Anthropogenic Release of Per- and Polyfluoroalkyl Substances into Surface Water Systems: Distribution Characteristics and Environmental Persistence Analysis

Abstract

In view of the issues including the incomplete identification of alternatives and difficulty in tracing pollution sources in PFAS pollution monitoring in surface water, this study took typical surface waters with intensive human activities as the object to perform PFAS screening. A nontarget analysis based on high-resolution mass spectrometry was developed, coupled with a modified solid phase extraction pretreatment method, to achieve the comprehensive screening of 12 legacy carboxylic acids and sulfonic acids, as well as 2 novel alternatives in water. Surface water samples were collected from typical functional areas of human activity to reveal the spatial differential distribution of PFAS concentrations. The long-chain PFASs showed a high detected concentration, among which PFOS, PFUnDA, and PFOA concentrations were especially high in urban complex pollution areas, while PFDA, PFOS, and PFOA were the main components in agricultural areas. The two exposure patterns showed a certain degree of differentiation, which may be related to different pollution sources. PFASs with a long carbon chain, especially chlorine-substituted sulfonic acid, are high-persistent-risk substances. This study provided the data basis for the prevention and control of PFAS pollution in surface water, and supported the treatment of emerging pollutants in the region.

1. Introduction

Per- and polyfluoroalkyl substances (PFASs) are a class of synthetic fluorinated organic compounds that have been widely used in industrial production and consumption since the 1950s [1]. Their unique structure makes PFASs hydrophobic, oleophobic, and thermally stable. Their main applications include industrial coatings, waterproof fabrics, food packaging materials, firefighting foam, and electronic devices [2,3]. However, the environmental and health risks of PFASs have gradually emerged. The extremely high energy of carbon–fluorine bond energies results in a long degradation time of up to hundreds of years, and they are able to be accumulated in living organisms and transmitted through the food chain [4]. Studies have shown that PFAS exposure is associated with liver damage, thyroid dysfunction, immune system suppression, reproductive developmental abnormalities, and increased risk of specific cancers [5,6]. Although several of the legacy homologs, including PFOA and PFOS, have been restricted by international conventions, a number of novel PFASs are still being produced and used, and their potential hazards continue to be of concern [3,7].

Human activities have brought PFASs into the environment, resulting in a widespread distribution. During the production process, PFASs are able to directly enter water bodies, soil, and the atmosphere through industrial wastewater and exhaust gas emission [7,8]. The use of consumer products including fluorine-containing foams and food packaging further accelerates the release and transformation of PFASs into the environment [9]. Sewage plants find it difficult to effectively remove PFASs, resulting in recycled water and sludge becoming a secondary source of pollution [3]. Study data show that PFASs have caused pollution on a global scale, and have been detected in polar bears, the terrestrial food chain from the Tibetan Plateau, urban drinking water, and atmospheric particulate matter [10,11,12]. The transport ability for long distances and distribution characteristics across an environmental medium make PFASs a typical emerging pollutant with global risks, and the long-term threat to ecosystems and human health needs to be systematically studied.

There are thousands of compounds in the PFAS homolog series, and with the phase-out of legacy PFASs, novel PFASs such as short-chain congeners, branched isomers, and perfluorinated alternatives have been continuously developed and put into use [3]. As a result, the detection coverage of existing methods is insufficient to cover the existed PFASs in the environment. Although the traditional method shows high sensitivity to target chemicals, it is difficult to analyze trace unknown pollutants, degradation products, and novel alternatives existing in the complex medium. In recent years, nontarget screening based on the development of high-resolution mass spectrometry (HRMS) has provided technical support for the identification of novel PFASs [3,12]. However, the separation capability, anti-interference ability, and processing efficiency still need to be improved for achieving high-throughput screening [13]. The establishment of powerful detection methods covering the accurate identification of PFASs, the quantification of multiple components, and toxicity effect analysis has become an important requirement for clarifying their environmental attribution, exposure risk, and control strategies.

In this study, we focused on surface water from two functional areas, which is the typical aquatorium significantly affected by human activities. The objectives of the present study were to (1) construct a nontarget analysis method based on HRMS for the screening and quantification of legacy and novel PFASs in water samples, (2) analyze the contribution of human activities to the exposure patterns of PFAS in different functional areas, and (3) perform a persistence assessment for PFASs to evaluate the hazards of pollution in surface water. This study can provide a data basis for the tracing and control of characteristic pollutants in surface water.

2. Materials and Methods

2.1. Study Area

The study area is located in a city in northern China, and two functional areas with an apparent influence of human activities are selected. The surface water of the study area one is located at a park in the city. The source of water includes the inflow of reclaimed water from the urban sewage treatment plant, the direct input of road traffic runoff, and domestic wastewater from surrounding residential areas. The hydrodynamic conditions of this area are weak, the pollutants may be intercepted for a long time, and a regional pollution accumulation effect exists. Sampling area two is located in the intensive agricultural area on the outskirts of the city, close to the contiguous farming area. The water quality is affected by the pollution from agricultural non-point sources, and the chemical fertilizers and pesticides used in agriculture may also enter the water through rain-wash or percolation. The hydrodynamic conditions in this area are relatively strong, with a certain flow capacity, and thus the pollutants have a wide range of migration and diffusion pattern.

2.2. Sample Collection and Pretreatment

The sampling sites were arranged at a horizontal distance of 1 m from the shoreline, and the sampling depth were 0.5 m below the water surface to ensure that the samples represented the characteristics of the physical and chemical mixed layer of the water in the research area. Water samples of the surface layer were collected with a polypropylene water sampler. Each sample’s volume was 1 L, and these were immediately transferred to polypropylene sample bottles, stored at 4 °C, away from light, and pretreatment was completed within 24 h. The water samples were filtered by a 1.2 μm glass fiber filter membrane, and then the filtered water samples were enriched through a tandem solid phase extraction system containing Oasis HLB (6 cc, 500 mg, Waters, Milford, MA, USA), WAX (6 cc, 150 mg, Waters, Milford, MA, USA), and WCX (6 cc, 150 mg, Waters, Milford, MA, USA). WAX, HLB, and WCX columns were eluted with 0.5% ammonia methanol, methanol, and 2% formic acid methanol, respectively. The eluent was concentrated to 1 mL under nitrogen and was filtered by a 0.22 μm PTFE filter to prevent the column from being blocked.

2.3. Instrumental Analysis

Samples were analyzed using an ultrahigh-performance liquid chromatography system coupled with an Orbitrap Exploris 480 mass spectrometer (UPLC-Orbitrap MS, Thermo Fisher Scientific Inc., Waltham, MA, USA) equipped with a heated-electrospray-ionization (H-ESI) source in negative mode. Chromatographic separation was carried out on a Waters ACQUITY UPLC BEH C18 column (2.1 mm × 150 mm, 1.7 μm, Waters, Milford, MA, USA). The column temperature was 40 °C, and the injection volume was 10 μL. The mobile phase A was water containing 2 mM of ammonium acetate, and mobile phase B was methanol. The gradient elution procedure is set as follows: 0–1 min, 10% of solvent B; 1–27 min, increased to 100% of solvent B; 27–31 min, 100% of solvent B; 31–31.5 min, decreased to 10% of solvent B; 31.5–35 min, 10% of solvent B. The flow rate was 0.3 mL/min.

2.4. Nontarget Screening Strategy

High-resolution mass spectrometry raw data were imported into MS-DIAL version 4.90 for nontarget analysis. MS1 tolerance was set to 0.01 Da and MS2 tolerance was set to 0.025 Da. Minimum peak height was set to an amplitude of 100 and a mass slice width of 0.1 Da in the peak detection process. Based on retention time and accurate mass, peak alignment among the samples was performed based on a retention time tolerance of 0.05 min and an MS1 tolerance of 0.015 Da to generate a feature matrix containing m/z, retention time, and peak area. Novel PFAS alternatives were annotated and identified by matching the public database and in-house library, combined with retention time information and characteristic fragments (e.g., m/z 118.9926 [C₂F₅⁻], m/z 168.9894 [C₃F₇⁻], m/z 268.9830 [C₄F₉⁻], etc.). External calibration curves were established by reference standards for legacy and novel PFASs with available reference standards purchased from Wellington Laboratories (Guelph, ON, Canada) and ANPEL Standard Technical Services (Shanghai, China).

2.5. Persistence Assessment

The chemical structures of the detected PFASs were described by the simplified molecular input line entry system (SMILES). Preprocessed SMILES files were imported into the BIOWIN 3 model in EPI Suite, which was established based on the estimated biodegradability of 200 chemicals. The occurrence of certain substructures in the molecular level of chemicals were analyzed by the model, and then the probable biodegradation half-life was estimated using the corresponding descriptors. BIOWIN 3 has been used to predict the persistence of PFASs and proved the reasonability of the application in previous studies [14,15].

2.6. Quality Control and Quality Assurance

Procedural blank samples were prepared using pure water treated with the same method of samples. The features that were not detected in the procedural blank or the intensity that was detected in samples were three-times higher than that in the procedural blank and were included for further analysis. PFASs labeled with the ¹³C isotope purchased from Wellington Laboratories were used as the internal standard to control the recovery. Materials that may cause PFAS contamination were carefully avoided during the pretreatment process.

3. Results

3.1. Detection and Identification of PFASs

In this study, fourteen legacy and novel PFASs were detected in surface water samples by combining nontarget screening and target quantification. The relative proportions of PFAS classes are shown in Figure 1. Legacy PFASs included perfluoroalkyl carboxylic acids (PFCAs) with the carbon of C4~C12, and perfluoroalkyl sulfonic acids (PFSAs) of C4, C6, and C8. Novel PFASs were 6:2 fluorotelomer sulfonate (6:2 FTS) and 8:2 chlorinated polyfluoroalkyl ether sulfonic acid (8:2 Cl-PFESA). The average of total concentration was 307.8 ng/L. Among the legacy PFASs, PFOS (30.4–124.7 ng/L), PFUnDA (28.1–94.9 ng/L), and PFOA (30.2–69.3 ng/L) were the dominant pollutants, accounting for 61.4%, indicating that long-chain PFASs were still the main pollutants exposed in this study area, and the potential input contribution may be fluorochemical consumer products such as coatings, surfactants, and additives. Novel alternatives including 6:2 FTS and 8:2 Cl-PFESA also showed a high response. The accurate m/z of 6:2 FTS was 426.9674, and the fragments of 79.95751 [SO₃⁻] and 80.9652 [HSO₃⁻] were consistent with the characteristic fragmentation rules of perfluorinated sulfonic acid. As an alternative to PFOS in the chrome plating industry, and the degradation product of fluorinated surfactants, 6:2 FTS has been detected in the environment [16]. The feature of 8:2 Cl-PFESA has the m/z of 630.8903, and the characteristic fragment of 450.9421 [C₈F₁₆ClO⁻]. This novel PFAS was the principal component and impurity of F-53B, which has been widely detected in water, air, fish, and humans [17]. The common feature of the two novel PFASs is that they contain special functional groups such as a sulfonic acid group and chlorine structure, and their environmental behavior including biodegradability and membrane permeability is significantly different from that of legacy PFASs.

3.2. Distribution Characteristics of Different Functional Areas

Quantification was achieved for legacy PFCAs and PFSAs, and the concentrations were shown in Figure 2. On the whole, the content of PFASs in urban mixed functional areas was higher than that in agricultural areas. The total PFAS concentration was 435.3 ng/L in urban areas and 180.2 ng/L in agricultural areas. The ratio of PFCAs to PFSAs in urban areas is 2.3. The concentrations of PFOS, PFUnDA, and PFOA were especially high, with an average value of 124.7, 94.9, and 69.3 ng/L. In addition, the concentrations of other long-chain PFASs including PFDA, PFDoDA, and PFNA were also high, which were 48.7, 39.5, and 31.7 ng/L, respectively. PFDA, PFOS, and PFOA were detected in high levels in the agricultural area, which were 31.5, 30.4, 30.2 ng/L, respectively, and the ratio of PFCAs to PFSAs was 3.3.

The two exposure patterns have a certain degree of differentiation, which may be related to different pollutant sources. The potential source of surface water in urban areas includes two parts. One of the sources is the discharge of reclaimed water from sewage treatment plants, which are composite treatment plants for both urban domestic sewage and industrial wastewater. The other source is the input of various urban activities, including surface traffic runoff and landfill leachate. PFAS sources in agricultural areas include the addition of PFASs as wetting agents or stabilizers in certain herbicides and pesticides. In addition, fluorine-containing mulch film and pesticide packaging are released into the environment after aging and degradation, and are washed by rain or filtered by irrigation water into ditches and enriched in water bodies [1].

3.3. Persistence Assessment of PFASs

The persistence of PFASs was assessed by predicting the biodegradation half-life directly from chemical structure (Figure 3). According to the descriptor meaning of the BIOWIN 3 output category [18], the highest value of PFASs in this study was 2.15, which was in the category of months. The values of PFBA and PFPeA were in the range of 1.75 to 2.25, indicating that the degradation time was several months. Most of the remaining PFASs were less than 1.75 and were classified as recalcitrant contaminants, the biodegradation of which was over several months. The persistence of PFASs containing the same functional group increased with the growth of the carbon chain. For each addition of a –CF₂– unit, the degradation capacity fraction decreased by about 0.32, indicating the risk of long-chain PFASs. Previous studies on non-human primates showed that the half-life of short-chain PFASs is shorter than that of long-chain PFASs [19]. In general, perfluorinated chains remain stable and persistent. For different classes of PFASs, PFSAs are more persistent than PFCAs with the same carbon number, with the score decreasing by 0.57, consistent with the trend in previous studies [20]. PFASs with Cl substitution were found to be harder for degradation and more persistent, corresponding to the reported half-life of Cl-PFAS in humans of 15.3 y in a previous study [21], which was thought to be one of the most biologically persistent PFASs.

4. Discussion

The pollution patterns of PFASs in surface waters are characterized by the coexistence of legacy long-chain PFAS residues and novel short-chain alternatives, and the differential distribution between urban and agricultural areas. In this study, the total PFAS concentration in urban areas was significantly higher than that in agricultural areas, and the concentration of long-chain PFCAs such as PFOA and PFNA was prominent, reflecting the essential differences in pollution history, industrial influence, and environmental behavior of the two functional areas.

In terms of types of PFASs, long-chain PFASs were still the main pollutants in the study region. Although wide use of PFOA and PFOS has been limited, they are persistent in the environment and related to extensive ecological risks due to their extremely high carbon–fluorine bond energy and long half-life [22]. In this study, long-chain (≥C8) perfluorinated compounds accounted for 82.5–93.9% of the total PFAS concentration in the water of the study area, and the proportion of PFOA and PFOS was 15.9–16.8% and 16.9–28.6%, respectively. PFOA and PFOS have been found to be the main PFASs in surface water in multiple studies [23,24], which is consistent with the result of this study, possibly due to their large-scale production and application, as well as the high resistance of biotransformation, which lead to their direct entry and occurrence in the surface water environment. On a longer time scale, the retention time of PFASs in long and short chains, as well as their distribution among water, sediment, and biota depend on their ability to bind to suspended particles and accumulate in sediments [25]. Long-chain PFASs are more easily decomposed into soil and sediment than they are in water due to their physical and chemical properties. Thus, short-chain PFASs can be diffused into the surface water far offshore under the action of water flow, while long-chain PFASs are mainly concentrated in the sediments near the pollutant emission source [26]. The high detection level of long-chain PFASs in this study may be due to the proximity of the sampling site to the nearshore emission source, so the influence of diffusion and distribution is weak.

The urban area and agricultural area of surface water are functional areas experiencing the strong intervention of human activities. The two regions show the commonality of pollution exposure mode due to the industrial application of fluorinated chemicals, and also show significant distinction in pollutant composition, migration path, and risk effect due to the differences in industrial attributes, emission characteristics, and environmental hydrological conditions. The thorough analysis of the internal mechanism of this pattern of similarity and distinction is important to the construction of an accurate pollution control system. Sewage treatment plants in urban areas, which receive domestic and industrial effluent, are one of the important reasons for the high concentration of PFASs in the surface water of the region. Industrial districts import high concentrations of PFASs into the environment as the main production source of fluorinated surfactants and perfluoropolymers, and intermediates and by-products due to incomplete synthesis reactions in the production process via which they may enter the surface water with wastewater. In some studies, industrial plants were considered the most important source of PFASs in the environment [27]. In a study of the Xiaoqing River and surrounding residents residing near a fluoropolymer production plant in China, the detected concentration of PFASs in the downstream river passing through the plant was several hundred times higher than that upstream [7]. A study conducted near a PFAS production plant in North Carolina, USA, showed that PFASs discharged from the facility could lead to the long-term contamination of groundwater and surface water, as well as impacts on downstream drinking water supplies [28].

The characteristics of PFAS pollution in agricultural areas are not only driven by local agricultural activities, but are also closely related to cross-media transmission. The contamination of the agricultural environment by PFASs may be related to irrigation with agricultural water and the use of biosolid modified soils [29]. In a study of an agricultural area, it was found that biosolids applied to an agricultural field were related to PFASs in rivers, among which PFOA was a dominating component [30], consistent with the present study. Another study in the agriculture-dominated watershed reported that long-chain PFASs were detected more frequently in surface water than in well water [31]. In this study, PFAS concentrations in agricultural areas were lower than in urban areas, whereas in previous studies, agricultural production areas in rural areas were considered to be potential background areas for PFAS pollution because the sources of PFASs in agricultural producing areas were thought to simply come from irrigation water, which includes mainly groundwater, and atmospheric precipitation [32]. However, according to the current detection results, the PFAS level in the agricultural area was higher than the theoretical value in the background area, indicating that modern agricultural activities have broken the natural background balance. Despite the low concentrations of short-chain PFASs in agricultural areas in this study, they may enter the food chain through crop absorption [33] and may still cause exposure risk.

PFOSs with a high detected level in this study and several long-chain PFCAs were found to be bioaccumulative in laboratory and field water environment tests [34,35]. For example, PFOS has been found to disrupt biological processes in aquatic species, leading to oxidative stress in microbial communities, damage to photosynthesis, altered pigment concentrations, impaired growth, and increased membrane permeability [36], thus threatening aquatic species and destabilizing ecosystems. It is worth noting that concentrations of short-chain PFAS were detected in the environment at higher levels than long-chain compounds in some cases [37]. As novel PFASs are increasingly adopted as alternatives, the risks in the environment require further attention and intensive monitoring.

Our study performed a comprehensive screening to analyze the distribution of legacy and novel PFASs, providing research data supplementation for pollution monitoring and management. This study has several limitations. For certain potential PFAS features, inadequate information was gained to achieve the complete identification of the structure. Meanwhile, for the novel PFASs which were difficult to obtain as reference standards, relative quantification rather than absolute quantification was adopted to measure the exposure level. In addition, more samples from different functional areas are expected to be analyzed in further studies.

5. Conclusions

Nontarget screening and quantitative analysis were adopted to detect fourteen legacy and novel PFASs in the surface water of urban and agricultural areas. The study revealed that legacy long-chain PFASs were still the dominant pollutants in the study region, among which PFOS, PFUnDA, and PFOA were the main pollutants, illustrating the long-term input and persistent contribution of legacy PFASs. The detection of novel alternative 6:2 FTS and 8:2 Cl-PFESA showed the potential risks of industrial substitution strategies. The total PFAS concentration in the urban area was significantly higher than that in the agricultural area. The ratio of PFCAs/PFSAs was 2.3 in the urban area, which was mainly derived from surface runoff input and the tailwater from sewage plants. The agricultural area was dominated by pesticide additives and sludge agricultural use, and the PFCAs/PFSAs ratio was 3.3. Long-chain PFASs and chlorinated PFASs were highly persistent due to their low biodegradability, further increasing their ecological risks. According to the research results, industrial source control and advanced wastewater treatment were required to be strengthened in urban areas, and in agricultural areas, the use of fluorine-containing pesticide additives was expected to be restricted. Meanwhile, it is urgent to incorporate novel PFASs into the monitoring system, and to formulate hierarchical management strategies based on chain length and functional group differences.