Environmental Pollution 281 (2021) 117025

Contents lists available at ScienceDirect

Environmental Pollution
journal homepage: www.elsevier.com/locate/envpol

Occurrence and abundance of poly- and perfluoroalkyl substances

(PFASs) on microplastics (MPs) in Pearl River Estuary (PRE) region:
Spatial and temporal variations*
Yao Cheng a, Lei Mai a, Xingwen Lu b, Zhe Li c, Ying Guo a, Da Chen a, Fei Wang a, *
a
School of Environment, Guangdong Key Laboratory of Environmental Pollution and Health, Jinan University, Guangzhou, 510632, China
b
School of Environmental Science and Engineering and Institute of Environmental Health and Pollution Control, Guangdong University of Technology,
Guangzhou, 510006, China
c
School of Engineering and Materials Science, Faculty of Science and Engineering, Queen Mary University of London, Mile End Road, London, E1 4NS, UK

a r t i c l e i n f o a b s t r a c t

Article history: Although the occurrence and distribution of various chemicals on microplastics (MPs) has been widely
Received 9 November 2020 studied, little was known about the concentrations of poly- and perfluoroalkyl substances (PFASs) on
Received in revised form MPs. In this study, MPs from eight rivers draining into Pearl River Estuary (PRE) region were collected
22 March 2021
and analyzed. Higher concentrations of PFASs on MPs (105e9.07
103 ng g1) were found in the drain
Accepted 24 March 2021
Available online 27 March 2021
outlets receiving wastewater from most urbanized cities with large population densities. On the other
hand, lower concentrations of PFASs on MPs (10.3e227.8 ng g1) were found in the drain outlets
receiving wastewater mostly from agricultural and forested areas. Specially, 8:2 disubstituted poly-
Keywords:
Microplastics
fluoroalkyl phosphates (8:2 diPAP) was detected with the highest frequency, in 92.5% of the samples.
PFAS Furthermore, a positive Spearman correlation was found between 6:2 disubstituted polyfluoroalkyl
Estuary region phosphates (6:2 diPAP) and perfluorotetradecanoic acid (PFTeDA) (rs ¼ 0.621, p ¼ 0.012), indicating they
Spatial variations might share similar sources. PFASs on MPs were found to vary significantly with different seasons. Higher
Temporal variations concentrations of PFASs on MPs were found in dry seasons, while lower concentrations were observed in
wet seasons. The results of this study should be important for the understanding of PFAS occurrence and
distribution on MPs and the partitioning mechanism of PFASs on MPs in estuary systems.
© 2021 Elsevier Ltd. All rights reserved.

1. Introduction 2011; Cai et al., 2017; Jambeck et al., 2015; Morritt et al., 2014;
Van Cauwenberghe et al., 2013)) and causing serious environ-
The annual global production of plastics has consistently mental and ecological problems. Microplastics mainly come from
increased in recent years, surpassing a mark of 300 million tons a primary sources (for example, facial cleaners, cosmetic prepara-
year (Andrady, 2017; Jambeck et al., 2015). Owing to their unique tions, scrubbers in air-blasting, and waste water from WWTP) and
properties including light-weight, versatility, strength and dura- secondary sources (degraded from larger plastics fragments as a
bility, plastics are widely used in our daily life, although they have result of chemical or physical processes) (Wang et al., 2016).
also led to a range of environmental problems (Andrady, 2011; The Pearl River Delta (PRD) in China is one of the most urban-
Cozar et al., 2014). Plastic particles with size less than 5 mm (also ized and industrialized regions in the world with an area of
referred to as “microplastics” (MPs)), in particular, is ubiquitously 453,700 km2 and over 200 million in population (Cheung and Fok,
distributed in the global environment (such as terrestrial environ- 2017). It was reported that 5.8 million tons of raw plastic materials
ments, marine environments, sediment systems, fresh water en- were utilized in Guangdong province (the location of PRD) in 2015
vironments and even atmospheric fallout processes (Andrady, (Cheung et al., 2018). The extensive use of plastics in this area
represents a substantial potential source of plastics discharged into
the Pearl River (Fan et al., 2019). It was reported that the majority
source of MPs in the ocean was derived from the use of plastics on
land via inland waterways, wastewater outflows and transport by
*
This paper has been recommended for acceptance by Baoshan Xing. wind or tides (Jambeck et al., 2015). Some studies show that the MP
* Corresponding author.
levels in inland water are higher than those in the marine
E-mail address: wf1984@jnu.edu.cn (F. Wang).

https://doi.org/10.1016/j.envpol.2021.117025
0269-7491/© 2021 Elsevier Ltd. All rights reserved.
Y. Cheng, L. Mai, X. Lu et al. Environmental Pollution 281 (2021) 117025

environment (Cheung and Fok, 2017; Zhao et al., 2014a). Therefore, 2.2. Field sampling
understanding of MP levels in estuary areas (such as: PRD area) is of
particular importance to specify their contributions of MPs to the All MP samples were collected from the water surfaces of eight
South China Sea. major river drain outlets of PRD (HM: Humen; JM: Jiaomen; HQ:
Since the size of MPs is typically only a few millimeters, they are Hongqimen; HE: Hengmen; MD: Modaomen; JT: Jitimen; HT:
therefore difficult to identify, making them easily ingested by Hutiaomen; YM: Yamen) (SI Fig. S1) during five discrete sampling
aquatic animals. Furthermore, MPs have larger sorption capacities events on 16-Jan, 19-Apr, 15-Jun, 11-Aug, and 8-Nov in 2018 (Wet
for hydrophobic organic chemicals (HOCs) than large plastics season: April to August; Dry season: September to March). For each
owing to their comparatively larger specific surface areas (Lee et al., sampling events, three samples were collected within 1 h during
2014). Consequently, MPs laden with chemicals may create risks via lower tide time. Samples of surface waters from eight major river
food-chain to aquatic biota and the ecological system (Andrady, drain outlets of PRD were collected on 8-Nov in 2018. The open
2011). The concentration of different HOCs on MPs in aquatic en- mouth (1.0 m height
0.5 m width) of manta trawl net was made of
vironments varies significantly. For example, the concentrations of silk, similar to that used for MP captures in previous study (Zhang
polybrominated diphenyl ethers (PBDEs), polycyclic aromatic hy- et al., 2017b). The manta trawl used for this study consists of a mesh
drocarbons (PAHs), bisphenol A (BPA), Polychlorinated biphenyls bag (0.33 mm) with a length of 3 m and a removable cod-end with
(PCBs), dichloro-diphenyl-trichloroethane and its metabolites the same mesh (Mai et al., 2018). The volume of the filtered water
(DDTs), and alkylphenols (OPs) on MPs from the open ocean and flow was measured by a digital flow meter (Hydro-Bios, Germany),
remote and urban beaches have been reported (Hirai et al., 2011). which was tied in the middle of the open-mouth of the manta trawl
Per- and polyfluoroalkyl substances (PFASs) are an emerging each time. The trawl was placed at the end of the sampling boat
pollutant containing the CnF2nþ1 group (Poothong et al., 2019; with a speed of approximately 2 knots for during the sampling
Wang et al., 2018). As an emerging contaminant that is widely time. To obtain enough MPs each time, the sampling time was
distributed in the environment, PFASs have been extensively controlled to be around 15 min. The detailed sampling conditions
studied about their environmental distributions and adverse effects are provided in SI Table S2. After each sampling activity, all the
on the environment and human health. Furthermore, PFAS levels in plastic-like debris captured by the trawl were transferred to a 1-L
the PRD areas are reported to be very high, as PFASs are not strictly brown bottle and the trawl was rinsed with local seawater to
regulated in China (Zhao et al., 2014b). Since PFASs affiliated with reduce cross contamination. For quality assurance and quality
plastics can be transported worldwide by currents and winds, control purposes, field blanks were dosed with commercially
PFASs laden on plastics may have adverse comprehensive effects on available MP particles (polyethylene was purchased from Good-
marine animals (Aliabad et al., 2019; Faure et al., 2015; Vaughan fellow Company (Cambridge, London)). Before each sampling ac-
et al., 2017; Veerasingam et al., 2016). However, little information tivity, the manta trawl was thoroughly rinsed with distilled water
was reported on the occurrence of PFASs on MPs (Liu et al., 2015). to remove possible contamination. After each sampling event, the
Therefore, it is important to investigate the distribution of PFASs on samples were transported to laboratories and stored at 4  C prior to
MPs in the PRD areas, in order to understand the fate and transport analysis.
of MPs and PFASs on MPs between inland water and marine envi-
ronments (Zhang et al., 2017a). 2.3. Sample pretreatment and polymer identification
In the present study, we have documented the occurrence and
composition of PFASs on MPs in the PRD region. The main aim of According to Mai et al.‘s methods with a minor modification,
this study is 1) to document the occurrence and distribution of samples from each bottles were first immersed in saturated sodium
PFASs on MPs in the surface water of eight PRD river drain outlets chloride solution (~1.2 g cm3), followed by floating the debris in
and 2) to reveal the spatial and temporal distribution characteris- 30% H2O2 overnight (approximately 12 h) under room temperature
tics of PFASs on MPs in the PRD region. (Mai et al., 2018). It has been reported that natural organic matter
can be sufficiently removed by H2O2 treatment without affecting
the physical shape of the target plastic particles (Liebezeit and
2. Materials and methods Dubaish, 2012). After processing, the samples were filtered
through 50 meshes’ stainless steel sieves (equivalent to pore sizes
2.1. Chemicals and reagents of 0.33 mm), and particles larger than 5 mm were removed and not
analyzed in this experiment. Any debris remaining on the surface of
Details of the 21 target PFASs standards (including C4-14 per- stainless steel sieves was collected and the visible plastics were
fluorinated carboxylic acids (PFCAs), C4, C6-8, C10 perfluorinated picked out by pre-cleaned tweezers with aid of a 40
optical mi-
sulfonic acids (PFSAs), C6, C8, C10 perfluoroalkyl phosphate croscope. Furthermore, any suspected MP debris was verified by
(PFPAs), and 6:2, 8:2 disubstituted polyfluoroalkyl phosphates FTIR analysis. Then the collected MPs were wrapped in aluminum
(diPAPs)) used in this study are shown in Table S1 of Supporting foil for freeze-drying. To identify polymer compositions of the
Information (SI). All PFAS standards were purchased from samples, an Attenuated total reflectance (ATR) Fourier transform
Wellington Laboratories Inc. (Guelph, ON, Canada). Mass labeled infrared spectrometer (FT-IR) (Shimadzu model 8300) was
standards (13C2-PFHxA, 13C2-PFOA, 13C2-PFDA (13C2-perfluoroo-n- employed to determine the predominant MP types. 8e10 samples
decanoic acid), 13C4-PFOS (13C4-perfluorooctane sulphonate), 13C4- from all plastics type/size categories were randomly taken for the
8:2 diPAP, 13C8-PFOA, and 13C8-PFOS) were also obtained from FTIR examination and each FTIR spectra of samples was compared
Wellington Laboratories Inc. (Guelph, ON, Canada). Methanol against standard reference polymer spectra from ATR library (SI
(HPLC-grade) was purchased from Fisher Scientific (Pittsburgh, PA). Table S3).
Ammonium acetate, ammonia solution (NH4OH, 25%), Sodium
chloride (NaCl) and Hydrogen peroxide solution (H2O2) were pro- 2.4. Sample extraction
vided by Sigma-Aldrich (ST. Louis, MO). Water was prepared with a
Milli-Q Advantage A10 system (Millipore, USA). Oasis WAX (3 cc, 60 The dried plastics were weighed by an analytical balance and
mg/6 cc, 150 mg) cartridges were purchased from Waters Co. transferred into glass vials and spiked with 5 ng of surrogate
(Milford, Ireland). standards (i.e., 13C2- PFHxA, 13C2-PFOA, 13C2-PFDA, 13C4-PFOS, and
2
Y. Cheng, L. Mai, X. Lu et al. Environmental Pollution 281 (2021) 117025

13
C4-8:2 diPAP) for each sample, then extracted consecutively 3 and cotton lab coats were worn during the entire sampling process
times for 15 min with 1% NH4OH in methanol in a 60  C sonication and experimental procedures in the laboratory analysis. Vessels
bath. The solvent and concentrated extracts were then combined and tools were rinsed three times with methanol until dryness and
under gentle nitrogen flow in a 35  C bath with a rotary evaporator wrapped with aluminum foil (PFAS-free) prior to use. No poly-
near dryness. Oasis WAX was conditioned with 4 mL 0.1% NH4OH in tetrafluoroethylene (PTFE) or other fluoropolymer materials was
methanol, 4 mL methanol and 4 mL ultrapure water. The concen- used during the sample collection, transportation and pretreat-
trates were rinsed 3 times with 5% methanol in water and loaded on ment. All solvents used in the experiment were HPLC grade; all the
cartridges at a flow rate of 1e2 drops per second. The cartridges glass vials were soaked in the alkaline overnight and baked in the
were washed with 4 mL of 25 mM ammonium acetate aqueous 500  C for 4 h to remove possible leftover compounds. Field blanks
solution and dried by a vacuum pump to achieve a better recovery, (n ¼ 8), procedural blanks (n ¼ 8) and matrix-spiked samples
and the dried cartridges were eluted with 4 mL methanol and 4 mL (n ¼ 3) of six different plastic particles (SI Fig. S3 and Fig. S4) were
0.1% NH4OH in methanol. The elutes were concentrated near dry- performed during MP extraction and possible background
ness and rinsed 3 times with methanol and transferred into sample contamination was monitored. Each field blank was prepared by
vials before further concentration to ~0.5 mL in vials. Internal placing 1 L of distilled water in a glass bottle and taken to the
standard (13C8-PFOA and 13C8-PFOS; 5 ng each) consisting of 1 mL of sampling site. No MP particles were found in the field blanks, only
methanol was added into each extract, followed by filtration few textile fibers were occasionally identified in the procedural
through a 0.2 mm Whatman inorganic membrane filter (Maid- blanks and excluded from statistical analysis. To evaluate the ac-
stone, UK) into sample vials prior to instrumental analysis. And the curacy and precision for target PFASs, matrix-spiked samples
detailed method of water samples was given in the SI Text S1. composed of three different spike levels of native PFAS homologues
were added to the MP samples. Procedural blanks were prepared
2.5. Instrument analysis for each sampling site, whose values were below the limit of
quantification (LOQ) except for PFOS and PFOA, which may be
Concentrations of PFASs were determined with liquid chroma- attributed to containers being contaminated by the process of
tography tandem mass spectrometry (HPLC-MS/MS) with an ultra- sampling, transportation and pretreatment (SI Table S5). And sol-
performance LC system (Shimadzu Corporation LC-30AD, Japan), vent blanks were run after 7e10 samples to evaluate the possible
equipped with a 50
21 mm Waters BEH C18 column (1.7 mm carryovers of the target chemicals between samples. Each con-
particle size), and coupled to quadrupole tandem mass spectrom- centration of samples was adjusted by the values of procedural
eters (AB Sciex API 5500 CA USA). The injection volume was blanks but not corrected by surrogate standard recoveries. The
controlled at 10 mL of each aliquot. The column temperature was mean matrix spike recoveries of PFASs were in the range of 55 ± 2%
maintained at 45  C, polyetheretherketone (PEEK) tubes was used to 279 ± 9% (SI Table S6). The recovery experiments on the potential
to replace PTFE and fluoropolymer materials in the instruments to loss of PFASs on MPs during MP washing process were performed
avoid possible sorption and leaching of PFASs. A gradient mobile and shown in Text S3. The results showed that the washing pro-
phase of 2 mM ammonium acetate in methanol (B) and Milli-Q cedure to remove organic debris from MPs had limited effects on
water (A) was used for the LC system. At a flow rate of 0.3 mL/ PFBA, PFOA, and PFOS on MPs. (SI Fig. S5). The LOD and LOQ were
min, the dual mobile-phase gradient began at 25% B; was held calculated at a signal to noise ratio (S/N) of 3 and 10, respectively.
constant for 1.5 min; changed to 50% B by 4 min, 75% B by 8 min, The concentration below the limit of detection (LOD) was divided
and 100% B by 10 min; remained constant until 13 min; returned to by square root of two. More details about quality assurance and
the original conditions by 14 min and then equilibrated for 4.1 min. quality control can be seen in the SI Test S2.
The total run time of each injection was 18.1 min. Data was
collected in full scan mode, and multiple reaction monitoring 3. Results
(MRM) mode with negative electrospray (ES) ionization was used
in this study. Ions were detected in selective ion monitoring mode 3.1. Occurrence and types of MPs in surface water
according to the m/z reported in SI Table S4. When possible, mul-
tiple daughter ions were monitored, but quantification was based The spatial distribution of MPs in surface water of the PRD re-
on a single product ion. And individual chromatograms of the gion was shown in Table 1, MPs were identified in all samples (3365
target PFASs was shown in SI Fig. S2. samples in total) ranging from 0.003 to 2.09 items m3 (average
0.106 items m3) (per volume of river water) or 0.001e5.68 mg m3
2.6. Statistical analysis (average 0.559 mg m3) (per gram of MPs). However, it was noted
that the greatest level of MPs (1547 particles, 2.09 items m3,
Statistical analysis was performed using the SPSS statistics 24.0 approximately two orders of magnitude higher than the average
software. In order to examine normality of our data, the Normality concentration) was found at the sampling site of HE in 16-Jan.
tests (ShapiroeWilk) were performed on possible normality prior Interestingly, fiber accounted for 94.2% of the total MPs. Such high
to statistical analyses. Moreover, normal Q-Q plot was adopted to level of fiber MPs might come from dropped fishing nets or ropes,
further verify our results. As a consequence, normal distribution as large proportions of fiber-shaped MPs were in a cluster state and
can’t be found in present data. Spearman’s correlation and non- fishing activities were relatively frequent before sample collection.
parametric analysis were used to examine possible correlations As a special case, this unusual concentration of MPs at HE (16-Jan)
among non-normally distributed data eventually. Kruskal-Wallis H was taken as an outlier and was omitted for the further calculation
test (pairwise comparisons) was performed to analyze the spatial in the following study. Concentrations of MPs in all samples range
and temporal distribution patterns of PFAS on MPs. In all cases, from 0.003 to 0.2 items m3 (average 0.028 items m3), while
p < 0.05 was recognized as statistically significant. relative higher concentration was found in, HE (samples from 16-
Jan and 19-Apr) and HM (samples from 19-Apr).
2.7. Quality assurance and quality control The spatial distribution characteristics of MPs varied greatly
among different sampling sites (Table 1). The MP concentrations
For the consideration on potential contamination of MPs from from eastern outlets (HM and HE except for HQ and JM) were
cross-contamination and atmospheric deposition, nitrile gloves generally higher than the western outlets (MD, JT, HT and YM)
3
Y. Cheng, L. Mai, X. Lu et al. Environmental Pollution 281 (2021) 117025

Table 1
Detected items and mass concentrations of microplastics in surface waters of the eight major riverine drain outlets in the pearl river delta region, China.

Concentration of MPs outlet 16-Jan 19-Apr 15-Jun 11-Aug 9-Nov

Number concentration (items m3) HM 0.041 0.534 0.063 0.020 0.037

JM 0.019 0.017 0.021 0.057 0.005
HQ 0.006 0.007 0.014 0.009 0.003
HE 2.091 0.584 0.017 0.031 0.014
MD 0.032 0.020 0.200 0.068 0.004
JT 0.004 0.003 0.050 0.009 0.006
HT 0.003 0.004 0.015 0.007 0.012
YM 0.016 0.014 0.093 0.064 0.018
Mass concentration (mg m3) HM 0.064 3.032 0.891 0.140 0.179
JM 0.224 0.120 0.201 0.492 0.011
HQ 0.001 0.008 0.039 0.019 0.001
HE 1.658 5.678 0.315 0.479 0.150
MD 0.103 0.028 1.714 4.741 0.005
JT 0.125 0.063 0.110 0.052 0.097
HT 0.078 0.070 0.071 0.031 0.103
YM 0.154 0.019 0.703 0.119 0.255

(Table 1). The number and mass concentration of MPs varied

greatly with the different months throughout a year (0.015 items
m3 or 0.3 mg m3 in January, 0.15 items m3 or 1.13 mg m3 in
April, 0.059 items m3 or 0.51 mg m3 in June, 0.033 items m3 or
0.76 mg m3 in August and 0.012 items m3 or 0.1 mg m3 in
November) (Table 1).
The characterization of MPs used in this study showed that
polypropylene (PP) (43.1%) and Polyethylene (PE) (39.1%) are two
dominant plastic types for MPs, followed by Polystyrene (PS)
(10.6%) and other polymer materials accounting for minor portion
(Fig. 1and SI Table S3), According to their morphology properties,
the guidance developed by Cheung et al. (2016) was adopted to
classify MPs (Cheung et al., 2016), which were categorized as
fragment (21.2%), fiber (53.1%), film (13.1%), foam (11.4%) and pellet
(1.27%) (Fig. 2). Overall, fiber accounts for the highest percentage of
the total MPs, which also one strong evidence of MP sources from
fishing nets or ropes.

3.2. Occurrence and composition of PFASs on MPs in surface water

Fig. 2. Relative abundance of MP with different shapes sampled in eight major drain
The concentration and detection frequency of PFASs on MPs
outlets of the PRD region.
were shown in Table 2. Twenty-one individual PFAS were quanti-
fied in collected MP samples. The detailed concentrations of PFASs
on MPs were provided in SI Table S7. The priority detection fre- Table 2
quencies of PFASs were in the order of 8:2 diPAP (92.5%), PFBA Concentrations (ng g1 dw) and detection frequencies (df) of PFASs on MPs.
P (N ¼ 40).
(82.5%), PFTeDA (80.0%), and PFHxS (77.5%). In general, the PFAS
1
concentration ranges from 3.11 ng g (HM in 11-Aug) to Compound Min max average median df (%)

FBA <LOQ 143.1 20.3 8.17 82.5

PFPeA <LOQ 3.39 0.39 <LOQ 45.0
PFHxA <LOQ 1.84 0.05 <LOQ 2.50
PFHpA <LOQ 18.3 1.63 <LOQ 42.5
PFOA <LOQ 1250 89.8 3.05 65.0
PFNA <LOQ 46.4 2.25 <LOQ 37.5
PFDA <LOQ 22.7 0.87 <LOQ 40.0
PFUdA <LOQ 0.96 0.06 <LOQ 32.5
PFDoA <LOQ 16.0 0.59 0.03 70.0
PFTrDA <LOQ 36.4 1.34 0.06 62.5
PFTeDA <LOQ 11.3 0.67 0.10 80.0
PFBS <LOQ 57.0 2.02 0.28 60.0
PFHxS <LOQ 354.2 15.4 1.80 77.5
PFHpS <LOQ 148.3 8.34 <LOQ 47.5
PFOS <LOQ 8238.2 449.1 56.44 70.0
PFDS <LOQ 10.9 0.57 0.02 62.5
FHxPA <LOQ 0.20 0.01 <LOQ 35.0
PFOPA <LOQ 2.73 0.30 0.11 75.0
PFDPA <LOQ 80.0 4.15 0.70 72.5
6:2diPAP <LOQ 83.9 3.72 0.45 60.0
8:2diPAP <LOQ 388.2 14.5 0.92 92.5
Fig. 1. Relative abundance of polymer type with different categories sampled in eight
major drain outlets of the PRD region. <LOQ: below the limit of quantification.

4
Y. Cheng, L. Mai, X. Lu et al. Environmental Pollution 281 (2021) 117025

P
Fig. 3. Spatial distribution of PFASs on MPs sampled in eight major drain outlets of the PRD region: Humen (HM); Jiaomen (JM); Hongqimen (HQ); Hengmen (HE); Modaomen
P
(MD); Jitimen (JT); Hutiaomen (HT) and Yamen (YM). All plotted columns are average values of PFASs concentrations on MPs collected from different months at the same site.

9.07
103 ng g1 (HQ in 16-Jan) with an average concentration of

616 ng g1 over the eight drain outlets. The individual PFASs con-
centrations were in the range of <LOQ to 8.24
103 ng g1 with an
average concentration of 29.3 ng g1.

3.3. Spatial and temporal variations of PFASs on MPs

The spatial variations of PFASs on MPs were shown in Fig. 3. The

P
PFAS concentrations varied greatly among eastern and western
outlets of the PRD region. The highest PFAS concentrations were
detected in the samples collected from HQ (in a range of
105e9.07
103 ng g1 with an average concentration of
2.31
103 ng g1) (Fig. 4). Temporal variations of PFASs on MPs
sampled in eight major rivers drain outlets were shown in Fig. 5.
P
Generally, higher PFAS concentrations were found in January and
P
November, whereas lower PFAS concentrations were found in
August. Fig. 6 shows the composition of PFASs on MPs, PFOS and
P
PFOA accounting for 67.8%e91.8% of PFAS on average. Individual
PFAS concentrations in surface water was provided in SI Table S8.
Fig. 4. The PFAS concentrations on MPs sampled in eight major drain outlets of the
P
PRD region. All plotted columns are average values of PFASs concentrations on MPs
4. Discussions collected from different months at the same site. Other PFASs (at low concentrations)
are categorized as “others”, “others” are shown in a small chart in Fig. 4.
4.1. Occurrence and types of MPs in surface water
urbanized cities with large population densities (Fig. 7), such as
The abundance of MPs in surface water are generally compa-
Guangzhou (14.5 millions) and Dongguan (8.3 millions) (Bureau,
rable to the findings of Mai et al. (2019), who reported that the
2001), while the western outlets receive wastewater from agri-
levels of MPs were in the range of 0.005e0.7 particles m3 and
cultural and forested areas with smaller population densities (SI
0.004e1.28 mg m3 for the number and mass concentrations in the
Fig. S6). Thus, excessively use of plastics in the big cities may have
eight major river outlets of the PRD (Mai et al., 2019). The highest
contributed to a larger amount of wastewater with plastic debris
abundance of MPs was found at HE, which may be related to high
entering eastern outlets. Meanwhile, wastewater from agricultural
proportion of density discharge of wastewater from upstream of
and forested areas contains a smaller number of plastic debris
West River.
entering western outlets. It was clear that HM had the greatest
The eastern outlets receive wastewaters from the most
5
Y. Cheng, L. Mai, X. Lu et al. Environmental Pollution 281 (2021) 117025

PRE regions.
In this study, PP (43.1%) and PE (39.1%) accounted for predom-
inant proportion in MP types, similar result was also found in Fei-
laixia reservoir in North River in PRD region (Tan et al., 2019),
where PP and PE were major compositions of MPs. The relative
abundance of MPs with different morphologies was shown in (SI
Fig. S2).

4.2. Occurrence and abundance of PFASs on MPs in surface water

In this study, PFOS and PFOA accounted for a relative higher

proportion. The predominance profile of PFOS and PFOA found in
surface water and sediment in the Pearl River Delta region, indi-
cating that the source (surface water and sediment) of PFASs is
essential for their occurrence in MPs (Chen et al., 2019a; 2019b).
According to one estimation study, textile treatment was recog-
nized as the largest source of PFOS contamination, accounting for
P
Fig. 5. Seasonal variations of PFASs on MPs sampled in eight major river drain
P
over 50% contribution in surrounding environment in this region
outlets of the PRD region. Orange chart bars represent PFASs in January; Pink chart (Xie et al., 2013b). By the end of 2017, the number of textile treat-
P P
bars represent PFASs in April; Yellow chart bars represent PFASs in June; Wheat
P P ments in Guangzhou (n ¼ 157) and Dongguan (n ¼ 136) accounted
chart bars represent PFASs in August, and Green chart bars represent PFASs in
August. (For interpretation of the references to colour in this figure legend, the reader for nearly one fifth of that in Guangdong province (n ¼ 1464)
is referred to the Web version of this article.) (Bureau, 2001). Guangdong province, the most developed region
specializing in manufacture, with 93% of the total PFOS amounts
from textile treatment and metal plating industries and ranking
second in the PFOS emissions in China. DiPAPs, as an important
class of polyperfluoroalkyl phosphate esters (mono, di, and tri-
PAPs), are precursors of PFCAs and widely used as coating mate-
rials. A positive Spearman correlation was found between 6:2 diPAP
and PFTeDA (rs ¼ 0.621, p ¼ 0.012), indicating that they are likely to
co-occur on MPs entering the estuary systems. To our knowledge,
this is the first study reporting the quantification of PAPs on MPs
collected from estuarine environment. Many studies have shown
that PAPs could be transformed to be PFASs in the body of rodents
(D’Eon J and Mabury, 2011). The finding of PAPs on MPs in estuarine
environments may suggest a possible indirect exposure routes of
PFASs towards aquatic animals.

4.3. Spatial variations of PFASs on MPs

Fig. 6. Composition profile of PFASs on MPs sampled in eight major drain outlets of the Spatially, the average MP concentrations varied greatly among
P
PRD region. All plotted columns are average values of PFASs concentrations on MPs sampling sites (Fig. 4). Higher concentrations predominance profile
collected from different months at the same site. Other PFASs (at low concentrations) of PFOS occurred at the four eastern outlets (HM, JM, HQ, and HE)
are categorized as “others” and refer to SI Table S7 for complete raw data.
than at all western outlets (JT, HT, and YM) except for JT, which has
the smallest water discharge among all outlets. The highest PFAS
concentrations were detected in the samples collected from HQ,
abundance of MP, as HM receives the flows from wastewater pro- while the lowest concentration (10.3e228 ng g1) of PFASs was
duced from Dongguan and Guangzhou. detected in MPs collected from MD. A Kruskal-Wallis H-test sug-
The average concentration of MPs sampled in the wet season gested significant greater concentrations of PFASs on MPs in HQ
(0.08 items m3 or 0.8 mg m3) was slightly higher than the than in MD (H ¼ 40; p ¼ 0.038), which may be attributed to HQ
samples in the dry season (0.014 items m3 or 0.2 mg m3), which receives the inflow of Pearl River and East River. As East River
may be attributed to the different water discharge amounts and combines three sub-tributaries running through Dongguan city,
hydrodynamic conditions which vary with the seasons (SI lots of industries (e.g., clothing, electronics, furniture
Table S9). The relatively higher number abundance was found in manufacturing and paper package making) are located with large
April and June, while the highest abundance mass concentration populations. There are therefore many possible sources of PFASs
was observed in April and August. The higher concentration of MP released into East River, in agreement with the close correlation
was related to larger runoff in the wet season, which may increase between the PFASs contamination levels in aquatic environments
the amount of MP debris entering into coastal systems. As and population densities (Kwok et al., 2015). While MD receive the
mentioned above, larger runoff may increase the amount of MP inflow of East River. West River mostly runs through agricultural
debris entering into coastal systems in the wet seasons. Similar and forested areas, with a lack of PFAS sources from daily life and
temporal trend of MPs was also reported by Cheung et al. (2018), industrial factories (Fig. 7 and Fig. S9). By the end of 2017, the
where higher MP concentrations were observed in the wet season. western region area (9.69
106 acres) was approximately two
It was further found that the release of larger number of MPs into times of eastern region area (4.79
106 acres) of the total sown
coastal systems often occurs after heavy rains in Hong Kong and area in Guangdong province (Bureau, 2001).
6
Y. Cheng, L. Mai, X. Lu et al. Environmental Pollution 281 (2021) 117025

Fig. 7. Picture shows the distribution of urbanization rates in Pearl River Delta. Red areas represent the urbanization rate. (For interpretation of the references to colour in this figure
legend, the reader is referred to the Web version of this article.)

4.4. Temporal variations of PFASs on MPs onto the MPs (Dong et al., 2019; Hartmann et al., 2017; Velzeboer
et al., 2014; Xu et al., 2019). As a type of emerging HOCs in the
P
The lowest concentration of PFASs was found in the August environment, lower PFAS concentration adsorbed on MPs may
(979.3 ng g1), which may be due to the dilution and dispersion result from greater dispersion and dilution effects of larger amount
effects caused by greater rainfall runoff from land. However, of river runoff and precipitation. Furthermore, the lower salinity
average PFAS levels on MPs in the wet season (April, June, and and higher temperature of water may also contribute to the lower
August) are higher than those in the dry season (January and PFASs concentration.
November) sampled at the HE, HM, YM and MD (places receiving
wastewaters from large cities) (Fig. 5), indicating that the discharge 4.5. Relative contributions of individual PFAS species on MPs
of industrial or domestic wastewater containing PFASs has much
stronger impact on their levels on MPs than dilution and dispersion Among all the analyzed PFASs in the previous studies (Naile
for those sampling sites. For PFASs on MPs, dilution of PFAS levels in et al., 2010; Pan et al., 2014; Senthilkumar et al., 2007; So et al.,
coastal water in the wet season could lead to desorption of PFASs 2004, 2007; Yang et al., 2011), PFOS and PFOA were also found to
from MPs to coastal water, leading to the decrease in PFAS levels on be the dominant compounds in the PRD region (Fig. 6), which is
MPs collected in the wet season. consistent with the current study. The source of PFASs is one
Although the precipitations are greater in June (390 mm) and important factor for PFAS distributions on MPs, but the occurrence
August (389 mm) than in January (112 mm), April (80 mm), and of PFASs in MPs is also affected by their partitioning to the MPs,
November (60 mm). A Kruskal-Wallis H-test suggested no signifi- which not necessary match with the composition in water or
P
cant greater concentrations of PFAS on MPs among different source. Similar conclusions were also reported that the partitioning
months (H ¼ 40; p ¼ 0.243). Due to limited study focused on process on natural organic sorbents (such as: MPs) play an
P
PFAS on MPs, a plausible explanation for this temporal pattern important role on the fate of PFASs in the water environment (Pan
P
was that PFAS on MPs result from multiple factors (i.e., pH, et al., 2014). The partition/distribution coefficients (Kd) value for
salinity, and temperature of the ambient aqueous environment). individual PFAS is shown in SI Table S10. Results showed that the
FTIR studies showed that PP (43.1%) and PE (39.1%) accounted for average Kd value of PFOS, PFOA and PFBA on obtained MPs were
more than 80% MPs, which are good sorbents for PFASs with low 112 L g1, 35.6 L g1, and 7.49 L g1, respectively (SI Table S10). There
polarizability. The MPs collected in this study are therefore ideal are limited data about the Kd values of PFAS on different materials.
samples for PFAS concentrations analysis. The pH, salinity and For example, one study showed that the Kd value of PFOA on HDPE
temperature of aqueous environments are potential factors influ- in seawater was around 64.1 L g1, while another study showed that
encing PFASs levels on MPs. pH may affect pollutants on MPs by on PS in freshwater was around 8.37 L g1. In addition, the Kd value
changing their surface charge. The pH value is found to be stable at of PFOS on PE, PS, PVC in freshwater was 32.8 L g1, 501 L g1, and
8.0 with a small variation in PRE throughout a year, which may have 101 L g1, respectively (Llorca et al., 2018; Wang et al., 2015). The
minor effects on PFAS sorption on MPs. The salinity of water was previous results showed that PFAS levels on MPs with different
greatly influenced by the amount of precipitation and river runoff polymers were different. The percentage of PFOS to total PFASs at
in PRE region. Lower salinity and higher temperature was generally each site was generally higher than PFOA (except for HM) (Fig. 6).
found in the wet season, which may reduce the adsorption of HOCs The polarity of PFOA is stronger than PFOS, and PE and PP has been
7
Y. Cheng, L. Mai, X. Lu et al. Environmental Pollution 281 (2021) 117025

widely recognized as nonpolar plastics. Thus, PFOS should have Declaration of competing interest
higher affinity towards PE and PP than PFOA. One previous study
had shown that PFASs with sulphonated group had stronger The authors declare that they have no known competing
sorption ability on plastics than those with same carbon chain financial interests or personal relationships that could have
length but carboxylic groups (Llorca et al., 2018). It was reported appeared to influence the work reported in this paper.
that the sources of PFOS-related chemicals in the PRD region
mainly came from domestic and industrial emissions including Acknowledgements
textile treatment, metal planting, fire-fighting, and semiconductor
manufacturing (Pan et al., 2014; Xie et al., 2013a, 2013b; Zhang This research was supported by the National Natural Science
et al., 2013). In addition, PFOA has been widely used in food- Foundation of China (Project No. 21637001), and Guangdong
packing processes (Xiao et al., 2012), and a large number of fac- (China) Innovative and Entrepreneurial Research Team Program
tories, of which 15e20% were electroplating factories and food- (No. 2016ZT06N258). It was also partially supported by the
packing process industrial facilities in Pearl River Delta (Bureau, Fundamental Research Funds for the Central Universities (No.
2013). 21617454).
The observation of higher PFBA concentration on MPs may be
due to the increasing production of N-methyl perfluorobutane Appendix A. Supplementary data
sulfonamidoethanol (N-MeFBSE) and N-methyl perfluorobutane
sulfonamide (N-MeFBSA) in the PRD region (Kwok et al., 2015). Supplementary data to this article can be found online at
These compounds are precursors of C4-based compounds (i.e., https://doi.org/10.1016/j.envpol.2021.117025.
PFBA and PFBS) and play important roles as substitutions of long-
chain C8 compounds (i.e., PFOS and PFOA), since the introduction Author statement
of restrictions on PFOS usage by the Stockholm Convention in 2009
(Kwok et al., 2015). In addition, strong positive Spearman correla- I have made substantial contributions to the conception or
tions were observed between PFDoA and PFDA (rs ¼ 0.782, design of the work; or the acquisition, analysis, or interpretation of
p ¼ 0.006), PFDoA and PFTeDA (rs ¼ 0.722, p ¼ 0.001) and PFDoA data for the work; and I have drafted the work or revised it critically
and PFOA (rs ¼ 0.835, p ¼ 0.001), which may be related to the for important intellectual content; and I have approved the final
degradation of fluorotelomer precursors of PFCAs through long version to be published; and I agree to be accountable for all aspects
distance of atmospheric and oceanic transportation and other of the work in ensuring that questions related to the accuracy or
common sources. K.Y. Kwok et al. reported that positive correla- integrity of any part of the work are appropriately investigated and
tions were found between PFOA and PFNA (r ¼ 0.753, p < 0.01), resolved. All persons who have made substantial contributions to
PFDA and PFUnDA (r ¼ 0.483, p < 0.01) in surface waters in China the work reported in the manuscript, including those who provided
South Sea, which was near to PRE region. They explained such editing and writing assistance but who are not authors, are named
correction as the degradation of volatile fluorotelomer precursors in the Acknowledgments section of the manuscript and have given
(i.e., 8:2 FTOH and 10:2 FTOH) through long-range atmospheric and their written permission to be named. If the manuscript does not
oceanic transport and direct releases of fluorinated polymer include Acknowledgments, it is because the authors have not
products (Kwok et al., 2015). A significant statistical correlation was received substantial contributions from nonauthors.
found between PFBS and PFOS (p < 0.05), which was attributed to
higher Kd and concentration of PFOS than PFBS in ambient envi- References
ronment. Apparently, PFHxA was only detected at JM in 19-Apr,
which may indicate the existence of a dependent source of PFHxA Aliabad, M.K., Nassiri, M., Kor, K., 2019. Microplastics in the surface seawaters of
chabahar bay, gulf of Oman (makran coasts). Mar. Pollut. Bull. 143, 125e133.
near the sampling site (Xie et al., 2013a; 2013b).
Andrady, A.L., 2011. Microplastics in the marine environment. Mar. Pollut. Bull. 62,
1596e1605.
5. Conclusions Andrady, A.L., 2017. The plastic in microplastics: a review. Mar. Pollut. Bull. 119,
12e22.
Bureau, G.S., 2001. Guang Dong Statistical Yearbook. China Statistics Press, Beijing.
The occurrence and distribution of PCBs, DDTs, and PAHs on MPs Bureau, G.S., 2013. China Statistical Yearbook. China Statistics Press, Beijing.
has been well documented, but little information has been reported Cai, L., Wang, J., Peng, J., Tan, Z., Zhan, Z., Tan, X., Chen, Q., 2017. Characteristic of
about PFAS contamination levels on MPs. As one type of persistent microplastics in the atmospheric fallout from Dongguan city, China: pre-
liminary research and first evidence. Environ. Sci. Pollut. Res. Int. 24,
and bio-accumulated chemical, the occurrence and distribution of 24928e24935.
PFASs on MPs is also important for their potential fate and transport Chen, L., Tsui, M.M.P., Lam, J.C.W., Hu, C., Wang, Q., Zhou, B., Lam, P.K.S., 2019a.
in ecological systems. The quantification of more than twenty Variation in microbial community structure in surface seawater from Pearl
River Delta: discerning the influencing factors. Sci. Total Environ. 660, 136e144.
PFASs on MPs in the current study suggested the potential trans- Chen, L., Tsui, M.M.P., Lam, J.C.W., Wang, Q., Hu, C., Wai, O.W.H., Zhou, B., Lam, P.K.S.,
portation and bioaccumulation of PFASs with MPs via the food 2019b. Contamination by perfluoroalkyl substances and microbial community
chain, serving as a potential pathway for the transport of PFASs structure in Pearl River Delta sediments. Environ. Pollut. 245, 218e225.
Cheung, P.K., Cheung, L.T.O., Fok, L., 2016. Seasonal variation in the abundance of
from water to biota. Specially, diPAPs were detected in most MPs marine plastic debris in the estuary of a subtropical macro-scale drainage basin
collected from estuarine environments, posing a potential indirect in South China. Sci. Total Environ. 562, 658e665.
exposure route of PFASs towards aquatic animals, as has been Cheung, P.K., Fok, L., 2017. Characterisation of plastic microbeads in facial scrubs and
their estimated emissions in Mainland China. Water Res. 122, 53e61.
proven to be transferred to PFASs in the body of animals. The spatial Cheung, P.K., Fok, L., Hung, P.L., Cheung, L.T.O., 2018. Spatio-temporal comparison of
variation study shows that PFAS levels on MPs have a strong link to neustonic microplastic density in Hong Kong waters under the influence of the
the economic establishment levels around the sampling locations, Pearl River Estuary. Sci. Total Environ. 628e629, 731e739.
Cozar, A., Echevarria, F., Gonzalez-Gordillo, J.I., Irigoien, X., Ubeda, B., Hernandez-
which further indicates that the control of PFASs and MPs should be
Leon, S., Palma, A.T., Navarro, S., Garcia-de-Lomas, J., Ruiz, A., Fernandez-de-
initialed through the restricted usage of PFAS- and MP-containing Puelles, M.L., Duarte, C.M., 2014. Plastic debris in the open ocean. Proc. Natl.
products. Higher concentrations of MPs were found in the wet Acad. Sci. U. S. A 111, 10239e10244.
P D’Eon J, C., Mabury, S.A., 2011. Exploring indirect sources of human exposure to
season, but higher concentrations of PFAS on MPs were observed
perfluoroalkyl carboxylates (PFCAs): evaluating uptake, elimination, and
in the dry season. Our findings indicate that the dilution effect biotransformation of polyfluoroalkyl phosphate esters (PAPs) in the rat. Envi-
could reduce the distribution of PFASs on MPs. ron. Health Perspect. 119, 344e350.

8
Y. Cheng, L. Mai, X. Lu et al. Environmental Pollution 281 (2021) 117025

Dong, X., Zheng, M., Qu, L., Shi, L., Wang, L., Zhang, Y., Liu, X., Qiu, Y., Zhu, H., 2019. Perfluorinated compounds in coastal waters of Hong Kong, South China, and
Sorption of tonalide, musk xylene, galaxolide, and musk ketone by micro- Korea. Environ. Sci. Technol. 38, 4056e4063.
plastics of polyethylene and polyvinyl chloride. Mar. Pollut. Bull. 144, 129e133. So, M.K., Miyake, Y., Yeung, W.Y., Ho, Y.M., Taniyasu, S., Rostkowski, P., Yamashita, N.,
Fan, Y., Zheng, K., Zhu, Z., Chen, G., Peng, X., 2019. Distribution, sedimentary r ecord, Zhou, B.S., Shi, X.J., Wang, J.X., Giesy, J.P., Yu, H., Lam, P.K., 2007. Perfluorinated
and persistence of microplastics in the Pearl River catchment, China. Environ. compounds in the Pearl River and yangtze river of China. Chemosphere 68,
Pollut. 251, 862e870. 2085e2095.
Faure, F., Demars, C., Wieser, O., Kunz, M., de Alencastro, L.F., 2015. Plastic pollution Tan, X., Yu, X., Cai, L., Wang, J., Peng, J., 2019. Microplastics and associated PAHs in
in Swiss surface waters: nature and concentrations, interaction with pollutants. surface water from the Feilaixia reservoir in the beijiang river, China. Chemo-
Environ. Chem. 12, 582. sphere 221, 834e840.
Hartmann, N.B., Rist, S., Bodin, J., Jensen, L.H., Schmidt, S.N., Mayer, P., Meibom, A., Van Cauwenberghe, L., Vanreusel, A., Mees, J., Janssen, C.R., 2013. Microplastic
Baun, A., 2017. Microplastics as vectors for environmental contaminants: pollution in deep-sea sediments. Environ. Pollut. 182, 495e499.
exploring sorption, desorption, and transfer to biota. Integrated Environ. Assess. Vaughan, R., Turner, S.D., Rose, N.L., 2017. Microplastics in the sediments of a UK
Manag. 13, 488e493. urban lake. Environ. Pollut. 229, 10e18.
Hirai, H., Takada, H., Ogata, Y., Yamashita, R., Mizukawa, K., Saha, M., Kwan, C., Veerasingam, S., Mugilarasan, M., Venkatachalapathy, R., Vethamony, P., 2016. In-
Moore, C., Gray, H., Laursen, D., Zettler, E.R., Farrington, J.W., Reddy, C.M., fluence of 2015 flood on the distribution and occurrence of microplastic pellets
Peacock, E.E., Ward, M.W., 2011. Organic micropollutants in marine plastics along the Chennai coast, India. Mar. Pollut. Bull. 109, 196e204.
debris from the open ocean and remote and urban beaches. Mar. Pollut. Bull. 62, Velzeboer, I., Kwadijk, C.J., Koelmans, A.A., 2014. Strong sorption of PCBs to nano-
1683e1692. plastics, microplastics, carbon nanotubes, and fullerenes. Environ. Sci. Technol.
Jambeck, J.R., Geyer, R., Wilcox, C., R, S.T., 2015. Plastic waste inputs from land into 48, 4869e4876.
the ocean. Mar. Pollut. Bull. 347, 768e771. Wang, F., Shih, K.M., Li, X.Y., 2015. The partition behavior of per-
Kwok, K.Y., Wang, X.H., Ya, M., Li, Y., Zhang, X.H., Yamashita, N., Lam, J.C., Lam, P.K., fluorooctanesulfonate (PFOS) and perfluorooctanesulfonamide (FOSA) on
2015. Occurrence and distribution of conventional and new classes of per- and microplastics. Chemosphere 119, 841e847.
polyfluoroalkyl substances (PFASs) in the South China Sea. J. Hazard Mater. 285, Wang, J., Tan, Z., Peng, J., Qiu, Q., Li, M., 2016. The behaviors of microplastics in the
389e397. marine environment. Mar. Environ. Res. 113, 7e17.
Lee, H., Shim, W.J., Kwon, J.H., 2014. Sorption capacity of plastic debris for hydro- Wang, Y., Yu, N., Zhu, X., Guo, H., Jiang, J., Wang, X., Shi, W., Wu, J., Yu, H., Wei, S.,
phobic organic chemicals. Sci. Total Environ. 470e471, 1545e1552. 2018. Suspect and nontarget screening of per- and polyfluoroalkyl substances in
Liebezeit, G., Dubaish, F., 2012. Microplastics in beaches of the East Frisian islands wastewater from a fluorochemical manufacturing park. Environ. Sci. Technol.
spiekeroog and kachelotplate. Bull. Environ. Contam. Toxicol. 89, 213e217. 52, 11007e11016.
Liu, S., Lu, Y., Xie, S., Wang, T., Jones, K.C., Sweetman, A.J., 2015. Exploring the fate, Xiao, F., Halbach, T.R., Simcik, M.F., Gulliver, J.S., 2012. Input characterization of
transport and risk of Perfluorooctane Sulfonate (PFOS) in a coastal region of perfluoroalkyl substances in wastewater treatment plants: source discrimina-
China using a multimedia model. Environ. Int. 85, 15e26. tion by exploratory data analysis. Water Res. 46, 3101e3109.
Llorca, M., Schirinzi, G., Martinez, M., Barcelo, D., Farre, M., 2018. Adsorption of Xie, S., Lu, Y., Wang, T., Liu, S., Jones, K., Sweetman, A., 2013a. Estimation of PFOS
perfluoroalkyl substances on microplastics under environmental conditions. emission from domestic sources in the eastern coastal region of China. Environ.
Environ. Pollut. 235, 680e691. Int. 59, 336e343.
Mai, L., Bao, L.J., Shi, L., Liu, L.Y., Zeng, E.Y., 2018. Polycyclic aromatic hydrocarbons Xie, S., Wang, T., Liu, S., Jones, K.C., Sweetman, A.J., Lu, Y., 2013b. Industrial source
affiliated with microplastics in surface waters of Bohai and Huanghai Seas, identification and emission estimation of perfluorooctane sulfonate in China.
China. Environ. Pollut. 241, 834e840. Environ. Int. 52, 1e8.
Mai, L., You, S.N., He, H., Bao, L.J., Liu, L.Y., Zeng, E.Y., 2019. Riverine microplastic Xu, P., Ge, W., Chai, C., Zhang, Y., Jiang, T., Xia, B., 2019. Sorption of polybrominated
pollution in the Pearl River Delta, China: are modeled estimates accurate? diphenyl ethers by microplastics. Mar. Pollut. Bull. 145, 260e269.
Environ. Sci. Technol. 53, 11810e11817. Yang, L., Zhu, L., Liu, Z., 2011. Occurrence and partition of perfluorinated compounds
Morritt, D., Stefanoudis, P.V., Pearce, D., Crimmen, O.A., Clark, P.F., 2014. Plastic in in water and sediment from Liao River and Taihu Lake, China. Chemosphere 83,
the Thames: a river runs through it. Mar. Pollut. Bull. 78, 196e200. 806e814.
Naile, J.E., Khim, J.S., Wang, T., Chen, C., Luo, W., Kwon, B.O., Park, J., Koh, C.H., Zhang, K., Xiong, X., Hu, H., Wu, C., Bi, Y., Wu, Y., Zhou, B., Lam, P.K., Liu, J., 2017a.
Jones, P.D., Lu, Y., Giesy, J.P., 2010. Perfluorinated compounds in water, sediment, Occurrence and characteristics of microplastic pollution in xiangxi bay of three
soil and biota from estuarine and coastal areas of Korea. Environ. Pollut. 158, gorges reservoir, China. Environ. Sci. Technol. 51, 3794e3801.
1237e1244. Zhang, W., Zhang, S., Wang, J., Wang, Y., Mu, J., Wang, P., Lin, X., Ma, D., 2017b.
Pan, C.G., Ying, G.G., Liu, Y.S., Zhang, Q.Q., Chen, Z.F., Peng, F.J., Huang, G.Y., 2014. Microplastic pollution in the surface waters of the Bohai Sea, China. Environ.
Contamination profiles of perfluoroalkyl substances in five typical rivers of the Pollut. 231, 541e548.
Pearl River Delta region, South China. Chemosphere 114, 16e25. Zhang, Y., Lai, S., Zhao, Z., Liu, F., Chen, H., Zou, S., Xie, Z., Ebinghaus, R., 2013. Spatial
Poothong, S., Padilla-Sanchez, J.A., Papadopoulou, E., Giovanoulis, G., Thomsen, C., distribution of perfluoroalkyl acids in the Pearl River of southern China. Che-
Haug, L.S., 2019. Hand wipes: a useful tool for assessing human exposure to mosphere 93, 1519e1525.
poly- and perfluoroalkyl substances (PFASs) through hand-to-mouth and Zhao, S., Zhu, L., Wang, T., Li, D., 2014a. Suspended microplastics in the surface water
dermal contacts. Environ. Sci. Technol. 53, 1985e1993. of the Yangtze Estuary System, China: first observations on occurrence, distri-
Senthilkumar, K., Ohi, E., Sajwan, K., Takasuga, T., Kannan, K., 2007. Perfluorinated bution. Mar. Pollut. Bull. 86, 562e568.
compounds in river water, river sediment, market fish, and wildlife samples Zhao, Y.G., Wan, H.T., Wong, M.H., Wong, C.K., 2014b. Partitioning behavior of
from Japan. Bull. Environ. Contam. Toxicol. 79, 427e431. perfluorinated compounds between sediment and biota in the Pearl River Delta
So, M., Taniyasu, S., Yamashita, N., Giesy, J., Zheng, J., Fang, Z., Im, S., Lam, P.K., 2004. of South China. Mar. Pollut. Bull. 83, 148e154.