initiated with a powdered natural sodium bentonite, with and without polymer modification. When the termination criteria were reached, the hydraulic conductivity was 5.74 × 10-11 m/s for the standard GCL and 4.42 × 10-11 m/s for the polymer-modified bentonite (PMB) GCL (Figure 2). After two pore volumes flow through the GCLs, concentrations of the three nationally regulated compounds (perfluorooctanoic acid [PFOA], perfluorooctanesulfonic acid [PFOS] and perfluorohexane sulfonate [PFHxS]) show that PFOS reduced from an initial concentration level of 0.24 ppb (0.24 µg/L) to below the level of reporting (LOR) in both GCL types (Figure 3). The PFOA concentration reduced from an initial level of 0.69 to 0.50 ppb (0.69 to 0.50 µg/L) in the standard GCL to 0.40 ppb (0.40 µg/L) in the PMB GCL. PFHxS concentrations reduced from an initial level of 0.94 to 0.71 ppb (0.94 to 0.71 µg/L) in the standard GCL to 0.56 ppb (0.56 µg/L) in the PMB GCL. This initial study demonstrated that the hydraulic performance does not seem to be negatively affected by PFAS within landfill leachate at these concentrations, and that some attenuation is provided by the standard and PMB GCLs. As only low pore volumes were passed through the GCLs in these preliminary investigations, it would be unreasonable to draw conclusions for long-term performance from this dataset.
GCL development for the attenuation of PFAS
The preliminary observations accelerated a need to develop a GCL that could be enhanced to further assist in the attenuation of PFAS. Options included modification to the synthetic components of the composite, such as an additional layer of adsorptive fabric or a unique blend of staple fibers in the cover geotextile layer, or additives that could be blended with the bentonite. The scope of the development included a requirement to ensure the enhanced GCL retained the mechanical and hydraulic properties of a traditional GCL, namely retention of interface and internal shear strength and maintenance of the low level of hydraulic conductivity expected from a secondary containment barrier.
Activated carbon (AC) is a carbonaceous adsorbent with a large internal surface area. It is commonly used to adsorb natural organic compounds and to remove taste, odor and synthetic organic chemicals in drinking water treatment systems. It has also been one of the most studied treatments for the removal of PFAS. AC works by the process of adsorption. Adsorption is the attachment or adhesion of atoms, ions and molecules (adsorbates) from a gaseous, liquid or solution medium onto the surface of an adsorbent such as AC. The porosity of AC offers a vast surface on which this adsorption can take place. There are a number of adsorption interaction mechanisms in relation to organic compounds, such as ligand exchange, electrodonor-acceptor interactions, oxidative coupling, hydrophobicity, Lewis acid-base reactions, H-bonds, electrostatic interactions and covalent bond formation (Kah et al. 2017). From these adsorption interaction mechanisms, there is strong consensus that both hydrophobicity and electrostatic interaction are the primary mechanisms involved in the adsorption of two of the main long-chain PFAS (PFOS and PFOA) onto AC (Goss 2008; Gagliano et al. 2020; Saeidi et al. 2020). For adsorption to be effective, it is important to match the adsorbate molecule size with the pore size and pore size distribution of the AC.
According to Zdravkov et al. (2007), AC pores are roughly classified into three groups: macropores (>50 nm diameter), mesopores (2–50 nm diameter) and micropores (<2 nm diameter). Micropores generally contribute a major part of the internal surface area. The larger macro- and mesopores are regarded as the highways into the carbon particle. The combined surface area of AC is typically between 16,953 and 50,859 square yards per ounce (500 and 1,500 m2/g) (Bansal and Meenakshi 2005). A higher porosity can mean a higher pollutant loading capacity, but it is highly dependent on the pore size distribution. For example, a high level of micropores may result in a large surface area but may not have enough entry points and transport pathways for larger molecules to take advantage of this. On the other hand, too many macropores and not enough meso- and micropores may result in a lower adsorption capacity. When looking at PFAS, generally a good balance of each size category results in optimized adsorption.
AC can be manufactured from virtually any organic material. Because of their high carbon contents, however, wood, coconut shells and coal are the most used raw materials. Activation may be carried out by chemical means or, more commonly, by high-temperature steam > 1,292°F (> 700°C) in a controlled atmosphere. There are important differences between AC produced from wood, coconut shells and coal. Wood-based AC generally has a higher level of macropores whereas coconut-based AC has a higher level of micropores. Coal-based AC is comprised of a good balance of micro- (small), meso- (medium) and macro- (large) pores, making it a solid contender for efficient adsorption of certain organic molecules, such as PFAS (Nowicki et al. 2014).
Coals are classified according to their age. Lignite, or brown coal, is the youngest type of coal followed by subbituminous, bituminous and anthracite. The origin and age of the coal is a key factor in producing an AC with a very high specific surface area and a good particle size distribution (PSD). Bituminous coal-based AC has been shown to have superior performance in trapping a wide range of perfluorinated compounds over other AC varieties (McNamara et al. 2018).
One of the earlier GCL modification trials included a 3.24 ounces per square yard (110 g/m2), 15.75 mil (0.4 mm) thick AC cloth placed within the GCL beneath the geotextile cover layer. This data demonstrated that the addition of an AC interface attenuated the PFOA, PFOS and PFHxS further than what was observed in the initial investigations with the standard and PMB GCLs. PFOS again was reduced to less than the LOR, PFOA reduced to 0.25 ppb (0.25 µg/L) and PFHxS reduced to 0.32 ppb (0.32 µg/L) concentrations (Figure 4). The hydraulic conductivity at termination of the analysis was 4.29 × 10-11 m/s, which was to be expected as this was the same PMB as the initial study with the addition of a permeable AC cloth.
An AC powder was sourced that had the specific set of unique properties shown to perform with certain PFAS compounds. This was blended with the PMB bentonite at increasing percentages to optimize the PFAS attenuation and hydraulic performance (Figure 5), as unlike the permeable AC cloth, an inverse relationship exists between increasing nonswelling powder content and decreasing hydraulic performance. Historic trials indicated that blending other minerals with a specific bentonite by up to 25% by weight would still result in a low level of hydraulic conductivity. Preliminary data collected highlighted a blend of 0.20 pounds per square foot (1 kg/m2) of powdered AC with 0.82 pounds per square foot (4 kg/m2) of select bentonite returned an acceptable level of both hydraulic and PFAS attenuation performance. All three compounds of specific interest (PFOS, PFOA and PFHxS) were attenuated to below the 0.1 ppb (0.1 µg/L) LOR after the same pore volumes of flow as the earlier trial data. Further research was then initiated with this powdered AC/bentonite blend, with the resulting product termed a hybrid geosynthetic clay liner (hGCL).
hGCL performance with Australian landfill leachate
The research and development facility undertaking this study frequently analyzes leachates, liquors and soil elutions for their chemical compatibility with a range of GCLs. Current leachates were screened for PFAS content, leading to a more rigorous analysis of the interaction between an hGCL with Leachate E, which contained a total PFAS level of ~83 ppb (~83 µg/L) (Figure 6). Leachate E was permeated through a PMB GCL with ~0.82 pounds per square foot (~4 kg/m2) bentonite and an hGCL with the same ~0.82 pounds per square foot (~4 kg/m2) bentonite blended with an additional 0.20 pounds per square foot (1 kg/m2) of a select grade of powdered AC. The PMB GCL had a stable hydraulic conductivity throughout its 290 days of permeation, with a final termination value of 1.77 × 10-11 m/s. After 330 days of permeation, the hGCL had a termination hydraulic conductivity value of 7.31 × 10-11 m/s. The effluent liquid from both permeability cells was analyzed for PFAS concentration throughout the duration of permeation (Figures 7 and 8).
As previously determined in other leachates, PFAS compounds passing through the PMB GCL generally remained at the initial concentrations, excluding perfluoropentanoic acid (PFPeA), which reduced from 37.6 to 0.94 ppb (37.6 to 0.94 µg/L), and perfluorobutanoic acid (PFBA), which demonstrated sporadic behavior, initially rising throughout the test period, then falling toward the final stages of analysis. These behaviors could be attributed to degradation reactions or analytes concentrating within the material, resulting in breakthrough when sorptive capacity is attained (Gates et al. 2020).
The hGCL effluent data for Leachate E show that the PFAS compounds were well attenuated, and for the regulated PFOS, PFOA and PFHxS compounds, attenuated values were below the NEMP 2.0 guidance requirements. Once more, PFBA gave the most observable breakthrough, as with some of the other shorter chain compounds in other trials conducted.
Ongoing hGCL development
The current hGCL has demonstrated a good degree of resistance to migration of the three currently regulated long-chain PFAS compounds, but the regulations are evolving, and the next revisions are likely to encompass a wider range of PFAS compounds, at potentially lower maximum concentrations. The next iterations of the hGCL—which include the addition of other supplementary minerals, compounds and resins in various formulations and dosages—are under development. The current studies have been specifically focused on the attenuation of PFAS compounds, but it is entirely possible that other emerging contaminants may also be captured by a new generation of hGCLs currently in development. With GCLs already serving as the secondary barrier to leakage within current lining systems, as we understand more about the effects on the environment and human health from emerging contaminants, an hGCL is well positioned to become best practice design to polish any leakage emanating from our waste facilities.
Ryan Hackney is the Technical and R&D Laboratory Manager at GRID Geofabrics Australasia Pty Ltd. in Molendinar, Queensland, Australia.
Daniel Gibbs is the General Manager—Technical, Research and Innovation, GRID Geofabrics Australasia Pty Ltd. in Molendinar, Queensland, Australia.
All figures courtesy of the authors.
References
Bansal, R. C., and Meenakshi, G. (2005). Activated carbon adsorption, CRC Press, Boca Raton, Fla.
Bouazza, A. (2021). “Interaction between PFASs and geosynthetic liners: Current status and the way forward.” Geosynthetics International, 28(2), 214–223.
Gagliano, E., Sgroi, M., Vagliasindi, F. G., and Roccaro, P. (2020). “Removal of poly- and perfluoroalkyl substances (PFAS) from water by adsorption: Role of PFAS chain length, effect of organic matter and challenges in adsorbent regeneration.” Water Research, 171.
Gallen, C., Drage, D., Eaglesham, G., Grant, S., Bowman, M., and Mueller, J. F. (2017). “Australia-wide assessment of perfluoroalkyl substances (PFASs) in landfill leachates.” Jour. of Hazardous Materials, 331, 32–141.
Gates, W. P., MacLeod, A. J., Fehervari, A., Bouazza, A., Gibbs, D., Hackney, R., Callahan, D. L., and Watts, M. (2020). “Interactions of per- and polyfluoralkyl substances (PFAS) with landfill liners.” Advances in Environmental and Engineering Research, 1(4), 40.
Goss, K. U. (2008). “The pKa values of PFOA and other highly fluorinated carboxylic acids.” Environmental Science and Technology, 42(2), 456–458.
Kah, M., Sigmund, G., Xiao, F., and Hofmann, T. (2017). “Sorption of ionizable and ionic organic compounds to biochar, activated carbon and other carbonaceous materials.” Water Research, 124, 673–692.
McNamara, J. D., Franco, R., Mimna, R., and Zappa, L. (2018). “Comparison of activated carbons for removal of perfluorinated compounds from drinking water.” Jour. American Water Works Association, 110(1), E2–E14.
Nowicki, H., Schuliger, W., Nowicki, G., and Sherman, B. (2014). “Evaluation of activated carbon performance.” Water Conditioning and Purification Magazine. www.wcponline.com/2014/06/17/evaluation-activated-carbon-performance.
Saeidi, N., Kopinke, F. D., and Georgi, A. (2020). “Understanding the effect of carbon surface chemistry on adsorption of perfluorinated alkyl substances.” Chemical Engineering Jour., 381.
Zdravkov, B. D., Čermák, J. J., Šefara, M., and Janků, J. (2007). “Pore classification in the characterization of porous materials: A perspective.” Central European Jour. of Chemistry, 5(2), 385–395.