Characterization of Per- and/or Polyfluoroalkyl Substances (PFASs) in Reclaimed Water in Three Arizona Communities

Abstract

Per- and polyfluoroalkyl substances (PFASs) are a class of manufactured organic chemicals that are widely employed for their heat-, oil-, and water-resistant properties. Studies have shown that the bioaccumulation of PFASs in living organisms and their related health effects are sufficient for classifying them as a group of toxicants worthy of great concern and further study. While PFASs travel through the air and soil, their contamination of water pathways proves to be the most common route for exposure. We analyzed PFASs from three different wastewater treatment plants (WWTPs) throughout Arizona to show that, despite treatment efforts, they persist as contaminants in water sources. Using U.S. Environmental Protection Agency method 1633, seasonally obtained field samples were prepared for analysis through liquid chromatography–tandem mass spectrometry. A total of 24 samples were taken at different stages of the treatment process to assess the proficiency of the removal processes during remediation. Duplicate samples were each taken from Tucson’s WWTP and Flagstaff’s WWTP before and after chlorination, and from three sites in Yuma County, upstream effluent, downstream effluent, and WWTP, before chlorination. From the samples obtained in Yuma, perfluorooctanesulfonic acid, perfluorooctanoic acid, and perfluorohexanesulfonic acid were detected but at levels below their limits of quantification. PFBS was detected at the Yuma and Tucson WWTP at levels up to 4.52 ng/L and 73.53 ng/L, respectively. The samples obtained from Flagstaff’s WWTP were below the instrument level of detection and, therefore, characterized as non-detects.

1. Introduction

Per- and/or polyfluoroalkyl substances (PFASs) are synthesized organic compounds that are utilized in a variety of industries for their ability to resist degradation in heat, oil, and water [1]. There are thousands of different types of PFASs [2], each composed of a carbon chain saturated with fluorine atoms—one of the strongest chemical bonds due to differences in electronegativities—and a functional group usually consisting of a carboxylic acid, sulfonic acid, sulfonamide, alcohol, or phosphonate group [3]. The amphiphilic properties of PFASs result from a polar hydrophilic head and a hydrophobic carbon-fluorine tail, which give them a high capacity to adsorb to many types of surfaces [4]. PFASs are resistant to degradation, and they bioaccumulate in both biological organisms and the environment [5], resulting in severe health implications. The most researched PFASs, perfluorooctanoic acid (PFOA) and perfluorooctane sulfonic acid (PFOS), have been linked to cancer, adverse reproductive effects [5], hormone and immune system suppression, and developmental issues [1] at levels as low as 4.0 ng/L [2].

Concerningly, studies have shown that almost all people in the United States have detectable levels of PFASs in their blood [1]. The main pathway for exposure is the direct consumption of, specifically, contaminated drinking water from sources such as groundwater, surface water, and snow [6,7].

This study analyzes reclaimed water and characterizes the different PFASs present at three wastewater treatment plants (WWTP) in the U.S. state of Arizona. Positioned in the North American Southwest, a geographic region characterized by an arid climate and low annual rainfall [8], the use of reclaimed water is of critical importance, especially as the area is experiencing unprecedented warming [9], unpredicted weather patterns that lead to megadroughts [9], and a fast growing population [10]. Three sites in Arizona were sampled: a tribe located in the southwestern region of Arizona in Yuma County, hereafter referred to as the Southwest Arizona Tribe, Tucson in Pima County, and Flagstaff in Coconino County. Samples were obtained at different stages of the treatment process, i.e., before and after chlorination, to assess the efficiency of PFAS removal through reclamation processes. As classified by the Arizona Department of Environmental Quality, both Pima County and Flagstaff produce A+-grade reclaimed water [11,12], which is achieved by undergoing a secondary treatment to obtain a nitrogen concentration of less than 10 ug/mL, fewer than 23 coliform organisms per 100 mL of sample, and a turbidity no greater than 2 Nephelometric turbidity units [13]. It is then distributed via purple pipe systems to their corresponding areas. Conversely, the Southwest Arizona Tribe manages a wastewater reclamation facility that utilizes an oxidation ditch system. This water is not categorized under a graded system.

While drinking water has been the main target of PFAS research, the presence of PFASs in reclaimed water is less studied. However, as WWTPs have been shown to contribute to PFAS release into the environment [14], data collection regarding PFAS types and their corresponding quantities can help determine areas of high risk. Additionally, the detection of PFASs in wastewater effluent has been associated with their presence in the drinking water they supply [15]. While two of the total three sites in this study discharge their effluent into a water source used for drinking water, i.e., from the Rio de Flag, the WWTP in Flagstaff, to the Colorado River and from the Tucson WWTP to groundwater, the goal of this study is to identify the possible sources of PFASs.

2. Materials and Methods

To mitigate the risk of analytes adsorbing to materials with a high affinity for PFASs, polypropylene containers, and glassware were used for all standard, sample, and extraction preparations. Brand names and/or catalog numbers are included for illustration only and do not imply endorsement of the products.

2.1. Sampling and Preparation Materials

Samples were collected in 500 mL high-density polyethylene (HDPE) Nalgene wide-mouth sampling bottles with a screw cap and top. ACS-grade 28–30% ammonium hydroxide (NH4OH; CAS 1336-21-6) and ACS-grade 96+% formic acid (CH2O2; CAS 64-18-6) were used with universal pH indicator strips (non-bleeding) and disposable borosilicate Pasteur pipettes (5 ^3/4”) to ensure a pH between 6 and 7.

Prior to extraction, the total suspended solids in each sample were determined using the following materials: glass fiber filters (1 µm, 47 mm), a porcelain Büchner funnel, and a 125 mL Pyrex Büchner flask. The filters were desiccated in ceramic crucibles and a Fisher Scientific Isotemp Programmable Multi Furnace (Waltham, MA, USA).

2.2. Extraction and Carbon Cleanup Materials

Before extraction, the samples were spiked with an aliquot of Mass-Labeled PFAS Extraction Standard Solution/Mixture (EIS), which was obtained from Wellington Laboratories (Guelph, ON, Canada). The standard contains 24 different analytes in a methanol/isopropanol (1%)/water (<1%) solvent and four molar equivalents of sodium hydroxide to prevent the conversion of carboxylic acids to their corresponding methyl esters.

The extraction procedure was performed on a manual vacuum extraction manifold with a VisiprepTM (Supelco, Bellefonte, PA, USA) large-volume sampler designed for use with SPE cartridges. The cartridges used were OASIS^® WAX 6 cc with a polymeric reverse-phase absorbent and an amine content of 0.45 (meq/g). The SPE cartridges were packed to the halfway mark with silanized glass wool (CAS#: 65997-17-3) before high-purity, LC/MS-grade methanol (CH₃OH, CAS#: 67-56-1), 28–30% ammonium hydroxide (NH4OH, CAS#: 1336-21-6), and 96+% formic acid (CH2O₂, CAS#: 64-18-6) were used to condition the SPE reservoirs. Pipettes ranging in volume between 5 mL and 10 mL were used to deliver the reagents to the SPE cartridges. The eluted analyte was collected in 15 mL sterile polypropylene tubes with CentriStarTM caps.

After extraction, 25 µL of glacial acetic acid (CH₃COOH; CAS 64-19-7) was delivered to the samples via a 100 µL micropipette. To absorb interfering organics, 10 mg SupelcleanTM ENVI-CarbTM SPE (Bayswater, Australia) activated carbon (CAS 7782-42-5) was added to the eluates and hand shaken for no more than five minutes, followed by vortexing for 30 s. A Fisherbrand Model 614B centrifuge (Waltham, MA, USA) was used to mix the samples with the activated carbon for 10 min at 3155 revolutions per minute. Analyte was filtered through 5 mL polypropylene luer-lock syringes fit with nylon syringe filters (25 mm filter, 0.22 µm) in order to remove the activated carbon from the eluate. The eluate was added to a clean polypropylene collection tube containing a 50 µL aliquot of Mass-Labeled PFAS Injection Standard Solution/Mixture (NIS) obtained from Wellington Laboratories (Guelph, ON, Canada). The NIS contains the following compounds: 13C3-PFBA, 13C2-PFHxA, 113C4-PFOA, 13C5-PFNA, and 13C2-PFDA. The samples were homogenized in a Daigger Vortex Genie 2TM (Vernon Hills, IL, USA).

2.3. Analysis Materials

Prior to analysis, the samples were transferred to 1.0 mL polypropylene autosampler vials with polypropylene caps. The samples were quantified using liquid chromatography–tandem mass spectrometry (LC-MS/MS) on an Agilent 6460 triple quadrupole mass spectrometer operating in negative-ion mode (Santa Clara, CA, USA). The HPLC system utilized a 150 mm Phenomenex Gemini C18 analytical column and a 50 mm C18 Phenomenex Guard column (Washington, DC, USA). The data were quantified using Skyline Targeted Mass Spec Environment software.

The instrument was calibrated using solutions made from single-analyte solutions of PFAS molecules: perfluorooctanoic acid (PFOA), perfluorooctanesulfonic acid (PFOS), perfluorohexanoic acid (PFHxS), perfluorohexanesulfonic acid (PFHxA), perfluorobutanoic acid (PFBA), perfluorobutane sulfonic acid (PFBS), and perfluorobutanesulfonic acid (PFOS). During analysis, the mobile phases were composed of 20 mM ammonium acetate solution and LC/MS-grade methanol. Ammonium acetate (NH4CH3CO2, CAS#: 631-61-8) was dissolved in nanopure water to formulate the 20 mM ammonium acetate solution.

2.4. Study Area

Three different locations in Arizona were sampled for PFAS analysis: the Southwest Arizona Tribe in Yuma County, Tucson in Pima County, and Flagstaff in Coconino County. Each location is unique in terms of the applications of and implications for its reclaimed water.

To respect the identity and privacy of the Southwest Arizona Tribe, limited information about the sample locations obtained from their reservation will be provided. However, it can be noted that the tribe manages a WWTP that uses an oxidation ditch system, where the effluent is used for irrigation of recreational spaces [16]. Additionally, as of 2022, the tribe has developed a plan to monitor the condition of the treated wastewater and implement a program to direct it toward environmental restoration efforts [17].

The city of Tucson, located in Pima County in the southern part of Arizona (Figure 1), can be characterized as a hot desert by its lack of rainfall—averaging just 11.92 inches per year—and arid climate [18]. Consequently, the area frequently experiences periods of drought [19]. Supporting a growing population of over half a million, Tucson has placed a priority on innovating a system that will support its water demand in more sustainable ways [19]. In the mid-1980s, Tucson became one of the first cities in the United States to use reclaimed water for non-potable use [20,21]; it currently constitutes 10% of its total yearly water supply [22]. Utilizing a 200-mile system of purple pipes supplied by several of Pima County’s water reclamation facilities, the city recycles and delivers between 14,000- and 20,000-acre feet per year for agricultural irrigation, groundwater recharge, ecological restoration, and watering golf courses [20]. While most of the non-potable reclaimed water is released for use in the city (95%), the other 5% is discharged into the Santa Cruz River [23]. Despite a growing population, the implementation of reclaimed water, among other measures, has led to a decrease in water consumption in the city over the past two decades [24]. Tucson is a pioneer of recycled water and, as such, demonstrates an approach toward innovating ways to tackle water scarcity.

Flagstaff is a city in northern Arizona (Figure 2) with a vastly different climate from Yuma and Tucson. Sitting at 7000 feet in elevation, Flagstaff receives an average of 23.4 inches of rain and 77 inches of snow per year [26]. It has a population of 75,907 [27] and is serviced by two WWTPs that together generate 10 million gallons of water per day [28]. Totaling 20% of all water use in Flagstaff, the reclaimed water from these plants is used to supply the city’s golf courses, cemeteries, parks, and artificial snow making for the Arizona Snowbowl Ski Resort [26]. Effluent from both WWTPs is released into the Rio de Flag, an ephemeral river channel engineered to mitigate floods in the area [28]. With its headwaters at the northern side of the San Francisco Peaks, the Rio de Flag flows southeast until it reaches the Little Colorado River—an inlet for the Colorado River that supplies water to seven different states [29]. Samples were collected from the Rio de Flag Water Reclamation Plant before and after being treated with chlorine.

2.5. Sample Collection

Three different locations were sampled in Yuma County on 16 August 2023 and 16 November 2023. At each destination, two samples were acquired—each treated as an individual sample but hereafter referred to as a duplicate—for a total of six samples. Additionally, a field blank (FB1) containing 500 mL of nanopure water in an HDPE sample bottle was brought to each location. At each sample site, the cap on the FB1 bottle was removed so that the water was exposed to the air for 10 s, and then the bottle was recapped. The same FB1 was brought to all three locations.

The first samples were taken from a pipe at the Southwest Arizona Tribe’s Wastewater Facility immediately before chlorination. For the other two locations, water samples were retrieved directly from the Lower Colorado River. To gather the upstream effluent, clean sample bottles were dipped into the river until completely filled with water and capped tightly. Because the water was actively flowing, the samples were taken from the surface of the water. The downstream effluent was obtained from a pool of standing water that formed an eddy on the side of the river. The sample bottles were dipped about 2 inches below the surface of the water, filled to the brim, and then capped. It was important to not gather samples from the top of the water line to obtain a more homogenized sample because PFASs, especially PFOA and PFOS, tend to aggregate and form a layer at the interface of water and air [17].

Two water samples were gathered from Tucson on 11 September 2023 and 4 January 2024. The clean sample bottles were uncapped, then held under a purple pipe at the Water and Energy Sustainable Technology Center until full (500 mL), and recapped. A field blank (FB2) containing 500 mL of nanopure water was brought to the location, exposed to the air for 10 s, and then closed again.

Four samples were acquired from the Rio de Flag WWTP on 29 September 2023 and 12 March 2024. The purpose of these water collections was to compare the difference in PFAS levels before and after being disinfected with ultraviolet (UV) radiation. The first two samples were taken from a large reservoir of reclaimed water after being exposed to UV light. As the water level was several feet below the walkway platform, a bucket attached to a long pole was utilized to retrieve the water below. After the bucket was filled with water, it was brought back up to the platform, and its contents were carefully poured into the 500 mL sample bottles and capped. To obtain reclaimed water prior to UV treatment, two bottles were held under a purple pipe until full and then capped. As with Yuma and Tucson, the field blank (FB3) containing 500 mL of nanopure water was brought to the WWTP and exposed to the air for 10 s. After each sampling trip, all samples were immediately placed in a cooler with ice for transportation back to the laboratory. Once in the lab, they were relocated to a refrigerator to be maintained at 0–6 °C.

2.6. Sample Preparation

The preparation of the reclaimed water samples followed EPA method 1633, allowing for an analysis of per- and polyfluoroalkyl substances (PFASs) in aqueous, solid, biosolids, and tissue samples using LC-MS/MS [19]. Solid-phase extraction (SPE) using weak anion exchange (WAX) cartridges was used to concentrate the water samples from their initial volume of 500 mL to 5 mL. Before extracting, the total suspended solids (TSSs) in each sample were determined to assess whether additional steps would be required during the sample preparation process. To evenly mix the particulate throughout the sample, the bottles were inverted several times. Then, using a Büchner funnel, a Büchner flask, and vacuum pressure, 10 mL of each sample was filtered through a pre-weighed glass fiber filter. The aliquot was delivered with a 10 mL micropipette. Between each new sample examination, the Büchner funnel was rinsed with nanopure water three times. After filtering 10 mL, filters were placed in ceramic crucibles, dried in a Fisher Scientific Isotemp Programmable Multi Furnace at 103–105 °C for an hour, and then weighed again. The following equation was used to calculate the TSSs:

$$TSS\left(mg\right)={\displaystyle \frac{weight of aliquot after drying \left(mg\right) -weight of filter \left(mg\right)}{0.01L}}\times 0.5L$$

(1)

A TSS content greater than 50 mg indicated that the sample would congest the cartridge. For such samples, two WAX cartridges were prepared before beginning the extraction. The steps after elution differed from the standard procedure and are discussed in the following section.

Between each use, the crucibles were cleaned with three hot water rinses, followed by three nanopure water bath rinses. They were then soaked in a 1% Citranox soap bath for one hour and rinsed three more times with nanopure water. Lastly, they were put in a 10% nitric acid bath for one hour, followed by three nanopure water rinses.

The samples, including the field blank, were spiked with 50 µL EIS solution directly into the sample bottle. The mass of each sample was recorded, and the samples were verified and adjusted, if necessary, to ensure a pH between 6.0 and 7.0. A pH in this range ensures an effective uptake capacity by the charged resin in the SPE cartridges.

Three different concentrations of formic acid were made from the 96% formic acid stock solution: 1 L of 0.3 M, 1 L of 0.1 M, and 0.1 L of 50% (v/v). For the 0.3 M formic acid solution, 11.79 mL of the stock solution was mixed with 988.21 mL of nanopure water. The 0.1 M solution was made with 3.93 mL of the 96% stock solution and 996.07 mL of nanopure water. Lastly, the 50% (v/v) solution was made by combining 50 mL of the stock solution with 50 mL of nanopure water. All three formic acid solutions had a shelf life of two years.

For 1% (v/v) methanolic ammonium hydroxide, 3.3 mL of the 30% ammonium hydroxide stock solution was added to 97 mL of methanol. This was replaced once every month. A 1:1 0.1 M formic acid/methanol solution was made with 100 mL of 0.1 M formic acid and 100 mL of methanol. This had a shelf life of two years. All solutions were kept in polypropylene bottles and stored at room temperature.

2.7. Extraction

Prior to extraction, the samples were stored in the dark at 0–6 °C and extracted within 28 days of the sampling date. This was carried out to stabilize the PFAS in the example matrix and mitigate the risk of its interconversion during storage [31].

An SPE WAX cartridge containing silanized wool was set up in the vacuum manifold and placed in the fume hood. For samples with more than 50 mg of TSSs, two cartridges were assembled in anticipation of the first one becoming clogged; however, only one was used at a time. After being placed in the vacuum manifold, the cartridge was conditioned with 15 mL 1% (v/v) methanolic ammonium hydroxide and then 5 mL 0.3 M formic acid with no vacuum. The wash was discarded. Preventing the reservoir from becoming dry throughout the conditioning, we carefully poured the sample directly into the cartridge. Depending on the sample matrix, the vacuum was typically held between 5 and 12 mmHg to allow the sample to pass through the reservoir at a speed of 5 mL/min, not letting the cartridge become dry at any point. The eluate was discarded appropriately.

After the entire bottle (500 mL) was filtered, the sample bottle was air-dried in the fume hood for several minutes. The bottle was then rinsed twice with 5 mL nanopure water and poured into the cartridge. This was followed by a rinse of the bottle with 5 mL 1:1 0.1 M formic acid/methanol. The vacuum was used for these rinses, and the waste was discarded. The cartridge was then dried for 15 s by maintaining the vacuum pressure and drawing air through. Next, the waste collection container in the vacuum manifold was replaced by a clean polypropylene centrifuge tube to collect the eluent. The sample bottle was rinsed with 5 mL of 1% methanolic ammonium hydroxide and transferred to the reservoir with a glass pipette. We used vacuum pressure to pull the solution through the cartridge and into the collection tube. The sample bottle was then air-dried for several minutes and weighed (with the cap) in grams.

2.8. Carbon Cleanup

If the eluates in the collection tubes were not processed immediately after being extracted, then they were stored in the dark at 0–6 °C. Post-extraction steps were carried out according to Section 2.2. As activated carbon is also an effective agent for removing PFASs from water sources, it was critical to limit its exposure time with the samples during this step [32]. The samples were stored at 0–6 °C until further analysis.

In the event where two SPE cartridges were used (because of elevated TSS levels), the preceding steps leading to the filtering of the eluate through the syringe filter remained consistent. Instead of filtering the solutions into a collection tube already containing NIS, the eluates were both filtered into the same empty collection tube. Afterward, 50 µL of NIS solution was pipetted into the tube and vortexed until mixed. Then, 350 µL of the filtered extract was transferred to a 1 mL polypropylene microvial, and its level was marked. Again, 350 µL of extract was added to the same microvial. The contents were placed in a 40 °C water bath, and, using a constant stream of nitrogen gas to displace the surface tension, they were evaporated and concentrated until they reached the level of the mark. The microvial was stored at 0–6 °C until ready for the LC-MS/MS analysis.

2.9. Analysis

Prior to analysis, the samples were vortexed to ensure that they were properly mixed. The samples were transferred to polypropylene autosampler vials via a 500 μL micropipette. The sample vials were then capped with a polypropylene cap and loaded onto the autosampler for analysis. The instrument was calibrated using solutions made from single-analyte solutions of PFAS molecules: perfluorooctanoic acid (PFOA), perfluorooctane sulfonic acid (PFOS), perfluorohexanoic acid (PFHxA), perfluorohexane sulfonic acid (PFHxS), perfluorobutanoic acid (PFBA), and perfluorobutane sulfonic acid (PFBS). Stock solutions were prepared in 1000 mL polypropylene bottles. The bottles were filled with 1000 mL of distilled and deionized (DDI) water, followed by the appropriate quantity of PFAS salt or solution (10 mg of PFHxA, PFHxS, PFOA, PFOS salts, 5.5 μL PFBS, and 6.1 μL PFBA) [33]. The exact concentration of the solutions was calculated based on the measured mass added (if the mass differed from 10 mg), correcting for the molecular weight of the salt form versus that of the molecular weight of the acid form for each PFAS to provide the final true concentrations in the acid form. The solutions were sonicated for 30 min prior to a series of dilutions that culminated in a final 2 µg/mL stock solution for each PFAS.

To prepare the calibration standards, the stock solution was diluted until the calibration solutions reached concentrations between 5 and 1000 ng/L). These calibration standards were analyzed using 1/x weighted linear calibration curves based on the relative response of each compound/internal standard set. All working solutions were stored in the dark at 4 °C for no more than one month.

The samples from datasets two and three were analyzed on an Agilent 6460 triple quadrupole mass spectrometer operating in negative-ion mode via liquid chromatography–tandem mass spectrometry. The sample introduction system was polytetrafluoroethylene (PTFE)-free to minimize sample contamination. The HPLC system utilized a 150 mm Phenomenex Gemini C18 analytical column and a 50 mm C18 Phenomenex Guard column. Each PFAS analysis included a compound-matched, mass-labeled internal standard at a concentration of 0.4 µg/L.

3. Results

The results of the reclaimed water sampling conducted in Arizona are reported in the following section based on the sampling locations.

3.1. Yuma County/Southwest Arizona Tribe

The results of the fall 2023 sample set collected from the Southwest Arizona Tribe are presented in Figure 3. All six samples that were obtained showed the presence of PFASs when analyzed on the LC-MS/MS. The only sample free of contamination was the corresponding field blank, as shown by the lack of detection by the instrument. Perfluoropentanoic acid (PFPeA) was detected in samples taken from the wastewater reclamation facility (WRF) and upstream effluent. The exact values cannot be provided because a PFPeA calibration standard was not used. As such, no limit of quantification (LOQ) could be obtained, and the data for this PFPeA could only be determined qualitatively. Similarly, PFOS was detected in both downstream effluent samples, and PFHxS was detected in both upstream effluent samples. However, both LOQs for PFOS and PFHxS exceeded the sample values. Therefore, the results could not be determined quantitatively.

The second sample set collected from the Southwest Arizona Tribe during winter 2023–2024 was analyzed on an LC-MS/MS. The data for this sample set are shown in Figure 4. Similarly to the samples that were collected in fall 2023, all six samples showed the presence of PFASs, except for the field blank. At the WRF, PFBS and PFOA were detected. PFBS was found in both samples, with an average concentration of 4.58 ng/L. PFOA was only observed in one of the samples and at a concentration well below the LOQ at 21.7 ng/L. PFOS was detected in the downstream and upstream effluents at concentrations averaging less than 1 ng/L, levels much lower than the LOQ. PFPeA was detected at the WRF, but because the instrument could not be calibrated for PFPeA, a quantitative analysis cannot be provided. As PFOS and PFOA were detected at concentrations below their corresponding LOQs, the exact concentrations cannot be reported with certainty.

3.2. Flagstaff

The Flagstaff Water Reclamation Plant was sampled in fall 2023 and winter 2023–2024, with four samples obtained from each sampling trip. Although both sample sets were analyzed, including a field blank from each of the sampling dates, PFAS compounds were detected at levels below the instrument limit of detections (LODs) for each compound tested (

Table 1. LC-MS/MS LODs (Agilent 6460 triple quadrupole mass spectrometer).

LC-MS/MS LODs
Compound	LOD (ng/L)
PFOA	1.2
PFOS	0.9
PFHxA	1.1
PFHxS	1.3
PFBA	0.8
PFBS	0.6

$\mathrm{L}\mathrm{O}\mathrm{D}=\mathrm{s}\times {\mathrm{t}}_{\left(\mathrm{n}-\mathrm{1,1}-\mathsf{\alpha }=0.99\right)}$ (s = standard deviation of replicate analyses, n = number of replicates, t = t-value for the 99% confidence level with n − 1 degrees of freedom).

).

3.3. Tucson

Water samples from Tucson were obtained in fall 2023 and winter 2023–2024. While both datasets were analyzed and characterized for PFASs, only the samples collected in fall 2023 showed levels of PFASs. Both PFBS and PFPeA were detected in the two water samples that were analyzed. PFBS was found at concentration levels of 59.3 ng/L and 73.5 ng/L (Figure 5), falling below the United States Environmental Protection Agency’s (EPA) lifetime health advisory level of 2000 ng/L [34]. PFPeA was also detected; however, as there was no calibration for the analyte, no quantitative data can be provided.

4. Discussion

The presence of PFASs in reclaimed water samples may be attributed to the wastewater that WWTPs process. As suggested by the name, wastewater is the collection of aqueous waste that comes from residential, commercial, and municipal sectors. Reclaimed water is the culmination of many [35] waste streams, making it difficult to locate the exact origin of PFASs. Possible PFAS wastewater sources can include the disposal of bathing water that may contain PFASs from cosmetics and other personal products and water contaminated from non-stick cookware [36,37]. Additionally, PFASs are usually excreted from the body in urine, making toilet water another pathway for contamination [35]. Sources of PFASs may also originate from surface runoff containing engine oils or vehicle lubricants or irrigation water containing pesticides. Because reclaimed water represents all used water in a given area, it is difficult to identify the exact source from which PFAS contamination in wastewater originates.

PFASs were detected in the water samples obtained from Tucson and Yuma County, indicating a deficiency in their water reclamation plants to remove PFASs from wastewater streams. Conversely, PFASs were classified as non-detections in the samples collected from Flagstaff. Although unlikely, these differences may be attributed to the different processes that the corresponding water treatment plants utilize. The Southwest Arizona Tribe employs an oxidation ditch system, a form of water treatment that is typically used by smaller communities with a limited number of resources [38]. The presence of legacy PFASs, such as PFOS, in its effluent is of concern because it may indicate the limitations of the system. While the WWTPs in Tucson and Flagstaff utilize the same systems and produce grade A+ water, Flagstaff differs in that it employs an additional ultraviolet (UV) radiation disinfection step. Tucson only utilizes chlorination to disinfect the wastewater, and PFBS and PFPeA were detected periodically. As such, it can be hypothesized that the non-detects of PFAS from the Rio de Flag may be attributed to the additional UV disinfection step. However, it is unlikely that this is the case, as WWTPs employing similar technologies have demonstrated the presence of PFASs in their treated wastewater [39]. Rather, it can be hypothesized that there was a lack of PFASs in the wastewater prior to treatment.

Unlike Tucson and Yuma, much of the water that Flagstaff residents use comes from local aquifers, fed by groundwater and annual rainfall and snowmelt [30]. Conversely, a major source of the water that Tucson and Yuma County receive comes from the Colorado River [20], whose headwaters are hundreds of miles north of these cities. As PFASs may accumulate as they move downstream, it is possible that the Tucson and Yuma County WWTPs receive higher loads of PFASs and are, therefore, more likely to exhibit higher concentrations of the contaminants in their treated wastewater [40]. Furthermore, it is possible that the lack of PFAS detections in the Flagstaff samples may not represent the true characterization of the wastewater effluent produced by the Rio de Flag. After these data were obtained, the Flagstaff, Arizona Sustainability Office awarded a community-based organization with a grant in 2025 to test the efficacy of PFAS removal using biochar [41]. Whether this research is being conducted based on recent findings of PFAS from the WWTP or on hypotheses of the contaminants’ presence has not yet been disclosed. In either case, this study prompts further research into the efficiency of WWTPs to identify factors that enhance greater PFAS removal from wastewater effluent.

The uncertainty surrounding the exact origins of PFASs, along with the efficacy of WWTPs in removing PFASs from wastewater, raises concerns regarding the applications of reclaimed water. Without adequate precautions and removal processes, the presence of PFASs in recycled water allows for a continued cycle of contamination. Studies have shown that the accumulation of PFASs in drinking water sources is correlated with the reclaimed water sources that supply those reservoirs [15]. Areas with high levels of PFASs in the drinking water are strongly associated with the number of wastewater treatment plants that contribute effluent back into the watershed [15]. Additionally, reclaimed water that is applied for land treatment, such as agricultural use, is more likely to cause the accumulation of PFASs as it moves through the environment. This occurs through the accumulation of PFASs in the soil and groundwater, resulting in the increased possibility of PFAS uptake in food crops [15]. While contaminated water provides a major source of contamination, studies suggest that food consumption is one of the main pathways for PFAS exposure in humans [15].

In Pima County, recycled water is distributed for agricultural irrigation; evidence of any PFAS in the effluent may be of concern because it allows for the accumulation of the compound in the food crops that it supplies and thus exacerbates an exposure route for human consumption. Moreover, reclaimed water is used for groundwater recharge [20], which consequently presents another pathway for contaminating the water table. Although detected at levels lower than its lifetime health advisory level of 2000 ng/L [34], the quantification of PFBS in the Tucson fall 2023 water samples is an area of interest because it can provide guidance for remediation efforts to fully remove all PFASs in the water.

In Yuma County, although characterized at levels lower than the LOQ, it must be noted that PFOA and PFOS, deemed the most toxic types of PFASs, were each found at concentrations close to the EPA’s maximum contaminant level goals at 4.0 ng/L [34]. The presence of these compounds is worth greater attention, especially as PFOS has a high bioaccumulation factor and accumulates as it ascends the food chain [42]. Studies have shown the presence of perfluoroalkyl acids (PFAAs)—including PFOS and PFOA—in the tissues of various terrestrial and aquatic organisms far from immediate sources of PFASs, indicating the movement of the toxins through food webs [42].

As the Southwest Arizona Tribe plans to use the reclaimed water for ecological restoration efforts [18], the implications of releasing this class of chemicals back into the environment without proper remediation may have severe health consequences for the ecosystem that they supply. PFOA and PFOS have been shown to disrupt the normal function of cell membranes in algae, consequently suppressing their ability to grow [42]. Moreover, PFAAs can have negative reproductive effects, act as endocrine disruptors, and impair immunotoxic responses in zooplankton, fish, and mammals accordingly [42]. Similarly, Tucson uses its reclaimed water for environmental restoration projects [20], and the presence of PFBS and PFPeA, although less studied, may also be of concern for species that interact with these compounds.

Furthermore, when reclaimed effluent enters bodies of water such as river systems, PFASs can be transported to other areas. Depending on the chain length and polarity of the compound, the mobility of PFASs increases in water matrices [43]. The contamination of water is also affected by the partitioning of short-chain PFASs out of sediments and into the water table [42]. In Yuma County, PFHxS was detected in the upstream effluent of the Lower Colorado River. Although it has a smaller bioaccumulation factor than PFOS, it has a longer half-life in water matrices [42] and is more persistent in the environment in this regard. Because water from the Colorado River is delivered to California and Mexico after it flows through Yuma, PFHxS likely remains in the water table as it is distributed to those areas. Consequently, the addition of PFASs to the river system has implications for downstream communities.

5. Conclusions

This study characterized PFASs in water reclamation plants in three different locations across Arizona. Positioned in a climate that is increasingly facing the threats of climate change and water scarcity, the use of reclaimed water has become a critically important adaptation for meeting water demands throughout the state. However, as treated wastewater is repurposed for agricultural irrigation, groundwater recharge, and ecological restoration, among other uses, the presence of PFASs in water provides a pathway for the toxic compounds to accumulate in the water and soil. The implications of this are the uptake of PFASs into food crops that are watered with the reclaimed water, resulting in a cyclic manner of contamination, which ultimately leads to humans’ increased exposure to the compounds. When applied to environmental restoration initiatives, PFASs can also have negative health implications for the organisms that inhabit those areas, potentially resulting in effects opposite to those intended and decreasing the ecological resiliency of the targeted site.

In fall 2023, PFHxS was detected in the upstream effluent at concentrations lower than the LOQ. Additionally, in fall 2023, water samples taken from the downstream effluent from the Southwest Arizona Tribal land contained PFOS at levels lower than the LOQ but likely higher than the EPA’s health advisory limit of 0.02 ng/L. Again, in winter 2023–2024, PFOS was detected in both the upstream and downstream effluents at concentrations exceeding the EPA’s limit but lower than the LOQ. In winter 2023–2024, PFBS was detected in the samples taken from the Southwest Arizona Tribe’s WWTP at levels lower than the LOQ. Although PFOA was also found, it was only found in one of the samples and again at concentrations lower than the LOQ. PFPeA was detected in fall 2023 and winter 2023–2024 at the Southwest Arizona Tribe’s WWTP. It was also detected in the upstream effluent samples obtained in fall 2023. As we were unable to be calibrate for PFPeA on the LC-MS/MS used for analyzing the samples, the PFPeA data remain qualitative. Furthermore, as the PFOS, PFOA, PFBS, and PFHxS concentrations are lower than their corresponding LOQs, data from this set cannot be reported with absolute certainty but may be used as guidelines.

The samples from Tucson’s water reclamation facility in fall 2023 showed the presence of PFBSs, with an average concentration of 66.4 ng/L. PFPeA was detected; however, no quantitative data can be provided. PFASs were not detected in any of the samples collected from Flagstaff’s Rio de Flag wastewater plant in fall 2023 and winter 2023–2024.

To assess where PFASs originate from and the efficiency of their removal in wastewater treatment, further studies must be conducted. Characterizing PFASs at different stages in the water treatment process, including the different sites from which the wastewater is collected, may help direct removal strategies to prevent the movement and cycling of PFASs throughout the environment. Additionally, further studies assessing whether seasonal variations play a role in the presence of certain PFASs in the water table would be beneficial for understanding the degradation, persistence, and behavior of PFASs in the environment.