Insight into the Impacts and Removal Pathways of Perfluorooctanoic Acid (PFOA) in Anaerobic Digestion

Abstract

Perfluorooctanoic acid (PFOA) that accumulates in wastewater and excess sludge interact with the anaerobes and deteriorate the energy recovery and pollutants removal performance in the anaerobic digestion (AD) system. However, the interaction between PFOA and microbial metabolism in the AD systems remains unclear. This study aimed to clarify the effects and mechanism of PFOA on the AD process as well as the removal pathways of PFOA in an AD system. The results showed that the methane recovery efficiency was inhibited by 7.6–19.7% with the increased PFOA concentration of 0.5–3.0 mg/L, and the specific methanogenesis activity (SMA) was inhibited by 8.6–22.3%. The electron transfer system (ETS) was inhibited by 22.1–37.3% in the PFOA-containing groups. However, extracellular polymeric substance (EPS) gradually increased due to the toxicity of PFOA, and the ratio of protein to polysaccharide shows an upward trend, which led to the formation of sludge aggregates and resistance to the toxic of PFOA. The PFOA mass balance analysis indicated that 64.2–71.6% of PFOA was removed in the AD system, and sludge adsorption was the main removal pathway, accounting for 36.1–61.2% of the removed PFOA. In addition, the anaerobes are proposed to have the potential to reduce PFOA through biochemical degradation since 10.4–28.2% of PFOA was missing in the AD system. This study provides a significant reference for the treatment of high-strength PFOA-containing wastes.

1. Introduction

Perfluorinated compounds (PFCs) have been widely used as synthetic substances since the middle of the 19th century and they play a major role in industry, energy, and manufacturing [1]. PFCs synthetic products mainly include surfactants, paints, surface hydrophobic agents, textiles, and so on [2,3]. According to the survey results, the most common perfluorinated compounds in wastewater discharge in China include perfluorooctanoic acid (PFOA), perfluorooctane sulfonic acid (PFOS), and perfluorobutane sulfonic acid (PFBS). The concentration of PFBS in wastewater in a WWTP of Wuhan city, Hubei Province, reaches 1400 ng/L [4]. The highest PFOA content in Shanghai, Dalian, Guangzhou, and other urban sewage treatment plants is more than 6000 ng/L, far higher than that of the United States (9–24 ng/L), Japan (6.6–142.1 ng/L), and other developed countries [5]. PFOA is one of the most harmful organic pollutants because of its physiological toxicity and carcinogenicity [6,7,8]. Updated research results have shown that PFOA has been found in both human and animal bodies, and has irreversible toxic effects on mammalian organs. In one study, PFOA was found to lower blood sugar and increase insulin in mice [9]. PFOA (15–288 ng/L) has also been found in human breast milk [10]. Zhang et al. (2021) found that the concentration of PFOA in drinking water sources ranges from 12–128 ng/L after the investigation of water sources of The Yangtze River in China [11]. In addition, the daily intake of PFOA for adults and children will be 0.54 and 0.82 (ng/kg bw/day), respectively [12]. There is no doubt that this is a great harm to the human body. The world’s largest PFC manufacturer, 3M of the United States, stopped the production of PFOS in 2003 because of the potential threat to the environment and living organisms [13].

As a persistent organic pollutant, PFOA causes damage to the atmosphere, soil, and water [14]. The fluoride ion in PFOA has high electronegativity, replacing the C–H bond to form a highly stable C–F bond, with a bond energy of up to 485 kJ/mol, and is extremely difficult to break down [15]. Therefore, PFOA has strong stability and is not easy to degrade. The refractory characteristics of PFOA have led to continuous enrichment in excess sludge, with the highest concentration of 4.78 mg/kg detected in the residual dry sludge of sewage treatment plants in central and eastern China, which affects the reuse and recovery of sludge [16]. The anaerobic digestion (AD) treatment technology with the characteristics of less sludge production and high organic matter removal and resource recovery attracts much attention [17]. However, anaerobic bacteria are easily inhibited by heavy metals, antibiotics, and other substances, and hinder the AD process [18,19]. Similarly, PFOA is highly toxic and can reduce microbial activity to a certain extent. Previous studies found that microbial abundance in activated sludge decreased under PFOA exposure [20]. Meanwhile, PFOA was found to inhibit transcription, amino acid transport, and metabolism of microorganisms [21]. In a 40-day semi-continuous experiment, the presence of PFOA significantly reduced the COD removal rate in the AD process [22]. Microbial community analysis showed that the diversity of microbial community decreased in the PFOA-containing groups, and the relative abundance of Proteobacteria decreased by 5.59%, thus reducing the ability of sludge to degrade organic matter. Wang et al. (2021) found that the presence of PFOA had a negative impact on acidification and methane production; 3–60 ug/g TS of PFOA reduced the methane production by 11.1–19.2% [23]. PFOA inhibited acidification and methanation process by reducing the relative abundance of acid-producing and methanogenic bacteria.

At present, many studies have been carried out on how to remove PFOA from wastewater. Yang et al. (2020) confirmed that PFOA removal by aerobic granular sludge was mainly by adsorption. The removal rate was 32.0–36.4%, but no aerobic biodegradation was observed [24]. Previous studies have suggested that there is no conclusive evidence that PFOA can be degraded under aerobic conditions [25]. In contrast, Huang et al. (2022) found that anaerobic bacteria could degrade PFOA and the removal rate of AD system reached 63% [26]. Meester et al. (2004) compared the aerobic and anaerobic biodegradation of PFOA, with the conclusion that the aerobic process almost did not degrade, and the anaerobic process almost completely removed PFOA [27]. In summary, the AD process can effectively remove PFOA, but the interaction between AD and PFOA still needs to be further studied.

This study evaluated the effect of PFOA on AD metabolism by analyzing the effects of different concentrations of PFOA on volatile fatty acid (VFAs), methane production, electron transfer system (ETS), and extracellular polymeric substances (EPS). By measuring the change in PFOA content in sludge and mixed liquid, the removal pathways of PFOA were analyzed. Furthermore, it provides technical reference and scientific basis for the application of the AD process in the treatment of PFOA-enriched wastes.

2. Materials and Methods

2.1. Substrate and Inoculum

The anaerobic digestion sludge (ADS) used in this study was taken from the continuous operational AnMBR reactor, and the initial sludge concentrations were 17.1 g/L (total solid, TS) and 10.9 g/L (volatile solid, VS). The chemical oxygen demand (COD) concentration was controlled to be 2500 ± 20 mg/L with glucose as the carbon source, and the optimal element ratio was approximately C:N:P = 100:5:1. No PFOA was found in the original ADS. The PFOA of analytical pure grade used in the study was purchased from Sigma-Aldrich (Saint Louis, MO, USA). PFOA reserve solution with a concentration of 0.012 g/L was prepared and stored in a 4 °C refrigerator for later use.

2.2. The Batch Test Design and Operation

A serum bottle with a total volume of 120 mL and a working volume of 100 mL was used as the AD reaction system (ADS 70 mL and substrate 30 mL). To simulate the enrichment of PFOA in different scenarios, six PFOA concentration gradients (0.0, 0.5, 1.0, 1.5, 2.0, and 3.0 mg/L) were set. The initial pH was adjusted to 7.0 ± 0.1 with 1 mol/L HCl or NaOH to eliminate the influence of pH on the experiment. After completion of each serum bottle unit, the samples were aerated with 98% nitrogen for 5 min to create an anaerobic environment.

2.3. Determination Indexes and Analysis Methods

TS, VS, total COD (TCOD), and soluble COD (SCOD) were determined according to the standard methods [28]. During the entire experiment, gas production was measured daily, and a gas chromatograph (Panna A60, JiangSu, China) was used to analyze the gas components. After centrifugation and dilution (10,000 rpm, 5 min), the VFAs produced in the reaction were determined by gas chromatography (Panna, A91 Plus, Shanghai, China). The electron transport capacity of the anaerobic sludge ETS was determined by the method of Maurines-Carboneill et al. (1998) [29]. The EPS extraction method was derived from Deng et al. (2014) [30], and polysaccharides (PN) and proteins (PS) were measured by a visible light spectrophotometer (Hach-DR3900, Loveland, CO, USA) to characterize the changes in EPS. The extraction and determination of PFOA were carried out by Wang et al. (2022), and the concentration of PFOA in water samples and sludge was accurately determined by adding an internal standard (50 ng/L PFOA) [31].

2.4. Data Analysis Method

The cumulative methane production in the reaction process based on methane-producing potential energy was fitted with the Gomperz model, and the fitting formula was based on Li et al. (2022) [32], as shown in Equation (1).

$$P {=\mathit{P}}_0·\mathit{exp}\left\{-\mathit{exp}\left[{\mathit{R}}_{\mathit{max}}·\frac{\mathit{e}}{{\mathit{P}}_0}{·\mathit{t}}_0- \mathit{t}+1\right]\right\}$$

(1)

where P is the actual methane production, mL; P₀ is the maximum methane production, mL; R_max is the maximum rate of methane production, mL/d; T₀ is the methane-producing delay time, d; T is methanogenesis duration, d; and e is the natural log.

The calculation formula of the AD specific methanogenic activity (SMA) is shown in Equation (2).

$$SMA=\frac{{\mathit{P}}_{\mathit{CH}4}}{{\mathit{VS}}_0}$$

(2)

where SMA is the specific methanogenic activity of AD, mL/g·VSS; P_CH4 is the cumulative methane production of each reactor with a concentration gradient of PFOA, mL; and VS₀ is the initial VS of sludge, g.

In this study, a logistic model was used to fit the inhibition rate of PFOA on methane production [33]. The calculation of the methane production inhibition rate is shown in Equation (3).

$$q_I=\frac{{\mathit{P}}_0-{ \mathit{P}}_{\mathit{e}}}{{\mathit{P}}_0}$$

(3)

where q_I is the inhibition rate of PFOA on AD methane production at each gradient level, %; P₀ is the actual methane production of the blank group, mL; and P_e is the methane production of each experimental group, mL.

Langmuir and Freundlich thermodynamic adsorption models were used to fit the process of AD sludge adsorption of PFOA, as shown in Equations (4) and (5) [34].

$$\mathrm{Langmuir} \mathrm{model}: \frac{C_e}{q_e}=\frac{1}{{\mathit{q}}_{\mathit{m}}\mathit{b}}+\frac{{\mathit{C}}_{\mathit{e}}}{{\mathit{q}}_{\mathit{m}}}$$

(4)

$$\mathrm{Freundlich} \mathrm{model}: ln{q}_e{=\mathit{lnK}}_{\mathit{F}}+\frac{1}{\mathit{n}}{\mathit{lnC}}_{\mathit{e}}$$

(5)

where C_e represents the concentration when adsorption reaches dynamic equilibrium, mg/L; Q_e stands for equilibrium adsorption capacity, mg/L; Q_m represents the maximum adsorption capacity of sludge, mg/g; K_F and b are two model constants; and 1/n is the heterogeneity factor.

The degradation rate of PFOA is calculated with Equation (6).

$$E_{PFOA}=\frac{{\mathit{PF}}_0-{ \mathit{PF}}_{\mathit{S}}-{ \mathit{PF}}_{\mathit{W}}}{{\mathit{PF}}_0}$$

(6)

where E_PFOA is the degradation rate of PFOA, %; PF₀ is the content of PFOA in each reactor, mg; PF_S is the content of PFOA in sludge, mg; and PF_W is the content of PFOA in supernatant, mg.

Since glucose is the only carbon source for COD and there is no hydrolysis process in glucose conversion, only acidification and methanation process efficiency are discussed in this study. The acidification efficiency and methanation efficiency of the reaction process are shown in Equations (7) and (8).

$$E_{Acid}=\frac{{\mathit{COD}}_{\mathit{VFAs}}-{ \mathit{COD}}_{\mathit{CH}4}}{{\mathit{COD}}_0}$$

(7)

$$E_{C{H}_4}=\frac{{\mathit{P}}_{{\mathit{CH}}_4}}{\mathit{P}}$$

(8)

where E_Acid and E_CH4 are the acidification efficiency and methanation efficiency, respectively, %; COD_VFAs and COD_CH4 are the corresponding values of VFA content and methane production in each stage, mg/L. COD₀ is the initial total COD in the reactor, mg/L. P_CH4 and P are the actual methane production and theoretical methane production, mL.

3. Results and Discussion

3.1. Effects of PFOA on Energy Recovery

As shown in Figure 1I, the AD system was rapidly and continuously inhibited by PFOA. Comparing the methane production in the blank group, the methane production in the PFOA group decreased by 7.6%, 14.3%, 15.5%, 18.6%, and 19.7%, respectively. Under the stress of PFOA concentration gradient (0.5, 1.0, 1.5, 2.0, and 3.0 mg/L), as shown in Figure 1II, with the increase in PFOA concentration, the inhibition effect on methanogenesis gradually increased, but the inhibition rate of methane production gradually approached the maximum value of 20% based on the fitting results of logistic model. Therefore, the AD system in the PFOA exposure condition has a certain tolerance to PFOA. From the actual methane-producing rate data in Figure 1III, the SMA of the PFOA group at a concentration of 3.0 mg/L decreased by 7.8 mL/d·gVSS, and the inhibition rate reached 22.3% compared with the blank group. The fitting results of the logistic model indicated that the inhibition rate of methane production in PFOA groups had the same trend as that of SMA. The results showed that PFOA in the system reduced the methane-producing capacity of sludge and the potential load capacity of the reactor. At the same time, AD can withstand high concentration of PFOA and has a certain tolerance to the toxicity of PFOA, which will not lead to system collapse.

Acidogens in the AD system degrade organic matter into propionic acid, butyric acid, acetic acid, and other organic acids, and methanogens further degrade acetic acid to methane [35]. Therefore, the determination of the content of various organic acids in the process of AD, combined with methane production, can be applied to explain the changes in acidogenesis and methanogenesis rate of the reaction system. The concentration of VFAs during the reaction is shown in Figure 2I. On the first day of reaction, the concentration of VFAs in PFOA groups was slightly lower than that in the blank group, indicating that the acidification rate of the AD process decreased under PFOA stress. From the second day of the experiment, the VFA concentration in the blank group was lower, indicating that the utilization rate of VFAs in the blank group was faster and the methane-producing efficiency was higher. In addition, it was found that the acidification efficiency (Figure 2II) and methanation efficiency (Figure 2III) in the PFOA groups were always lower than those in the blank group, and the trend was relatively stable. On the first day of the experiment, the acidification efficiency of the PFOA groups was only 3.39–4.52% lower than that of the blank group. During the entire process, the methanation efficiency and cumulative methane production trend were similar, but the methanation efficiency was 21% lower than that of the blank group at the highest PFOA level of 3.0 mg/L. Jiao et al. (2022) also demonstrated that PFOA inhibited the acidification and methanation stages of AD and had no significant effect on the hydrolysis stage [36]. These results indicated that the stress of PFOA affected the acidification and methanogenesis stages, and reduced the energy recovery efficiency.

3.2. Process and Performance of PFOA Removal in AD

The concentration of PFOA in the supernatant decreased rapidly in a short time and reached the lowest value within 24 h in all the groups (Figure 3I). In previous studies, it was found that the removal rate of PFOA in wastewater by aerobic activated sludge is 32–36.4%, and the main method of removal is the adsorption of PFOA by sludge. In order to explain the reasons for the decrease of PFOA concentration in the reactors, Langmuir and Freundlich thermodynamic adsorption models were used in this research to fit the adsorption process. As shown in

Table 1. After fitting Langmuir and Freundlich adsorption models, the R² of PFOA adsorbed by sludge and the saturated adsorption capacity of sludge to PFOA.

PFOA Concentration (mg/L)	0.5	1.0	1.5	2.0	3.0
Langmuir model R2	0.7926	0.6713	0.8221	0.8632	0.9079
Freundlich model R2	0.9789	0.9898	0.9858	0.9902	0.9796
Saturated adsorption capacity (ug/g·VSS)	38.5182	84.8912	122.1567	161.8340	239.5228

, the fitting correlation coefficient of the Freundlich model (R² > 0.97) was higher than that of the Langmuir model (R² > 0.67). The results indicated that the decrease in PFOA concentration was mainly caused by the multimolecular layer physical adsorption, and the fitting curves of PFOA absorbed by sludge within 24 h are shown in Figure 3II. The fitting results of the Freundlich model also showed that the saturated adsorption capacity of sludge to PFOA in each reactor (0.5, 1.0, 1.5, 2.0, and 3.0 mg/L PFOA) was 38.52, 84.89, 122.16, 161.83, and 239.52 ug/g·VSS. It is also found in Figure 3I that along with the AD reaction, PFOA adsorbed by sludge was desorbed again, increasing the concentration of PFOA in the supernatant. The results can be explained by the dynamic balance of adsorption and desorption in the AD system [37]. The adsorption process of PFOA by sludge can be divided into two stages. According to various physical and biochemical interactions of PFOA in the AD system, its removal pathways can be divided into two stages, as shown in Figure 3II. Stage 1 is the main process of rapid adsorption of PFOA by sludge within 24 h, and stage 2 is the sludge adsorption and desorption process. Some studies have shown that the enrichment of PFOA in sludge is related to temperature, pH, and hydrophilicity of the adsorbent. Among them, the pH of the solution changes the surface charge by affecting the functional groups in the sludge, zeta point position, and other factors and then affects the adsorption efficiency. In the study of Zhou et al. (2010), the adsorption capacity of activated sludge decreased with the pH increase of the solution. When the pH increased from 3 to 9.5, the adsorption capacity of PFOA decreased from 85% to 42% [38]. In this study, as shown in Figure 2I, VFAs were continuously degraded as the reaction progressed, and the pH in the system gradually recovered. Therefore, it can be speculated that the main reason for PFOA desorption can be attributed to the recovery of pH, so that the adsorbed PFOA was released again.

At the end of the reaction, the PFOA mass balance in the AD systems was analyzed, and the content of PFOA in solid and liquid phases of the system is shown in Figure 4. The results show that with the increase of PFOA concentration, the total removal rate of PFOA from the liquid phase did not fluctuate much, basically within the range of 60–70%. However, the mass balance results indicated that except for the PFOA in supernatant and sludge, 10.4–28.2% of PFOA disappeared from the AD process. It is reasonable to speculate that the AD process has a certain biodegradation of PFOA. To illustrate the performance of each concentration gradient of PFOA in the AD systems, the content and proportion of PFOA in each stage of the AD system are listed in

Table 2. The proportion and content of PFOA in supernatant and sludge in each reactor.

PFOA Concentration (mg/L)		0.5	1.0	1.5	2.0	3.0
PFOA in sludge	Proportion (%)	61.2	47.7	47.0	43.4	36.1
	Content (mg/L)	0.307	0.477	0.705	0.868	1.084
PFOA in supernatant	Proportion (%)	28.4	29.9	32.5	31.23	35.7
	Content (mg/L)	0.142	0.299	0.488	0.624	1.072
Others	Proportion (%)	10.4	22.4	20.5	25.37	28.2
	Content (mg/L)	0.053	0.223	0.307	0.507	0.843

. When the concentration of PFOA is 0.5 mg/L, the content of PFOA in the sludge phase is 0.31 mg/L, accounting for 61.2% of PFOA removal. However, at 3 mg/L, although the content of PFOA in sludge reaches 1.08 mg/L, the proportion of the total PFOA is only 36.1%. Combined with Figure 4 and Table 2, it was found that the higher the concentration of PFOA in the system, the stronger the biodegradation effect of AD system for PFOA. Therefore, the experimental results further illustrate the removal pathways of the AD system for different PFOA concentrations, that is, when the PFOA concentration is low, the removal pathway is mainly realized by adsorption, while when the PFOA concentration is high, the removal mainly depends on the combined action of biodegradation and sludge adsorption. Cao et al. (2022) showed that the average PFOA removal rate of anaerobic granular sludge was 52–55%, which was mainly caused by adsorption. However, it was found that PFOA could be degraded to C₇F₁₅ by acid decarboxylation [22]. Shahasavari et al. (2021) described the possible pathways of PFOA biodegradation [39]. The degradation process mainly involves p450 enzyme, benzoyl-CoA reductase (BCR), and other key enzymes that can promote PFOA degradation and release F from the C–F bond. Tiedt et al. (2017) showed that enzymatic defluorination represented by enyl-CoA hydrase can effectively solve the problem of high stability of C–F, and thus degrades PFOA [40]. Of course, the process and mechanism of AD degradation of PFOA need to be further confirmed by subsequent experiments.

3.3. Inhibition Mechanism of Anaerobic Digestion under PFOA Exposure

As mentioned previously, PFOA is toxic to microorganisms. In the AD system, to resist environmental toxicity, anaerobes secrete more EPS as a barrier to protect themselves from damage [41]. Furthermore, studies have shown that the ratio of PN/PS in EPS reflects the surface charge and hydrophilicity of sludge in the system [42]. In this study, the higher the concentration of PFOA exposed in the AD system, the more upwards the trend of PN/PS, indicating that anaerobic microorganisms further formed aggregates to resist the toxicity of the external environment [43]. The EPS content of each group after the experiment was measured, as shown in Figure 5. The ratio of PN/PS in PFOA groups increased from 4.93 to 6.95, which was increased by 14.67–45.82%. The results showed that proteins secreted by EPS played a leading role in the anti-toxicity of PFOA. The PN content in the EPS of the PFOA groups increased compared with that of the blank group, and it increased with increasing PFOA concentration. Proteins secreted by microbial cells promote sludge flocculation and agglomeration, and limit the conversion of organic matter to methane, resulting in further inhibition [44]. Therefore, PFOA inhibits AD methanogenesis not only through inhibiting microbial activity, but also through multiple effects of sludge toxic stress response. In addition, EPS in the AD system can effectively improve the adsorption capacity of microorganisms, promote the formation of microbial aggregates, and further improve the resilience of the AD system [45].

In the AD system, interspecific electron transfer from the acidogens to the methanogens is necessary for methane production. Under toxic environmental conditions, the response of microorganisms in the electron transfer process fluctuates greatly, which directly affects the methane production. At the end of the experiment, the ETS of sludge mixture was measured; as shown in Figure 6, the ETS of the blank group was 386.36 mg/(g·h). The PFOA groups were suppressed by a gradient of 22.06–37.38%, showing an overall downwards trend, and the results showed that PFOA had a direct negative effect on the interspecific electron transfer in the AD system. The electronic transfer is the basic process for organic matter degradation in AD. The decrease of ETS indicated that the electronic transfer ability between bacteria decreased under the stress of PFOA, thus inhibiting the growth of microorganisms [46]. The experimental results corresponded to the inhibition of methane recovery efficiency by PFOA, which mutually confirmed the inhibition of PFOA on sludge activity.

4. Conclusions

The effects of PFOA on energy recovery efficiency and microbial activity in an AD system as well as the removal process and mechanisms of PFOA were discussed in this study. The results showed that PFOA had a negative effect on the AD process. The maximum inhibition rate of PFOA on methane production was approximately 20%, indicating that AD has a certain tolerance to PFOA. At the same time, SMA of sludge under high concentration of PFOA was inhibited by 22.3%, EPS showed anti-toxic reaction, and ETS decreased by 22.06% to 37.38%. Therefore, the presence of PFOA will inhibit the activity of anaerobic bacteria and reduce energy recovery efficiency. During the AD treatment, approximately 64.2–71.6% of PFOA was removed. Sludge adsorption was the main pathway for PFOA removal. Meanwhile, 10.4–28.2% PFOA was missing in AD systems, probably due to biodegradation. PFOA has high biological resistance, and physical and biochemical removal of PFOA requires additional economic input. Combined with the advantages of AD energy recovery, rational use of AD to remove PFOA will be a good choice.