Regeneration of per- and polyfluoroalkyl substance-laden granular activated carbon using a solvent based technology

Highlights

• Laboratory contaminated and spent field GAC can be regenerated for future use.

• Ethanol with 0.5% (v/v) NH₄OH was the optimum performing solvent regenerant.

• PFAS-laden GAC can be regenerated multiple times using a solvent-based technology.

• Similar PFAS sorption and better PFAS removal observed from solvent regenerated GAC.

• PFAS-containing regenerant solvent can be concentrated for destruction or disposal.

Abstract

Per- and polyfluoroalkyl substances (PFAS) are a large class of chemicals widely used for many commercial and industrial applications and have resulted in contamination at sites across globally. Pump-and-treat systems, groundwater extraction, and ex situ treatment using granular activated carbon (GAC) are being implemented, either in full or pilot scale, to treat PFAS-impacted groundwater and drinking water. The only current method of regenerating spent GAC is to reactivate it at temperatures greater than 1000 °C, which requires large amounts of energy and is quite expensive. This research focused on development and demonstration of an effective GAC regeneration technology using a solvent-based method for PFAS-laden GAC used in water treatment. Two different organic solvents (ethanol and isopropyl alcohol) with 0.5% and 1.0% ammonium hydroxide (NH₄OH) as a base additive were tested to determine the most effective regenerant solution to remove PFAS from the contaminated GAC. Based on column tests using laboratory-contaminated GAC with perfluorooctanoic acid (PFOA) and perfluorooctanoic sulfonate (PFOS), the solvent-base mix (SBM) of ethanol with 0.5% NH₄OH was found to be the optimum performing regenerant solution. The GAC life span assessment showed that solvent-regenerated GAC performed similar to virgin GAC without losing its optimal performance of PFAS sorption. Further, the solvent-regenerated GAC showed optimal performance even after four cycles of solvent regenerations tested using the optimum SBM. Average percent removal in laboratory-contaminated GAC using the optimum SBM was 65% and 93% for PFOS and PFOA, respectively. Four field-spent GAC samples were also regenerated using the optimum SBM. Percent removal from these samples was found to be in range of 55%–68%. The type of GAC used, level of contamination and type of PFAS present, water type and quality, and the presence of co-contaminants may have influenced the removal capacity. Distillation experiments have shown that it is feasible to concentrate the spent solvent prior to disposal, which reduces the amount of PFAS-contaminated solvent waste produced in regeneration cycles.

Introduction

Per- and polyfluoroalkyl substances (PFAS), including perfluorooctane sulfonate (PFOS) and perfluorooctanoic acid (PFOA), have been widely used globally for many applications such as lubricants, adhesives, stain and soil repellents, paper coatings, pharmaceuticals, insecticides, cosmetics, food packaging, and fire-fighting foams (Prevedouros et al., 2006). There is significant public concern over the widespread detections of PFAS in environmental media including surface water, groundwater, soil, biota and air, coupled with uncertainties about health risks to humans and the environment from these types of compounds (Conder et al., 2008; Gagliano et al., 2020; Houde et al., 2011). Although some PFAS compounds with known human health risks have been voluntarily phased out (PFOA and PFOS), legacy contamination remains, and additional PFAS compounds have been introduced with limited understanding of their health risks (Janousek et al., 2019; Wang et al., 2013). The U.S. Environmental Protection Agency (EPA) has established a drinking water lifetime health advisory limit at 70 ng/L for individual or combined concentrations of PFOS and PFOA while no established maximum contaminant level (MCL) has been set to regulate the acceptable level of these and other PFAS compounds in drinking water. At the same time, multiple states have issued state-specific drinking water guidelines ranging from <2 ng/L to 400 ng/L (Califonia, 2019; ITRC, 2020; NJDEP, 2019; USEPA, 2016). To date, in the United States, PFAS chemicals have not been listed as hazardous substances, therefore requirements under the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA) process to monitor and regulate contamination in water, soil, and sediment do not currently apply. In the absence of federal regulation, PFAS site management activities focus on identifying the presence of PFAS in soil, groundwater, and drinking water with limited remediation efforts.

Most conventional remediation techniques are reportedly ineffective in destroying PFAS. The strong carbon-fluorine bonds of PFAS make them chemically and physically stable compounds and resistant to chemical, physical, and biological degradation (Khan et al., 2019; Kucharzyk et al., 2017; Merino et al., 2016). Sorption by carbon is found to be an effective ex situ technique to remove various PFAS from water matrices (Carter and Farrell, 2010; Ochoa-Herrera and Sierra-Alvarez, 2008; Senevirathna et al., 2010a, 2010b; Siriwardena et al., 2019a, 2019b; Takagi et al., 2011; Zhang et al., 2011). The use of activated carbon to sorb PFAS from water has been studied by numerous groups (Carter and Farrell, 2010; Ochoa-Herrera and Sierra-Alvarez, 2008; Senevirathna et al., 2010a, 2010b; Takagi et al., 2011; Xiao et al., 2017; Yu et al., 2009; Zhi and Liu, 2015). The porous structure of granular activated carbon (GAC) and the high surface area for sorption distinguishes GAC as an effective contaminant removal media for many water-soluble contaminants such as fuel oil, solvents, polychlorinated biphenyls (PCBs), dioxins, and PFAS (Merino et al., 2016; Siriwardena et al., 2019a, 2019b; Vecitis et al., 2009). GAC is currently used to remove PFAS in water treatment facilities. It is reportedly more than 90% effective at removing long-chain PFAS, but is less effective for short-chain PFAS and precursors (Crone et al., 2019; Merino et al., 2016; Ross et al., 2018; Vecitis et al., 2009). Use of GAC to remove PFAS has been found to be beneficial due to its ability to be reactivated and the added benefit of the simultaneous removal of other water contaminants (Forrester, 2019).

When GAC reaches its PFAS contaminant sorption capacity, the treatment system will not remain effective unless the spent GAC is removed and replaced with clean GAC. The life span of GAC is dependent on several factors including the type of contaminant of concern, contaminant level, other co-contaminants, and water quality. In some cases, it can last for several years until contaminants begin to break through the GAC bed and exceeding the required treatment specifications (Health, 2018; Matthis and Carr, 2019). When PFAS water concentrations are high, the lifecycle of GAC can be short, resulting in the frequent need to replace the spent GAC with virgin GAC. Disposal or reactivation/replacement of PFAS-spent GAC is the main challenge associated with use of GAC in water treatment facilities. Regular replacement of spent GAC in the treatment system with virgin GAC is an expensive process (McNamara et al., 2018). Spent GAC is either disposed or reactivated by undergoing a high temperature incineration process before reuse, both of which involve high costs (Matthis and Carr, 2019).

Heating the GAC to high temperatures (800 to > 1000 °C) in furnaces to thermally destroy the sorbed PFAS is the common method used for reactivation of PFAS-spent GAC (Matthis and Carr, 2019; McNamara et al., 2018); however, no standard operating procedure has been established for this method. Moreover, the physical and chemical properties of the reactivated GAC change with each reactivation cycle due to exposure to high temperature conditions which can affect the sorption behavior and sorption capacity of the reactivated GAC (Matthis and Carr, 2019). Other conditions that may affect the sorption behavior of the reactivated GAC include loss of sorption sites due to irreversible sorption (once sorbed, a molecule cannot desorb or partition away from the surface), a decrease in the available volume percent of the sorption pores, and 5%–10% carbon loss during each reactivation cycle (Matthis and Carr, 2019). Because of this loss of sorption sites, the addition of virgin GAC to reactivated GAC is required prior to returning the reactivated GAC to service in order to obtain predetermined GAC performance specifications (Matthis and Carr, 2019).

Challenges attributed to thermally reactivating PFAS-spent GAC have led to solvent regeneration of GAC being investigated as a possible alternative. Solvent regeneration of ion exchange (IX) resins has been studied extensively (Carter and Farrell, 2010; Deng et al., 2010; Conte et al., 2015; Zaggia et al., 2016; Gagliano et al., 2020), however, very limited documentation on activated carbon solvent regeneration exists in the peer-reviewed literature. A small number of literature-reported studies did demonstrate solvent regeneration of PFAS compounds from a variety of carbon sources including powder activated carbon (PAC) (Punyapalakul et al., 2013), bamboo-derived activated carbon (BdAC) (Du et al., 2015), and reactivated coconut shell-based GAC (R-CAC) (Du et al., 2016). Each of these studies was carried out as a batch experiment which is different than the flow-through system envisioned for activated carbon solvent regeneration in a field application. These regeneration experiments included various mixtures of salt, base, and/or organic solvents to obtain a range of ionic strengths with the goal of aiding in the removal of PFAS from different sorbents. The primary sorption mechanisms involved with these sorbents may be through hydrophobic affinity of PFAS carbon-fluorine chain with the sorbent. In addition to that, sorption is likely facilitated by the electrostatic interaction between the negatively charged head group of PFAS and the positively charged sites of the sorbents (Woodard et al., 2017). Overall, conventional sodium salt solutions were not effective in regenerating PFAS-contaminated sorbents due to the lower solubility of PFAS in higher ionic strength media (Conte et al., 2015; Deng et al., 2015; Carter and Farrell, 2010), whereas incorporation of an organic solvent to the salt solution increased the regeneration efficiency drastically by weakening the hydrophobic interactions between PFAS and the sorbent (Deng et al., 2010, 2015; Zaggia et al., 2016).

An example of solvent based regeneration was applied to regenerate IX resins conducted in column experiments was found in Carter and Farrell (2010), wherein they attempted to regenerate anion exchange (AE) resins loaded with PFOS and perfluorobutane sulfonic acid (PFBA) using various concentrations of sodium chloride (NaCl) or sodium hydroxide (NaOH) at different temperatures, which resulted in minimal removal of the PFAS compounds. In a batch experiment, Conte et al. (2015) used NH₄Cl and NH₄OH as a substitute of NaCl and NaOH, due to higher water solubility of PFAS ammonium salts compared to sodium salts, improved IX resin regeneration was observed. Further, when NH₄OH was mixed with organic solvents showed better removal efficiencies compared to using only NH₄OH or NH₄Cl or mixtures of these two chemicals in the absence of organic solvent (Conte et al., 2015).

The technology described here builds from these previous studies by incorporating polar organic solvents mixed with a base additive, which is consistent with the understanding that hydrophobic interactions are the dominant sorption mechanism of PFAS to carbon while electrostatic interactions between anionic PFAS and carbon surfaces also occurs (Du et al., 2014; Johnson et al., 2007; Siriwardena et al., 2019a). Organic solvents have been reported as being effective at disrupting the hydrophobic interactions of PFAS (Woodard et al., 2017) and increased pH can make the sorbent surface more negatively charged, creating an electrostatic repulsion that weakens the attraction of PFAS anions (Du et al., 2014) and improves PFAS removal efficiency. To the best of our knowledge, this is the first study of solvent regeneration of PFAS-spent GAC using a column study that was conducted with the goal of future scale-up for onsite water treatment.

Current research is focused on development and demonstration of an effective GAC regeneration technology using a solvent-based method for PFAS-laden GAC used in water treatment (Patents US16/830,210 and PCT/US20/24774). This included the identification and optimization of various solvent and base mixtures (SBMs) to determine the most effective solvent system for removal of PFOA and PFOS from laboratory-contaminated GAC. A sorption study using a small-scale column test was conducted to investigate the PFAS sorption behavior of new (virgin) GAC and regenerated GAC as well as an assessment of the ongoing efficacy of GAC regenerated multiple times. Four samples with PFAS spent-GAC from different groundwater and drinking water treatment facilities were regenerated using the optimized SBM and distillation of spent PFAS-impacted solvent was carried out to demonstrate the proof of concept for on-site regeneration and solvent recycling. Finally, the results were compared with existing literature-reported methods on chemical regeneration of PFAS-saturated sorbents, experimental conditions and regeneration efficiency.