Fluorescent Dyes for Visualizing Microplastic Particles ......68 different types of plastics, it is necessary to incorporate dyes that yield non-overlapping colors. 69 The textile

TSpace Research Repository tspace.library.utoronto.ca

Fluorescent Dyes for Visualizing Microplastic

Particles and Fibers in Laboratory-based Studies

Evan G. Karakolis, Brian Nguyen, Jae Bem You, Chelsea M. Rochman,

and David Sinton

Version Post-print/Accepted Manuscript

Citation (published version)

Karakolis EG, Nguyen B, You JB, Rochman CM, Sinton D. Fluorescent Dyes for Visualizing Microplastic Particles and Fibers in Laboratory-Based Studies. Environmental Science & Technology Letters. 2019 May 22.

Copyright/License This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0

International License. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Publisher’s Statement

This document is the Accepted Manuscript version of a Published Work that appeared in final form in Environmental Science & Technology Letters, copyright © American Chemical Society after peer review and technical editing by the publisher. To access the final edited and published work see https://pubs.acs.org/doi/abs/

10.1021/acs.estlett.9b00241.

How to cite TSpace items

Always cite the published version, so the author(s) will receive recognition through services that track citation counts, e.g. Scopus. If you need to cite the page number of the author manuscript from TSpace

because you cannot access the published version, then cite the TSpace version in addition to the published version using the permanent URI (handle) found on the record page.

This article was made openly accessible by U of T Faculty. Please tell us how this access benefits you. Your story matters.

Fluorescent Dyes for Visualizing Microplastic Particles and 1

Fibers in Laboratory-based Studies 2

Evan G. Karakolisa#, Brian Nguyena#, Jae Bem You a, Chelsea M Rochmanb* and David Sintona* 3

aDepartment of Mechanical and Industrial Engineering and Institute for Sustainable Energy, University of Toronto, 4

5 King’s College Road, Toronto, ON, Canada, M5S 3G8. 5

bDepartment of Ecology and Evolutionary Biology, University of Toronto, 25 Willcocks St, Toronto, ON, Canada, 6

M5S 3B2 7

#Equally contributing author 8

*Co-corresponding author 9

*E-mail: [email protected]; [email protected] 10

Abstract 11

Observing microplastics in manipulative experiments is of paramount importance for 12

understanding the fate of microplastics in the environment, organisms and food webs. Labeling 13

microplastics with fluorescent dyes is a useful tool in laboratory experiments for tracking 14

microplastics. However, literature using fluorescence-based detection is heavily biased toward the 15

use of polystyrene and polyethylene microspheres, potentially due to their commercial availability. 16

Consequently, much less is understood about the fate of non-spherical morphologies and other 17

types of plastics common in the environment. Presented here is a heat-mediated microplastic 18

dyeing protocol that facilitates the stable incorporation of inexpensive commercially available 19

fluorescent disperse dyes directly into the polymer structure for use in laboratory-based studies. 20

We demonstrate this microplastic labelling approach is compatible with a wide variety of plastic 21

types (Polystyrene (PS), Low Density Polyethylene (LDPE), High Density Polyethylene (HDPE), 22

Polyvinylchloride (PVC), Polypropylene (PP), Polyethylene Terephthalate (PET), 23

Polyacrylonitrile (PAN)), and is customizable with different colors and fluorescent spectra. The 24

stability of the fluorescence intensity of the labelled plastics was measured over 72h and compared 25

to an existing microplastic dying approach using Nile Red. We found the dyeing approach was 26

more fluorescently stable for PS, HDPE, PET, PVC, and PAN than for LDPE and PP. The dyes 27

are also more robust to 4M KOH digestion and exposure to mineral oil than Nile Red. Finally, the 28

cost of preparing microplastics with the technique shown here is half of the cost of Nile Red. This 29

new plastic dyeing method represents a low-cost, versatile approach enabling laboratory-based 30

experiments with different polymer types and shapes using existing fluorescent microscopy 31

tracking techniques. This will help provide a more representative knowledge of the interactions of 32

microplastics with organisms. 33

Introduction 34

Observing microplastics in experiments, particularly when assessing their fate, is of paramount 35

importance. Existing studies that observe microplastics using fluorescence microscopy are heavily 36

biased by the use of two polymer types and a single particle shape (Fig. 1). The commercial 37

availability of fluorescently-labeled polystyrene and polyethylene microspheres and the ease of 38

synthesizing spherical particles of controlled size, make these plastics easily accessible to 39

researchers which could contribute to their disproportionate use. However, fibers and fragments 40

are significantly more abundant than microspheres – making microspheres an ill-fitting 41

representation of environmental exposure scenarios1. Furthermore, evidence shows that plastic 42

Figure 1. Summary of plastic type and shape used in microplastic exposure studies from two

literature reviews of papers published in all years through November 26th, 2017. The first search

using Web of Science with keywords: plastic debris and microplastic. The second search using

Scopus with keywords: microplastic* OR marine debris AND ingest* OR consum* OR eat*. (A)

Particles smaller than 300 µm (B) Particles larger than 300 µm. See Supporting Information for

more details.

morphology likely has an effect on impact2,3. For example, microplastic morphology was shown 43

to affect ingestion and mortality in daggerblade grass shrimp (Palaemonetes pugio)2, and egestion 44

time in Hyalella azteca3. Morphology may also affect the transport and fate of microplastics in the 45

environment. 46

A review of the microplastic exposure literature suggests that environmental relevance in terms of 47

plastic types and shapes may have been compromised due to the convenience of commercially 48

fabricated particles and utility of fluorescent visualization in laboratory experiments (See Figure 1 49

and Supporting Information). Figure 1 illustrates the number of laboratory exposure studies which 50

have considered the effects of ingestion of microplastics on marine organisms, sorted by size and 51

shape. Studies using polystyrene and polyethylene microspheres dominate other abundant polymer 52

types and shapes, particularly in the smaller size range (< 300µm). One factor contributing to this 53

may be the need for fluorescently labeled particles and the commercial availability of labelled 54

polystyrene or polyethylene spheres. The possible link between polystyrene/polyethylene spheres 55

and fluorescent labeling is demonstrated by the fact that 62% (29 out of 47) of the studies which 56

used polystyrene and polyethylene spheres smaller than 300 µm (Figure 1) also used fluorescently 57

labeled particles, suggesting that a need for fluorescence-based detection may have contributed to 58

their choice of microplastic. 59

What is needed is an approach to fluorescently label any plastic type or shape, enabling the study 60

of various plastic morphologies and polymer types while still using the fluorescence imaging 61

identification techniques preferred in current scientific work. Recently, studies have shown that 62

Nile Red dye is a promising approach for tagging plastics via hydrophobic interaction in exposure 63

experiments4 as well as in environmental samples.5–7 Nile Red can bind to plastic and fluoresces 64

in a range of colors from yellow to deep red depending on the surface hydrophobicity of the plastic. 65

However, the emitted color of Nile Red can be affected by the change in hydrophobicity of plastics 66

due to the surface contamination from the environment. Therefore, to clearly distinguish the 67

different types of plastics, it is necessary to incorporate dyes that yield non-overlapping colors. 68

The textile industry has motivated the development of many dyes for fabrics. Azo dyes render 69

colors on both natural and synthetic textile products8–10. These dyes contain aromatic amines11,12 70

in their structure, so they are regarded as attractive fluorescent compounds. Therefore, the 71

commercial textile dyes hold the potential to be applied as fluorescent labels for microplastics due 72

to their simple staining procedure as well as possible fluorescence emission. 73

Introduced here is an alternative approach to purchasing fluorescently labeled spheres or 74

fluorescently labeling plastics with Nile Red for microplastic research. Commercially available 75

disperse dyes (iDye) marketed as textile dyes were identified as candidates compatible with a 76

variety of plastics and acting as fluorophores with spectral characteristics compatible with 77

common fluorescent filter sets. Furthermore, this technique allows spectral customization of the 78

fluorescent characteristics of the dyed plastics. The approach is simple, versatile and customizable 79

to users’ needs and enables the study of more environmentally relevant range of microplastic 80

shapes and types. 81

Materials and Methods 82

Plastic Particles 83

The plastics used in this study are summarized in Table 1 below. The source and method by which 84

they were fragmented/prepared is summarized in Table S3. In this study a fragment is considered 85

to be a non-spherical, non-fibrous particle of irregular shape (See figure 2C). 86

Table 1. Summary of polymer types, shapes, and size ranges of microplastics. Fragment sizes are longest dimension 87 of fragment. Sphere size are diameter. Fiber sizes are diameter / length of fiber. Glass transition and melting point 88 temperatures were obtained from PerkinElmer13 and Sigma-Aldrich14. 89

90

Dye Fluorescence Testing 91

The fluorescence of a wide range of dyes was tested (see Supplemental Information for list of dyes 92

tested) to select the three chosen dyes: iDye pink (pink dye), iDye blue (blue dye), and Rit 93

DyeMore Kentucky Sky (kentucky dye). Various iDye Poly (Jacquard) and Rit DyeMore synthetic 94

fiber dyes were tested to determine if they fluoresced in common microscopy fluorescence filter 95

ranges. We considered 3 fluorescence ranges: green (Ex: 470/22 Em: 525/50), red (Ex: 531/40 96

Em: 593/40), and far-red (Ex: 628/40 Em: 685/40). We determined that pink dye fluoresced in the 97

red range, blue dye in the far-red range, and kentucky dye in the green and red ranges. These three 98

fluorescent spectra can be used simultaneously to label three different polymer types within a 99

sample (Figure S3). See the supplemental information for visual depictions of unique fluorescent 100

signatures (Figure S3). 101

Plastic Dyeing 102

Polymer Type Shape Size Range Glass

Transition Melting Point

Polystyrene Fragment 100-300 µm 90 to 100 °C 240 °C

Polystyrene Sphere 13.5-16.5 µm 90 to 100 °C 240 °C

Low-Density

Polyethylene

Fragment 100-200 µm -130 °C 85 to 125 °C

High-Density

Polyethylene

Sphere 10-90 µm -125 °C 130 to140 °C

Polyethylene

Terephthalate

Fragment 50-500 µm 70 to 80 °C 245 to 265 °C

Polyethylene

Terephthalate

Fiber 30-60 µm /

150-5000 µm

70 to 80 °C 245 to 265 °C

Polyvinylchloride Fragment 50-300 µm 81 °C 227 °C

Polypropylene Fragment 500-4000 µm -20 to -5 °C 165 to 176 °C

Polyacrylonitrile Fiber 20-50 µm /

300-3000 µm

125 °C 319 °C

Dry plastic particles were added to a 100 mg/mL iDye Poly (Jacquard) powder in DI water solution 103

(or 50% v/v with DI water for Rit DyeMore liquid dye) at a concentration of 50 mg plastic particles 104

and 25 mg plastic fibers per 10 mL dye solution. Plastics were then heated at 70°C in vials for 2h 105

in darkness, then removed from the vial by pouring through a filter and rinsed three times, each 106

time resuspending them in fresh DI water then pouring through a filter. After 3 rinses they were 107

stored in DI water. The dyeing temperature of 70°C was chosen to stay conservatively below the 108

melting point of LDPE, the plastic with the lowest melting point (85°C, see Table S3), and was 109

kept the same for all plastics. The heating step facilitates the diffusion of dye molecules into the 110

polymer matrix, which remain in place after cooling.15 The cost of iDye Poly, at the time of this 111

study, was ~$0.20 USD/g and the cost of Rit DyeMore, at the time of this study, was ~$0.09 112

USD/mL equating to ~$4.03 USD/g and ~$1.79 USD/g plastic coated respectively (at 2018 113

December USD to CAD conversion rates, 1 USD =1.34 CAD). Particles were also labelled with 114

Nile Red for comparison. For Nile Red dyeing, dry plastic particles were added to a 100 µg/mL 115

Nile Red in DI water solution at a concentration of 50 mg plastic particles and 25 mg plastic fibers 116

per 10 mL solution. To prepare this solution, 1 mg of Nile Red was first dissolved in 1 mL of 117

acetone, then the solution was added to 10 mL of DI water. The protocol for dyeing Nile Red, 118

including the concentration of Nile Red in solution, was selected based on existing microplastic 119

dyeing protocols5. Vials were left for 2h in darkness, then rinsed three times as described above, 120

then stored in DI water. The cost of Nile Red, at the time of this study, was $417.91 USD/g (Sigma 121

- 72485) equating to $8.36 USD/g plastic coated. 122

Fluorescence Stability Testing 123

Labelled particles were placed in one of eight conditions of varying water chemistry, light 124

conditions, temperature, organism presence and solution polarity (See table S4 for summary of 125

conditions). The baseline treatment, which other treatments were compared to, was: 35 ppt salinity, 126

dark, 24 °C, no organisms, and no KOH or Mineral Oil. In all other treatments, one parameter was 127

changed. For example, the effect of salinity was tested by reducing the salinity to freshwater levels 128

of 0.6 ppt, but leaving other parameters constant (dark, 24 °C, no organisms, and no KOH or 129

Mineral Oil). The potential for photobleaching was tested with a 12h:12h light:dark treatment with 130

light intensity of 5 μmol photons m−2 s−1. The effect of low temperatures was tested by exposing 131

plastics to 8 °C saltwater. The effect of non-ingestion organism exposure was tested by exposing 132

each plastic to 24h post-hatching juvenile Artemia salina. We note that the size of the plastic 133

particles tested relative to A. salina precluded ingestion and digestion, however the presence of 134

organisms and defecated gut enzymes in feces was tested. The effect of potassium hydroxide 135

digestion was tested by exposing the plastics to 4M KOH (Potassium Hydroxide pellets purchased 136

from BioShop, reagent grade) for 72 h. The temperature of the solution was not measured in the 137

KOH treatment, but there was no biogenic matter in the solution to react, and the wellplates that 138

housed each treatment were kept in 24 °C ambient conditions. The effect of polar liquids such as 139

fats and oils were tested by exposing plastics to mineral oil (Light - Bioshop). 140

The particles were tested over 72h to determine their relative fluorescence stability over time. Tests 141

were completed in 48-well plates filled with 300 µL of a given treatment fluid. Cold treatments 142

were left in a refrigerator at 8 °C. See supplemental methods section for details on A. salina 143

culturing. Nile Red and pink dye were tested for all plastic types and for all conditions. Blue dye 144

and kentucky dye were not tested since they appeared toxic (see results section). 145

For each treatment there were three replicates (i.e. three wells) each with one or more pieces of 146

plastics per replicate well. Fluorescence microscopy images (16-bit) were taken initially, then 147

every 24h up to a final image 72h after initial exposure. For each plastic type the same imaging 148

settings (light intensity and exposure time) were used for all dyes, but different imaging settings 149

were used for different plastic types. Images were analyzed using ImageJ image analysis software. 150

The fluorescence intensity for a plastic was determined by measuring the average intensity over 151

the area of that plastic, then subtracting the mean intensity of the background area (area with no 152

plastic). The intensity for a given well was the average intensity of the plastics within that well, 153

and the intensity for a given treatment and a given timepoint was determined by averaging the 154

intensities of the replicate wells. 155

Ingestion Stability Testing 156

The robustness of the microplastic dyes to organism ingestion and digestion was tested by 157

exposing 24h post-hatching A. salina to small microplastic particles (PS spheres 13.5 - 16.5 µm) 158

dyed by each of the three dyes (pink dye, blue dye, and kentucky dye). This ingestion stability 159

experiment was different than the organism exposure treatments described in the stability 160

experiments above. In this experiment, plastics were small enough to be easily ingested. Ingestion 161

experiments took place in the wells of a 24 well plate with 2 mL of saltwater, 500 plastics, and 162

three A. salina per well. For each dye there were 12 replicates (i.e. 12 wells). Fluorescence 163

microscopy images (16-bit) of plastics were taken in-situ within each well using an EVOS FL 164

Auto Imaging System with wellplate insert. Images were taken initially, then every 24h up to a 165

final image 72h after initial exposure. Images were taken at a random location in each well with 3 166

or more plastics per image. We assumed all plastics were ingested for two reasons. First, the 167

filtration rate of a single A. salina of the size and age used in this experiment (size: 600-1,000µm, 168

age: 24h post-hatching, approximate filtration rate: 0.2 mL/h16), is such that entire volume of the 169

well will have been filtered more than 14-times per day by the three A. salina in each well. Second, 170

the polystyrene plastics used in this exposure are not buoyant, so they settled on the bottom of the 171

well where the A. salina feed. For these two reasons it is likely most or all of the plastics were 172

ingested during the 96h exposure. 173

Images were analyzed using ImageJ image analysis software. The fluorescence intensity for a 174

given replicate well was the average intensity of the plastics in the image taken for that well, and 175

the intensity for a given treatment and timepoint was determined by averaging the intensities of 176

each of the 12 replicate wells per dye. Fluorescence intensity was determined as described in the 177

Fluorescence Stability Testing section above. 178

Dye Toxicity Testing 179

The toxicity of each dye was tested on 24h post-hatching A. salina over 96h. A. salina was selected 180

as a model organism due to ease of cultivation and the past use of Artemia sp. as biological 181

organisms in toxicity testing17–20. Furthermore, a recent study showed that acute toxicity test 182

results using the second and third instar A. salina nauplii (24h after hatching) had significant 183

interspecies correlations with results of other common biological model organisms used in toxicity 184

testing including fish (Danio rerio) and zooplankton (Daphnia magna)18, meaning toxicity tests 185

on A. salina may be indicative of toxicity to other zooplankton and potentially fish. This being 186

said, we strongly suggest that for each new organism used with these dyes, a new toxicity test 187

should be conducted to determine species specific toxicity. 188

Tests were completed in 6-well plates with each well containing 10 mL of saltwater, one A. salina, 189

500 plastic particles, and 10,000 Thallasoira weissfloggii brown algae cells. After 48h an 190

additional 10,000 T. weissfloggii cells were added to replenish the food supply. The plastics used 191

here were 13.5-16.5µm PS spheres (Polybead #18328-5). There were three treatments and two 192

controls: One treatment for each dye (pink dye, blue dye, and kentucky dye), a control with 193

uncoated polystyrene spheres, and a control with no plastics (algae only). There were 6 organisms 194

per replicate (one per well in a six well plate) and 4 replicate plates per treatment. The number of 195

living organisms on each replicate plate was determined after 24h, 48h, 72h, and 96h. Then, the 196

survival ratio on each replicate plate, defined as number of living organisms per plate divided by 197

six, was calculated. The average survival fraction between the 4 replicate plates was also 198

determined for each treatment and control. 199

Results and Discussion 200

Presented here is a novel method of fluorescently tagging microplastics in laboratory experiments 201

through relatively inexpensive commercially available fabric dyes (available at ~$1.79-4.03 202

USD/g plastic coated compared to ~ $8.36 USD/g plastic coated for Nile Red). The pink dye’s 203

Figure 2. Examples of plastic types and morphologies dyed with different dyes (A) Different morphologies: Spheres

(15-150 µm), Fragments (50-300 µm) and Fibers (30-60 µm / 150-5000 µm) dyed with pink dye (B) Different

Fluorophores: Green (kentucky dye), Red (pink dye), Far-red (blue dye). Pseudo colors are applied to different

fluorescent channels for differentiation purposes. (C) Different Polymer Types dyed with pink dye: Polyethylene

Terephthalate (PET) fragments (50-500 µm), High-Density Polyethylene (HDPE) spheres (15-150 µm),

Polyvinylchloride (PVC) fragments (50-300 µm), Low-Density Polyethylene (LDPE) fragments (100-200 µm),

Polypropylene (PP) fragments (500-4000 µm), Polystyrene (PS) fragments (100-300 µm), Polyester (PET) fiber (30-

60 µm / 150-5000 µm), Polyacrylonitrile (PAN) fiber (20-50 µm / 300-3000 µm). Scale bars are 550 µm.

applicability with different polymer types was tested along with compatibility with various 204

morphologies (Figure 2). The pink dye results in relatively strong fluorescent signal under these 205

conditions with most of the polymer types tested except for LDPE and PP. For LDPE, the 206

fluorescent signal was weak and much lower than for the Nile Red comparison case (Figure S2a-207

d), while the signal for PP was also weak but similar to that of Nile Red (Figure S2e-f). This poor 208

staining for LDPE and PP may be due to impurities or additives in the plastics which may affect 209

dye staining. That being said, staining could potentially be improved by increasing dye 210

concentration, staining temperature, and/or staining duration. Nevertheless, useable levels of 211

fluorescence were observed in all cases and optimization of labelling conditions for LDPE and PP 212

could potentially improve the weaker signals in those cases. The textile dyes performed extremely 213

well for PS, PET, and PVC fragments, HDPE microspheres, PET, and PAN fibers with several 214

times more signal compared to Nile red in most cases (Figure S1-S2). These results indicate the 215

new dyeing method is usable since its fluorescence intensity matches or is better than the 216

established ambient-temperature Nile Red approach. We note that the heat-mediated dyeing 217

procedure used for this new method could improve the dyeing of plastics with Nile Red, so the 218

fluorescence intensity comparison here indicates only that the new dyeing method presented here 219

has higher fluorescence (in most plastics tested), than the established method for ambient 220

temperature Nile Red dying5, but does not indicate the dyes presented here have better performance 221

than Nile red in general. 222

223

Importantly, the pink dye was stable in conditions relevant to exposure studies including different 224

environmental conditions, KOH digestion, and non-polar (mineral oil) environments for at least 225

72h. This stability is attributed to the absorption of dyes into the polymer matrix. It has been known 226

that small molecules such as dyes can be trapped in the polymer matrix when exposed to the glass 227

transition temperature (Tg) of the polymer21,22. Therefore, when the plastics are heated during the 228

staining process, the dyes can become trapped in the polymer matrix enhancing the stability against 229

strong solvents such as KOH and mineral oil. However, for some of the fibers it appears KOH and 230

non-polar solutions degraded the intensity over time (Figure S1k,l,o,p). Still, in most cases, the 231

labeled plastics can be fluorescently visualized throughout many steps of an experiment, for 232

example a plastic particle translocated into tissues with high lipid content or an experimental state 233

study in a wastewater treatment plant. 234

Moreover, the ability to use fluorescent dyes in different channels (Figure 2b) enables labeling 235

various polymers in multiple channels to distinguish them simultaneously in exposure studies. In 236

the environment, many types of plastics are present while in most experiments to date only a single 237

type of plastic is used. In addition, since the dye absorbs into the polymer matrix of the plastic, 238

this approach could be used in concert with techniques that rely on coating the plastics5,7,23, or in 239

studies where surface-adsorbed biofilms24–26 or chemical pollutants27–29 are of interest. 240

The compatibility of the dyes with organism ingestion was tested by exposing A. salina to 13.5-241

16.5 µm PS microbeads dyed with each of the three dyes. The pink dye, which fluoresced in the 242

red range (Ex: 531/40 Em: 593/40), appears to be non-toxic over 96h to one-day-old A. salina 243

when compared to the uncoated PS control and no-plastic control, all having greater than 79% 244

survival (Figure 3a). In contrast, kentucky dye and blue dye which fluoresce in the green (Ex: 245

470/22 Em: 525/50) and far-red (Ex: 628/40 Em: 685/40) respectively, both appear to be toxic to 246

the A. salina. The blue dye treatment had a survival of 58%, while the kentucky dye had a survival 247

of 33% (Figure 3a). All three dyes appear to be fluorescently stable after ingestion by A. salina. 248

This is confirmed by measuring the average fluorescence intensity of particles ingested by A. 249

salina (Figure 3b-d), and by visually observing the fluorescence of particles in the gut of A. salina 250

(Figure 3e-g). Overall it appears that although dyes are fluorescently stable even when being 251

ingested by A. salina, two out of the three dyes are likely toxic to this organism. This is consistent 252

Figure 3. Compatibility with organism exposure (A) Mean (±S.E.M.) survival fraction of 24h post-hatching Artemia

salina exposed to plastics with different coatings over 96h exposure period (n=4). (B-D) Mean (±S.D.) fluorescence

intensity of dyed polystyrene spheres (13.5-16.5 µm) after being exposed to 24h post-hatching Artemia salina over

72h (n=12). (B) Rit DyeMore Kentucky Blue. (C) iDye Poly Pink. (D) iDye Poly Blue. (E-G) Images of Artemia

salina after 72h with fluorescent channels for respective dyes overlaid using pseudo-colors. (E) Rit DyeMore

Kentucky Blue. (F) iDye Poly Pink. (G) iDye Poly Blue. Scale bars are 200 µm. Images in (E-G) were brightness-

and contrast-adjusted for clarity

with previous literature that shows some azo dyes are toxic to Daphnia magna30. However, the 253

pink dye appears to be non-toxic and thus can be considered for broader use. 254

Finally, we found that the dyes did not fluoresce when dispersed in water. This means that these 255

dyes could be used to allow visualization of plastics in environmental samples treated with the 256

dye, similar to the way Nile Red is currently used5. This could be explored in future work. 257

Conclusion 258

In this study, commercially available dyes, which have the ability to absorb into the polymer matrix 259

of common plastic types were identified and tested for their applicability in microplastics exposure 260

and fate studies. Dyes were selected that exhibited fluorescence spectra compatible with filter sets 261

commonly used in fluorescence microscopy. The protocol for using these dyes is straightforward 262

and low in cost. The availability of multiple colors of dye enables experiments that simultaneously 263

assess different kinds or shapes of plastics, each labelled in a particular fluorescent channel. It is 264

expected that these dyes will provide an inexpensive complement to dyes which adsorb to plastic 265

surfaces – similar to Nile Red. Although two of the dyes proved toxic to A. salina (kentucky dye 266

– green, and blue dye – far-red), they may be applicable to experiments where organism survival 267

is secondary, such as short-term ingestion, trophic transfer, or translocation studies. The non-toxic 268

dye (pink dye – “red”) appears to be most compatible with typical organismal microplastic 269

exposure experiments where fate and toxicity are measured. In addition, all dyes may be very 270

applicable when measuring the fate of various types, shapes or sizes of microplastics in a treatment 271

plant or environmental matrix in a short-term study. For a longer study, further work is required 272

to measure the stability over longer time-scales. 273

ASSOCIATED CONTENT 274

Supporting Information 275

Summary table for ingestion literature from all years through Nov-26, 2017 (Table S1), 276

Summary table for microplastic exposure literature from all years through Nov-26, 2017 (Table 277

S2), Summary of polymer types, shapes, size ranges, source, and preparation technique of 278

microplastics used in experiments (Table S3), Summary of commercial dyes tested (Table S4), 279

Summary of treatments and levels for water quality stability testing (Table S5), iDye and Nile 280

Red fluorescence intensity for different plastics under different conditions over time (Figure S1 281

and S2), fluorescent microscope images for each dye in each fluorescent channel (Figure S3), 282

iDye and Nile Red fluorescence intensity for different plastics under different light and 283

temperature conditions over time (Figure S4), summary of data for fluorescence intensity 284

stability experiments with PS fragments (Table S6), summary of data for fluorescence intensity 285

stability experiments with LDPE fragments (Table S7), summary of data for fluorescence 286

intensity stability experiments with HDPE spheres (Table S8), summary of data for fluorescence 287

intensity stability experiments with PET fragments (Table S9), summary of data for fluorescence 288

intensity stability experiments with PVC fragments (Table S10), summary of data for 289

fluorescence intensity stability experiments with PP fragments (Table S11), summary of data for 290

fluorescence intensity stability experiments with PET fibers (Table S12), summary of data for 291

fluorescence intensity stability experiments with PAN fibers (Table S13). 292

ACKNOWLEDGMENTS 293

This work was supported through a Strategic Grant from the Natural Science and Engineering 294

Research Council of Canada, the E.W.R. Steacie Memorial Fellowship (DS), and the Canada 295

Research Chairs Program. BN gratefully acknowledges funding from Ontario Graduate 296

Scholarships, the Queen Elizabeth II Graduate Scholarships in Science & Technology, NSERC 297

CGS Scholarships and the MEET, NSERC CREATE Program. EK thanks Sabrina T. Barsky for 298

editing the manuscript and acknowledges funding from Ontario Graduate Scholarships, and NSERC 299

CGS Scholarships. Ongoing infrastructure support from the Canadian Foundation for Innovation 300

and operational support through the NSERC Discovery and Discovery Accelerator Programs is also 301

gratefully acknowledged. J.B.Y would like to acknowledge the support by Basic Science Research 302

Program through the National Research Foundation of Korea (NRF) funded by the Ministry of 303

Education (2018R1A6A3A03012768). 304

305

REFERENCES 306

(1) Phuong, N. N.; Zalouk-Vergnoux, A.; Poirier, L.; Kamari, A.; Châtel, A.; Mouneyrac, C.; 307

Lagarde, F. Is there any consistency between the microplastics found in the field and those 308

used in laboratory experiments? Environ. Pollut. 2016, 211, 111–123. 309

(2) Gray, A. D.; Weinstein, J. E. Size- and shape-dependent effects of microplastic particles 310

on adult daggerblade grass shrimp (Palaemonetes pugio). Environ. Toxicol. Chem. 2017, 311

36 (11), 3074–3080. 312

(3) Au, S. Y.; Bruce, T. F.; Bridges, W. C.; Klaine, S. J. Responses of Hyalella azteca to acute 313

and chronic microplastic exposures. Environ. Toxicol. Chem. 2015, 34 (11), 2564–2572. 314

(4) Cole, M. A novel method for preparing microplastic fibers. Sci. Rep. 2016, 6 (1), 34519. 315

(5) Maes, T.; Jessop, R.; Wellner, N.; Haupt, K.; Mayes, A. G. A rapid-screening approach to 316

detect and quantify microplastics based on fluorescent tagging with Nile Red. Sci. Rep. 317

2017, 7 (November 2016), 1–10. 318

(6) Shim, W. J.; Song, Y. K.; Hong, S. H.; Jang, M. Identification and quantification of 319

microplastics using Nile Red staining. Mar. Pollut. Bull. 2016. 320

(7) Erni-Cassola, G.; Gibson, M. I.; Thompson, R. C.; Christie-Oleza, J. A. Lost, but Found 321

with Nile Red: A Novel Method for Detecting and Quantifying Small Microplastics (1 322

mm to 20 μm) in Environmental Samples. Environ. Sci. Technol. 2017. 323

(8) El-Zawahry, M. M.; Mashaly, H. M. Acid dyeing of linen fabrics using gemini cationic 324

surfactants. J. Text. Inst. 2010, 101 (7), 645–652. 325

(9) Hallas, G.; Choi, J.-H. Synthesis and properties of novel aziridinyl azo dyes from 2-326

aminothiophenes - Part 2: Application of some disperse dyes to polyester fibres. Dye. 327

Pigment. 1999, 40 (2–3), 119–129. 328

(10) Sadeghi-Kiakhani, M.; Gharanjig, K.; Arami, M. Study on dyeing and fastness properties 329

of wool-polyester blend fabrics using novel mono azo-naphthalimide dyes. J. Text. Inst. 330

2014, 105 (1), 52–58. 331

(11) Pinheiro, H. M.; Touraud, E.; Thomas, O. Aromatic amines from azo dye reduction: 332

Status review with emphasis on direct UV spectrophotometric detection in textile industry 333

wastewaters. Dye. Pigment. 2004. 334

(12) Yoshino, J.; Kano, N.; Kawashima, T. Fluorescent azobenzenes and aromatic aldimines 335

featuring an N-B interaction. Dalton Transactions. 2013. 336

(13) PerkinElmer. MP, Tg, and Structure of Common Polymers. Perkin Elmer. 337

(14) Sigma-Aldrich. Polymer Properties Thermal Transitions of Homopolymers: Tg and MP. 338

52–53. 339

(15) Slark, A. T.; Hadgett, P. M. The effect of specific interactios on dye transport in polymers 340

above the glass transition. Polymer (Guildf). 1999, 40 (14), 4001–4011. 341

(16) Reeve, M. R. The Filter-Feeding of Artemia. J. Exp. Biol. 1963, 40 (1), 207 LP-214. 342

(17) Libralato, G.; Prato, E.; Migliore, L.; Cicero, A. M.; Manfra, L. A review of toxicity 343

testing protocols and endpoints with Artemia spp. Ecol. Indic. 2016, 69, 35–49. 344

(18) Kalčíková, G.; Zagorc-Končan, J.; Žgajnar Gotvajn, A. Artemia salina acute 345

immobilization test: a possible tool for aquatic ecotoxicity assessment. Water Sci. 346

Technol. 2012, 66 (4), 903–908. 347

(19) Svensson, B.-M.; Mathiasson, L.; Mårtensson, L.; Bergström, S. Artemia salina as test 348

organism for assessment of acute toxicity of leachate water from landfills. Environ. Monit. 349

Assess. 2005, 102 (1), 309–321. 350

(20) Dvorak, P.; Benova, K.; Vitek, J. Alternative Biotest on Artemia franciscana. In 351

Ecotoxicology; 2012. 352

(21) Law, K. Y.; Loutfy, R. O. Spectroscopy of dyes in polymer matrixes: dual fluorescence of 353

p-[(dialkylamino)benzylidene]malononitrile. Macromolecules 1981, 14 (3), 587–591. 354

(22) Victor, J. G.; Torkelson, J. M. On Measuring the Distribution of Local Free Volume in 355

Glassy Polymers by Photochromic and Fluorescence Techniques. Macromolecules 1987. 356

(23) Karakolis, E. G.; Nguyen, B.; You, J. B.; Graham, P. J.; Rochman, C. M.; Sinton, D. 357

Digestible Fluorescent Coatings for Cumulative Quantification of Microplastic Ingestion. 358

Environ. Sci. Technol. Lett. 2018, 5 (2), 62–67. 359

(24) Allen, A. S.; Seymour, A. C.; Rittschof, D. Chemoreception drives plastic consumption in 360

a hard coral. Mar. Pollut. Bull. 2017. 361

(25) Vroom, R. J. E.; Koelmans, A. A.; Besseling, E.; Halsband, C. Aging of microplastics 362

promotes their ingestion by marine zooplankton. Environ. Pollut. 2017, 231, 987–996. 363

(26) Savoca, M. S.; Tyson, C. W.; McGill, M.; Slager, C. J. Odours from marine plastic debris 364

induce food search behaviours in a forage fish. Proc. R. Soc. B Biol. Sci. 2017. 365

(27) Holmes, L. A.; Turner, A.; Thompson, R. C. Adsorption of trace metals to plastic resin 366

pellets in the marine environment. Environ. Pollut. 2012, 160 (1), 42–48. 367

(28) Avio, C. G.; Gorbi, S.; Milan, M.; Benedetti, M.; Fattorini, D.; D’Errico, G.; Pauletto, M.; 368

Bargelloni, L.; Regoli, F. Pollutants bioavailability and toxicological risk from 369

microplastics to marine mussels. Environ. Pollut. 2015. 370

(29) Batel, A.; Linti, F.; Scherer, M.; Erdinger, L.; Braunbeck, T. Transfer of benzo[a]pyrene 371

from microplastics to Artemia nauplii and further to zebrafish via a trophic food web 372

experiment: CYP1A induction and visual tracking of persistent organic pollutants. 373

Environ. Toxicol. Chem. 2016, 35 (7), 1656–1666. 374

(30) Bae, J.-S.; Freeman, H. S. Aquatic toxicity evaluation of new direct dyes to the Daphnia 375

magna. Dye. Pigment. 2007, 73 (1), 81–85. 376

377