Particle size and co-presence of PFOA alter the transport of microplastics in saturated natural sediment

ABSTRACT

The transport of microplastics (MPs) with two different sizes in the absence and presence of perfluorooctanoic acid (PFOA) in sediment were systematically investigated in this study. Smaller size of MPs exhibited great transport mobility, while larger size of MPs significantly retained in sediment via electrostatic repulsion and straining mechanisms. MPs transport was insignificantly affected by influent pH values (4.7 and 6.0), likely due to the great buffer capacity of sediment. Effect of PFOA on MPs transport in sediment varied with MPs size and influent pH conditions. PFOA induced more negatively charged MPs surface and greater MPs transport at influent pH of 4.7, and this accelerating effect was more significant for MPs with greater size. In contrast, PFOA have little effect on the transport of both sized MPs at influent pH of 6.0. This work improved our understanding of MPs transport with the co-presence of PFOA in sediment environment.

KEYWORDS:

• Microplastics
• perfluorooctanoic acid
• transport
• particle size
• sediment

1 Introduction

Microplastics (MPs) are emerging environmental pollutants with diameters less than 5 mm, which have received increased attention worldwide. Owing to the wide application, low-recycling rates and poor waste management, a significant portion of the plastics would inevitably enter into the environment and ubiquitously present in various environment matrices, including surface water, sediment, soil and biotas[1–4]. Previous studies showed that MPs can transfer through the food chain and induce adverse impacts on the health of biota and human [5,6]. Thus, understanding the transport of MPs in aquifer medium was crucial to assess the environmental risk and develop remediation strategies for MPs in the environment.

To date, some studies have explored the transport of MPs in aquifer medium, and found that MPs transport was strongly affected by solution hydrochemical conditions (e.g. ionic strength, cation type, pH, and dissolved organic matter), flow conditions (e.g. flow rate and flow orientation), and physiochemical properties of aquifer media (e.g. clay mineral, porous medium particle size) [7–12]. For example, Dong et al. [7] found polyethylene glycol terephthalate (PET) MPs mobility was enhanced with decreasing electrolyte concentration, increasing pH, and increasing HA concentration in saturated quartz sand. Ling et al. [11] reported high flow rate, low salinity, high water saturation, and low temperature would facilitate the mobility of polystyrene nanoplastics (PSNPs) in quartz sand. Most of these studies employed well-defined porous media, often homogeneous quartz sand. However, the physiochemical properties of natural aquifer media, such as sediment, are far more complicated than well-defined quartz sand. Unfortunately, the transport of MPs in natural sediments, which was likely to be a final accumulation place and a long-term sink for MPs in the environment [3,13], were still largely unknown.

MPs are likely to interact with other substances such as organic pollutant, surfactants and nanoparticles present in natural environment, and thus their fate and transport in environment would be altered [14–19]. For instance, Hu et al. [14] showed that the presence of naphthalene decreased the mobility of PSNPs in the sand column, because of the charge-shielding effect. Zhao et al. [19] found that the presence of tetracycline slightly inhibited MPs mobility in K⁺ solutions, but facilitated it in Ca²⁺ solutions. Perfluorooctanoic acid (PFOA) is an emerging pollutant of particular concern that ubiquitously present in various environmental media, including sediment [20,21]. As an anionic surfactant, PFOA may act as a highly surface-active organic to influence the transport of MPs in the environment. To the best of our knowledge, only Rong et al. [22] studied the effect of PFOA on the transport of different electrically charged MPs in quartz sand, and found that PFOA inhibited negatively charged MPs transport, while enhanced positively charged MPs transport in quartz sand. However, the potential effect of PFOA on the transport of MPs in natural sediment, especially whether the effect of PFOA on MPs transport varied in response to MPs size, was still unclear.

The overarching objective of this work is to explore the transport of MPs and the potential effect of PFOA on MPs transport in natural sediment. Laboratory saturated column experiments were conducted to examine the transport behavior of MPs of different sizes with the absence and presence of PFOA under two pH conditions. The underlying mechanisms that govern the transport of MPs under different conditions were also proposed and discussed.

2 Materials and methods

2.1 MPs solution

Fluorescent polystyrene MPs suspensions (10 g/L) with two different diameters were purchased from Baseline ChromTech Research Centre (Tianjin, China), and were abbreviated as PS-MPs-1 (small diameter) and PS-MPs-2 (large diameter), respectively. PFOA (96% purity) was purchased from the Macklin Biochemical (Co., Ltd. Shanghai). MPs suspensions (10 mg/L) with the absence and presence of PFOA (0 and 10 mg/L) were prepared at two different pH values (pH = 4.7 and 6.0). These two pH values were well within the range of the reported groundwater pH values (~2.2 − 11.32) in the natural environment [23,24]. Zeta potentials and hydrodynamic diameters of MPs at two influent pH values were measured using a ZetaPALS (Brookhaven Instruments Corporation, NY, U.S.A.).

2.2 Transport media

The sediment collected at Taihu Lake (China) was air-dried, homogenized and sieved to a size range of 0.125–0.22 mm. Sediment has had a relatively high silt content (86.80%), and low sand (4.10%) and clay contents (9.00%). The organic content of sediment was determined to be very low (0.933%). The pH of sediment was determined to be 7.2. Quartz sand purchased from Unimin Corporation (MN, U.S.A.) was used as control porous media. The sand was sieved to a same size range of 0.125–0.22 mm with sediment. Prior to use, the sand was sequentially treated by tap water, 10% nitric acid (v:v), and deionized (DI) water to remove metal oxide and other impurities [25]. Zeta potentials of sediment and quartz sand under different experimental conditions were measured using the ZetaPALS.

2.3 Column experiments

Systematic column experiments were conducted using quartz sand and sediment as the transport medium (Table 1). Sediments were packed into columns at a 1:3 (m/m) sediment to quartz sand mass ratio according to Fisher-Power and Cheng [26], in order to maintain sufficient hydraulic conductivity and ensure particulates at a low level so as to not clog column stoppers. The sediment and quartz sand were mixed uniformly by shaking the dry materials before packing. All column experiments were performed in duplicate.

Table 1. Zeta potential and hydrodynamic diameter of MPs and porous media.

Material	CPFOA (mg/L)	Influent pH	Hydrodynamic diameter (μm)	Zeta potential (mV)
PS-MPs-1	0	4.7	0.13 ± 0.00	−16.10 ± 1.32
	10		0.13 ± 0.00	−18.73 ± 0.64
	0	6.0	0.13 ± 0.00	−21.20 ± 0.66
	10		0.13 ± 0.00	−22.20 ± 0.14
	0	7.2	0.13 ± 0.03	−25.20 ± 0.61
	10		0.13 ± 0.00	−27.52 ± 2.21
PS-MPs-2	0	4.7	3.20 ± 0.16	−0.48 ± 0.05
	10		2.99 ± 0.08	−7.81 ± 1.89
	0	6.0	1.47 ± 0.18	−4.13 ± 2.12
	10		1.37 ± 0.23	−6.00 ± 1.79
	0	7.2	1.40 ± 0.41	−11.68 ± 2.83
	10		1.32 ± 0.36	−13.46 ± 2.35
Sediment	0	4.7	NA	−15.90 ± 0.41
	10			−16.33 ± 1.06
	0	6.0	NA	−16.76 ± 0.53
	10			−16.80 ± 0.48
	0	7.2	NA	−23.23 ± 0.45
	10			−23.24 ± 0.50
Quartz sand	0	4.7	NA	−2.60 ± 1.24
	10			−2.54 ± 1.13
	0	6.0	NA	−4.80 ± 1.28
	10			−5.47 ± 1.06
	0	7.2	NA	−12.76 ± 1.26
	10			−12.37 ± 1.24

NA: not available.

Column experiments were conducted in polypropylene columns with length of 12 cm and inner diameter of 2.5 cm. Stainless steel screens of 100-mesh were used at both ends of the column to support the solid and distribute flow. A peristaltic pump (BT100-1F, Longer Pump, Hebei, China) was used to apply working solutions in an up-flow mode. The columns were uniformly dry packed and slowly saturated with DI water at a flow rate of 0.2 mL/min. The porosity of the saturated column was determined gravimetrically to be 0.38–0.39 for all columns. Nonreactive tracer tests were carried out with a solution of 25 mg/L KNO₃ to determine water flow characteristics and column performance (detailed in S1 and Figure S1). The tracer results showed that both the sand and sediment media were well-packed in the columns without notable signs of ‘wall effects’ or preferential flow paths.

After packing, the columns were pre-equilibrated with electrolyte solution (1.5 mM NaCl) at different influent pH values (4.7 and 6.0) for 12 h to achieve chemical and hydrodynamic stabilization. MPs suspensions with different sizes with the absence or presence of PFOA at desired influent pH values were then injected for three pore volumes (PVs) into columns at a constant flow rate of 0.5 mL/min before influent was switched back to 1.5 mM NaCl background solution at the same pH value. Effluent samples were collected every 6 minutes using a fraction collector (BS–100A, Shanghai Huxi analytical instrument co., LTD, China) and monitored for pH. Result in Figure S2 showed that sediment had a great buffer capacity, causing the effluent pH values stabilized at 7.21 ± 0.21 during the experiment. MPs concentration in effluents were determined by using a water fluorescence spectrometer (Cary eclipse fluorescence spectrophotometer, Agilent, U.S.A.). The standard curves of MPs were shown in Figure S3, and the results showed the concentration of MPs had a strong linear relationship with fluorescence intensity (R² >0.999).

2.4 Statistical treatment of results

Statistical treatment based on t-test was performed to determine the difference between the averages of the results obtained under different experimental conditions. Statistically significant difference was performed using the one-way analysis of variance (ANOVA) by the SPSS 25.0. Probabilities lower to 0.05 (p < 0.05) were considered as statistically significant.

3 Results and discussion

3.1 Hydrodynamic diameters and zeta potentials of MPs and transport media

The hydrodynamic diameters and zeta potentials of MPs and transport media under influent pH (4.7 and 6.0) and effluent pH (7.2) are illustrated in Table 1. The hydrodynamic diameters of PS-MPs-1 were stabilized as 0.13 ± 0.00 μm under all tested conditions, suggesting the negligible effects of pH and PFOA. Similarly, pH was found to have almost no significant impact on the aggregation of PSNPs in the pH range of 5–9 previously [27,28]. The presence of PFOA also had subtle effect on the hydrodynamic diameters of PS-MPs-2. With PFOA concentration increasing from 0 to 10 mg/L, the hydrodynamic diameters of PS-MPs-2 only slightly ranged from 2.99 ± 0.08 μm to 3.20 ± 0.16 μm at pH of 4.7, 1.37 ± 0.23 μm to 1.47 ± 0.18 μm at pH of 6.0, and 1.32 ± 0.36 μm to 1.40 ± 0.41 μm at pH = 7.2. However, the hydrodynamic diameters of PS-MPs-2 were strongly affected by pH values. As pH values increased from 4.7 to 7.2, the hydrodynamic diameters of PS-MPs-2 significantly decreased from 2.99 ± 0.08 ~ 3.20 ± 0.16 μm to 1.32 ± 0.36 ~ 1.40 ± 0.41 μm.

Zeta potentials of both sized MPs were all negatively charged under all tested experimental conditions. PS-MPs-2 (−13.46 ± 2.35 mV~−0.48 ± 0.05 mV) were less negatively charged than PS-MPs-1 (−27.52 ± 2.21 mV~−16.10 ± 1.32 mV), indicating smaller size of MPs responded to more negatively charged surface. This result was consistent with previous studies [29–31]. The presence of PFOA caused more negatively charged MPs surface, and these shifts were more significant for PS-MPs-2 at lower pH value. The presence of PFOA caused zeta potentials of PS-MPs-2 to decrease from −0.48 ± 0.05 mV to −7.81 ± 1.89 mV at pH of 4.7, but only slightly decrease from −4.13 ± 2.12 mV to −6.00 ± 1.79 mV at pH of 6.0. Previous studies have also reported anionic surfactant can increase negative charge of nanoparticles, MPs and colloids [22,32–35]. In addition, higher pH induced more negatively charged surface for both sized MPs, consistent with previous studies [29,36,37].

Sediment and quartz sand were negatively charged under all experimental conditions. Sediment (−23.24 ± 0.50 mV~−15.90 ± 0.41 mV) had more negatively charged surface than quartz sand (−12.76 ± 1.26 mV~−2.54 ± 1.13 mV). Sediment had a strong buffer capacity, and the effluent pH values in sediment columns can be buffered closer to the natural pH of sediment (~7.2) whether influent pH values were 4.7 or 6.0 (Figure S2). This thus caused the deprotonation of sediment and more negatively charged surface than quartz sand [38,39]. Besides, the strong buffer capacity of sediment may also reduce the impact of influent pH on the surface charge of sediment. Conversely, the surface charge of quartz sand was significantly affected by pH (pH = 4.7: −2.60 ± 1.24 mV~−2.54 ± 1.13 mV; pH = 7.2: −12.76 ± 1.26 mV~−12.37 ± 1.24 mV), consistent with previous studies [40]. The presence of PFOA had little effect on the surface charge of both sediment and quartz sand under all tested pH values. For example, zeta potentials of sediment and quartz sand only changed from −16.33 ± 1.06 mV to −15.90 ± 0.41 mV, and to −2.60 ± 1.24 mV to −2.54 ± 1.13 mV at pH of 4.7, respectively.

3.2 Transport behavior of individual MPs

The breakthrough curves (BTCs) of two sized MPs in sediment under different influent pH conditions (4.7 and 6.0) are presented in Figure 1a and b. BTCs are plotted as normalized effluent concentrations (C/C₀) versus PVs. The experimental conditions and result summary are presented in Table 2. PS-MPs-1 had a great transport mobility in sediment under both influent pH conditions, with effluent recoveries exceeding 93.59 ± 1.49% (Figure 1a). This was likely due to the negatively charged surface of both PS-MPs-1 and sediment, causing great electrostatic repulsion between them [15,32,33]. In contrast, a great reduction in the breakthrough concentration was observed for PS-MPs-2 with effluent recoveries of only 25.68 ± 2.99%~29.03 ± 0.52%, suggesting great retention of PS-MPs-2 during the transport in sediment (Figure 1b). Pore-throat straining has been documented to be an important mechanism responsible for the transport of colloids and nanoplastics in previous studies [11,41,42]. The ratio typically used to determine the possibility of straining is d_p/d₅₀, where d_p is the particle diameter and d₅₀ is median grain size. Previous studies reported that minimum values of d_p/d₅₀ ranged from 0.003 to 0.02 where straining has been observed in angular silica sand [41,43]. In this study, the d_p/d₅₀ values for PS-MPs-2 were determined to be 0.008 ~ 0.018 in sediment. This indicated a possibility of pore-throat straining, which might contribute to the great retention of PS-MPs-2 in sediment. In addition, the less negatively charged surface of PS-MPs-2 (−6.00 ± 1.79 mV~−0.48 ± 0.05 mV) than PS-MPs-1 (−22.20 ± 0.14 mV~−16.10 ± 1.32 mV) would lead to reduced electrostatic repulsion between PS-MPs-2 and sediment surface, and may also cause inhibited transport of PS-MPs-2. Similarly, Brewer et al. [44] suggested that larger plastic particles (110 nm) were less mobile than smaller plastic nanoparticles (50 nm) in a saturated soil. Shaniv et al. [12] found a higher elution value for the 50 nm PSNPs (90%) than that for 110 nm PSNPs (∼45%) during transport in soil.

Figure 1. Breakthrough curves of MPs in saturated sediment (a and b) and quartz sand (c and d) under two different pH conditions (pH = 4.7 and 6.0), IS = 1.5 mM NaCl. All data show the means of replicates (n = 2).

Table 2. The effluent recovery of MPs during transport in sediment and quartz sand.

Medium	MPs	CPFOA (mg/L)	Influent pH	Reffluent (%)
Sediment	PS-MPs-1	0	4.7	93.59 ± 1.49
		10		98.66 ± 0.82
		0	6.0	96.81 ± 0.86
		10		101.21 ± 2.58
	PS-MPs-2	0	4.7	25.68 ± 2.99
		10		37.29 ± 2.65
		0	6.0	29.03 ± 0.52
		10		31.59 ± 0.70
Quartz sand	PS-MPs-1	0	4.7	0.10 ± 0.02
			6.0	0.06 ± 0.04
	PS-MPs-2	0	4.7	0.51 ± 0.02
			6.0	0.45 ± 0.14

R_effluent: Effluent recovery.

The transport behaviors of both sized MPs in sediment were overall not affected by the tested influent pH conditions (pH = 4.7 and 6.0, Figure 1). With influent pH increasing from 4.7 to 6.0, no significant changes in effluent recoveries were observed for both sized MPs (p > 0.05, PS-MPs-1: 93.59 ± 1.49%~96.81 ± 0.86  %; PS-MPs-2: 25.68 ± 2.99%~29.03 ± 0.52%). As shown in Figure S2, the effluent pH can be buffered and remained approximately constant to 7.21 ± 0.21 under both influent pH conditions. The pore water in sediment would change along the column length from pH values of 4.7/6.0 to 7.2, which might strongly weaken the influence of influent pH on MPs transport in sediment. In order to test the speculation, column experiments regarding the transport of two sized MPs in quartz sand (without buffer capacity) under two influent pH conditions were also conducted. As shown in Figure 1c and d, almost all MPs were retained in quartz sand columns under both tested pH values, with effluent recoveries only to be 0.10 ± 0.02%~0.51 ± 0.02% (Table 2). The greater retention of MPs in quartz sand can probably be attributed to the less negatively charged sand surfaces and consequently less electrostatic repulsion between them. Similarly, Fisher-Power and Cheng. [26] reported that effluent pH was buffered closer to the natural pH of sediment, and can strongly affect the transport of nTiO₂ via electrostatic interaction. Wu et al. [45] also found the strong buffering capacities of soils, which can influence surface charges of both MPs and soil minerals to affect their retention and transport in soil. Thus, it can be concluded that the strong buffer capacity of sediment strongly weakened the influence of influent pH on the transport of MPs in sediment columns.

3.3 Effect of PFOA on the transport of MPs in sediment

As shown in Figure 2, the effect of PFOA on the transport of MPs in sediment varied with MPs size and influent pH. When influent pH was 4.7, the presence of PFOA caused increases in BTCs peak values and effluent recoveries for both sized MPs (Figure 2a and c). With PFOA concentration increasing from 0 to 10 mg/L, the peak values of BTCs and effluent recoveries slightly increased from 1.01 to 1.02 and 93.59 ± 1.49% to 98.66 ± 0.82% for PS-MPs-1, while obviously increased from 0.35 to 0.58 and 25.68 ± 2.99% to 37.29 ± 2.65% for PS-MPs-2 (p < 0.05, Table 2). This indicated that the presence of PFOA enhanced the transport of both sized MPs in sediment at influent pH value of 4.7, and this effect was more significant for MPs with greater size. PFOA is an anionic surfactant that comprise a hydrophobic fluorocarbon chain and a hydrophilic head-group. Previous studies documented that hydrophobic tail chains of surfactants can be attached to the hydrophobic surface of MPs via hydrophobic interaction, while hydrophilic head groups of surfactants are connected to the water phase [46]. MPs can be enveloped by PFOA, and the outwards hydrophilic end of PFOA would lead to more negatively charged MPs surface. As discussed in 3.1, the presence of PFOA caused more negatively charged surface for both sized MPs at influent pH of 4.7. Hence, the electrostatic repulsion between MPs and sediment can be increased with the presence of PFOA, which facilitated the transport of MPs. Similar effects of hydrocarbon surfactants on the transport of colloids and MPs were also reported in some previous studies [15,47]. In addition, previous studies suggested the co-presence of contaminants (e.g. tetracycline) can favor the aggregation of MPs particles [19]. MPs aggregates with greater average hydrodynamic diameter were more likely to be retained in porous media [17]. However, the presence of PFOA had negligible effect on the hydrodynamic diameter of both sized MPs in this study (Table 2), indicating that PFOA cannot induce MPs aggregation under tested conditions. This suggested that the enhanced transport of MPs with the presence of PFOA was not attributed to the unchanged MPs sizes. Analogously, Rong et al. [22] found that PFOA rarely affected the size of micro-polystyrene modified with negatively charged carboxyl groups, and thus cannot contribute to the decreased transport of CMPs transport.

Figure 2. Breakthrough curves of MPs of two different sizes in the absence and presence of PFOA (C_PFOA = 0 and 10 mg/L) under two pH conditions (pH = 4.7 and 6.0) in saturated sediment, IS = 1.5 mM NaCl. All data show the means of replicates (n = 2).

In comparison with the enhanced MPs transport induced by PFOA at influent pH of 4.7, PFOA had no effect on both sized MPs transport at influent pH of 6.0 (Figure 2 b and d). The effluent recoveries exceeded over 96.81 ± 0.86% for PS-MPs-1, and only ranged from 29.03 ± 0.52% to 31.59 ± 0.70% for PS-MPs-2 with the absence and presence of PFOA (p > 0.05, Table 2). This was consistent with the minimum changes in both zeta potentials and hydrodynamic diameters of both sized MPs and sediment between the absence and presence of PFOA at influent pH of 6.0 (detailed in 3.1). These led to almost unchanged electrostatic interaction between MPs and sediment, and then insignificant effect of PFOA on MPs transport at influent pH of 6.0.

4 Conclusions

The transport of MPs and the effect of PFOA on MPs transport in sediment were systematically investigated in this study. Results showed that the size of MPs significantly influence their transport in sediment. Larger size of MPs resulted in greater retention during transport in sediment via pore-throat straining and electrostatic attraction. The great buffer capacity of sediment strongly reduced the effects of influent pH on the transport of MPs in sediment. The presence of PFOA can facilitate the transport of MPs at influent pH of 4.7, but had no effect on MPs transport at influent pH of 6.0. These findings highlight the importance of considering MPs size and solution pH when assessing the transport potential of MPs with the co-presence of PFOA in sediment.

Acknowledgment

This work was partially supported by the National Natural Science Foundation of China (42007114), the National Natural Science Foundation of Jiangsu Province (BK20200817), and the Open Foundation of State Environmental Protection Key Laboratory of Soil Environmental Management and Pollution Control (MEESEPC202305).

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/26395940.2024.2308116.

Additional information

Funding

This work was supported by the National Natural Science Foundation of China [42007114]; Open Foundation of State Environmental Protection Key Laboratory of Soil Environmental Management and Pollution Control [MEESEPC202305]; National Natural Science Foundation of Jiangsu Province [BK20200817].