Marine Pollution Bulletin xxx (2016) xxx–xxx

MPB-08124; No of Pages 8

Contents lists available at ScienceDirect

Marine Pollution Bulletin

j ourna l homepage: www.e lsev ie r .com/ locate /marpo lbu l

Identification and quantification of microplastics using Nile Red staining

Won Joon Shim ⁎, Young Kyoung Song, Sang Hee Hong, Mi JangOil and POPs Research Laboratory, Korea Institute of Ocean Science and Technology, 41 Jangmok-1-gil, Geoje 53201, Republic of KoreaDepartment of Marine Environmental Sciences, Korea University of Science and Technology, Gajeong-ro, Daejeon 34113, Republic of Korea

⁎ Corresponding author at: Oil and POPs Research GScience and Technology.

E-mail address: wjshim@kiost.ac.kr (W.J. Shim).

http://dx.doi.org/10.1016/j.marpolbul.2016.10.0490025-326X/© 2016 Elsevier Ltd. All rights reserved.

Please cite this article as: Shim, W.J., et al., I(2016), http://dx.doi.org/10.1016/j.marpolbu

a b s t r a c t

a r t i c l e i n f o

Article history:Received 21 July 2016Received in revised form 15 October 2016Accepted 18 October 2016Available online xxxx

We investigated the applicability of Nile Red (NR), a fluorescent dye, for microplastic analysis, and determinedthe optimal staining conditions. Fivemg/LNR solution in n-hexane effectively stainedplastics, and theywere eas-ily recognized in green fluorescence. The NR staining method was successfully applied to micro-sized polyethyl-ene, polypropylene, polystyrene, polycarbonate, polyurethane, and poly(ethylene-vinyl acetate), except forpolyvinylchloride, polyamide and polyester. The recovery rate of polyethylene (100–300 μm) spiked topretreated natural sand was 98% in the NR stating method, which was not significantly (p b 0.05) differentwith FT-IR identification. The NR stainingmethodwas suitable for discriminating fragmented polypropylene par-ticles from large numbers of sand particles in laboratoryweathering test samples. Themethod is straightforwardand quick for identifying and quantifying polymer particles in the laboratory controlled samples. Further studies,however, are necessary to investigate the application of NR staining to field samples with organic remnants.

© 2016 Elsevier Ltd. All rights reserved.

Keywords:Nile Redstainingidentificationmicroplastics

1. Introduction

Microplastics have been recognized as important sources of marinecontamination and in recent years there has been increasing concernabout their levels in the environment (Shim and Thompson, 2015).They are widely distributed, from lakes to the open ocean (Imhof etal., 2013; Eriksen et al., 2014), in surface water and deep sea sediments(Song et al., 2015b; Woodall et al., 2014), and in various organismsthrough the trophic levels (Boerger et al., 2010; Murray and Cowie,2011; Van Franeker et al., 2011; De Witte et al., 2014; VanCauwenberghe and Janssen, 2014). Levels ofmicroplastics have been re-ported to have significantly increased over recent decades (Thompsonet al., 2004; Claessens et al., 2011). The density of microplastics general-ly increases with decreasing their size in beaches and surface waters(Martins and Sobral, 2011; Lee et al., 2013; Cozar et al., 2014). However,the isolation and subsequent identification of smaller microplastics be-come more difficult.

Microplastics are usually visually identifiedwith amicroscope. How-ever, the method has low reliability, especially for small, transparentand (or)fiber-type particles (Song et al., 2015a; Lenz et al., 2015). A Fou-rier Transform infrared (FT-IR) or Raman spectroscope equippedwith amicroscope has generally been used for chemical identification ofmicro-scale polymers, including qualitative confirmation of polymertypes (Song et al., 2014; Van Cauwenberghe et al., 2013). It is theoreti-cally possible to detect microplastics tens of microns in size using

roup, Korea Institute of Ocean

dentification and quantificatil.2016.10.049

micro-FT-IR and Raman spectroscopes (Lenz et al., 2015), but repeatedtrials are often required to obtain reliable spectra of very small andweathered plastic particles. Both the FT-IR and Raman spectroscopymethods require expensive instruments and are time-consumingwhen a large number of microplastic particles are present on the filterpaper. This is the case not only for environmental sampleswith complexmatrices but also for quantification of controlled laboratory experimen-tal samples. Therefore, it is becoming increasingly important to developalternative methods to facilitate identification of microplastics for fieldmonitoring and laboratory studies on the toxicity, accumulation,weathering, etc. of microplastics.

Plastics are comprised of hydrocarbons derived from petroleum,natural gas or biomass, and are hydrophobic. Lipophilic dyes can beused to visualize microplastics under a fluorescence microscope(Andrady, 2011). Nile Red (9-diethylamino-5H-benzo[α]phenoxazine-5-one; hereafter, NR), a fluorescent dye, has been used to stain neutrallipids in biological samples (Greenspan and Fowler, 1985) and can beused to stain synthetic polymers in polymer chemistry (Jee et al.,2009). NR is a hydrophobic fluorophore that specifically binds to neutrallipids and is strongly fluorescent only in the presence of a hydrophobicenvironment (Greenspan et al., 1985). We therefore decided to investi-gate the applicability of NR staining to the identification ofmicroplasticsamong other interfering particles, as an alternative method to the con-ventional microscopic and spectroscopic identification of syntheticpolymers.

We examined NR staining in 11 synthetic polymers. The NR carryingsolvents, optimum NR concentration, and excitation and emissionwavelengths for stained synthetic polymers were determined. In addi-tion, theNR stainingmethodwas comparedwith the FT-IR spectroscopy

on of microplastics using Nile Red staining, Marine Pollution Bulletin

2 W.J. Shim et al. / Marine Pollution Bulletin xxx (2016) xxx–xxx

method for the qualitative and/or quantitative identification ofmicroplastics in the spiked andweathering test samples in the laborato-ry and samples collected in the field.

2. Materials and Methods

2.1. Solvent test

A solvent that does not degrade the black polycarbonate (PC) filterpaper used for subsequent microscopic and FT-IR identification shouldbe selected to dissolve and carry NR for the staining of microplastics.NR is highly soluble and strongly fluorescent in a wide range of organicsolvents (Sitepu et al., 2012). A total of eight solvents –methanol, etha-nol, ethyl acetate, acetonitrile, dichloromethane, toluene, n-Hexane, andcyclohexane,−were tested. A few drops of each solvent were added toblack PC filter papers; the color change and deformation of shape of thefilter papers were observed.

2.2. Excitation and emission wavelengths

For convenience in the subsequent analyses, three major commonlight filters generally used in a fluorescent microscope to identifystained microorganisms were tested. Each NR-stained synthetic poly-mer was tested with three excitation and emission wavelengths: blue(excitation wavelength (ex.): 365 nm; emission wavelength (em.):445 nm), green-yellow (ex.: 450–490; em: 515–565 nm), and orange-red (ex.: 534–558; em.: N 590 nm) under a fluorescent microscope(Axio Scope A1, Zeiss, Jena, Germany).

2.3. Staining of different polymer types

NR (Sigma N-3013) was purchased as a dark purplish-red powderfrom Sigma-Aldrich (Seoul, Korea). Four micro-sized synthetic polymerparticles – low-density polyethylene (LDPE; size range b 200 μm), high-density polyethylene (HDPE; 80–200 μm), polyurethane (PU; b

100 μm), and poly(ethylene-vinyl acetate) (PEVA; 100–400 μm) –were purchased from Abifor, Ltd. (Zürich, Switzerland). Pre-productionresin pellets of polypropylene (PP), polyvinylchloride (PVC), and PCpurchased from a local market in Busan, Republic of Koreawere groundusing sand paper to make micro-sized particles. Expanded polystyrene(EPS) and polyethylene-tere-phthalate (PET) were ground into pow-dered samples from spherules taken from EPS floats and PET bottles, re-spectively. Polyester (PES) and polyamide (PA) fiber samples wereobtained by cutting a cloth and a stocking, respectively. The color ofLDPE, HDPE, PP, PU, and PEVA were translucent, and PC, PES and PAwere transparent, while EPS was white and PVC was black. Each poly-mer typewas confirmedwith a Fourier transform infrared (FT-IR) spec-troscope equipped with a microscope (Thermo Nocolet 6700 andContinuμm, Thermo Scientific, Waltham, USA) in attenuated total re-flection (ATR) mode (Song et al., 2014). Among the polymers tested,the NR staining conditions were examined first with the three commonplastic debris types, LDPE, PP and EPS, and themethodwas subsequent-ly applied to the other polymer types. NR staining of all the polymers didnot affect their identification with FT-IR spectroscopy.

Powder from each polymer was placed on separate black PC filterpapers and stained with 200 μL NR working solution. The particles andfilter paper surface were washed with 100 μL n-hexane, and the filterpaperswere air-dried for about 5min and examined under afluorescentmicroscope. None of visible damagewas detected in the plastics with n-hexane.

2.4. Recovery test

The NR stainingmethodwas tested with LDPE-spiked sand samples.Natural sand was collected from Heungnam Beach in Geoje, Korea andsieved to retain sand particles 63–1000 μm in size. Light particles

Please cite this article as: Shim, W.J., et al., Identification and quantificati(2016), http://dx.doi.org/10.1016/j.marpolbul.2016.10.049

including plastics mixed in the sand were removed by density separa-tion with saturated NaCl solution, repeating the density separationuntil no more particles were visible. The cleaned sand was pre-combusted overnight at 450 °C to remove the remaining plastics and or-ganic matter. Irregularly-shaped LDPE particles 0–200 μm in lengthwere sieved to select particle sizes between 100 and 300 μm. Then,about 1 mg of sieved LDPE was spiked to 50 g of the pretreated sandin a 500-mL separation funnel and mixed by shaking for several mi-nutes. Approximately 200 mL distilled water was added to the separa-tion funnel, and the funnel was vigorously shaken by hand for 1 minand left to stand for 6 h to allow the sand to settle. The supernatant inthe funnel was transferred to a pre-cleaned beaker, and the same sepa-ration procedure was repeated twice. The collected supernatant was fil-tered through black PC membrane filter paper (47-mm Ø, 1.2-μm poresize). The dried LDPE particles on the filter paper were stained usingthe above-described method. The stained LDPE particles were countedunder a fluorescence microscope and collected to measure their totalweight to calculate the LDPE recovery rate. The same samples were an-alyzed using a FT-IR spectroscope to compare the LDPE counts obtainedfrom the two methods.

2.5. Staining of weathering test samples

PP pellets that had been exposed to ultraviolet light under a metalhalide lamp in a chamber for 6 months were mixed with the pretreatednatural sand in a 250-mL amber bottle using three replicates and tum-bled in a roller mixer at 35 rpm for 2months to determine the fragmen-tation of the PP to micro-sized particles. After 2 month of mechanicalabrasion, the sand containing the PP particles was mixed thoroughlywith 330 mL saturated NaCl solution, vigorously shaken by hand for1 min, and then left to stand for 10 min. This process was repeatedthree times, and the supernatant was filtered using a black PC mem-brane filter (1.2 μm; 47-mm Ø). The filter was dried in a glass Petridish at room temperature and stained with NR solution. Then, the oc-currence and characteristics of staining of the PP particles with NRwere determined.

2.6. Staining of field samples

Sand samples were collected from Heungnam Beach from a0.5 × 0.5 m quadrat of 5-cm depth, using a stainless scoop, through a1-mm sieve (Tyler sieve; Cisa, Barcelona, Spain). The sieved sand sam-ples were density-separated with saturated NaCl solution using theabove-described method, and the supernatants were collected in a 1-Lbeaker and left to stand until the overlying water appeared clear. Thesupernatant was filtered using a black PC membrane filter (1.2 μm;47-mm Ø). The filter paper was soaked with hydrogen peroxide(H2O2) solution (35%) for 1 d to remove natural organic matter. Then,thefilter was dried in a glass Petri dish at room temperature and stainedwith NR solution. The occurrence and characteristics of staining of theplastic particles with NR were qualitatively determined.

Floating microplastics were sampled at three stations along thesouthern coast of Korea using aManta trawl (0.4 × 0.195-m rectangularopening, 3m long; 330-μmmesh net) (Kang et al., 2015). After recover-ing the net, it was carefully washed from the outside with pre-filteredseawater into a cod-end bucket. The contents of the bucket were trans-ferred to a 1-L sample bottle. In the laboratory, the 1-L sample waspoured in a 0.3-mm stainless steel sieve and rinsed with distilledwater to remove the salt. The sieved samples from each sieving weretransferred into beakers using a spatula and minimally rinsed with dis-tilledwater. All samples were dried in a drying oven at 60 °C for 24 h. Toremove the organic matter, the wet peroxide oxidation method wasfollowed, using aqueous 0.05 M Fe(II) solution and 35% H2O2 (Masuraet al., 2015). After oxidation of the natural organic material, the solutionwas transferred in a glass funnel to the density separation, adding NaCland settled overnight. The supernatant was filtered using a PC

on of microplastics using Nile Red staining, Marine Pollution Bulletin

3W.J. Shim et al. / Marine Pollution Bulletin xxx (2016) xxx–xxx

membrane filter (5 μm; 47-mm Ø). The plastics collected on the filterwere stained with NR and identified under a fluorescence microscope.In addition, they were also separately identified using FT-IRspectroscopy.

3. Results

3.1. Optimum staining conditions for PE, PP, and EPS

It is important to select a suitable solvent to carry NR, for stainingwithout degrading the black PC filter paper on which microplastics areidentified using a microscope or FT-IR spectrometer. Among the eightsolvents tested, the black PC filter paper was highly degraded by ethylacetate, acetonitrile, dichloromethane, and toluene, and its color waschanged by methanol, ethanol and cyclohexane (Fig. S1). However, n-hexane did not cause any significant degradation of the filter paper,and therefore it was selected as our NR working solution. However, itwas necessary to prepare theNR stock solution in acetone (0.05 g/L) be-cause of the low solubility of NR in n-hexane, and it was stored in atightly sealed amber glass container at 4 °C. This stock solution was di-luted 10 and 100 times with n-hexane to make NRworking solutions of0.5 and 5 mg/L, respectively.

Three excitation and emission wavelengths were investigated foridentification of theNR-stained plastics under a fluorescentmicroscope.The NR-stained LDPE, PP and EPS particles were visible in both greenand red fluorescence (Fig. 1), but there was no response in blue ateach excitation and emission wavelength. Green fluorescence waswell-recognized for all three polymer types without background stain-ing. Although redfluorescence could be recognized from the plastic par-ticles, staining of the background prevented clear identification ofpolymer particles on the filter paper. Therefore, green fluorescencewas used for subsequent analyses.

Two NR working solutions, 0.5 mg/L and 5 mg/L in n-hexane, wereinvestigated for staining PE, PP and EPS particles. Three polymer

Fig. 1. Photos under amicroscope (left) and a fluorescencemicroscopewith excitation and emisfor the Nile Red stained low density (a) polyethylene (LDPE), (b) polypropylene (PP) and (c)legend, the reader is referred to the web version of this article.)

Please cite this article as: Shim, W.J., et al., Identification and quantificati(2016), http://dx.doi.org/10.1016/j.marpolbul.2016.10.049

particles were stained with 0.5 mg/L NR solution, but their green fluo-rescence was too weak for clear identification (Fig. 2a–c). However,stainingwith 5mg/LNR solution provided sufficiently bright greenfluo-rescence for PE, PP and EPS particles (Fig. 2d–f).

3.2. Staining other polymers

Eight other polymers were stainedwith 5mg/L NRworking solutionin n-hexane and examined in green fluorescence on black PC filterpaper. HDPE, PC, PU, and PEVA were effectively stained and identifiedunder the given staining condition (Fig. 3a–d). However, the groundblack-colored PVC particles only appeared as a dim green glow (Fig.3e), and PES stained as a barely detectable dim glow (Fig. 3f), likewisefor PET and PA, which were observable with blue fluorescence withoutNR staining (Fig. 3g–i).

3.3. Recovery test

The LDPE particles (100–300 μm) that had been spiked to thepretreated natural sand were analyzed using the NR staining (n = 3)and FT-IR spectroscopy methods to determine and compare the recov-ery rates for both methods. The NR-stained LDPE particles were distinctand easily identifiable from other interfering particles on the filter pa-pers (Fig. S2). The recovery rate of the spiked LDPE particles was98 ± 3% on a weight basis with the NRmethod, which was very similarto the rate (96 ± 7%) obtained using the FT-IR spectroscopy method(Table 1). The number of recovered LDPE particles was 153 ± 4 forthe NR method, and 150 ± 3 for FT-IR method, which were not signifi-cantly (p b 0.05; Student's t-test) different. The NR stainingmethodwasfaster (b 1 h) for counting the LDPE particles than the FT-IR method (N2 h), in which each individual particle has to be counted separatelyusing an attenuated total reflection (ATR) probe. Control samples(n = 3) of the pretreated sands only had 44, 45, and 66 NR-stained

sionwavelength of 534–558 and 515–565 nm(middle) and 534–558 and N590 nm(right)expanded polystyrene (EPS). (For interpretation of the references to colour in this figure

on of microplastics using Nile Red staining, Marine Pollution Bulletin

Fig. 2. Comparison of green fluorescence from (a) low density polyethylene (LDPE), (b) polypropylene (PP) and (c) expanded polystyrene (EPS) stainedwith 5mg/L Nile Red solution and(d) LDPE, (e) PP and (f) EPS with 50 mg/L solution. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

4 W.J. Shim et al. / Marine Pollution Bulletin xxx (2016) xxx–xxx

particles (Fig. S2), but all of these particles had a maximum length ofb50 μm and they were all discernible with the spiked LDPE particles.

3.4. Application to laboratory weathering test samples

Ultraviolet light exposure and subsequentmechanical abrasion of PPpellets with sand resulted in a complex mixture of particles even afterdensity separation (Fig. 4a). The similar irregular shapes, sizes and tex-tures of both the sand and PP particles prevented identification of the PPparticles both under the microscope and with the micro-FT-IR spectro-scope.We quantitatively identified only several tens of PP particleswithmultiple trials with an ATR probe for several hours. However, when thesample was stained with NR, fragmented micro-sized PP particles were

Fig. 3.Nile Red stained (a) high density polyethylene, (b) polycarbonate, (c) polyurethane, (d)and non-stained (g) polyester (h) polyethylene tere-phthalate and (i) polyamide in blue fluorreferred to the web version of this article.)

Please cite this article as: Shim, W.J., et al., Identification and quantificati(2016), http://dx.doi.org/10.1016/j.marpolbul.2016.10.049

spotted and easily recognized in green fluorescence (Fig. 4b). These NR-stained particles were then confirmed as PP using FT-IR spectroscopy.The mean total number (± SD) of NR-stained PP particles in three rep-licates was 1658 ± 386 particles/bottle after subtracting number of theNR-stained particles found in the control samples (52 ± 12 particles/bottle). The fragmented PP particles b300 μm in maximum lengthaccounted for 98.7% of the total number of particles.

3.5. Application to field samples

Theparticles in thefield sand sampleswere stained and qualitativelyidentified in the laboratory using the samemethodused for the polymertests. The microplastic extraction from the sediment samples using

poly(ethylene-vinyl acetate), (e) polyvinyl chloride and (f) polyester in green fluorescenceescence. (For interpretation of the references to colour in this figure legend, the reader is

on of microplastics using Nile Red staining, Marine Pollution Bulletin

Table 1Recovery of the spiked low density polyethylene particles (100–300 μm) to the pretreatedsand quantified with Nile Red staining and fluorescent microscope method and FTIR-ATRmethod.

Replicate No. No. of particlesWeight (mg)Recovered / spiked

Recovery rate(%)

Nile Red staining method1 153 0.972 / 1.020 952 150 0.953 / 0.990 963 157 1.010 / 1.000 101Mean ± SD 153 ± 4 0.978 ± 0.029 / 1.003 ± 0.015 98 ± 3

FTIR-ATR method1 156 0.998 / 1.107 982 150 0.898 / 1.010 893 147 1.018 / 1.000 102Mean ± SD 150 ± 3 0.971 ± 0.064 / 1.009 ± 0.009 96 ± 7

5W.J. Shim et al. / Marine Pollution Bulletin xxx (2016) xxx–xxx

saturated NaCl solution in this study had limitation to recover rela-tively dense polymers such as polyvinylchloride and polyesters. PE,PP and EPS were mostly detected polymers in the sediment samples.We examined the co-staining of residual natural organic particles byNR after H2O2 treatment. Although the particles were separated bydensity and treated with H2O2, some inorganic and organic particlesremained on the filter paper with the microplastics. Plant debrisremnants were not stained by NR, but a NR-stained micro-sized par-ticle was well-recognized (Fig. 5a) and confirmed as EPS in subse-quent FT-IR analysis. An amphipod carapace remaining on the filterpaper was also not stained by NR (Fig. 5b). Some organic particlessuch as fish egg shells (Fig. S3) were not fully removed with H2O2

treatment, and they were slightly stained with NR showing dimgreen fluorescence (Fig. 5c). However, unidentifiable transparentplastic particles were revealed with NR staining and clearly observedunder a fluorescent microscope (Fig. 5c).

Three neuston net samples collected from three different baysalong the southern coast of Korea were quantitatively analyzed andcompared using both the FT-IR spectroscopy and NR stainingmethods. The total counts on the filter papers from three sampleswere 13, 3, and 71 for FT-IR spectroscopy, and 14, 3, and 97 for NR(Table 2). The numbers of microplastics were fairly similar for bothmethods in samples collected from Deukryang and Gwanyang, butthe NR method counted about 1.4 times more particles than the FT-IR method in the sample collected from Busan. After observationand counting of the NR-stained particles, some of the stained parti-cles were identified by spectroscopy (Fig. 6). In the three field sam-ples, PE, PP, EPS, PES, and PU particles were effectively stained withNR, but some particles with strong green fluorescence were notidentified as synthetic polymers (Fig. S4).

Fig. 4. Identification of polypropylene fragments (arrows) produced by six-month UV exposure(b) a fluorescence microscope after Nile Red staining. (For interpretation of the references to c

Please cite this article as: Shim, W.J., et al., Identification and quantificati(2016), http://dx.doi.org/10.1016/j.marpolbul.2016.10.049

4. Discussion

4.1. NR staining condition

Staining of microplastics with NR was performed on a filter paper inthe final stage of observation. Staining on filter paper can increase thecontact frequency of NR molecules and microplastic surfaces, and re-quires less NR solution compared to staining in an aqueous phase. In ad-dition, the volume of the NR solution added to the sample can be easilycontrolled. However, organic NR-carrying solvents may cause de-color-ation and/or physical damage of themembrane filter papers, which pre-vents subsequent microscopic and FT-IR identification of microplasticson the filter paper. Many microplastics from water and sediment aretransparent or white in color, whichmake their visual identification dif-ficult on white filter paper (Song et al., 2015a). The complex glass fibermatrix and cushion produced by filters made of thicker fibers (e.g., GF/F) also interfere so that it is difficult to obtain a clear spectrum of smallmicroplastic particles using a micro-FT-IR spectroscope with an ATRprobe. A thin black PC filter paper can provide a flat surface anddark background for identification of microplastics on the paper.Microplastics should be stained with NR in organic solvent on a PC filterpaper without damaging the paper. In this study, NR working solutionin n-hexane diluted 10 times from acetone-based stock solution wassufficiently effective for staining various synthetic polymers withoutde-coloration or physical damage on the black PC filter paper. Whilepolar solvents such as acetone, ethanol, ethylene glycol and dimethylsulfoxide are applied to facilitate penetration of the NR dye into cellsand to reduce the background staining of the cell membrane (Chen etal., 2011), they are not necessary for microplastic samples. Due to therelatively polar nature of NR molecules compared to plastics, thepartitioning of NR molecules from the carrier solvent to plastics can bemore facilitated in non-polar carrier solvents such as n-hexane than inpolar solvents.

4.2. NR staining of different polymer types

Small differences in the excitation and emission wavelengths of NR,in neutral lipid staining in biological samples and polymers, have beenreported depending on the polarity of the solvent and target lipid andthe staining conditions (Greenspan and Fowler, 1985; Dutta et al.,1996). Yellow-red fluorescence has been used mainly for lipid stainingin various biological samples (Diaz et al., 2008). The peak emission fluo-rescence shifted from red to green with increasing hydrophobicity ofthe solvent and target staining lipid (Greenspan et al., 1985). In thisstudy, green emission fluorescence was sufficiently clear to enable ef-fective identification of LDPE, HDPE, PP, EPS, PC, PU, and EVA after NRstaining. It was reported that NR showed fluorescent peak at 576 nm

and subsequent two-month mechanical abrasion with sands under (a) a microscope andolour in this figure legend, the reader is referred to the web version of this article.)

on of microplastics using Nile Red staining, Marine Pollution Bulletin

Fig. 5. Identification of microplastics density separated from beach sand samples under amicroscope (left) and a fluorescence microscope (right) after Nile Red staining: (a) anexpanded polystyrene (EPS) particle with plant remnants, (b) an amphipod carapaceand (c) a polyethylene particle among organic remnants. (For interpretation of thereferences to colour in this figure legend, the reader is referred to the web version ofthis article.)

6 W.J. Shim et al. / Marine Pollution Bulletin xxx (2016) xxx–xxx

in n-hexane solution (Elsey et al., 2007). Yellow-red fluorescence wasdetected with the NR-stained polymers, but background staining ofthe filter paper prevented clear identification of microplastics. The in-tensity of green and red fluorescence varied among the stained poly-mers. There were significant differences in the absorption andemission band positions and intensities as a function of the polarity ofthe medium, which can be attributed to the twisted intramolecularcharge transfer state of the NR molecule in polar media (Dutta et al.,1996). It has been shown that (poly)methyl methacrylate (PMMA)and (poly)vinyl alcohol (PVA) have different peak emission wave-lengths with NR staining (Dutta et al., 1996). Relatively less dense PE,PP and PS, which are dominant polymer types in microplastics foundon beaches and in surface water (Hidalgo-Ruz et al., 2012), were effec-tively stained with NR in the present study. More dense PES and PAmicrofibers, which have high crystallinity, were only slightly, or not atall, stained with NR (Figs. 3f–i and 6e) in this study. These microfibers,which include acrylic, have been reported to dominate sea bottom

Table 2Comparison of microplastic abundances in the surface water samples quantified by FT-IRand Nile Red staining methods.

Sample

MicroplasticsTotal counts per filter paper /abundance (particles/m3)

FT-IR / NR ratioFT-IR Nile Red

Deukryang 13 / 0.64 14 / 0.59 1.07Gwangyang 3 / 0.10 3 / 0.10 1.00Busan 71 / 2.72 97 / 3.72 0.73

Please cite this article as: Shim, W.J., et al., Identification and quantificati(2016), http://dx.doi.org/10.1016/j.marpolbul.2016.10.049

sediment, water column and biota (Browne et al., 2010; Lusher et al.,2013; Woodall et al., 2014). In our study, polymers of large surfacearea, with a mesoporous structure of expanded foam (Fig. 1c) or irregu-lar shape produced by sand grinding (Fig. 1b), tended to be stainedbrightly with NR compared to more spherical PE (Fig. 1a). The color ofplastics can influence the intensity of emission. In addition, theweathering of plastics by photo-oxidation can increase the polarity ofthe plastic surface over time, and subsequent physical weathering canchange the surface roughness and stainable surface area. Further studiesare required to determine how the peak emission wavelength and in-tensity of fluorescence emission varies among different polymers, andaccording to crystallinity, weathering status, shape and color withinthe same polymer.

4.3. Application to laboratory controlled samples

In studies on the accelerated weathering of plastics, it is essential toidentify and quantify the microplastic fragments that are produced.Large numbers of micro- and nano-sized particles were observed onthe filter paper under a microscope and scanning electron microscope.Most of thesewere identified as inorganic particles frompretreated nat-ural sand (data not shown) that could not be separated by density.These inorganic particles greatly outnumbered the fragmented PP parti-cles interfering in identification using FT-IR spectroscopy. Our NR stain-ing method was effective in selectively staining PP fragments on filterpapers and enabled their quantification among a very large number ofinorganic particles. Thus, the NR staining method is useful formicroplastic identification in controlled samples in laboratory experi-ments. The method could provide an effective and convenient meansof identifying microplastics mixed with inorganic materials that inter-fere with their identification in conventional methods. Unknown, verysmall NR-stained particles were found in the pretreated sand samplesin the recovery test in our study (Fig. S2). These particles are thoughtto be remnants of previously existing plastic or organic particles innatural sand, which remain even after pretreatment or contaminationduring sample processing. Therefore, it is necessary to examine NR-stained control samples together with the target samples in laboratoryexperiments.

4.4. Usefulness of NR staining

Many of the plastic-like particles found in sediment- and organic-material-rich neuston net samples are of unknown type, and it is diffi-cult to identify each individual plastic-like particle using microscopyor spectroscopy (Song et al., 2015a). When the particle size falls below200 μm, the effort required to identify them increases significantly dueto interference from the matrix in which they are situated. The differ-ence in the size of microplastics between field monitoring (e.g., N300 μm) and laboratory toxicity tests (e.g., b 100 μm) presents a chal-lenge for the assessment of the ecological risks of microplastics (Shimand Thompson, 2015). Improvements in the current methods of identi-fication are required for effective monitoring of microplastics using 50-and 80-μmmesh nets and bulk water sampling (Song et al., 2014). Fur-thermore, plastic particles that are covered by other non-plastic solidson a particle-overloadedfilter paper can also bemissed in counting pro-cedures (Fig. 5c). In this scenario, NR staining would be useful to spothidden microplastics and help for further spectroscopic analysis.When a fluorescent filter is installed on an FT-IR microscope, spectro-scopic identification can be confirmed for the same particles immedi-ately after fluorescence microscopy. A combination of fluorescencemicroscopy after NR staining followed by FT-IR confirmation would re-duce the likelihood of missingmicroplastics in the identification of fieldsamples, as well as the time needed to check every plastic-like particleby spectroscopy.

on of microplastics using Nile Red staining, Marine Pollution Bulletin

Fig. 6. Identification of various microplastics from neuston net samples under a microscope (left), a fluorescence microscope (middle) and a FT-IR (right) after Nile Red staining: (a-b)polyethylene, (c) polypropylene (polyethylene:polypropylene), (d) polystyrene, (e) polyester, and (f) polyurethane. (For interpretation of the references to colour in this figurelegend, the reader is referred to the web version of this article.)

7W.J. Shim et al. / Marine Pollution Bulletin xxx (2016) xxx–xxx

4.5. Limitation of NR staining

One of the main limitations in applying the NR staining method tofield samples is the co-staining of natural organicmaterial. It is thereforeimportant to remove natural lipids andorganicmatter from the samplesbefore NR staining. Some organicmatter and natural polymerswere notfully oxidized by H2O2 in this study and this has also been reportedusing other chemical treatments, such as strong acids and bases, and en-zymes (Claessens et al., 2013; Cole et al., 2014). Some of the organicremnants of plants and animals whose origins were identifiable fromtheir shapewere not stained at all orwere only slightly stained, showinga dim glow in our study. However, some remnant particles were effec-tively stained with NR, but could not be identified with microscopyand spectroscopy (Fig. S4). These particleswere thought to be refractory

Please cite this article as: Shim, W.J., et al., Identification and quantificati(2016), http://dx.doi.org/10.1016/j.marpolbul.2016.10.049

organic matter, although the possibility that they were unknown newlysynthetized polymers not included in the FTIR spectrum library cannotbe ruled out. In addition, some highly weathered polymers are also dif-ficult to match with the spectrum library. Although in our study the NRstaining and FT-IR methods produced comparable microplastic abun-dances within a factor of two in neuston net samples, NR-stained un-identified particles increase the risk of overestimating microplasticabundances in environmental samples. Furthermore, the number,shape, and type of NR-stained organic remnants can vary from sampleto sample. There is currently nomethod for removing organic materialsother than plastics from environmental samples that can guaranteecomplete elimination of “false” NR staining natural particles. Thus, atpresent, it is not recommended to use the NR staining method alonefor quantifying microplastics in field samples.

on of microplastics using Nile Red staining, Marine Pollution Bulletin

8 W.J. Shim et al. / Marine Pollution Bulletin xxx (2016) xxx–xxx

Acknowledgement

The authors thank Jung-Hoon Kang at Korea Institute of Ocean Sci-ence and Technology for providing the neuston net samples. Thisstudy was supported by research projects titled “Environmental RiskAssessment of Microplastics in theMarine Environment” from theMin-istry of Oceans and Fisheries, Korea.

Appendix A. Supplementary data

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.marpolbul.2016.10.049.

References

Andrady, A.L., 2011. Microplastics in the marine environment. Mar. Pollut. Bull. 62,1596–1605.

Boerger, C.M., Lattin, G.L., Moore, S.L., Moore, C.J., 2010. Plastic ingestion by planktivorousfishes in the North Pacific Central Gyre. Mar. Pollut. Bull. 60, 2275–2278.

Browne, M.A., Galloway, T.S., Thompson, R.C., 2010. Spatial patterns of plastic debris alongestuarine shorelines. Environ. Sci. Technol. 44, 3404–3409.

Chen, W., Sommerfeld, M., Hu, Q., 2011. Microwave-assisted Nile red method for in vivoquantification of neutral lipids in microalgae. Bioresour. Technol. 102, 135–141.

Claessens, M., DeMeester, S., Van Landuyt, L., De Clerck, K., Janssen, C.R., 2011. Occurrenceand distribution of microplastics in marine sediments along the Belgian coast. Mar.Pollut. Bull. 62, 2199–2204.

Claessens, M., Van Cauwenberghe, L., Vandegehuchte, M.B., Janssen, C.R., 2013. New tech-niques for the detection of microplastics in sediments and field collected organisms.Mar. Pollut. Bull. 70, 227–233.

Cole, M., Webb, H., Lindeque, P.K., Fileman, E.S., Halsband, C., Galloway, T.S., 2014. Isola-tion of microplastics in biota-rich seawater samples and marine organisms. Sci.Rep. 4, 4528.

Cózar, A., Echevarria, F., Gonzalez-Gordillo, J.I., Irigoien, X., Ubeda, B., Hernandez-Leon, S.,Palma, A.T., Navarro, S., Garcia-de-Lomas, J., Ruiz, A., Fernandez-de-Puelles, M.L.,Duarte, C.M., 2014. Plastic debris in the open ocean. Proc. Natl. Acad. Sci. U. S. A.111, 10239–10244.

De Witte, B., Devriese, L., Bekaert, K., Hoffman, S., Vandermeersch, G., Cooreman, K.,Robbens, J., 2014. Quality assessment of the bluemussel (Mytilus edulis): Comparisonbetween commercial and wild types. Mar. Pollut. Bull. 85, 146–155.

Diaz, G., Melis, M., Batetta, B., Angius, F., Falchi, A.M., 2008. Hydrophobic characterizationof intracellular lipids in situ by Nile Red red/yellow emission ratio. Micron 39,819–824.

Dutta, A.K., Kamada, K., Ohta, K., 1996. Spectroscopic studies of nile red in organic solventsand polymers. J. Photochem. Photobiol. A 93, 57–64.

Elsey, D., Jameson, D., Raleigh, B., Cooney, M.J., 2007. Fluorescent measurement ofmicroalgal neutral lipids. J. Microbiol. Methods 68, 639–642.

Eriksen, M., Lebreton, L.C.M., Carson, H.S., Thiel, M., Moore, C.J., Borerro, J.C., Galgani, F.,Ryan, P.G., Reisser, J., 2014. Plastic pollution in the world's oceans: More than 5 tril-lion plastic pieces weighing over 250,000 tons afloat at sea. Plos One 9, e111913.

Greenspan, P., Fowler, S.D., 1985. Spectrofluorometric studies of the lipid probe, Nile Red.J. Lipid Res. 26, 781–789.

Please cite this article as: Shim, W.J., et al., Identification and quantificati(2016), http://dx.doi.org/10.1016/j.marpolbul.2016.10.049

Greenspan, P., Mayer, E.P., Fowler, S.D., 1985. Nile red: a selective fluorescent stain for in-tracellular lipid droplets. J. Cell Biol. 100, 965–973.

Hidalgo-Ruz, V., Gutow, L., Thompson, R.C., Thiel, M., 2012. Microplastics in the marineenvironment: A review of the methods used for identification and quantification. En-viron. Sci. Technol. 46, 3060–3075.

Imhof, H.K., Ivleva, N.P., Schmid, J., Niessner, R., Laforsch, C., 2013. Contamination of beachsediments of a subalpine lake with microplastic particles. Curr. Biol. 23, R867–R868.

Jee, A.-Y., Park, S., Kwon, H., Lee, M., 2009. Excited state dynamics of Nile Red in polymers.Chem. Phys. Lett. 477, 112–115.

Kang, J.-H., Kwon, O.Y., Lee, K.-W., Song, Y.K., Shim, W.J., 2015. Marine neustonicmicroplastics around the southeastern coast of Korea. Mar. Pollut. Bull. 96, 304–312.

Lee, J., Hong, S., Song, Y.K., Hong, S.H., Jang, Y.C., Jang, M., Heo, N.W., Han, G.M., Lee, M.J.,Kang, D., Shim, W.J., 2013. Relationships among the abundances of plastic debris indifferent size classes on beaches in South Korea. Mar. Pollut. Bull. 77, 349–354.

Lenz, R., Enders, K., Stedmon, C.A., Mackenzie, D.M.A., Nielsen, T.G., 2015. A critical assess-ment of visual identification of marine microplastic using Raman spectroscopy foranalysis improvement. Mar. Pollut. Bull. 100, 82–91.

Lusher, A.L., McHugh, M., Thompson, R.C., 2013. Occurrence of microplastics in the gastro-intestinal tract of pelagic and demersal fish from the English Channel. Mar. Pollut.Bull. 67, 94–99.

Martins, J., Sobral, P., 2011. Plastic marine debris on the Portuguese coastline: A matter ofsize? Mar. Pollut. Bull. 62, 2649–2653.

Masura, J., Baker, J., Foster, G., Arthur, C., 2015. Laboratory methods for the analysis ofmicroplastics in themarine environment: Recommendations for quantifying synthet-ic particles in waters and sediments. NOAA Technical Memorandum NOS-OR&R-48.

Murray, F., Cowie, P.R., 2011. Plastic contamination in the decapod crustacean Nephropsnorvegicus (Linnaeus, 1758). Mar. Pollut. Bull. 62, 1207–1217.

Shim, W.J., Thompson, R.C., 2015. Microplastics in the ocean. Arch. Environ. Contam.Toxicol. 69, 265–268.

Sitepu, I.R., Ignatia, L., Franz, A.K., Wong, D.M., Faulina, S.A., Tsui, M., Kanti, A., Boundy-Mills, K., 2012. An improved high-throughput Nile red fluorescence assay for estimat-ing intracellular lipids in a variety of yeast species. J. Microbiol. Methods 91, 321–328.

Song, Y.K., Hong, S.H., Jang, M., Kang, J.H., Kwon, O.Y., Han, G.M., Shim, W.J., 2014. Largeaccumulation of micro-sized synthetic polymer particles in the sea surfacemicrolayer. Environ. Sci. Technol. 48, 9014–9021.

Song, Y.K., Hong, S.H., Jang, M., Han, G.M., Rani, M., Lee, J., Shim,W.J., 2015a. A comparisonof microscopic and spectroscopic identification methods for analysis of microplasticsin environmental samples. Mar. Pollut. Bull. 93, 202–209.

Song, Y.K., Hong, S.H., Jang, M., Han, G.M., Shim, W.J., 2015b. Occurrence and distributionof microplastics in the sea surface microlayer in Jinhae Bay, South Korea. Arch. Envi-ron. Contam. Toxicol. 69, 279–287.

Thompson, R.C., Olsen, Y., Mitchell, R.P., Davis, A., Rowland, S.J., John, A.W.G., McGonigle,D., Russell, A.E., 2004. Lost at sea: Where is all the plastic? Science 304, 838.

Van Cauwenberghe, L., Janssen, C.R., 2014. Microplastics in bivalves cultured for humanconsumption. Environ. Pollut. 193, 65–70.

Van Cauwenberghe, L., Vanreusel, A., Mees, J., Janssen, C.R., 2013. Microplastic pollution indeep-sea sediments. Environ. Pollut. 182, 495–499.

Van Franeker, J.A., Blaize, C., Danielsen, J., Fairclough, K., Gollan, J., Guse, N., Hansen, P.L.,Heubeck, M., Jensen, J.K., Le Guillou, G., Olsen, B., Olsen, K.O., Pedersen, J., Stienen,E.W.M., Turner, D.M., 2011. Monitoring plastic ingestion by the northern fulmarFulmarus glacialis in the North Sea. Environ. Pollut. 159, 2609–2615.

Woodall, L.C., Sanchez-Vidal, A., Canals, M., Paterson, G.L.J., Coppock, R., Sleight, V., Calafat,A., Rogers, A.D., Narayanaswamy, B.E., Thompson, R.C., 2014. The deep sea is a majorsink for microplastic debris. R. Soc. Open Sci. 1, 140317.

on of microplastics using Nile Red staining, Marine Pollution Bulletin