Science of the Total Environment 694 (2019) 133798

Contents lists available at ScienceDirect

Science of the Total Environment

journal homepage: www.elsevier.com/locate/scitotenv

Characterization of microplastics and the association of heavy metals

with microplastics in suburban soil of central China
Yanfei Zhou a,c, Xiaoning Liu a,⁎, Jun Wang b,⁎
a
Key Laboratory of Aquatic Botany and Watershed Ecology, Wuhan Botanical Garden, Chinese Academy of Sciences, Wuhan 430074, China
b
College of Marine Sciences, South China Agricultural University, Guangzhou 510642, China
c
University of Chinese Academy of Sciences, Beijing 100049, China

H I G H L I G H T S G R A P H I C A L A B S T R A C T

• Soils were polluted by microplastics

• Small size microplastics occupied a
higher proportion
• Microplastic in soil contained different
levels heavy metals
• Microplastics and heavy metals in soils
posed higher risks.

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

Article history: Microplastics (MPs) are ubiquitous in the aquatic and terrestrial environment and can absorb other kinds of pol-
Received 19 May 2019 lutants on surfaces due to strong hydrophobicity and higher specific surface area. This study investigated the oc-
Received in revised form 23 July 2019 currence and distribution of MPs and the interaction of MPs and heavy metals in three different land types of
Accepted 4 August 2019
central China. Results showed that the abundance of MPs ranged from 2.2 × 104 to 6.9 × 105 particle·kg−1,
Available online 05 August 2019
and 81.7% of particles measured ranged from 10 to 100 μm. The smaller sized particles made up the greater pro-
Editor: James Chapman portion of MPs in the study area. In woodland, the abundance of MPs (4.1 × 105 particle·kg−1) was significantly
higher than that in vegetable plots (1.6 × 105 particle·kg−1) and vacant land (1.2 × 105 particle·kg−1). The most
Keywords: prevalent form of MP particles was fragments (53%), followed by fiber (15.2%). Polyethylene, polypropylene,
Microplastics pollution polystyrene, polyamide, and polyvinyl chloride were commonly found at each sampling site, with polyethylene
Soil environment as the dominant form (36.1%). It was inferred that industrial production activities, litter, and sewage sludge might
Heavy metals be the main sources of the MPs pollution. Moreover, MP particles in the soil environment contained different
Synergic pollution levels of heavy metals, including Cd, Cr, Pb, Ag, Cu, Sb, Hg, Fe, and Mn. The concentration of heavy metals in
MPs, particularly Cd, Pb, Mn, and Hg, was closely related to the extent of heavy metal contamination in the soil
environment, inferring that MPs in the soil can be vectors for transporting heavy metals. The synergistic effect
of MPs and heavy metals might pose greater risks to soil organisms and soil safety.
© 2019 Elsevier B.V. All rights reserved.

⁎ Corresponding authors.
E-mail addresses: [email protected] (X. Liu), [email protected] (J. Wang).

https://doi.org/10.1016/j.scitotenv.2019.133798
0048-9697/© 2019 Elsevier B.V. All rights reserved.
2 Y. Zhou et al. / Science of the Total Environment 694 (2019) 133798

1. Introduction MPs and heavy metals can pose a risk of synergetic pollution to the en-
vironment, resulting in potential adverse impacts on organisms. For in-
Plastic products have been widely used since the first modern plastic stance, chronic toxic impacts and enhanced metal bioaccumulation in
was developed in 1907 for its versatile and cheap characteristics (Cole fish was induced by co-exposure to MPs and heavy metals (Lu et al.,
et al., 2011). However, plastic wastes pose a serious threat to the ecolog- 2018). In addition, the swimming velocity and resistance time of fish
ical environment due to the slow degradation rate (Cole et al., 2011; were repressed through exposure to synergetic pollution of MPs and
Machado et al., 2018). Once discarded in environment, plastic waste heavy metal (Barboza et al., 2018).
would be disintegrated into small plastic particles, ultimately forming So, over the past few decades, there are evidences of widespread, se-
microplastics (MPs) with the particle size ≤5 mm (Horton et al., 2017; rious, and persistent heavy metal pollution in soils. The objective of this
Jambeck et al., 2015; Song et al., 2017; Thompson et al., 2004). It is gen- study was to investigate the level of MPs abundance and the contents of
erally recognized that there are two categories of MPs: primary MPs and heavy metals in MP particles in three different land types (vacant land,
secondary MPs. Primary MPs are manufactured plastic within the mi- woodland, and vegetable plots) in central China. We compared the
crometer size range, such as microbeads and abrasive scrubbers used characteristics of MPs in the three land use types. We analyzed the cor-
in cosmetic and personal products (Anderson et al., 2016; Hodson relations of heavy metal content levels in MPs with the heavy metal
et al., 2017). Secondary MPs result from constant abrasion and content in soils to elucidate their synergetic effect on soil organisms
weathering of plastic products (Alimi et al., 2018; Song et al., 2017). and soil quality as well.
There are reports of MPs occurring both in marine and freshwater envi-
ronments (Hendrickson et al., 2018; Peeken et al., 2018; Porter et al., 2. Methods and materials
2018; Toumi et al., 2019). Peeken et al. (2018) reported that the concen-
tration of MPs in Arctic sea ice was up to 1.2 × 107 N·m−3. Hurley et al. 2.1. Study sites and sample collection
(2018) found that the concentration of MPs in the river bed was approx-
imately 5.2 × 106 particle·m−2. Recent publications revealed the ad- Twenty-four sampling sites were selected from the suburb of
verse impacts of ingestion MPs by mammals, fish, and invertebrates Wuhan City, central China, in December 2018. The samples were
(Barboza et al., 2018; Hodson et al., 2017; Jin et al., 2019; Lu et al., taken from three different land types, including vacant land (s4, s6, s7,
2018). The latest study had shown that gut microbiota dysbiosis and s9, s12, s21, s24), woodland (s3, s8, s11, s13-s16), and vegetable plots
metabolic disorders of mice were caused by ingestion of MPs (Jin (s1, s2, s5, s10, s17-s22). Three mixed soil samples were collected at a
et al., 2019). depth of 5 cm at each sampling site with a steel shovel, and each
Although MPs are also ubiquitous in the terrestrial environ- mixed soil sample was divided into 5 subsamples. A schematic of the
ment, relatively few studies focused on the occurrence and fate mixed sampling and geographic coordinates are depicted in Supple-
of MPs in the soil environment. The soil environment can act as mentary Fig. 1 and Table 1, respectively. All debris larger than 5 mm
temporary or permanent sinks for MPs, mostly originated from (e.g. large plastic debris, wood debris, and stones) were manually re-
anthropogenic activities (Liu et al., 2019; Machado et al., 2018). moved. Soil samples were immediately transferred into an aluminum
It was reported that 90% of Swiss floodplain soils had been con- box and stored at 4 °C until analysis.
taminated by MPs (Scheurer and Bigalke, 2018). Recently, in crop-
land of south China, the concentration of MPs was up to 4.3 × 104 2.2. Soil analysis and extraction of MPs
particle·kg −1 , predominately thought to have originated from
plastic film mulching and application of sewage sludge (Zhang Approximately 100 g soil was separated from each mixed soil
and Liu, 2018). In general, possible sources of MPs in the soil en- sample and ground to pass through a 0.150 mm sieve after air drying
vironment include plastic litter, application of sewage sludge, at room temperature. Soil total carbon (TC) was measured using a
plastic film mulching, flooding and street runoff, and atmospheric Vario TOC cube (Elementar), and pH was determined using a pH-
fallout (Bläsing and Amelung, 2018). Owing to its small size, MPs meter (Mettler Toledo FE20) at a ratio of water and soil is 2.5:1.
can accidentally be ingested by organisms, inducing deleterious Soil water content (WC) was measured by stoving method (dried
impacts (Huerta et al., 2016; Zhang and Liu, 2018). Huerta et al. at 105 °C for 24 h). Soil pH and moisture content were measured im-
(2016) found that the weight, growth rate, and mortality of mediately after returning the samples to the laboratory. The remain-
earthworms were influenced after exposure to artificial soil con- ing part of the mixed soil sample was dried at 65 °C in an oven to
taining 28% MPs (with size ≤150 μm) in the litter layer of topsoil. constant weight, and then a total of three 20 g subsamples of soil
Furthermore, Rodriguez-Seijo et al. (2017) showed that a soil were extracted from each mixed soil sample. Before implementing
environment with MPs concentration N125 mg·kg−1 caused histo- density separation, each soil sample was treated with 30 mL mixture
pathological and gut epithelium damage in earthworms. solution of 30% KOH: NaClO to digest organic matter (Cole et al.,
There are growing concerns around the levels of MPs in the soil en- 2014; Enders et al., 2017). Small plastic debris was thought to be in-
vironment, even in remote areas (Horton et al., 2017; Rillig, 2018; corporated into soil aggregates, so ultrasonic treatments are neces-
Scheurer and Bigalke, 2018; Zhang and Liu, 2018). Meanwhile, this sary to finish the separation of MPs from soil aggregates before
new anthropogenic stressor not only poses direct adverse impacts on density fraction (Bläsing and Amelung, 2018). Ultrasonic bath with
soil animals and soil microbes but may also form synergetic pollution energy input at 60 J·mL−1 for 20 min has been found to be efficient
with other contaminants, such as heavy metals (Holmes et al., 2014; to distract soil aggregates for better extraction of MPs (Imhof et al.,
Rillig, 2018). There is recent evidence that, after being subjected to a se- 2012; Löder and Gerdts, 2015). Afterwards, the digestion process
ries of changes in the environment, MPs can absorb organic pollutants was carried out at 50 °C for 48 h, and the solutions of soil and diges-
and metals on their surfaces due to their small size, strong hydrophobic- tion mixture were manually shaken for 30 s every 2 h.
ity, and larger surface area to volume ratio (Holmes et al., 2014; Vedolin After completing the digestion process, mixtures were centrifuged at
et al., 2018). Ashton et al. (2010) noted that cadmium (Cd) and lead 4000 G for 5 min, and the supernatants were collected. Subsequently,
(Pb) accumulated on the surface of plastic pellets collected from 30 mL saturated NaCl solution (ρ = 1.19 g·cm−3) was added into the
beaches. Vedolin et al. (2018) concluded that metal accumulation on centrifuge tubes, and the samples were re-centrifuged at 4000 G for
plastic particles (collected from beaches) was higher than that on virgin 5 min. The first density separation was repeated 3 times, and the super-
pellets. However, there is a risk that heavy metals adsorbed on MPs natants were collected in each round. The second density fractionation
could be released into the surrounding environment again (Brennecke with ZnCl2 (ρ = 1.55 g·cm−3) was carried out, as the density of certain
et al., 2016; Massos and Turner, 2017; Song et al., 2017). Inevitably, synthetic polymers, such as polyvinyl chloride (PVC, ρ = 1.38 g·cm−3),
 Y. Zhou et al. / Science of the Total Environment 694 (2019) 133798 3

Table 1
Detail information of soil microplastics in previous studies from other regions in comparison to this study.

Region Location specification Abundance Composition (%) Size (μm) and shape distribution (%) Extraction method Reference

Swiss Floodplain soil 593 particle·kg−1 (50 PE, 88%; PP, b1%; 125–500, 70%; 500–1000, 15%; Density separation: 27% NaCl (ρ = Scheurer
mg·kg−1) PS, 3%; PVC, 2%; 1000–2000, ≈3%; 2000–3500, ≈10%; 1.2 g.cm−3); organic digestion: 65% and
others, 7.5% 3500–5000, ≈8.5% HNO3, 70 °C, 48 h Bigalke
(2018)
China Vegetable plot Mean 18,760 – Size: 50–1000, 95%; N1000, 5% Organic digestion: 35% H2O2 and Zhang
(southern basin of particle·kg−1, range Shape: Fiber, 95%; plastic fragment, 4.0%; 10% FeSO4, 50 °C, 0.5 M NaOH; and Liu
Dian Lake, China) from 7100 to 42,960 film, 3.7%; plastic string, 0.3% Density separation: saturated NaI (ρ (2018)
particle·kg−1 = 1.8 g.cm−3)
China Farmland (suburbs of 84.7 ± 13.2 items·kg−1 PP, 50.5%; PE, b1000, 48.8% (shallow) and 59.8% (deep) Density separation: saturated NaCl Liu et al.
Shanghai, China) (0–3 cm), and 65.7 ± 43.3%; PES, 6.1% Shape: Shallow: fiber, 53.3%; fragment, (ρ = 1.19 g.cm−3); organic (2018)
13.9 items·kg−1 (3–6 37.6%; film, 6.7%; pellet, 2.21%; Deep: digestion: 30% H2O2, 50 °C, 72 h
cm) fiber, 37.6%; film, 33.8%; fragment, 28.3%;
pellet, 0.3%
Australia Soil of industrial area Range from 300 PVC, PE, PS – A method based on Pressurized Fuller
(Sydney) mg·kg−1 to 67,500 Fluid Extraction (PFE) and
mg·kg−1 (0.03–6.7%) Gautam
(2016)
China Coastal soil Mean 740.1 N·kg−1, PE, PP, PS, PEU Size:b50, 1.8%; 50–100, 4.2%; 100–250, Density separation: saturated NaCl Zhou
−3
(Shandong province, varied from 1.3 to 28.9%; 250–500, 15.8%, 500–1000, 8.9%; (ρ = 1.19 g.cm ), and saturated et al.
east China) 14,712.5 N·kg−1 1000–5000, 40% NaI (ρ = 1.6 g.cm−3) (2018)
Shape: pellet, 0.1%; foam, 27.8%;
fragment, 1.1%; flake, 69%; film 0.2%;
fiber, 1.0%; sponge, 0.8%
China Woodland, vegetable Mean 2.2 × 105 PE, 36.1%; PP, Size:10–50, 46.1%; 50–100, 35.3%; Organic digestion: 30% KOH:NaClO This
plot, vacant land of particle·kg−1, range 11.5%; PS,8.9%; 100–500, 18.1%, N500, 0.2% Density separation: NaCl (ρ = 1.19 study
(Wuhan city, central from 2.2 × 104 to 6.9 × PA,17.3%; PVC, Shape: fragment, 52%; bead, 14%; fiber, g.cm−3) and ZnCl2 (ρ = 1.55 g.
China) 105 particle·kg−1 8.5%; others, 17.7% 13.8%; foam, 9.7%; film 1.6% cm−3)

polyethylene terephthalate (PBT, ρ = 1.34–1.39 g·cm−3), and produced by microfiltration. Prior to use, all plastic labware was soaked
polylactic acid, (PLA, ρ = 1.21–1.43 g·cm−3), is higher than that of sat- in an HCl (10% v/v) solution for 24 h, and thoroughly rinsed with ultra-
urated NaCl. This extraction process was repeated a few times until pure water (Brennecke et al., 2016). Analytical grade chemical reagents
there were no particles were observed in the supernatant. In the filtra- used in this study were purchased from Sinopharm Chemical Reagent
tion process, each beaker was flushed with saturated NaCl to rinse ad- Co., Led, SCR, China, and all solutions were filtered using 0.45 μm mem-
hering materials. The same procedure was performed on five control brane filtration.
samples without adding soil.
2.4. Heavy metal analysis
2.3. Pellet analysis
The method for the extraction of heavy metals from MPs was based
Suspected plastic particles from filters extracted from all samples on previous studies (Ashton et al., 2010; Dobaradaran et al., 2018). Four
were identified by size and shape and counted using a stereomicroscope hundred plastic particles (within a size range of 200–5000 μm) were
(Nikon, SMZ25) (Zhou et al., 2018). The plastic debris was classified by randomly selected from each sampling site for the extraction of metals.
shape into fragment, bead, ball, fiber, foam, and film (Liu et al., 2018; All plastic particles were placed in an ultrasonic bath to further remove
Zhou et al., 2018). Five categories of the size of MPs were recorded: adhered materials. Aqua regia was made by mixing 2 M HCl and 3 M
N500 μm, 200–500 μm, 100–200 μm, 200–500 μm, 50–100 μm, and HNO3 in a ratio of 3:1, and a diluted solution of 20% aqua regia was pre-
10–50 μm (Peeken et al., 2018; Zhou et al., 2018). All suspected particles pared. The MP particles were transferred into centrifuge tubes, and then
were counted and classified under 1× lens with an objective of 8× mag- 2.5 mL of 20% aqua regia solution was added into each tube and shaken
nification. In addition, photographs of MPs were captured with a visual at 150 rpm for 24 h at 25 °C. Five control blank samples were prepared
objective of 10× magnification using a high-resolution camera (Nikon, to ensure precision and accuracy. After digestions, the samples were re-
SMZ25). For ascertaining the type of synthetic polymer, a DXR2 covered from the centrifuge tubes and stored at 4 °C until analysis.
Raman microscope (Thermo Fisher Scientific, USA) was used to identify Following the MP extraction process, 0.1 g samples of dried and
polymeric compositions. The Raman microscope was set up as follows: ground soil were placed into digestion tubes and treated with 4 mL
number of sample scans 3, exposure times 15, background exposure HNO3 and 2 mL HF, and then the samples were heated in a micro-
times 512, laser 532 nm, laser energy 8 mW. Each suspected MP was wave oven at 150 °C for 2 h. After drying, 50 mL distilled ionized
manually located with 10× lens of DXR2 Raman (examples of the poly- water was added, and the digestion solution was filtered and stored
mer Raman spectra are shown in Supplementary Fig. 4). All obtained at 4 °C. Inductively Coupled Plasma-Mass Spectrometry (ICP-MS)
Raman spectra were compared with the libraries database via OMNIC (Thermo Scientific X Series 2) was employed to analyze the concen-
(Thermo Fisher Scientific, USA) with at least 50% matching score for tration of Cd, chromium (Cr), Pb, silver (Ag), copper (Cu), antimony
spectra. Following this, the total MP particle numbers were recalculated, (Sb), mercury (Hg), and manganese (Mn) in the samples. Iron (Fe)
and the typical surface morphology of MPs was obtained using Quanta concentration in the remaining digestive solution was analyzed
250 scanning electron microscopes (SEM) (Fig. 3i and j). using inductively coupled plasma-optical emission spectrometry
Every calculated value presented in this study was deduced from the (ICP-OES, Perkin Elmer Optima 8000).
value of the control. Very small sized MP particles (down to ~10 μm)
could be distinguished using a stereomicroscope (Nikon, SMZ25) 2.5. Data treatment and statistical analysis
under 1× lens with an objective of 10× magnification (Supplementary
Fig. 1b). To prevent contamination during the sample treatment pro- Richness (N) and Shannon-Wiener (H′) were calculated using the
cess, all labware was thoroughly cleaned using ultrapure water Vegan package in R. One-Way ANOVA data were tested for normality
4 Y. Zhou et al. / Science of the Total Environment 694 (2019) 133798

(Shapiro-Wilk) and homogeneity of variances (Levene's test). Logarith- words, large values of H′ would be representative of greater diversity.
mic transformation and rank cases (Blom) were performed in advance if As illustrated in Fig. 2b, values of N varied from 7 to 17 among all sam-
the assumptions of parametric tests could not be fulfilled. One-way pling sites, and values of H′ ranged from 1.4 to 2.4. The N and H′ in the
ANOVA was used to determine significant differences in MP-related woodland were statically higher than those in the vegetable plot and va-
data among groups (post-hoc Turkey's multiple comparison test). Sig- cant land, with the average values of 14.6 ± 1.7 and 2.2 ± 0.1,
nificance levels of 5% and 1% were set in the relevant analysis. Calcula- respectively.
tion of Spearman rank correlations was carried out with non-
transformed data.
3.3. Shape and size distribution

3. Results As shown in Fig. 3a–h, six different shapes of MP particles were

recorded, and fragment, bead, ball, and fiber were observed at most
3.1. MP abundance sites. The predominant shapes, fragments and fibers, made up
51.7% and 15.2%, respectively (Fig. 4a). Fragments were predominant
Overall, the average MP abundance at all sampling sites in vacant land (50.5%), woodland (55.4%), and vegetable plots
varied from 2.2 × 104 to 6.9 × 10 5 particle·kg−1, with an aver- (49.9%). In addition, the percentage of fibers in the vegetable plot
age of 2.2 × 10 5 particle·kg −1 (Fig. 1a). The average MP abun- (20.8%) was significantly higher than that in vacant land (10.4%)
dance in woodland, vegetable plots, and vacant land ranged from and woodland (12.1%). However, there were no differences observed
9.6 × 10 4 –6.9 × 10 5 , 4.3 × 10 4 –6.2 × 10 5 , and 2.2 × 10 4 –2 × 10 5 in the composition of fragment, bead, ball, foam, and film among the
particle·kg −1 , respectively. MP abundance in woodland was three land types (Supplementary Table 3). The size distribution of
significantly higher than that in the vegetable plot and vacant MP particles is presented in Fig. 4b and five categories were classified
land (P b 0.05, Fig. 1b), but no significant difference was found according to particle size (N500 μm, 200–500 μm, 100–200 μm,
between the vegetable plot and vacant land (Fig. 1b). Site s8, s20, 200–500 μm, 50–100 μm, and 10–50 μm). The MP particles size b50
and s21, in woodland, vegetable plot, and vacant land, showed μm and 100 μm accounted for 46.1% and 81.3%, respectively. In
the highest MP abundance, with the average values of 6.9 × 10 5 , total, 99.8% of plastic particles were smaller than 500 μm. Most of
6.2 × 105, and 2 × 105 particle·kg−1, respectively (Fig. 1a). the MP particles observed in our study were classified into the
small size categories.
3.2. MPs composition
3.4. Heavy metals content of MPs and soil
As shown in Fig. 2a and b, 17 different synthetic polymer types were
identified based on Raman spectra (shown in Supplementary Fig. 3). In All target metals were measured both in MPs and the soil at each
total, the proportion of polyethylene (PE), polyamide (nylon 6, PA), and sampling site, and the results are presented in Fig. 5. The mean con-
polypropylene (PP) were 36.1%, 17.3%, and 11.5%, respectively. Further- tent of Cd, Cr, Pb, Ag, Cu, Sb, Hg, Fe, and Mn in MP particles was 0.6,
more, the MP composition at each site was variable, but PE, PP, polysty- 14.2, 13.1, 0.2, 13.7, 0.5, 0.1, 3367.4, and 14.5 μg·g−1, respectively,
rene (PS), PA, and polyvinyl chloride (PVC) were commonly found at while the mean content of Cd, Cr, Pb, Ag, Cu, Sb, Hg, Fe, and Mn
each sampling site. No significant differences were detected among contents of soil was 0.5, 74.9, 23.9, 1.5, 21.2, 1.2, 0.2, 37,330.1, and
the different land types (Supplementary Table 3). Another 12 types of 121.5 μg·g−1, respectively. In some cases, total mean content levels
polymer were exclusively found in partially sampled sites. The of target metals in the MP particles were greater than those in the
dominant polymer type at all sampling sites was PE, which accounted soils. For instance, Cd, which ranged from 0.1 to 1.2 μg·g −1 , was
for 38%, 34.6%, and 35.7% of the total MP composition in vacant land, higher in the MP particles than in soils, with the exception of site
woodland, and vegetable plot, respectively. The polymer richness s6, s9, s12, s14, s18, s22, and s23. In this study, the variance analysis
(N) indicates the diversity of synthetic polymer types at a sampling showed that there were no other obvious differences between the
site, and the Shannon-Wiener index (H′) reflects the equitability or content levels of target metals in MP particles in the different land
evenness of polymer distribution in the polymer types. In other types (Supplementary Table 4), except for Cd, Cr, and Mn.

Fig. 1. (a) Microplastic abundance (particle·kg−1) of different sampling sites; (b) one-way NOVA results of microplastic abundance between woodland, vegetable plot, vacant land.
Asterisk (*) indicate a significant level of 0.05. (n = 3).
 Y. Zhou et al. / Science of the Total Environment 694 (2019) 133798 5

Fig. 2. (a) Average % composition of polymers in each sample site of vacant land, woodland and vegetable plot. (PC, PIB, PAI, PVDF, PAN, PET, EVA, PE-PP, PMMP, ABS, PU and PES were
summarized as others, abbreviations for synthetic polymers were shown in Supplementary Table 2). (b) Polymer richness (N) and the Shannon-Wiener index (H′). Steps indicate the
name of sampling sites. (n = 3).

3.5. Correlations of metal contents and microplastic particles contents in soils (Supplementary Table 4). Further linear regres-
sion analysis revealed that, to some degree, the content of Cd,
The correlation analysis showed that the contents of Cd, Pb, Pb, and Hg in MPs would be positively influenced by their con-
Hg, and Mn in MP particles were strongly correlated with their tents in soil. The content of Mn in the soil would increase with

a b c d

e f g h

i j

Fig. 3. Optical microscope image of typical MP particles with different shapes: fragment (a), bead (b), fiber (c–d), ball (e–f), fragment (f), foam (g), film (h). (scale bar = 200 μm). SEM
image of typical surface character of PE ball (j) and PP line (k), by using Quanta 250 scanning electron microscopes.
6 Y. Zhou et al. / Science of the Total Environment 694 (2019) 133798

Fig. 4. (a) The distribution of MP particles by shape. (b) The distribution of MP particles by particle size (μm). Box-whiskers plot of percentage (%) shares the quantity of polymer of
different size classes of all sample sites. The boundaries of the box closest and farthest to zero indicate the 25th percentiles and 75th percentiles, respectively. Black lines within each
box mark the median. Whiskers (error bars) above and below the box mean the 90th and 10th percentiles, respectively. Red lines and black dots indicate the mean and the quantity
distribution of size and shape in each sample sites, respectively. Black dots on the right side of each box mean specific proportion in each sampling site. (n = 3). (For interpretation of
the references to color in this figure legend, the reader is referred to the web version of this article.)

an increase in Mn content in MPs (Supplementary Fig. 2). The frequently discharged into the surroundings from a nearby chemical
content levels of Cd (r = 0.490⁎, P = 0.02), Cr (r = 0.470⁎, P = fiber factory, textile factory, and cement mill. Furthermore, as demon-
0.03), and Hg in MP particles were positively correlated with strated in Fig. 1a, site s8, s20, and s21 showed the highest MP abundance
total particle numbers (Supplementary Table 3). In particular, in woodland, vegetable plot, and vacant land, respectively. Site s8,
the content of Hg in MPs was strongly positively correlated with which had the highest total MP abundance in this study, located in an
the total particle numbers (r = 0.588⁎⁎, P b 0.001). industrial area with a nearby chemical fiber factory (Supplementary
Table 1). A plastics factory, concrete company, electronics factory, and
4. Discussion colleges (with relatively high population) were found in close proximity
to site s20 and s21. This indicates that industrial production processes
4.1. MP abundance increased MP abundance in soil.
It has been reported that MP pollution is pervasive in suburban soil,
The abundance of MPs observed in this work was higher, compared and the industrial activities greatly enhanced the abundance of MPs in
to the results in previous studies. Liu et al. (2018) found that MP abun- soil (Fuller and Gautam, 2016; Liu et al., 2018; Scheurer and Bigalke,
dance in farmland soil was 84.8 items·kg−1, while Zhang et al. (2018) 2018; Zhang and Liu, 2018; Zhou et al., 2018). In general, the contami-
reported MP abundance up to 4.3 × 104 particle·kg−1 in cropland. Re- nation level of MPs in the soil is related to adjacent pollution sources,
cently, Zhou et al. (2018) showed MP abundance in coastal soil was plastic waste, and the population density at the sampling site (Fuller
740.1 particle·kg−1. Population, plastic wastes, plastic film mulching, and Gautam, 2016; Scheurer and Bigalke, 2018). According to previous
and sewage sludge application at sampling sites were closely correlated assessments, street runoff, littering, irrigation with wastewater (for
with soil MP abundance (Liu et al., 2018; Scheurer and Bigalke, 2018). the vegetable plot), and atmospheric fallout can lead to accumulation
Most sampling sites in this study had large amounts of plastic litter of MPs in the soil environment (Bläsing and Amelung, 2018; Nizzetto
and plastic film debris (Supplementary Fig. 3). This might explain the et al., 2016). Soil is regarded as a possible temporary or permanent
higher MP abundance recorded in this study. Previously, Fuller and sink of MPs (Nizzetto et al., 2016; Peeken et al., 2018). Hence, MPs in
Gautam (2016) demonstrated that higher MP abundance was related soil should be given more attention, as MPs present an ecological risk
to the proximity of sampling sites to industrial areas. In this study, and might be ingested by soil organisms (Massos and Turner, 2017;
nearly half of the soil samples were obtained from industrial areas Rillig, 2018; Rillig et al., 2019).
with obvious pollution sources (Supplementary Table 1) which could
explain the higher MP abundance. In addition, the extraction and iden- 4.2. MPs composition
tification method used (approximately down to 10 μm) might be one of
the possible reasons for the difference in MP abundance (Fuller and The results relating to MP composition are in accordance with
Gautam, 2016; Liu et al., 2018; Rodrigues et al., 2018; Scheurer and previous studies (Eo et al., 2018; Peeken et al., 2018; Scheurer and
Bigalke, 2018). Bigalke, 2018). As shown in previous studies, the proportion of PE
Moreover, MP abundance varied across the different land use types. reached up to 88% in floodplain soil and accounted for 43% in farm-
In particular, MP abundance in woodland was significantly higher than land soil (Rodriguez-Seijo et al., 2017; Zhang and Liu, 2018). In this
that in vegetable plots and vacant land. This result might be attributed study, the proportion of PE at site s2 and s5 were 50.1% and 41.3%,
to the different pollution sources at the sampling sites. Most sampling respectively, which might be caused by the higher use of plastics
sites (except site s13) in the woodland were taken from an industrial composed of PE. Meanwhile, PE particles could easily be migrated
area, where wastewater, sewage sludge, and exhaust gas were and transferred in the soil due to the lighter density (Machado
 Y. Zhou et al. / Science of the Total Environment 694 (2019) 133798 7

Fig. 5. Heavy metal contents in MP particles and in soils (μg·g−1) of each sampling site, black and red line indicate metal concentration extract from soil and MP particles,
respectively. (n = 3). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

et al., 2018; Rodriguez-Seijo et al., 2017). The high proportion of PE varied in different studies (Fuller and Gautam, 2016; Liu et al., 2018;
could also be related to greater plastic film mulching and sewage Peeken et al., 2018). Possible sources of MPs in soil could include
sludge application (Zhou et al., 2018), as discarded plastic film de- waste and exhaust gas discharged from relative factories, plastic
bris were found in the vegetable plots, and sewage sludge was usu- mulching, and sewage sludge application (Fuller and Gautam, 2016;
ally applied to woodland as a fertilizer. In contrast to previous Hurley et al., 2018; Scheurer and Bigalke, 2018).
studies (Fuller and Gautam, 2016; Liu et al., 2018; Peeken et al.,
2018; Scheurer and Bigalke, 2018), PA had the second highest pro- 4.3. Shape and size distribution
portion in this study (17.3%). The discharge of domestic wastewater
and various manufacturing and processing industries have in- A higher proportion of fragments in sediment or the aquatic envi-
creased the prevalence of PA in the environment (Liu et al., 2019). ronment were reported in many previous studies (Eo et al., 2018;
For instance, a relatively higher proportion of PA was recorded at Fischer et al., 2016; Horton et al., 2017; Rodrigues et al., 2018;
site s17 (23.5%) and s22 (25.3%). This might be correlated with Sighicelli et al., 2018). For instance, Eo et al. (2018) showed that the per-
wastewater irrigation as ponds or ditches containing wastewater centage of fragment particles reached up to 97.6% in beaches of South
were observed at these sampling sites (Fig. 2a and Supplementary Korea. Fragment particles were the predominant shape type recorded
Fig. 3e). in this study. Most fragments were identified as PE or PP in previous
The values for N and H′ in the woodland were statically higher than work (Rodrigues et al., 2018), suggesting that high proportions of PE
those in the vegetable plots and vacant land. This implies that industrial and PP were closely related to the high proportion of fragments. Simi-
production process can introduce a greater diversity of polymers into larly, the high percentage of fragment particles recorded in this work
the soil. Most sampling sites in this study partially reflect the composi- might be due to the relatively higher proportions of PE and PP. Fiber par-
tion of widely produced polymers, and other identified polymer types ticles have been commonly found in wastewater and sewage sludge
8 Y. Zhou et al. / Science of the Total Environment 694 (2019) 133798

(Nizzetto et al., 2016; Zhang and Liu, 2018). Irrigation with wastewater was a strong positive correlation between Cd content in MP particles
and application of sewage sludge were considered the main source of fi- and in the soil. The high Cd content in MP particles might be related
bers in agricultural soil (Scheurer and Bigalke, 2018). This may explain to Cd pollution levels in the soil. Furthermore, The Cd content in MPs
the higher proportion of fibers in the vegetable plot. Microbeads are was positively correlated with the number of total MP particles (Supple-
prevalent and widespread owing to their use in facial cleaners, tooth- mentary Table 5). Therefore, as Wang et al. (2018) suggested, the inher-
paste, and detergents (Hurley et al., 2018; Liu et al., 2019). Wastewater ent load metals in plastic might be an explanation for the higher Cd
plays a pivotal role in distributing of small MPs, such as microbeads content in MPs.
(Hurley et al., 2018). Thus, the relatively high proportion of beads
might be closely associated with street runoff and litter containing pri- 4.5. Correlations of metal content and plastic particles
mary MPs.
Most MP particles observed in this study were small size categories. There is a strong correlation between metal content in MP particles
Generally, the size distribution of MP particles recorded in this study is and soils, which indicates that the variability of metal content levels in
in accordance with previous studies (Scheurer and Bigalke, 2018; Zhou the extraneous solid may be important in the variability of metal con-
et al., 2018). Scheurer and Bigalke (2018) found that, on average, 84% of tent levels in MP particles (Holmes et al., 2014). The MP particles con-
plastic particles in floodplain soils were smaller than 500 μm. Zhou et al. tain metals, incorporated into the synthetic polymer during the
(2018) reported that on average 49.7% of plastic particles in coastal soils manufacturing process, and metals could be released into the ambient
were smaller than 500 μm. Very small size MPs (even nanoplastics) environment (Fries et al., 2013; Sighicelli et al., 2018). Our results
could occur due to physical abrasion or biodegradation of plastic show that the content levels of Cd, Cr, and Hg in MP particles were pos-
waste (Hurley et al., 2018; Song et al., 2017). MPs in the soil environ- itively correlated with the total number of MP particles (Supplementary
ment are an emerging threat. Hence, it is of great importance to assess Table 3). This may indicate that heavy metal content in MP particles is
the potential hazardous effects induced by soil MP contamination, espe- related to MP abundance.
cially for smaller sized particles (b50 μm), since those particles might Plastic polymers are regarded as relatively inert to water-soluble
easily be ingested accidentally by organisms and accumulate in food metal ions. To date, there are few studies on the interaction between
webs (Hurley et al., 2018; Zhang and Liu, 2018). More importantly, MPs and heavy metal pollutants in soil. Laboratory experiments
smaller MP particles tend to provide a greater specific surface area for (Fischer et al., 2016; Horton et al., 2017; Rillig et al., 2019; Sighicelli
accumulation and transfer of other pollutants, such as organic pollut- et al., 2018) and environmental monitoring (Nizzetto et al., 2016;
ants, human pathogens, and heavy metals (Imran et al., 2019; Tang Tang et al., 2018) have demonstrated that plastic particles could adsorb
et al., 2018; Vedolin et al., 2018). metals from surroundings. Metal content in MPs might come from the
inherent load, which was added into the synthetic polymer as stabi-
4.4. Heavy metal content of MPs and soil lizers, pigment or catalysts (Nizzetto et al., 2016; Tang et al., 2018).
The metals extracted from MPs are thus either a product of adsorption
The mean content of Cr, Pb, Ag, Sb, Cu, Hg, Fe, and Mn in soils was or inherent in MP particles. This study shows that MPs in soil have vary-
greater than that of MP particles, which is in agreement with previous ing levels of heavy metals, and some metal content in MPs is greater
studies (Ashton et al., 2010; Dobaradaran et al., 2018; Wang et al., than in the surrounding soils. Moreover, the heavy metal content in
2017). Soil particulate matter may provide a greater specific surface MPs was strongly influenced by the content levels in the soil. This im-
area than MP particles (usually a few tens of cm2·g−1 geometric surface plies that MPs in the soil environment might act as vectors for transfer-
area) (Ashton et al., 2010; Holmes et al., 2014). The higher metal con- ring heavy metal pollutants. Inevitably, the synergetic pollution of MPs
tent in the soil might also be attributed to the high porosities and den- and heavy metals might pose an ecological risk and lead to potential ad-
sities of charged surface sites in soil organic material which facilitates verse impacts on soil organisms. Further studies are warranted to inves-
the accumulation of metals (Ashton et al., 2010). Overall, the results in- tigate the mechanism of metal adsorption on MP particles in the soil
dicate that soil MP particles may accommodate different levels of heavy environment, as well as the harmful effects induced by synergetic
metals, as found in previous studies (Dobaradaran et al., 2018; Holmes pollution.
et al., 2014; Wang et al., 2017; Zhang and Liu, 2018). Furthermore, the
mean Fe content in MP particles and in soils were 3367.4 and 37,330.1 5. Conclusion
μg·g−1, respectively, whereas, Dobaradaran et al. (2018) found Fe con-
tent in MPs and sediments were 531 and 3045 μg·g−1, respectively. The The abundance of MP particles ranged from 2.2 × 104 to 6.9 × 105
content of other heavy metals was similar to the results of previous re- particle·kg−1, with an average of 2.2 × 105 particle·kg−1. The MP parti-
search. For example, in this study, the Mn content in MPs and in soils cles b100 μm accounted for 81.7%, suggesting that smaller MP particles
were 14.5 and 121.4 μg·g−1, respectively, and in previous study, were play a dominant role in the study area. The predominant polymer was
found to be 32.2 and 126.5 μg·g−1, respectively (Dobaradaran et al., PE, accounting for 36.1%, followed by PA (17.3%). Fragments are the
2018). This suggests that the content of heavy metals in MP particles most prevalent form of MP particles, contributing 53%, followed by
might be influenced by local levels of heavy metal pollution. fiber (15.2%). The MP abundance and shape of particles displayed obvi-
There are many factors that influence the process of accumulation of ous differences across different land types. In particular, MPs abundance
metals on plastic particles (Holmes et al., 2014; Massos and Turner, in woodland was much higher than that in the vegetable plot and va-
2017). Brennecke et al. (2016) showed that metal adsorption capacity cant land. In addition, MP particles in the soil environment also had
depends on polymer type, and the accumulation of Cu in PVC particles varying levels of heavy metals, including Cd, Cr, Pb, Ag, Cu, Sb, Hg, Fe,
was significantly higher than in PS. Moreover, Holmes et al. (2014) re- and Mn. The strong correlation between heavy metal content in MP par-
vealed that metal adsorption capacity was related to environmental ticles and soils implies that the heavy metal content in MPs might be
conditions. For instance, Pb adsorption became relatively constant closely related to the extent of heavy metal contamination in the soil en-
when the pH was higher than 6, while Cd adsorption increased with in- vironment. There was also a strong correlation between the metal con-
creasing pH (Holmes et al., 2014). In our result, the content of Cd in MP tent in MPs and the total number of MP particles, suggesting that MP
particles in 17 of the 24 sampling sites was higher than that in soils, and abundance influences the heavy metal content. Based on these results,
most of the sites were in an industrial area. Wang et al. (2016) reported it is likely that MPs in the soil environment can be vectors for transfer-
that titanium in plastic pellets exceeded the extraneous solids in part of ring heavy metals. The synergetic pollution of MPs and heavy metals
sampling sites. This result might be correlated with the higher soil pH of might present ecological risks and pose potential adverse impacts on
the 17 industrial area sampling sites (Supplementary Table 1). There soil organisms.
 Y. Zhou et al. / Science of the Total Environment 694 (2019) 133798 9

Declaration of Competing Interest identify the knowledge gaps and future research priorities. Sci.Total Environ. 586,
127–141. https://doi.org/10.1016/j.scitotenv.2017.01.190.
Huerta Lwanga, E., Gertsen, H., Gooren, H., Peters, P., Salanki, T., van der Ploeg, M., et al.,
None declared under financial, general, and institution competing 2016. Microplastics in the terrestrial ecosystem: implications for Lumbricus terrestris
interest (Oligochaeta, Lumbricidae). Environ. Sci. Technol. 50, 2685–2691. https://doi.org/
10.1021/acs.est.5b05478.
Hurley, R., Woodward, J., Rothwell, J.J., 2018. Microplastic contamination of river beds sig-
Acknowledgments nificantly reduced by catchment-wide flooding. Nat. Geosci. 11, 251–257. https://doi.
org/10.1038/s41561-018-0080-1.
Imhof, H.K., Schmid, J., Niessner, R., Ivleva, N.P., Laforsch, C., 2012. A novel, highly efficient
This work was grateful to the Funding Project of National Key Re- method for the separation and quantification of plastic particles in sediments of
search and Development Program of China (2018YFD0900604), Guang- aquatic environments. Limnol. Oceanogr. Methods (7), 524–537 https://doi.org/
dong Province Universities and Colleges Pearl River Scholar Funded 10.4319/lom.2012.10.524.
Imran, M., Das, K.R., Naik, M.M., 2019. Co-selection of multi-antibiotic resistance in
Scheme (2018), and Sino-Africa Joint Research Center, Chinese Acad-
bacterial pathogens in metal and microplastic contaminated environments: an
emy of Sciences (Y623321K01). emerging health threat. Chemosphere 215, 846–857. https://doi.org/10.1016/j/
chemosphere.1028.10.114.
Appendix A. Supplementary data Jambeck, J.R., Geyer, R., Wilcox, C., Siegler, T.R., Perryman, M., Andrady, A., 2015. Plastic
waste inputs from land into the ocean. Science 347, 768–771. https://doi.org/
10.1126/science.1260352.
Supplementary data to this article can be found online at https://doi. Jin, Y.X., Lu, L., Tu, W.Q., Luo, T., Fu, Z.W., 2019. Impacts of polystyrene microplastic on the
org/10.1016/j.scitotenv.2019.133798. gut barrier, microbiota and metabolism of mice. Sci.Total Environ. 649, 308–317.
https://doi.org/10.1016/j.scitotenv.2018.08.353.
Liu, M.T., Lu, S.B., Song, Y., Lei, L.L., Hu, J.N., Lv, W.W., 2018. Microplastic and mesoplastic
References pollution in farmland soils in suburbs of Shanghai, China. Environ. Pollut. 242,
855–862. https://doi.org/10.1016/j.envpol.2018.07.051.
Alimi, O.S., Budarz, J.F., Hernandez, L.M., Tufenkji, N., 2018. Microplastics and Liu, X.N., Yuan, W.K., Di, M.X., Li, Z., Wang, J., 2019. Transfer and fate of microplastics dur-
nanoplastics in aquatic environments: aggregation, deposition, and enhanced ing the conventional activated sludge process in one wastewater treatment plant of
contaminant transport. Environ. Sci. Technol. 52, 1704–1724. https://doi.org/ China. Chem. Eng. J. 362, 176–182. https://doi.org/10.1016/j.cej.2019.01.033.
10.1021/acs.est.7b05559. Löder, M.G.J., Gerdts, G. Methodology used for the detection and identification of
Anderson, A.G., Grose, J., Pahl, S., Thompson, R.C., Wyles, K., 2016. Microplastics in microplastics-a critical appraisal. In: Bergmann, Melanie, Gutow, Lars, Klages, Michael
personal care products: exploring perceptions of environmentalists, beauticians (Eds.), Marine Anthropogenic Litter. Springer International Publishing, Cham. 2015;
and students. Mar. Pollut. Bull. 113, 454–460. https://doi.org/10.1016/j/ 201–227. doi:https://doi.org/10.1007/978-3-319-16510-3_8.
marpolbul.2016.10.048. Lu, K., Qiao, R.X., An, H., Zhang, Y., 2018. Influence of microplastics on the accumulation
Ashton, K., Holmes, L., Turner, A., 2010. Association of metals with plastic production pel- and chronic toxic effects of cadmium in zebrafish (Danio rerio). Chemosphere 202,
lets in the marine environment. Mar. Pollut. Bull. 60, 2050–2055. https://doi.org/ 514–520. https://doi.org/10.1016/j.chemosphere.2018.03.145.
10.1016/j.marpolbul.2010.07.014. Machado, A.A.D., Kloas, W., Zarfl, C., Hempel, S., Rillig, M.C., 2018. Microplastics as an
Barboza, L.G.A., Vieira, L.R., Guilhermino, L., 2018. Single and combined effects of emerging threat to terrestrial ecosystems. Global Change Bio 24, 1405–1416.
microplastics and mercury on juveniles of the European seabass (Dicentrarchus https://doi.org/10.1111/gcb.14020.
labrax): changes in behavioural responses and reduction of swimming velocity and Massos, A., Turner, A., 2017. Cadmium, lead and bromine in beached microplastics. Envi-
resistance time. Environ. Pollut. 236, 1014–1019. https://doi.org/10.1016/j. ron. Pollu. 227, 139–145. https://doi.org/10.1016/j.envpol.2017.04.034.
envpol.2017.12.082. Nizzetto, L., Futter, M., Langaas, S., 2016. Are agricultural soils dumps for microplastics of
Bläsing, M., Amelung, W., 2018. Plastics in soil: analytical methods and possible sources. urban origin? Enviro. Sci. Technol. 50, 10777–10779. https://doi.org/10.1021/acs.
Sci. Total Environ. 612, 422–435. https://doi.org/10.1016/j.scitotenv.2017.08.086. est.6b04140.
Brennecke, D., Duarte, B., Paiva, F., Cacador, I., Canning-Clode, J., 2016. Microplastics as Peeken, I., Primpke, S., Beyer, B., Gutermann, J., Katlein, C., Krumpen, T., et al., 2018. Arctic
vector for heavy metal contamination from the marine environment. Estuar. Coast. sea ice is an important temporal sink and means of transport for microplastic. Nat.
Shelf S. 178, 189–195. https://doi.org/10.1016/j.ecss.2015.12.003. Commun. 9, 12. https://doi.org/10.1038/s41467-018-03825-5.
Cole, M., Lindeque, P., Halsband, C., Galloway, T.S., 2011. Microplastics as contaminants in Porter, A., Lyons, B.P., Galloway, T.S., Lewis, C., 2018. Role of marine snows in microplastic
the marine environment: a review. Mar. Pollut. Bull. 62, 2588–2597. https://doi.org/ fate and bioavailability. Environ. Sci. Technol. 52, 7111–7119. https://doi.org/
10.1016/j.marpolbul.2011.09.025. 10.1021/acs.est.8b01000.
Cole, M., Webb, H., Lindeque, P.K., Fileman, E.S., Halsband, C., Galloway, T.S., 2014. Isola- Rillig, M.C., 2018. Microplastic disguising as soil carbon storage. Environ. Sci. Technol. 52,
tion of microplastics in biota-rich seawater samples and marine organisms. Sci. 6079–6080. https://doi.org/10.1021/acs.est.8b02338.
Rep. 4, 8. https://doi.org/10.1038/srep04528. Rillig, M.C., Machado, A.A.D., Lehmann, A., Klumper, U., 2019. Evolutionary implica-
Dobaradaran, S., Schmidt, T.C., Nabipour, I., Khajeahmadi, N., Tajbakhsh, S., Saeedi, R., et tions of microplastics for soil biota. Environ. Chem. 16, 3–7. https://doi.org/
al., 2018. Characterization of plastic debris and association of metals with 10.1071/en18118.
microplastics in coastline sediment along the Persian Gulf. Waste Manag. 78, Rodrigues, M.O., Abrantes, N., Goncalves, F.J.M., Nogueira, H., Marques, J.C., Goncalves,
649–658. https://doi.org/10.1016/j.wasman.2018.06.037. A.M.M., 2018. Spatial and temporal distribution of microplastics in water and sedi-
Enders, K., Lenz, R., Beer, S., Stedmon, C.A., 2017. Extraction of microplastic from biota: ments of a freshwater system (Antua River, Portugal). Sci. Total Environ. 633,
recommended acidic digestion destroys common plastic polymers. ICES J. Mar. Sci. 1549–1559. https://doi.org/10.1016/j.scitotenv.2018.03.233.
74, 326–331. https://doi.org/10.1093/icesjms/fsw173. Rodriguez-Seijo, A., Lourenco, J., Rocha-Santos, T.A.P., da Costa, J., Duarte, A.C., Vala,
Eo, S., Hong, S.H., Song, Y.K., Lee, J., Lee, J., Shim, W.J., 2018. Abundance, composition, and H., et al., 2017. Histopathological and molecular effects of microplastics in
distribution of microplastics larger than 20 mu m in sand beaches of South Korea. En- Eisenia andrei Bouche. Environ. Pollut. 220, 495–503. https://doi.org/10.1016/j.
viron. Pollut. 238, 894–902. https://doi.org/10.1016/j.envpol.2018.03.096. envpol.2016.09.092.
Fischer, E.K., Paglialonga, L., Czech, E., Tamminga, M., 2016. Microplastic pollution in Scheurer, M., Bigalke, M., 2018. Microplastics in Swiss floodplain soils. Environ. Sci.
lakes and lake shoreline sediments - a case study on Lake Bolsena and Lake Technol. 52, 3591–3598. https://doi.org/10.1021/acs.est.7b06003.
Chiusi (central Italy). Environ. Pollut. 213, 648–657. https://doi.org/10.1016/j. Sighicelli, M., Pietrelli, L., Lecce, F., Iannilli, V., Falconieri, M., Coscia, L., et al., 2018.
envpol.2016.03.012. Microplastic pollution in the surface waters of Italian Subalpine Lakes. Environ. Pollut.
Fries, E., Dekiff, J.H., Willmeyer, J., Nuelle, M.T., Ebert, M., Remy, D., 2013. Identification of 236, 645–651. https://doi.org/10.1016/j.envpol.2018.02.008.
polymer types and additives in marine microplastic particles using pyrolysis-GC/MS Song, Y.K., Hong, S.H., Jang, M., Han, G.M., Jung, S.W., Shim, W.J., 2017. Combined effects of
and scanning electron microscopy. Environ. Sci. Process. Impacts. 15, 1949–1956. UV exposure duration and mechanical abrasion on microplastic fragmentation by
https://doi.org/10.1039/c3em00214d. polymer type. Environ. Sci. Technol. 51, 4368–4376. https://doi.org/10.1021/acs.
Fuller, S., Gautam, A., 2016. A procedure for measuring microplastics using pressurized est.6b06155.
fluid extraction. Environ. Sci. Technol. 50, 5774–5780. https://doi.org/10.1021/acs. Tang, G.W., Liu, M.Y., Zhou, Q., He, H.X., Chen, K., Zhang, H.B., et al., 2018. Microplastics
est.6b00816. and polycyclic aromatic hydrocarbons (PAHs) in Xiamen coastal areas: implications
Hendrickson, E., Minor, E.C., Schreiner, K., 2018. Microplastic abundance and composition for anthropogenic impacts. Sci. Total Environ. 634, 811–820. https://doi.org/
in western Lake Superior as determined via microscopy, Pyr-GC/MS, and FTIR. Envi- 10.1016/j.scitotenv.2018.03.336.
ron. Sci. Technol. 52, 1787–1796. https://doi.org/10.1021/acs.est.7b05829. Thompson, R.C., Olsen, Y., Mitchell, R.P., Davis, A., Rowland, S.J., John, A.W.G., et al., 2004.
Hodson, M.E., Duffus-Hodson, C.A., Clark, A., Prendergast-Miller, M.T., Thorpe, K.L., Lost at sea: where is all the plastic? Science 304, 838. https://doi.org/10.1126/
2017. Plastic bag derived-microplastics as a vector for metal exposure in terres- science.1094559.
trial invertebrates. Environ. Sci. Technol. 51, 4714–4721. https://doi.org/ Toumi, H., Abidli, S., Bejaoui, M., 2019. Microplastics in freshwater environment: the first
10.1021/acs.est.7b00635. evaluation in sediments from seven water streams surrounding the lagoon of Bizerte
Holmes, L.A., Turner, A., Thompson, R.C., 2014. Interactions between trace metals and (Northern Tunisia). Environ. Sci. Pollut. Res. Int. https://doi.org/10.1007/s11356-019-
plastic production pellets under estuarine conditions. Mar. Chem. 167, 25–32. 04695-0.
https://doi.org/10.1016/j.marchem.2014.06.001. Vedolin, M.C., Teophilo, C.Y.S., Turra, A., Figueira, R.C.L., 2018. Spatial variability in the con-
Horton, A.A., Walton, A., Spurgeon, D.J., Lahive, E., Svendsen, C., 2017. Microplastics in centrations of metals in beached microplastics. Mar. Pollut. Bull. 129, 487–493.
freshwater and terrestrial environments: evaluating the current understanding to https://doi.org/10.1016/j.marpolbul.2017.10.019.
10 Y. Zhou et al. / Science of the Total Environment 694 (2019) 133798

Wang, J.D., Peng, J.P., Tan, Z., Gao, Y.F., Zhan, Z.W., Chen, Q.Q., et al., 2017. Microplastics in Zhou, Q., Zhang, H.B., Fu, C.C., Zhou, Y., Dai, Z.F., Li, Y., et al., 2018. The distribution
the surface sediments from the Beijiang River littoral zone: composition, abundance, and morphology of microplastics in coastal soils adjacent to the Bohai Sea
surface textures and interaction with heavy metals. Chemosphere 171, 248–258. and the Yellow Sea. Geoderma 322, 201–208. https://doi.org/10.1016/j.
https://doi.org/10.1016/j.chemosphere.2016.12.074. geoderma.2018.02.015.
Zhang, G.S., Liu, Y.F., 2018. The distribution of microplastics in soil aggregate fractions in
southwestern China. Sci. Total Environ. 642, 12–20. https://doi.org/10.1016/j.
scitotenv.2018.06.004.