Process Safety and Environmental Protection 141 (2020) 9–17

Contents lists available at ScienceDirect

Process Safety and Environmental Protection

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

Removal behavior of microplastics using alum coagulant and its

enhancement using polyamine-coated sand
Nirmal Kumar Shahi a , Minsoo Maeng b , Donghyun Kim a , Seok Dockko a,∗
a
Department of Civil and Environmental Engineering, Dankook University, 152, Jukjeon-ro, Suji-gu, Yongin-si, Gyeonggi-do 16890, South Korea
b
Department of Environmental and Safety Engineering, Ajou University, 206, World cup-ro, Yeongtong-gu, Suji-gu, Yongin-si, Gyeonggi-do 16499, South
Korea

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

Article history: In recent years, microplastics (MPs) smaller than 100 ␮m have been reported to be abundant in bulk
Received 18 February 2020 water samples from drinking water treatment plants (WTPs). In this study, the removal behavior of MPs
Received in revised form 27 April 2020 for different sizes (10–100 ␮m), shape and surface morphology were investigated using alum coagulant
Accepted 11 May 2020
and alum combined with cationic polyamine-coated (PC) sand. For identification and quantification, MPs
Available online 17 May 2020
were stained with Nile Red dye and observed under a laser scanning fluorescent microscope. The removal
of MPs increases with increasing doses of alum up to 30 mg L–1 (70.7%). Further increases in the doses
Keywords:
resulted in a sharp decrease in the removal of MPs. The analysis for different sizes showed lower removal
Microplastics
Orthokinetic flocculation
of smaller MPs (10–30 ␮m) for all doses of alum. PC sand (500 mg L–1 ) combined with 20 mg L–1 of alum
Cationic polyamine-coated sand dose showed the highest removal (92.7%) of MPs. The removal was enhanced by 26.8% when compared
Physical characteristic with alum alone. The removal of MPs followed the behavior of elongated-rough (ER) > elongated-smooth
Fluorescent technology (ES) > spherical-rough (SR) > spherical-smooth (SS) and was supported by a flocculation kinetic study.
This study revealed that size, shape and surface morphology of MPs play important roles in the removal
of MPs from drinking WTPs.
© 2020 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

1. Introduction use of plastics. These discarded plastics are very difficult to remove
once they have entered the aquatic environment (Shim et al., 2016).
The tiny particles of plastic less than 5 mm are defined as MPs are heterogeneous in nature which differ in size, shape
microplastics (MPs) (Cole et al., 2011; Lin et al., 2018; Di and Wang, and chemical composition. MPs with different compositions have
2018) and can be divided into two categories based on their ori- been detected and identified in aquatic environments, including
gin, primary MPs and secondary MPs. Primary MPs are larger than MPs such as acrylics, polyamides, polyesters, polyethylenes (PE),
2.5 mm and are defined as MPs originally manufactured from the polypropylenes (PP) and polystyrenes (Di et al., 2018; Pivokonsky
raw plastic materials and contained in textiles, medicines, cosmet- et al., 2018; Lei et al., 2018). Different plastics have their own chem-
ics, blasting media, cleaning products, personal care products such ical compositions and polymer additives. The additives that are
as facial products and body scrubs, among others (Andrady, 2011; used are often composed of harmful substances which are very
Lambert and Wagner, 2018; Simon et al., 2018; Lee and Kim, 2018). hazardous (organic and heavy metal pollutants) to nature and may
Secondary MPs are MPs that are derived from larger fragments leach at very fast rates during the aging process of plastic debris
or large plastic debris of MPs by mechanical degradation, photo (Kowalski et al., 2016; Lei et al., 2018). However, depending on
or ultraviolet (UV) degradation, biodegradation, thermal-oxidative the physical, chemical and biological conditions of the receiving
degradation, hydrolysis and other mechanisms (Lin et al., 2018; Lee environments, the concentrations of leachable compounds vary
and Kim, 2018). Eriksen et al. (2013) reported that the majority of (Lambert and Wagner, 2018). The additives found in plastics are
MPs found in aquatic environments are of the secondary-MPs type polybrominated diphenyl ethers, phthalate and bisphenol A and
and would increase in abundance with increasing production and are known as endocrine disruptors that can mimic, compete with
and disrupt the synthesis of endocrine hormones. Besides this,
MPs are reported to act as excellent vectors or carriers of organic
pollutants and heavy metals which, upon ingestion, cause harm
∗ Corresponding author.
to aquatic organisms (Cole et al., 2011). Because of large surface
E-mail address: [email protected] (S. Dockko).
areas and strong hydrophobicity, MPs adsorb persistent organic

https://doi.org/10.1016/j.psep.2020.05.020
0957-5820/© 2020 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
10 N.K. Shahi et al. / Process Safety and Environmental Protection 141 (2020) 9–17

pollutants, i.e., polycyclic aromatic hydrocarbons, polychlorinated Table 1

Synthetic water characteristics.
biphenyls (Mizukawa et al., 2013; Lee and Kim, 2018) and heavy
metals such as Fe, Al, Mn, Zn, Cu, Pd and Ag (Ashton et al., DOC (mg L–1 ) UV254 (cm–1 ) Turbidity (NTU) pH Temperature (◦ C)
2010). However, in freshwater environments, MPs are supposed 2.4 ± 0.03 0.3 14.3 ± 0.5 7 ± 0.5 20 ± 2
to adsorb other emerging contaminants such as pharmaceuticals,
personal care products, flame retardants and other industrial chem-
icals (Lambert and Wagner, 2018). The presence of pharmaceutical
products such as antibiotics, antidepressants and beta-blockers in which corresponds to 15 mmol L–1 (Ma et al., 2019b). Additional
surface water has been reported (Khetan and Collins, 2007). The application of polyacrylamide (PAM) to the coagulation of PE parti-
research study by Razanajatovo et al. (2018) shows adsorbing of cles was also investigated. The applied PAM concentrations greatly
pharmaceutical products such as the antibiotic sulfamethoxazole, exceeded the maximum allowable dose of 1 mg L–1 (World Health
beta-blocker propranolol (PRP) and the antidepressant sertraline Organization, 2003). There is not much information available in the
(SER) on the surface of polyethylene MPs, and desorption of PRP literature about the problems resulting from MPs and their removal
and SER from MPs within 48 h. This indicates the potential risk of characteristics from drinking water sources based on size, shape
bioaccumulation of these pharmaceutical products once ingested and surface morphology.
by microorganisms in aquatic environments and in humans. How- The main purpose of this study was to investigate the removal
ever, the toxicological effects of MPs and the risks associated with behavior of MPs using a common coagulant, i.e., alum, in the
them are not fully understood and described. Further, the sorp- drinking water treatment process, and the enhanced removal by
tion of persistent organic pollutants and pharmaceutical products polyamine-coated (PC) sand. First, this study compared conven-
varies with the MPs nature, i.e., pristine and aged MPs. During the tional coagulant, alum, and alum coupled with PC sand (cationic
aging process of MPs, the molecular weight has been decreased and polymer additive to enhance floc formation) for the removal of
form oxygen-containing functional groups (−OH, C O and C OH) MPs, where the MPs varied in size, shape and surface morphol-
(Hüffer et al., 2018; Liu et al., 2019). The oxygen-containing func- ogy. Second, the study investigated the removal behavior of MPs as
tional groups form hydrogen bonds with the water molecules and characterized by their size (10–100 ␮m), their shape based on their
make sorption less favorable. Also, aging decreases the hydropho- aspect ratio (AR) (spherical and elongated) and surface morphol-
bicity and increase the polarity on MPs. Thus, sorption coefficients ogy (rough or smooth). Finally, the study compared the effect of
of organic pollutants (aliphatic and aromatic compounds) by aged these physical properties on coagulation and flocculation kinetics.
MPs were reported lower than pristine MPs (Hüffer et al., 2018). A novel Nile Red (NR) staining method to identify and quantify MPs
In recent years, the focus of research on MPs has been directed using fluorescent technology, combined with image analysis, was
toward freshwater environments. Several studies revealed the used for this research study.
presence of microscopic plastics debris, so-called MPs in freshwater
systems such as lakes, river beaches, surface water and sediments 2. Materials and method
of rivers and reservoirs (McCormick et al., 2016; Ivleva et al., 2017;
Lambert and Wagner, 2018; Pivokonsky et al., 2018). Particles with 2.1. Materials
sizes smaller than 100 ␮m are found in abundance in bulk water
samples (Su et al., 2016; Di and Wang, 2018). The number of iden- PE (0.91 g cm–3 with sizes smaller than 500 ␮m) was purchased
tified MPs in a liter of water ranges from less than 10 to more from Thermo Fisher Scientific, Shore Rode, Heysham, LA3 2XY,
than 4000 particles (Wang et al., 2018; Pivokonsky et al., 2018; England (Alfa Aesar) for use in this study. In this study, the sizes
Luo et al., 2019; Mintenig et al., 2019). This discrepancy resulted of the MPs ranged from 10 to 100 ␮m and the MPs were sepa-
from the different methodologies used for sampling, their anal- rated by stainless steel mesh. Humic acid (HA) was purchased from
ysis and the lower limit of the particle sizes (Li et al., 2018b; Sigma-Aldrich, Spruce Street, St. Louis, MO 63103 USA. The sand
Silva et al., 2018). Interestingly, the majority of the research stud- was collected from a river and sieved using stainless steel mesh to
ies showed higher abundance of MPs with the smaller size, e.g., obtain microsized particles (90–125 ␮m). The cationic polyamine
81–92% of MPs were observed to be 1–10 ␮m (Pivokonsky et al., (50%) polymer additive was gifted by Hansol Chemical, Korea. The
2018), 96% of MPs to be 4–20 ␮m (Triebskorn et al., 2019), 61% NR “N-3013” was purchased from Sigma-Aldrich (Seoul, Korea).
of MPs to be 10–300 ␮m (Leslie et al., 2017), more than 80% of The hydrophilic polycarbonate (PCTE) membrane black filter with
MPs to be 50–2000 ␮m (Wang et al., 2017) and 73% of MPs to be a pore size of 5 ␮m and diameter 47 mm was purchased form Ster-
50–500 ␮m (Yuan et al., 2019). In most of the studies, PE and PP litech corporation, USA.
are reported to be found abundantly in surface water (Triebskorn
et al., 2019), and are characterized as low-density MPs whose den-
sity is less than that of water, i.e., 0.91 and 0.94 g cm–3 , respectively. 2.2. Synthetic water and PC sand
The research study by Pivokonsky et al. (2018) shows the numbers
of MPs in the treated water from three drinking water treatment A stock solution of HA with the concentration of 1 g L–1 was pre-
plants (WTPs) with several techniques and methods located in the pared using deionized (DI) water and HA and was stored at 4 ◦ C in a
Czech Republic approximately 70–83% lower than source water. dark place. This stock solution was used for the preparation of syn-
Further, a study by Ma et al. (2019a, 2019b) under laboratory con- thetic water solution that resembled natural water. To maintain the
ditions used large-sized MPs which exceed the particle size for the background turbidity, kaolin was added. The characteristics of the
coagulation process. The PE particles used were in the range of synthetic water are listed in Table 1. Sand collected from a river was
0.5–5 mm. The coagulation process, in general, is used for aggre- washed several times with DI water, dried in an oven at 60 ◦ C for 1 h
gation of dissolved or colloidal particles (Ngo et al., 1995; Bai and transferred to a muffle furnace at 550 ◦ C for 2 h to ignite organic
and Tien, 1997; Pivokonsky et al., 2011). The removal of PE was solid waste present. The sand was then treated with 30% hydrogen
achieved 13.27% ± 2.19% by FeCl3 ·6H2 O coagulant at 112 mg L–1 Fe peroxide (H2 O2 ) for oxidation of the remaining organic pollutants
dose. The Fe dose used was extremely higher than that of nor- for 24 h. Next, the sand was washed several times with DI water
mally used in drinking WTPs (less than 20 mg L–1 Fe) (Shen et al., and kept in the oven at 60 ◦ C for 12 h. Finally, the required amount
2020). In addition, the removal of PE was achieved 36.27% ± 1.06% of sand was mixed with an optimum dose of cationic polyamine
by alum (AlCl3 ·6H2 O) coagulant at 405 mg L–1 Al (extremely high), and kept for 24 h at 23 ◦ C.
 N.K. Shahi et al. / Process Safety and Environmental Protection 141 (2020) 9–17 11

2.3. Characterization of MPs and PC sand meaning low solubility. A stock solution of 50 mg L–1 NR was there-
fore prepared in acetone and kept in an amber bottle at 4 ◦ C. The
The physical characteristics of MPs such as size, shape and sur- working solution of 5 mg L–1 was prepared by diluting the stock
face morphology were investigated through image analysis (see solution with n-hexane (10 times) and keeping it in a tightly sealed
Section 2.5). The size was measured based on Feret’s measurement amber bottle at 4 ◦ C. For staining, 5 mL of the supernatant was pipet-
(Fig. S1) where the longest distance between any points along the ted from the collected whole sample (after vigorous mixing) and
selected boundary is defined as Feret’s diameter (dF , ␮m) and the filtered through a hydrophilic PCTE membrane filter. The collected
minimal Feret’s distance which is calculated after considering all MPs on the surface of the membrane filter were stained with 200 ␮L
possible orientations from 0◦ to 180◦ is called the minimal Feret’s working NR solution. Then, the filter with the stained MPs was left
diameter (Min df , ␮m) (Mazzoli and Favoni, 2012). The MPs were to air dry for 5 min and the measurements were carried out.
divided into different Min df size categories ranging from 10 to
100 ␮m (10–30, 30–50, 50–70 and 70–100 ␮m). The shape of the 2.5.2. Microscopic imaging
MPs was measured using the parameter, AR, which is defined as Imaging was performed using an upright confocal laser scanning
the ratio of the longest to shortest diameter (Li et al., 2018a), as microscope (CLSM) (LSM700, Zeiss, Jena, Germany) equipped with
shown in the Eq. (1). The particles with AR value ranging from 1 to a high-resolution AxioCam microscope camera. The fluorescence
2.4 are considered to be spherical, from 2.4 to 5 are considered to measurement of the NR-stained MPs was performed at a mag-
be elongated (Sieving et al., 2003; Li et al., 2018a) and those with nification of 10× in green fluorescent protein with an excitation
AR values higher than 5 are considered to have fiber shapes. The wavelength of 488 nm and emission wavelength of 539 nm.
surface morphology of the MPs was measured based on the rough-
ness factor or the convexity of the particle which is defined as the
2.5.3. Image analysis
ratio of actual perimeter to the convex perimeter and is calculated
The particle recognition and quantification of the fluorescent
as shown in Eq. (2) (Janoo, 1998).
image of the MPs, based on their size, shape and surface mor-
dF phology, was carried out using ImageJ 1.8.0 (National Institutes of
AR = (1)
Min df Health, USA). The image analysis was carried out as per the steps
illustrated in Fig. S2 (available with the online version of this paper).
p
R= (2) The visualization of the image was carried out in the arrays of
pc 1024 × 1024 pixels resolution with a pixel size of 0.3 ␮m2 . The cap-
where R is the roughness factor or convexity, p is the actual perime- tured images were exported in tagged image file format. Because
ter and pc is the convex perimeter in ␮m of MPs. For smooth the captured images contained background noise, the image bright-
particles, R will approach the value of 1 and 0 for rough parti- ness and contrast were adjusted. Under this condition, the digitized
cles. Besides this, the zeta (␨) potentials in mV of MPs, sand and PC images were binarized, i.e., converted to black and white with a gray
sand were measured using NanoPlus, a zeta/nanoparticle analyzer value of 256. This often helps to enhance the image features for
(ELSZ-1000, USA). visualization. The gray value of 256 needs to be processed in such
a way to allow quantification in reduced gray values. This process
2.4. Experimental setup for settling process can be accomplished with the process called thresholding in which
the values are reduced to 0 and 1 for black and white, respectively.
During the test, predetermined doses of coagulant and floccu- Finally, the measurements were taken, exported to a spreadsheet
lation additives were added. The coagulated water sample pH was file and interpreted.
adjusted to 7 by the addition of 0.1 mol L–1 NaOH and HCl. A jar test
was performed with rapid mixing (200 rpm) for 1 min, subsequent 2.6. Theory of coagulation and flocculation kinetics
slow mixing (50 rpm) for 10 min and settling for 30 min for coag-
ulation, flocculation and sedimentation, respectively. The removal As per the traditional colloid chemistry theory, i.e., Derjaguin,
efficiency of MPs was tested for alum doses of 10, 20, 30, 40 and Landau, Verwey and Overbeek, the ␨ potential is the primary fac-
50 mg L–1 respectively. To enhance the removal of MPs and opti- tor that affects the stability of the colloids/particles. However, the
mize the alum dose, the tests were conducted for 0.5, 1 and 2 mg L–1 shape, size and surface texture of colloids/particles in the water
cationic polyamine additive. The cationic polyamine additive was treatment process have much to do with their stability. Thus,
added immediately after rapid mixing. From this, the optimum dose the shape, size and surface texture factors of the MPs cannot be
of cationic polyamine from the above experiment was used to coat neglected. Particles with the same volume but with different mor-
the sand. The test was performed with different mass concentra- phology factors have different rate constants of coagulation and
tions of PC sand, i.e., 300, 400 and 500 mg L–1 . The turbidity was depend in their action diameter, i.e., dF . For both orthokinetic and
analyzed using a turbidity meter (TURBI200ND, Hach Co., USA). The prokinetic coagulation reactions, the morphology factors have a
DOC was analyzed with a total organic carbon analyzer (TOC-L, Shi- great impact on the reaction rate constants (Jiang and Guan, 2006).
madzu Co., Japan). UV254 was measured using a UV spectrometer In traditional theories, the particles are considered to be spher-
(DR5000, Hach Co., USA). ical. For MPs larger than 1 ␮m the rate constant for the coagulation
reaction can be calculated as:
2.5. Identification and quantification of MPs kt = 2Min df G/3
3
(3)

2.5.1. Staining of MPs where kt (m3 s–1 ) is traditionally the orthokinetic rate constant and
The first and foremost step is the selection of the appropriate G (s–1 ) is the velocity gradient.
solvent to perform NR staining of the MP particles. Solvents such as The study conducted by Jiang and Guan (2006) showed that par-
ethyl acetate, acetonitrile, dichloromethane and toluene are found ticles with the same volume but with different shapes and surface
to degrade hydrophilic PCTE membrane filters, while methanol, morphologies displayed different rate constants for the coagulation
ethanol and cyclohexane cause decolorization of hydrophilic PCTE and flocculation reactions. Different shape and surface morphol-
membrane surfaces. The solvent n-hexane did not show any dam- ogy have different action radii. Because Eq. (3) does not provide for
age to the filters (Shim et al., 2016) but it did not dissolve the NR, shape and surface morphology factors, a volume expansion factor
12 N.K. Shahi et al. / Process Safety and Environmental Protection 141 (2020) 9–17

Fig. 2. ␨ potential value of sand, PS, PE and PC sand. Data for the PS from Dong et al.
(2020).

morphology (roughness factor or convexity) was carried out. As

mentioned in the methodology section above, if the value for the
roughness factor approaches 1, the surfaces of the particles will
be smooth and if approaches 0, the surface of the particles will
be rough. Data acquired from the image analysis showed similar
results for the surface morphology distribution as for size distribu-
tion. In Fig. 1b, 52% of the MPs were found to have smooth surfaces
whereas 48% had rough surfaces. These two characteristics, i.e.,
shape and surface morphology, directly affect the size of the parti-
cles which is directly linked with coagulation and the flocculation
Fig. 1. Physical characteristics of MPs with (a) shape distribution and (b) surface kinetics reaction. The overall characterization of MPs based on the
morphology for Min df values. combined shape and surface morphology is illustrated in Fig. S5.
The highest percentage distribution was observed for the spherical-
() is introduced, which is the ratio of volume (V1 ) of the sphere rough (SR) MPs, followed by spherical-smooth (SS) MPs. There is
with dF to the actual volume (V0 ) of the spherical particle Fig. S1. no significant difference in the distribution for the elongated MPs
 3 where MPs with elongated-smooth (ES) and elongated-rough (ER)
 = V1 /V0 = dF /Min df (4) characteristics were found to be the lowest.

The incorporation of Eq. (4) in Eq. (3) gives:

3.1.2. ␨ potential of MPs and PC sand
km = 2Min df3 G/3, (5) The primary mechanism involved in flocculation is charge neu-
tralization. Thus, the surface charge measurement was analyzed
where km (m3 s–1 ) is the modified orthokinetic rate constant with using NanoPlus, a zeta/nanoparticle analyzer. The analysis was car-
the shape and surface morphology factor. ried out for MPs, sand and PC sand, as illustrated in Fig. 2. The
␨ potential value for PE (this study) and PS (Dong et al., 2020)
3. Results and discussion was found to be –14.2 and –26.8 mV, respectively. Because MPs
have negatively charged surfaces, and their removal is reported to
3.1. Characteristics of MPs and PC sand be low in drinking water, it may be assumed hypothetically that
the cationic surfactant additive might enhance the removal of the
3.1.1. Physical characteristics of MPs MPs. Thus, the sand mixed with cationic polyamine showed a high
For shape, size and surface morphology analysis of MPs, the positive ␨ potential value of 67.2 mV whereas sand showed a low
images were acquired using CLSM, as illustrated in Fig. S3. From negative value of –45.8 mV.
the image analysis using ImageJ, of the total 208 MPs in the 5 mL of
synthetic raw water sample, 10% diameter (d10 ), median diameter 3.2. Removal behavior of MPs by conventional coagulant “alum”
(d50 ), and 90% diameter (d90 ) MPs was found to be 22.4, 12.4 and
76.3 ␮m, respectively (Fig. S4). The diameter was calculated based 3.2.1. Effect of different doses of alum on the removal of MPs
on the Feret’s analysis for this study because indirect advanced In drinking WTPs, the dose of alum used is normally less than
methods such as scattering, diffracted light or laser assume the par- 20 mg/L as Al (Ayekoe et al., 2017; Ma et al., 2019a). However,
ticles to be spherical, which is not predominantly the case in the the presence of MPs was reported in the effluent from the drink-
real natural environments for the particles. The study conducted by ing WTPs (Pivokonsky et al., 2018). Thus, to understand fully the
Mozzolo and Favoni (2012) showed that the equivalent diameter removal behavior of MPs, different doses of alum were used, and
values calculated are in the same value range as Min df . Similarly, the results are shown in Fig. 3. The removal of MPs increased with
the shapes of the MPs were analyzed, and the results show that increasing doses of alum up to 30 mg L–1 and was found to be 70.7%.
spherical MPs dominate the elongated ones. As shown in Fig. 1a, of However, the second-highest removal was observed for 20 mg L–1
the total MPs, 73% were spherical and 27% were elongated. Most with 65.9% followed by 40 mg L–1 (48.8%). For the coagulant doses of
of the MPs with small size, and in particular smaller than 30 ␮m, 20 mg L–1 and 30 mg L–1 , the restabilization of the MP particles does
are found in the spherical zone. Further, an analysis of the surface not occur, which could be because the increased rate of aluminum
 N.K. Shahi et al. / Process Safety and Environmental Protection 141 (2020) 9–17 13

Fig. 4. Removal behavior of different size MPs for different doses of alum at pH
Fig. 3. Removal of MPs for different doses of alum at pH 7 ± 0.5 and temperature
7 ± 0.5 and temperature 20 ± 2 ◦ C. The size of the MPs is given in minimal Feret’s
20 ± 2 ◦ C.
diameter (Min df , ␮m).

hydroxide precipitation may compete with the adsorption of incip- range of 30–50 ␮m, followed by 70–100 ␮m and 50–70 ␮m. The
iently formed soluble aluminum(III) species (Stumm and O’Melia, MPs in the size range 10–30 ␮m showed low removal, a result
1968; Sun et al., 2019). For this condition, sweep flocculation pre- which has also been found in previous research (Pivokonsky et al.,
dominates and facilitates the formation of bigger flocs. Although 2018). In addition, the abundance, or the concentration of the
the removal of MPs for 30 mg/L is high compared with other doses particles, affects the removal rate. However, a low particle con-
of alum, the removal efficiency is low. Further increases in the dose centration was observed for MPs in the size range of 50–70 ␮m in
of alum, in particular at 50 mg L–1 , resulted in sharp decreases in the raw water and in contrast, the second-highest concentration was
removal rate. This might be attributed to the restabilization of the found after pretreatment with an alum dose of 10 mg L–1 . As per the
surface charge of the MPs because aluminum hydroxide precipita- orthokinetic reaction, the removal of the MPs must increase with
tion is not supposed to exceed the adsorption of incipiently formed increasing size and the results for the dose of 10 mg L–1 are not in
soluble alum species. However, the charge neutralization mecha- line with this theory. The dF affects the aggregation because in the
nism is supposed to be dominant for low doses of alum (0.5 mM, flow velocity gradient, the larger axis, i.e., the action diameter dF ,
i.e., 13.49 mg L–1 as Al) (Sun et al., 2019). The removal of MPs for always orientates itself toward the flow velocity’s direction (Jiang
10 mg L–1 and 50 mg L–1 showed no significant difference. Further, and Guan, 2006). The greater the value for dF , the greater will be the
from the ␨ potential value which was –14.2 mV for the MPs, the ratio between dF and Min df . This will lead to an increase in the rate
addition of the alum dose must lead to high coagulation and floc- constant for the orthokinetic flocculation reaction. As illustrated in
culation rates for MPs and facilitate their removal. However, the Fig. S7, the ratio of dF to Min df for MPs in the size range of 30–50 ␮m
results obtained were not in line with the charge neutralization was found to be greater for almost all particles than for MPs of size
mechanism, which is explained in the sweep flocculation section 50–70 ␮m. This may have caused the lower removal of 50–70 ␮m
above. Although charge neutralization is the primary factor, partic- MPs.
ularly for low doses of alum, other factors such as size, shape and
surface morphology factor may have played important roles in the 3.2.3. Effect of shape and surface morphology on the removal
removal mechanisms. A previous study showed that besides charge behavior of MPs
neutralization, the shapes of the particles also affect the coagula- Fig. 5a. clearly shows the effect of the shape of MPs on their
tion and flocculation reaction kinetics (Jiang and Guan, 2006). The removal. Although the abundance of spherical MPs is almost three
shapes of the particles directly affect the force distribution that times higher than for elongated MPs in raw water, the removal of
surrounds them, which ultimately influences the coagulation and spherical MPs is lower for all alum doses than for the removal of
flocculation process. elongated MPs. As per the study conducted by Sun et al. (2019),
particles with higher concentrations showed higher removal per-
3.2.2. Effect of size on removal behavior of MPs centages than with lower concentrations. This might be because
As shown in Fig. 4 and Fig. S6, the removal of the MPs in of the similar shape characteristics of the particles. However, in
the smaller size range 10–30 ␮m was observed to be the lowest. this study, different shape characteristics were found to have a
The highest removal of 52% was observed for 30 mg L–1 followed noticeable effect. With the increase in the alum dose, the removal
by 20 mg L–1 (48%). MPs in the size range 30–100 ␮m were not of elongated-shape MPs sharply increased up to an alum dose of
detected for the 30 mg L–1 alum dose. For 20 mg L–1 , 40 mg L–1 and 20 mg L–1 . For the elongated-shape MPs, the maximum removal
50 mg L–1 , MPs in the size range 30–50 ␮m were detected, and of 81.8% was observed for both 20 mg L–1 and 30 mg L–1 of alum
removal was found to be 83.3%. With the increase in the alum dose. Further increases in the dose to 40 mg L–1 resulted in a grad-
dose above 30 mg L–1 , a sharp decrease in the removal of MPs was ual decrease in the removal of the MPs to a point where the removal
observed, specifically for the smaller size range of 10–30 ␮m. From remained stable. In contrast, the spherical-shape MPs showed a
the results, we observed that the removal of MPs increased with an gradual increase in removal of MPs with increasing doses up to
increase in the size for the higher dose of alum where sweep floccu- 30 mg L–1 . Further increases in alum dose showed a sharp decrease
lation is dominant. In contrast, for an alum dose of 10 mg L–1 where in the removal of MPs. The maximum removal of 66.7% was
charge neutralization is dominant for doses less than 13.49 mg L–1 observed for 30 mg L–1 alum, with the minimum removal of 33.3%
(0.5 mM as Al), the results show higher removal of MPs in the size for 50 mg L–1 . A possible reason for these results might be related to
14 N.K. Shahi et al. / Process Safety and Environmental Protection 141 (2020) 9–17

Table 2
Values of variables and constants for the coagulation and flocculation kinetics.

Variables and constants Value Unit

Impeller constant (k) 2 Flat, 2 blades single paddle

Rotational speed (N) 0.8 rps (revolutions per second)
Impeller diameter (D) 7.6 × 10–2 m
Kinematic viscosity at 20 ◦ C (v) 1.1 × 10–6 m2 s–1
Volume of tank (V) 1 × 10–3 m3
Velocity Gradient (G) 67.9 s–1

An interesting observation was made for SS and SR MPs. The abun-

dance of both SS and SR MPs was the same in raw water and after
pretreatment with alum dose, the removal of SR MPs was higher
than for SS MPs. A study conducted by Hassas et al. (2016) showed
that the particles with rough surfaces displayed strong attachment
compared with smooth particles. The rough-surface MPs are more
nonsymmetrical than the smooth-surface MPs, which results in a
difference in the forces acting on the MPs. Moreover, the orienta-
tion of these nonsymmetric MPs can accelerate the agglomeration
among nonsymmetrical particles because the dF of the MPs usu-
ally align themselves along the flow velocity’s direction (Jiang and
Guan, 2006). This clearly showed the effects of shape and surface
morphology of the MPs in coagulation and flocculation kinetics.
Overall, the removal of nonsymmetric MPs is almost 30–40% higher
than for symmetric MPs. Considering the abundance and removal
of MPs in treated samples, the results indicate the removal behavior
as ER > ES > SR > SS.

3.3. Coagulation and flocculation kinetics

The values used for parameters and constants are listed in

Table 2. The values obtained for rate constants km and kt using Eqs.
(5) and (3), respectively, are shown in Fig. 6a. As can be seen, the val-
ues obtained for the rate constant of kt for the same Min df are the
same (it is hypothesized that all particles are perfectly spherical),
which does not agree with the results obtained above. However,
MPs with the same Min df are not all spherical. For this, MPs with
the same Min df were carefully segregated and found to have differ-
ent shapes and surface morphologies for different df . From Eq. (5),
Fig. 5. Removal behavior of MPs with (a) different shapes and (b) different shapes
combined with surface morphology at pH 7 ± 0.5 and temperature 20 ± 2 ◦ C. SS
the third power of the ratio of dF to Min df is directly proportional
is spherical-smooth, SR is spherical-rough, ES is elongated-smooth, and ER is to the rate constant. The results of km for the MPs with the same
elongated-rough. Min df but with different shape and surface morphology are shown
in Fig. 6b. The results follow the order ER > ES > SR > SS, which is in
line with the result obtained above. This suggests that the rate con-
their surface area. Compared with spherically shaped particles, the stant must be higher for nonsymmetric MPs than for symmetric
surface area of elongated MPs is larger for particles with the same MPs. The higher the rate constant for the flocculation, the higher
volume because the dF is larger. Because the surface area of the the removal rate will be. In summary, the size, shape and surface
elongated MPs is larger, adsorption and interception of the MPs to morphology of the MPs played a significant role in the coagulation
the Al(OH)s(am) particles or floc is supposed to be more pronounced and flocculation mechanism.
compared with spherical-shape MPs. This might be the reason that
leads to the higher removal of elongated MPs for 20 mg L–1 and 3.4. Removal behavior of MPs by PC sand combined with alum
30 mg L–1 alum. However, with further increases in the doses of
alum (40 mg L–1 and 50 mg L–1 ), aluminum hydroxide precipitation 3.4.1. Effect of different doses of PC sand combined with alum
is not supposed to exceed the adsorption of incipiently formed sol- As shown in Fig. 7, no significant improvement was observed
uble alum species, and in particular Al(OH)4, because of the high with the addition of 300 mg L–1 of PC sand for both 20 mg L–1 and
dose of alum. This might have caused restabilization of the MPs 30 mg L–1 alum dose. The optimum dose of cationic polyamine
which ultimately leads to the decline in the removal of the MPs. added was 1 mg L–1 to coat sand particles (Fig. S9). Further increases
Further, the combined effect of shape and surface morphology of in the dose of PC sand showed an increase in the removal of MPs.
MPs was studied, and the results are shown in Fig. 5b. For symmetric For 400 mg L–1 PC sand, the removal of MPs was increased by 7%
MPs, lower removal was observed for SS MPs, than for nonsymmet- and 15% for 20 and 30 mg L–1 alum, respectively. Similarly, PC sand
ric MPs, i.e., ER, ES and SR for both 20 mg L–1 and 30 mg L–1 alum. with 500 mg L–1 combined with alum showed the highest removal
For 30 mg L–1 , nonsymmetric MPs showed a removal above 80% with 92.7% and 90.2% for 20 mg L–1 and 30 mg L–1 , respectively.
while the removal for symmetric MPs was below 50%. The max- When compared with alum alone, the removal was increased by
imum removal of 86.7% was observed for SR MPs. Although the 26.8% for 20 mg L–1 and 19.5% for 30 mg L–1 . Based on these results,
removal percentage is higher, the MPs count was higher than for 20 mg L–1 of alum combined with 500 mg L–1 of PC sand showed
ES and ER MPs (Fig. S8) because of their abundance in raw water. better efficiency. The enhanced removal may be because of the
 N.K. Shahi et al. / Process Safety and Environmental Protection 141 (2020) 9–17 15

Fig. 6. Effect of shape and surface morphology of MPs on coagulation and floccula-
tion kinetics. (a) Modified orthokinetic rate constant (km ) vs traditional orthokinetic
rate constant (kt ) and (b) km with the same Min df and different shape and surface
morphology.

Fig. 8. Removal of MPs by PC sand. (a) Different sizes, (b) different shapes and
(c) shapes combined with surface morphology at pH 7 ± 0.5, temperature 20 ± 2 ◦ C
and alum 20 mg L–1 . SS is spherical-smooth, SR is spherical-rough, ES is elongated-
Fig. 7. Removal of MPs for different doses of PC sand combined with alum (20 and smooth, and ER is elongated-rough.
30 mg L–1 ) at pH 7 ± 0.5 and temperature 20 ± 2 ◦ C.

3.4.2. Effect of PC sand combined with alum on different size,

shape and surface morphology MPs
Owing to the better removal efficiency of PC sand, further inves-
cationic polyamine polymers used which enhance the bridging tigations were performed based on the size, shape and surface
effect between floc, particles and MPs, and the high specific gravity morphology of MPs. For a 20 mg L–1 alum dose, MPs within the
sand that results in improved settling of the suspending floc MPs. range of 10–50 ␮m were detected and the removal of 30–50 ␮m
Further, with the use of PC sand, the alum dose was decreased by and 10–30 ␮m sized MPs were 83.33% and 48%, respectively. How-
50%. ever, no MPs in the size range 30–100 ␮m were detected for both
16 N.K. Shahi et al. / Process Safety and Environmental Protection 141 (2020) 9–17

400 and 500 mg L–1 of PC sand (Figs. 8a and S10). The MPs detected References
were in the range of 10–30 ␮m. For smaller MPs in the size range
of 10–30 ␮m, the removal was enhanced by 40% with the addi- Andrady, A.L., 2011. Microplastics in the marine environment. Mar. Pollut. Bull. 62,
1596–1605, http://dx.doi.org/10.1016/j.marpolbul.2011.05.030.
tion of 500 mg L–1 PC sand. The removal achieved was 88% for Ashton, K., Holmes, L., Turner, A., 2010. Association of metals with plastic production
the smaller MPs. This might be because of strong attractive forces pellets in the marine environment. Mar. Pollut. Bull. 60, 2050–2055, http://dx.
between negatively charged MPs and positively charged PC sand. doi.org/10.1016/j.marpolbul.2010.07.014.
Ayekoe, C.Y.P., Robert, D., Gone, L.D., 2017. Combination of coagulation-flocculation
This leads to the attachment of the MPs to the surface of the sand. and heterogeneous photocatalysis for improving the removal of humic sub-
Further, the polymer resulted in the bridging between MPs and stances in real treated water from Agbôriver (Ivory-Coast). Catal. Today 281,
sand, making agglomeration strong, compact and heavier. Here, the 2–13, http://dx.doi.org/10.1016/j.cattod.2016.09.024.
Bai, R., Tien, C., 1997. Particle detachment in deep bed filtration. J. Colloid Interface
PC sand acts as a collector, which collects negatively charged MPs Sci. 186, 307–317, http://dx.doi.org/10.1006/jcis.1996.4663.
and flocs present. Further investigations were carried out based on Cole, M., Lindeque, P., Halsband, C., Galloway, T.S., 2011. Microplastics as contam-
the shape of the MPs. As illustrated in Fig. 8b, both 400 mg L–1 and inants in the marine environment: a review. Mar. Pollut. Bull. 62, 2588–2597,
http://dx.doi.org/10.1016/j.marpolbul.2011.09.025.
500 mg L–1 PC sand showed significant increases in the removal of Di, M., Wang, J., 2018. Microplastics in surface waters and sediments of the three
elongated MPs. For spherical MPs, the removal was not significantly Gorges Reservoir, China. Sci. Total Environ. 616-617, 1620–1627, http://dx.doi.
higher for 400 mg L–1 PC sand than with alum alone. The removal org/10.1016/j.scitotenv.2017.10.150.
Dong, C.D., Chen, C.W., Chen, Y.C., Chen, H.H., Lee, J.S., Lin, C.H., 2020. Polystyrene
of spherical MPs was increased by around 7% for 400 mg L–1 PC
microplastic particles: In vitro pulmonary toxicity assessment. J. Hazard. Mater.
sand. However, the removal of spherical MPs for 500 mg L–1 PC 385, http://dx.doi.org/10.1016/j.jhazmat.2019.121575.
sand increased significantly. The removal was increased by 30% and Eriksen, M., Mason, S., Wilson, S., Box, C., Zellers, A., Edwards, W., Farley, H., Amato, S.,
found to be 90%. On the other hand, for the PC sand 500 mg L–1 test, 2013. Microplastic pollution in the surface waters of the Laurentian great lakes.
Mar. Pollut. Bull. 77, 177–182, http://dx.doi.org/10.1016/j.marpolbul.2013.10.
elongated MPs were not detected, and the removal was increased 007.
by around 18% compared with alum alone. Besides this, surface Hassas, B.V., Caliskan, H., Guven, O., Karakas, F., Cinar, M., Celik, M.S., 2016. Effect of
morphology is another factor that affects the coagulation and floc- roughness and shape factor on flotation characteristics of glass beads. Colloids
Surf. A Physicochem. Eng. Asp. 492, 88–99, http://dx.doi.org/10.1016/j.colsurfa.
culation. There was significant removal of rough MPs compared 2015.12.025.
with smooth. Rough MPs were not detected for 500 mg L–1 PC sand Hüffer, T., Weniger, A., Hofmann, T., 2018. Sorption of organic compounds by aged
combined with alum. On the other hand, the removal was observed polystyrene microplastic particles. Environ. Pollut. 236, 218–225, http://dx.doi.
org/10.1016/j.envpol.2018.01.022.
to be 85.7% for smooth MPs. The removal was increased by around Ivleva, N.P., Wiesheu, A.C., Niessner, R., 2017. Microplastic in aquatic ecosystems.
25%. From the above results, the bridging mechanism is predomi- Angew. Chem. Int. 56, 1720–1739, http://dx.doi.org/10.1002/anie.201606957.
nant for rough particles, as compared with smooth particles. Thus, Janoo, V., Special Report 98–1 Cold Regions Research and Engineering Laboratory
1998. Quantification of Shape, Angularity and Surface Texture of Base Course
to understand this, the combined effect of shape and surface mor- Materials. US Army Corps of Engineers, Hanover, NH https://rosap.ntl.bts.gov/
phology was investigated, and the results are shown in Fig. 8c. For view/dot/14065.
PC sand 500 mg L–1 , no MPs other than SS MPs were detected. When Jiang, Z., Guan, Y., 2006. Flocculation morphology: effect of particulate shape and
coagulant species on flocculation. Water Sci. Technol. 53, 9–16, http://dx.doi.
compared with alum alone, the removal of SS MPs was enhanced
org/10.2166/wst.2006.202.
by around 27% and found to be 80%. From this, it can be concluded Khetan, S.K., Collins, T.J., 2007. Human pharmaceuticals in aquatic environment: a
that the surface morphology plays an important role in bridging, challenge to green chemistry. Chem. Rev. 107, 2319–2364, http://dx.doi.org/10.
interception and agglomeration of particles. 1021/cr020441w.
Kowalski, N., Reichardt, A.M., Waniek, J.J., 2016. Sinking rates of microplastics and
potential implications of their alteration by physical, biological, and chemical
4. Conclusion factors. Mar. Pollut. Bull. 109, 310–319, http://dx.doi.org/10.1016/j.marpolbul.
2016.05.064.
Lambert, S., Wagner, M., 2018. Microplastics are contaminants of emerging con-
The removal behavior of MPs, using common coagulants in cern in freshwater environment: an overview. In: Wagner, M., Lambert, S. (Eds.),
drinking WTPs, was investigated. To improve the removal of MPs, Freshwater Microplastics, Hdb. Env. Chem. Springer, pp. 1–23, 58.
Lee, H., Kim, Y., 2018. Treatment characteristics of microplastics at biological sewage
cationic PC sand was used and studied. The removal of MPs using treatment facilities in Korea. Mar. Pollut. Bull. 137, 1–8, http://dx.doi.org/10.
alum only was found to be in the average range. However, PC sand 1016/j.marpolbul.2018.09.050.
combined with alum significantly enhanced the removal of MPs Lei, L., Wu, S., Lu, S., Liu, M., Song, Y., Fu, Z., Shi, H., Raley-Susman, K.M., He, D.,
2018. Microplastic particles cause intestinal damage and other adverse effects
by 26.8% compared with alum alone. In addition, with the addi-
in zebrafish Danio rerio and nematode Caenorhabditis elegans. Sci. Total Environ.
tion of PC sand, the alum dose was reduced by 50%. Further, results 619-620, 1–8, http://dx.doi.org/10.1016/j.scitotenv.2017.11.103.
obtained based on MPs’ physical characteristics such as size, shape Leslie, H.A., Brandsma, S.H., van Velzen, M.J.M., Vethaak, A.D., 2017. Microplastic-
and surface morphology showed noticeable effects on coagulation sen route: field measurements in the Dutch river delta and Amsterdam canals,
wastewater treatment plants, North Sea sediments and biota. Environ. Int. 101,
and flocculation processes. The MPs with small-smooth-spherical 133–142, http://dx.doi.org/10.1016/j.envint.2017.01.018.
characteristics showed low removal by both alum and alum com- Li, J., Wang, K., Li, Z., Tu, J.P., Jin, G., Su, J., Zhai, B., 2018a. Mechanical tests, wear
bined with PC sand. simulation and wear particle analysis of carbon-based nanomultilayer coatings
on Ti6 Al4 V alloys as hip protheses. RCS Adv. 8, 6849–6857.
Li, J., Liu, H., Chen, J.P., 2018b. Microplastics in freshwater systems: A review on
Declaration of competing interest occurrence, environmental effects, and methods of microplastics detection.
Water Res. 137, 362–374, http://dx.doi.org/10.1016/j.watres.2017.12.056.
Lin, L., Zuo, L.Z., Peng, J.P., Cai, L.Q., Fok, L., Yan, Y., Li, H.X., Xu, X.R., 2018. Occurrence
The authors have no conflicts of interest to declare. and distribution of microplastics in an urban river: A case study in the Pearl
River along Guanghou City, China. Sci. Total Environ. 644, 375–381, http://dx.
doi.org/10.1016/j.scitotenv.2018.06.327.
Acknowledgement Liu, G., Zhu, Z., Yang, Y., Sun, Y., Yu, F., Ma, J., 2019. Sorption behavior and mechanism
of hydrophilic organic chemicals to virgin and aged microplastics in freshwa-
This research study is supported by National Research Founda- ter and sweater. Environ. Pollut. 246, 26–33, http://dx.doi.org/10.1016/j.envpol.
2018.11.100.
tion (NRF-2018R1A2B2003059). Luo, W., Su, L., Craig, N.J., Du, F., Wu, C., Shi, H., 2019. Comparison of microplastic
pollution in different water bodies from urban creeks to coastal waters. Environ.
Pollut. 246, 174–182, http://dx.doi.org/10.1016/j.envpol.2018.11.081.
Appendix A. Supplementary data Ma, B., Xue, W., Ding, Y., Hu, C., Liu, H., Qu, J., 2019a. Removal characteristics of
microplastics by Fe-based coagulants during drinking water treatment. J. Envi-
Supplementary material related to this article can be found, in ron. Sci. China 78, 267–275, http://dx.doi.org/10.1016/j.jes.2018.10.006.
Ma, B., Xue, W., Hu, C., Liu, H., Qu, J., Li, L., 2019b. Characteristics of microplastic
the online version, at doi:https://doi.org/10.1016/j.psep.2020.05.
removal via coagulation and ultrafiltration during drinking water treatment.
020. Chem. Eng. J. 359, 159–167, http://dx.doi.org/10.1016/j.cej.2018.11.155.
 N.K. Shahi et al. / Process Safety and Environmental Protection 141 (2020) 9–17 17

Mazzoli, A., Favoni, O., 2012. Particle size, size distribution and morphological Sieving, A., Wu, B., Mayton, L., Nasser, S., Wooley, P.H., 2003. Morphological
evaluation of airborne dust particles of diverse woods by Scanning Electron characteristics of total joint arthroplasty-derived ultra-high molecular weight
microscopy and image processing program. Power Technol. 225, 65–71, http:// polyethylene (UHMWPE) wear debris that provoke inflammation in a murine
dx.doi.org/10.1016/j.powtec.2012.03.033. model of inflammation. J. Biomed. Mater. Res. A 64, 457–464, http://dx.doi.org/
McCormick, A.R., Hoellein, T.J., London, M.G., Hittie, J., Scott, J.W., Kelly, J.J., 2016. 10.1002/jbm.a.10368.
Microplastic in surface waters of urban rivers: concentration, sources, and asso- Silva, A.B., Bostos, A.S., Justino, C.I.L., da Costa, J.P., Duarte, A.C., Rocha-Santos, T.A.P.,
ciated bacterial assemblages. Ecosphere 7, e01556, http://dx.doi.org/10.1002/ 2018. Microplastics in the environment: Challenges in analytical chemistry -
ecs2.1556. A review. Anal. Chim. Acta 1017, 1–19, http://dx.doi.org/10.1016/j.aca.2018.02.
Mintenig, S.M., Löder, M.G.J., Primpke, S., Gerdts, G., 2019. Low number of microplas- 043.
tics detected in drinking water from ground water sources. Sci. Total Environ. Simon, M., Alst, N.V., Vollertsen, J., 2018. Quantification of microplastic mass and
648, 631–635, http://dx.doi.org/10.1016/j.scitotenv.2018.08.178. removal rates at wastewater treatment plants applying Focal Plane Array (FPA)-
Mizukawa, K., Takada, H., Ito, M., Geok, Y.B., Hosoda, J., Yamashita, R., Saha, M., based Fourier Transform Infrared (FT-IR) imaging. Water Res. 142, 1–9, http://
Suzuki, S., Miguez, C., Frias, J., Antunes, J.C., Sobral, P., Santos, I., Micaelo, C., dx.doi.org/10.1016/j.watres.2018.05.019.
Ferreira, A.M., 2013. Monitoring of a wide range of organic micropollutants on Stumm, W., O’Melia, C.R., 1968. Stoichiometry of coagulation. J. Am. Water Works
the Portuguese coast using plastic resin pellets. Mar. Pollut. Bull. 70, 296–302, Assoc. 60, 514–539 https://www.jstor.org/stable/41265308.
http://dx.doi.org/10.1016/j.marpolbul.2013.02.008. Su, L., Xue, Y., Li, L., Yang, D., Kolandhasamy, P., Li, D., Shi, H., 2016. Microplastics
Ngo, H.H., Vigneswaran, S., Dharmappa, H.B., 1995. Optimization of direct filtration: in Taihu Lake, China. Environ. Pollut. 216, 711–719, http://dx.doi.org/10.1016/j.
experiments and mathematical models. Environ. Technol. 16, 55–63, http://dx. envpol.2016.06.036.
doi.org/10.1080/09593331608616245. Sun, H., Jiao, R., Xu, H., An, G., Wang, D., 2019. The influence of particle size and
Pivokonsky, M., Bubakova, P., Pivokonska, L., Hnatukova, P., 2011. The effect of global concentration combined with pH on coagulation mechanisms. J. Environ. Sci.
velocity gradient on the character and filterability of aggregates formed during China 82, 39–46, http://dx.doi.org/10.1016/j.jes.2019.02.021.
the coagulation/ flocculation process. Environ. Technol. 32, 1355–1366, http:// Triebskorn, R., Braunbeck, T., Grummt, T., Hanslik, L., Huppertsberg, S., Jekel, M.,
dx.doi.org/10.1080/09593330.2010.536786. Knepper, T.P., Krais, S., Müller, Y.K., Pittroff, M., Ruhl, A.S., Schmieg, H., Schür,
Pivokonsky, M., Cermakova, L., Novotna, K., Peer, P., Cajthaml, T., Janda, V., 2018. C., Strobel, C., Wagner, M., Zumbülte, N., Köhler, H.R., 2019. Relevance of nano–
Occurrence of microplastics in raw and treated drinking water. Sci. Total Environ. and microplastics for freshwater ecosystems: a critical review. Trac-Trends Anal.
643, 1644–1651, http://dx.doi.org/10.1016/j.scitotenv.2018.08.102. Chem. 110, 375–392, http://dx.doi.org/10.1016/j.trac.2018.11.023.
Razanajatovo, R.M., Ding, J., Zhang, S., Jiang, H., Zou, H., 2018. Sorption and des- Wang, W., Ndungu, A.W., Li, Z., Wang, J., 2017. Microplastics pollution in inland
orption of selected pharmaceuticals by polyethylene microplastics. Mar. Pollut. freshwaters of China: A case study in urban surface waters of Wuhan, China.
Bull. 136, 516–523, http://dx.doi.org/10.1016/j.marpolbul.2018.09.048. Sci. Total Environ. 575, 1369–1374, http://dx.doi.org/10.1016/j.scitotenv.2016.
Shen, M., Song, B., Zhu, Y., Zeng, G., Zhang, Y., Yang, Y., Wen, X., Chen, M., Yi, H., 2020. 09.213.
Removal of microplastics vis drinking water treatment: Current knowledge and Wang, W., Yuan, W., Chen, Y., Wang, J., 2018. Microplastics in surface waters of
future directions. Chemosphere 251, http://dx.doi.org/10.1016/j.chemosphere. Dongting Lake and Hong Lake, China. Sci. Total Environ. 633, 539–545, http://
2020.126612. dx.doi.org/10.1016/j.scitotenv.2018.03.211.
Shim, W.J., Song, Y.K., Hong, S.H., Jang, M., 2016. Identification and quantification of World Health Organization, 2003. Acrylamide in Drinking-water: Background Doc-
microplsatics using Nile Red staining. Mar. Pollut. Bull. 113, 469–479, http://dx. ument for Development of WHO Guidelines for Drinking-water Quality. World
doi.org/10.1016/j.marpolbul.2016.10.049. Health Organization https://apps.who.int/iris/handle/10665/75373.