Accepted Manuscript

Identification of microplastics using Raman spectroscopy: latest developments

and future prospects

Catarina F. Araujo, Mariela M. Nolasco, Antonio M.P. Ribeiro, Paulo J.A. Ribeiro-
Claro

PII: S0043-1354(18)30442-1

DOI: 10.1016/j.watres.2018.05.060

Reference: WR 13825

To appear in: Water Research

Received Date: 08 March 2018

Accepted Date: 31 May 2018

Please cite this article as: Catarina F. Araujo, Mariela M. Nolasco, Antonio M.P. Ribeiro, Paulo J.A.
Ribeiro-Claro, Identification of microplastics using Raman spectroscopy: latest developments and
future prospects, Water Research (2018), doi: 10.1016/j.watres.2018.05.060

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form.
Please note that during the production process errors may be discovered which could affect the
content, and all legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

1 Identification of microplastics using Raman

2 spectroscopy: latest developments and future prospects

3 Catarina F. Araujo1*, Mariela M. Nolasco1, Antonio M.P. Ribeiro1, Paulo J.A Ribeiro-Claro1*

4 1CICECO – Aveiro Institute of Materials, Departamento de Química, Universidade de Aveiro, 3810-

5 193 Aveiro, Portugal. *[email protected], *[email protected]

7 Abstract

9 Widespread microplastic pollution is raising growing concerns as to its

10 detrimental effects upon living organisms. A realistic risk assessment must stand on

11 representative data on the abundance, size distribution and chemical composition of

12 microplastics. Raman microscopy is an indispensable tool for the analysis of very small

13 microplastics (< 20 μm). Still, its use is far from widespread, in part due to drawbacks

14 such as long measurement time and proneness to spectral distortion induced by

15 fluorescence. This review discusses each drawback followed by a showcase of

16 interesting and easily available solutions that contribute to faster and better

17 identification of microplastics using Raman spectroscopy. Among discussed topics are:

18 enhanced signal quality with better detectors and spectrum processing; automated

19 particle selection for faster Raman mapping; comprehensive reference libraries for

20 successful spectral matching. A last section introduces non-conventional Raman

21 techniques (non-linear Raman, hyperspectral imaging, standoff Raman) which permit

22 more advanced applications such as real-time Raman detection and imaging of

23 microplastics.

1
 ACCEPTED MANUSCRIPT

24

25 Keywords

26 Small microplastics; Automation; Fluorescent tagging; Library

27 matching; Real-time analysis; Stimulated Raman Scattering;

28

29

30 1. Introduction

31

32 The advent of synthetic plastics ushered in a golden era of unprecedented

33 technological advancement and improved levels of material comfort, perfectly summed

34 in the classic DuPont slogan “Better Living Through Chemistry”. Yet, this euphoric

35 period of unbridled techno-optimism inevitably abated in tandem with growing

36 concerns regarding the harmful impacts of plastic waste, among them the plight of

37 widespread microplastic pollution in the biosphere. According to a recent estimate there

38 are currently more than 5 trillion plastic particles floating at sea, totalling circa 270 000

39 tonnes (Eriksen et al., 2014). As plastic degradation proceeds and each particle

40 fragments into ever smaller pieces, the total number of particles gallops upwards in an

41 exponential fashion – and so do the risks they pose to animal and human life (Avio et

42 al., 2017; Karami et al., 2016; Ogonowski et al., 2018; Revel et al., 2018; Wright and

43 Kelly, 2017).

44 Microplastics (MPs), defined as plastic particles ranging from 1 μm to 5 mm in

45 size, are now ubiquitously present in aquatic and terrestrial environments (Duis and

2
 ACCEPTED MANUSCRIPT

46 Coors, 2016; Horton et al., 2017b) – often finding their way into our food and drink

47 (Alexander et al., 2016; Karami et al., 2018; Schymanski et al., 2018). A realistic

48 assessment of the ill effects of MPs must commence with a representative, large scale

49 analysis of their abundance, size distribution and chemical composition (Jahnke et al.,

50 2017). Towards this end, chemistry provides a set of identification tools capable of

51 tackling the MP issue in its many facets (Hanvey et al., 2017; Shim et al., 2017; Silva et

52 al., 2018) such as vibrational spectroscopy, densitometry, differential scanning

53 calorimetry (DSC), gas chromatography-mass spectrometry (GC-MS) based methods

54 and the recently proposed technique hyperspectral imaging (Shan et al., 2018).

55 This review focuses on a particular characterization technique – Raman

56 spectroscopy – which is rapidly gaining ground in the analysis of small MPs. Raman

57 spectroscopy is a vibrational spectroscopy technique based on the inelastic scattering of

58 light that provides information upon the molecular vibrations of a system in the form of

59 a vibrational spectrum. The Raman spectrum is akin to a fingerprint of chemical

60 structure allowing identification of the components present in the sample.

61 Some of the advantages of Raman spectroscopy are shared by FT-IR techniques,

62 such as non-destructiveness, low sample amount requirement, possibility for high

63 throughput screening and environmental friendliness. These are relevant advantages

64 over other reported methods (i.e. DSC, Pyr-GC-MS) for MPs analysis and

65 characterization (Elert et al., 2017; Ribeiro-Claro et al., 2017). Indeed, vibrational

66 spectroscopy is a common choice for MP identification and has been recommended by

67 the European Union expert group on marine litter, who advocate that all suspected MPs

68 in the 1-100 μm size range should have their polymer identity confirmed by

69 spectroscopic analysis (Gago et al., 2016).

3
 ACCEPTED MANUSCRIPT

70 Compared with FTIR spectroscopy, Raman techniques show better spatial

71 resolution (down to 1 μm while that of FTIR is 10-20 μm), wider spectral coverage,

72 higher sensitivity to non-polar functional groups, lower water interference and narrower

73 spectral bands. On the downside, Raman spectroscopy is prone to fluorescence

74 interference, has an inherently low signal to noise ratio and might cause sample heating

75 due to the use of a laser as light source, leading to background emission occasionally

76 followed by polymer degradation. A more comprehensive explanation of the theoretical

77 principles of Raman spectroscopy, its strengths and weaknesses for MP analysis, as well

78 as instrumentation details can be found elsewhere (Ivleva et al., 2017; Ribeiro-Claro et

79 al., 2017; Silva et al., 2018; Smith and Dent, 2005).

80 Despite its many advantages, the identification of MPs through the use of

81 Raman spectroscopy is yet to attain the popularity of FT-IR techniques, although the

82 number of publications is steeply increasing. At this juncture (May 2018), a search on

83 ISI Web of Knowledge, ResearchGate and ScienceDirect yields a total of 71 original

84 articles on MP identification using Raman spectroscopy. Out of these, 86% employ

85 Raman microscopy (μ-Raman) – the classical Raman setup was only used in early

86 studies and since then discontinued. The surveyed works are listed and grouped into

87 handy categories, in Table 1. Additionally, 6 book chapters and 21 reviews cover the

88 topic with varying levels of detail, with one book chapter from (Ribeiro-Claro et al.,

89 2017) completely dedicated to it. Yet, the significant boom in μ-Raman detection of

90 MPs that took place since then warrants an updated digest of current limitations and

91 emerging solutions. Before delving into that, it is useful to demonstrate why Raman

92 spectroscopy is essential for the study of very small MPs, a domain where FTIR

93 techniques are wholly inadequate.

4
 ACCEPTED MANUSCRIPT

94 Table 1 – Comprehensive list of documents dealing with MP analysis through

95 Raman, as of January 2018

Reference guide for works on Raman identification of microplastics

in seawater, (Cozar et al., 2014; De Tender et al., 2015; Di and Wang, 2018; Enders

et al., 2015; Erni-Cassola et al., 2017; Frere et al., 2017, 2016; Ghosal et
freshwater and
al., 2018; Karlsson et al., 2017; Lares et al., 2018; Lenz et al., 2015;
wastewater
Lusher et al., 2014; Sujathan et al., 2017; Xiong et al., 2018; Yonkos et

al., 2014; Zettler et al., 2013; Zhang et al., 2017; Zhao et al., 2017, 2015)

in sediment (Ballent et al., 2016; Clunies-Ross et al., 2016; De Tender et al., 2015;

Di and Wang, 2018; Elert et al., 2017; Erni-Cassola et al., 2017; Fischer

et al., 2015; Horton et al., 2017a; Imhof et al., 2018, 2016, 2013, 2012;

Kaeppler et al., 2016; Karlsson et al., 2017; Lots et al., 2017; Maes et al.,

2017; Scheurer and Bigalke, 2018; Shan et al., 2018; Sruthy and

Ramasamy, 2017; Van Cauwenberghe et al., 2013; Xiong et al., 2018;

Young and Elliott, 2016; Zada et al., 2018; Zhang et al., 2017, 2016)

in aquatic organisms (Cole et al., 2013; Collard et al., 2017a, 2017b, 2015; Dehaut et al.,

2016; Enders et al., 2017; Galloway et al., 2017; Ghosal et al., 2018;

Goldstein and Goodwin, 2013; Halstead et al., 2018; Horton et al., 2018;

Karami et al., 2017b, 2016; Karlsson et al., 2017; Liboiron et al., 2018;

Murray and Cowie, 2011; Naidu et al., 2018; Remy et al., 2015; Van

Cauwenberghe et al., 2015; Van Cauwenberghe and Janssen, 2014;

Wagner et al., 2017; Watts et al., 2016, 2014; Xiong et al., 2018)

in food, drink and (Gündoğdu, 2018; Karami et al., 2017a, 2017c, 2018; Lei et al., 2017;

Muhlschlegel et al., 2017; Schymanski et al., 2018; Van Cauwenberghe

cosmetics
and Janssen, 2014; Wiesheu et al., 2016; Zada et al., 2018)

5
 ACCEPTED MANUSCRIPT

Model/Reference (Cai et al., 2018; Kaeppler et al., 2015; Karami et al., 2016; Ossmann et

al., 2017; Zada et al., 2018)

samples

Reviews and book (Andrady, 2017; Blaesing and Amelung, 2018; Crawford and Quinn,

2017; da Costa et al., 2016; Duis and Coors, 2016; Hanvey et al., 2017;
chapters
Hidalgo-Ruz et al., 2012; Horton et al., 2017b; Ivleva et al., 2017; Jiang,

2018; Klein et al., 2018; Li et al., 2017; Loeder and Gerdts, 2015; Lusher

et al., 2017; Mai et al., 2018; Miller et al., 2017; Qiu et al., 2016; Renner

et al., 2018; Ribeiro-Claro et al., 2017; Rocha-Santos and Duarte, 2015;

Rodríguez-Seijo and Pereira, 2017; Shim et al., 2017; Silva et al., 2018;

Syakti, 2017; Wu et al., 2018; Yu et al., 2018)

96

97 2. The importance of going small

98

99 It might be argued that, relative to μ-FTIR, the most attractive feature of μ-

100 Raman is higher resolution, which becomes especially relevant for identifying very

101 small MPs (< 20 μm) (Elert et al., 2017; Ivleva et al., 2017; Kaeppler et al., 2016)

102 otherwise undetectable using infrared techniques. As detailed in this section, there is an

103 urgent need for a fast and easily implementable monitoring tool capable of detecting

104 small MPs.

105 Plastic particles exposed to environmental stressors undergo continuous

106 fragmentation (photo-, thermal and biodegradation), so that the number of particles is

107 expected to increase steeply for smaller sizes (Andrady, 2017; Cozar et al., 2014).

108 Despite this fact, the smaller fraction of MPs has been consistently neglected in

109 quantification studies. Conkle et al. (Conkle et al., 2018) reviewed 41 surveys of aquatic

6
 ACCEPTED MANUSCRIPT

110 MPs in which plastic debris were collected using neuston nets. Circa 80% of these

111 studies used nets with relatively large mesh size (≥ 300 μm), thus entirely missing the

112 smaller MP fraction and leading to a severe underestimation of the actual MP load in

113 aquatic environments. Likewise, a long-term study (Beer et al., 2018) on MP

114 concentration in the Baltic Sea, spanning three decades, used a bongo net with mesh

115 size 150 μm and did not find a significant increase in MP abundance between 1987 and

116 2015. The results might have been different for the MP fraction below 150 μm. As

117 highlighted in a study by Enders et al. (Enders et al., 2015), among small MPs (<400

118 μm) collected in the Atlantic Ocean and identified using μ-Raman, 64% are under 40

119 μm in size, and their distribution in the range 10-100 μm follows a power law with a

120 scaling exponent of 1.96, as depicted in Figure 1.

121

122 [INSERT FIGURE 1 HERE]

123 Figure 1 - Particle size distribution of MPs collected in the Atlantic ocean (np=543). Dark grey bars

124 represent particles by length. Light grey bars show the size as geometric mean of length and width. Insets

125 show the smallest (left) and largest (right) MP particles found and confirmed via μ-Raman. Reprinted

126 from Marine Pollution Bulletin, 100, K. Enders, R. Lenz, C.A. Stedmon, T.G. Nielsen, Abundance, size

127 and polymer composition of marine microplastics ≥10μm in the Atlantic Ocean and their modelled

128 vertical distribution, 70-81, Copyright 2015, with permission from Elsevier.

129

130 These findings are supported by the conclusions of Erni-Cassola et al., (Erni-

131 Cassola et al., 2017, p.) who also used μ-Raman to identify MPs (<400 μm) from the

132 surface water of Plymouth Bay (UK), determining that the fraction of MPs smaller than

133 40 μm accounts for roughly 50% of the total population. Therefore, properly measuring

134 the abundance of MPs smaller than 40 μm is of paramount importance when estimating

135 the total MP load in the environment. Moreover, MP identification is increasingly

7
 ACCEPTED MANUSCRIPT

136 becoming a consumer health issue. In a recent statement (Alexander et al., 2016), the

137 European Food Safety Authority (EFSA) acknowledged the lack of legislation for

138 microplastics and nanoplastics as contaminants in food and the need to develop reliable

139 identification methods, especially for smaller MPs, which present a higher chance of

140 translocation across the gut barrier . Indeed, μ-Raman analysis has found MPs in edible

141 fish tissues (Collard et al., 2017a; Karami et al., 2017c, 2018), table salt (Gündoğdu,

142 2018; Karami et al., 2017a) and bottled water (Schymanski et al., 2018). The latter

143 study, by Schymanski and co-workers, analysed 38 brands of bottled water (plastic,

144 carton and glass containers) and found MP contamination in all of them, with 80% of

145 found particles belonging to the 5-20 μm size range. A comparison of FTIR and Raman

146 techniques for MP identification, performed by Kӓppler et al. (Kaeppler et al., 2016),

147 illustrates the superior performance of μ-Raman for detecting small MPs. While for

148 particles larger than 20 μm both techniques proved equally capable, for smaller particles

149 the detection success rate of μ-FTIR lagged behind. μ-Raman detected MPs as small as

150 5 μm, yet μ-FTIR missed all particles in the 5-10 μm size range and 40% of MPs in the

151 11-20 μm size range. Moreover, even though μ-FTIR did succeed in identifying some

152 MPs in the 11-20 μm range, the quality of the spectra suffered due to low signal to noise

153 ratio – a direct consequence of the size of the particle approaching that of the

154 instrument’s diffraction limit. This issue is illustrated by comparing the FTIR and

155 Raman spectra of a small (15-20 μm) polypropylene (PP) particle, as shown in Figure

156 2b, where a clear Raman spectrum (left) contrasts with a weak and noisy FTIR spectrum

157 (right).

158

159 [ INSERT FIGURE 2 HERE ]

160

8
 ACCEPTED MANUSCRIPT

161 Figure 2 – a) Raman image (left) and IR image (right) with false-colouring denoting the spectral intensity

162 in the 2780–2980 cm−1 range. B) Raman spectrum (left) and IR transmission spectrum (right) of particle 2

163 in comparison with a reference of polypropylene. Adapted by permission from Springer Nature

164 Analytical and Bioanalytical Chemistry, Analysis of environmental microplastics by vibrational

165 microspectroscopy: FTIR, Raman or both?, A. Käppler, D. Fischer, S. Oberbeckmann et al., Copyright

166 2016.

167

168 The efficiency of an ingenious post-filtration system implemented in the

169 wastewater treatment plant of Oldenburg (Germany) was evaluated by comparing the

170 amount of MPs present in wastewater samples collected upstream and downstream from

171 the filtration point (Mintenig et al., 2017). A removal efficiency of 100% was reported

172 for MPs > 500 μm and of 93% for those < 500 μm. Given that the filtration system

173 consists of pile fabric with a mesh size of 10-15 μm and that particles smaller than 20

174 μm could not be detected using FTIR imaging, a significant portion of very small MPs

175 might be passing through unnoticed. Being so, FTIR imaging studies should be

176 complemented by μ-Raman for properly accounting for MPs in the smaller size range.

177 However, μ-Raman is not used in the majority of MP identification studies (14%

178 according to a recent review (Renner et al., 2018)) in great part due to its

179 cumbersomeness (e.g. fluorescence masking) and long measurement time. Certainly, the

180 shortcomings of μ-Raman must be overcome prior to its large scale implementation.

181 Recently, a considerable effort has been made to optimize signal quality (section 3) and

182 automation routines (section 4) that enable faster and more reliable μ-Raman

183 identification of MPs.

184

185

186 3. Enhancing signal quality

9
 ACCEPTED MANUSCRIPT

187

188 Two commonly cited drawbacks of classical Raman scattering as a method for

189 MP analysis are the inherent weakness of the signal – only about 10-8 of the photons

190 bombarding the sample are actually translated to Raman signal (Borman, 1982) – and

191 its proneness to fluorescence interference – either intrinsic to the main constituent of the

192 MP or due to impurities such as colouring agents, biological material and degradation

193 products. A weak signal imposes the necessity of extending the integration time which,

194 in the best case scenario, increases the duration of the measurement, and in the worst

195 case scenario results in laser-induced degradation of the sample. Fluorescence results in

196 a raised baseline which, in the worst cases, completely overshadows the Raman signal.

197 There are many possible routes to minimize these problems. A radical approach is to

198 forego spontaneous Raman scattering in favor of nonlinear Raman techniques, which by

199 their very nature offer high signal to noise ratio while being free from the interference

200 of fluorescence (Borman, 1982). However, nonlinear Raman methods require expensive

201 equipment and expert user knowledge so that their implementation in MP analysis is

202 still incipient (Cole et al., 2013; Galloway et al., 2017; Watts et al., 2016, 2014; Zada et

203 al., 2018). The present section focuses instead on offering examples of recently

204 suggested good practices designed to minimize the weak signal and fluorescence

205 problems of spontaneous Raman spectroscopy.

206 In the specific context of MP analysis, an obvious way of reducing fluorescence

207 caused by organic debris – and, in some cases, by organic dyes - is the use of an

208 appropriate cleaning protocol to remove the contaminants. Among the pre-treatments

209 proposed for cleaning MPs are those employing acids, bases, oxidative and enzymatic

210 agents, as extensively discussed in a series of recent reviews (Blaesing and Amelung,

211 2018; Lusher et al., 2017; Miller et al., 2017, 2017; Renner et al., 2018; Silva et al.,

10
 ACCEPTED MANUSCRIPT

212 2018). Worthy of mention is the benchmark study of Dehaut et al. (Dehaut et al., 2016)

213 which assessed the efficiency and degradation effect of six digestion methods. The

214 authors concluded that digestion with nitric acid, at the time the official method

215 recommended by the International Council for the Exploration of the Sea (ICES), leads

216 to significant polymer degradation and consequent MP underestimation. Instead, the

217 authors recommended digestion in a 10% KOH solution. Similar conclusions were

218 reached by Enders et al. (Enders et al., 2017), who confirm the destructive effects of

219 nitric acid digestion and, in partnership with ICES, issued a new official

220 recommendation for MP cleaning using a mixture of 30% KOH:NaClO Alternatively,

221 an enzymatic digestion method employing proteinase K efficiently cleaned samples of

222 salmon viscera and fecal matter, spiked with MPs, without significant polymer

223 degradation (Karlsson et al., 2017).

224 Alas, even after thorough cleaning, some samples will exhibit a problematic

225 degree of fluorescence, mainly due to the presence of hard to remove colouring agents.

226 A commonly cited solution is photo-bleaching the sample, that is, placing it under the

227 laser for the time necessary for degrading the fluorescing agent. However, besides being

228 time consuming, this strategy cannot be applied to samples prone to photo-degradation

229 or pyrolysis – and even when applied it does not always work. A speedy solution for

230 enabling correct identification of fluorescing samples is proposed by Ghosal et al.

231 (Ghosal et al., 2018) who use an automated algorithm to remove the fluorescence

232 background and reveal the underlying polymer spectrum. A striking example of the

233 algorithm’s effectiveness is shown in Figure 3 depicting a plastic particle partially

234 covered in biofilm and Raman spectra collected at covered (red) and bare (black)

235 locations. Prior to background subtraction the spectrum of the covered surface is

236 saturated with the fluorescence signal which completely overshadows the characteristic

11
 ACCEPTED MANUSCRIPT

237 polymer peaks. After processing, the polymer peaks are clearly visible, allowing

238 polymer identification by a library matching software. The source code for the

239 algorithm, which includes a graphical user interface, is freely available at

240 https://github.com/michaelstchen/modPolyFit.

241

242 [ INSERT FIGURE 3 HERE ]

243

244 Figure 3 – Raman spectra of a polyethylene (PE) particle partially covered in biofilm. The characteristic

245 PE peaks readily apparent in the raw spectrum collected from the bare plastic surface (black line, upper

246 panel) seem to be absent from the raw spectrum of the biofilm-covered plastic surface (red line, upper

247 panel). After automated fluorescence correction, the spectral fingerprint of PE clearly emerges in the

248 processed spectrum of the biofilm-covered plastic surface (red line, lower panel). Adapted from

249 Environmental Pollution, 233, S. Ghosal, M. Chen, J. Wagner, Z. Wang, S. Wall, Molecular identification

250 of polymers and anthropogenic particles extracted from oceanic water and fish stomach – A Raman

251 micro-spectroscopy study, 1113-1124, Copyright 2018, with permission from Elsevier.

252 An additional source of fluorescence may be the substrate underneath the MPs

253 (i.e. the filter), as noted by Oßmann et al. (Ossmann et al., 2017) who compared the

254 performance of six commercial filters commonly used for μ-Raman analysis and custom

255 made polycarbonate substrates coated with three different metals. The filter offering the

256 best performance is made of polycarbonate coated with aluminum, which not only

257 minimizes fluorescence and burning of the sample but also optimizes optical contrast

258 between filter and MPs, thereby easing automated particle search.

259 As mentioned before, a drawback of Raman spectroscopy is the notoriously

260 weak intensity of Raman scattering, which requires relatively long acquisition times to

261 achieve a decent signal to noise ratio. One avenue for signal enhancement arises from

262 the development of improved detector systems – as illustrated by the comparison

12
 ACCEPTED MANUSCRIPT

263 between conventional and electron-multiplying charged coupled device detectors (CCD

264 vs. EM-CCD). Relative to conventional CCD detectors, EM-CCD detectors are

265 equipped with a multiplication register that acts prior to the readout – amplifying the

266 gain up to 1000 times (Andor Technology, 2007; Griffiths and Miseo, 2014). As a

267 result, EM-CCD detectors are expected to require less acquisition time to achieve the

268 same signal to noise ratio as conventional CCD detectors, rendering the former

269 especially relevant for fast Raman mapping and imaging applications (Griffiths and

270 Miseo, 2014). The power of EM-CCD detectors applied to Raman characterization of

271 MPs is clearly illustrated by work of Dieing and Hollricher, from which Fig. 4 has been

272 retrieved (Dieing and Hollricher, 2008). Figure 4a) depicts the Raman image of

273 contaminated poly(methyl methacrylate) (PMMA) on a glass slide obtained using a

274 classical CCD detector and with an integration time of 36ms/spectrum. The signal to

275 noise ratio is low, so that one can barely distinguish the vertical stripe in the center

276 (glass) from the side regions (PMMA+impurities). Using a EM-CCD detector, the same

277 signal to noise ratio is achieved in a tenth of the integration time (Figure 4b). By setting

278 the integration time at the same as that used for the CCD detector, a superior image is

279 obtained (Figure 4c), where not only the glass is clearly distinguished from PMMA, but

280 the streaks of impurities are also easily discernible.

281

282 [ INSERT FIGURE 4 HERE ]

283

284 Figure 4 - Raman Images of a glass slide, partially covered with a layer of PMMA and contaminant

285 (scheme d), obtained with: (a) back-illuminated CCD; (b and c) EMCCD. Scale bar:10 μm. Reprinted

286 from Vibrational Spectroscopy, 48 (1), T. Dieing, O. Hollricher, High-resolution, high-speed confocal

287 Raman imaging, 22-27, Copyright 2008, with permission from Elsevier.

13
 ACCEPTED MANUSCRIPT

288

289

290 4. Automated routines for Raman mapping

291

292 Until recently, the great majority of studies reporting MP load in environmental

293 samples relied solely upon visual inspection (usually under a microscope) to determine

294 the total number of plastic particles in the analyzed sample. The perils of such approach

295 have been highlighted in the impactful work of Lenz et al. (Lenz et al., 2015) who warn

296 that visual inspection alone leads both to false positives (e.g. paint particles) and false

297 negatives (e.g. darkly coloured MPs mistaken for naturally occurring particles). Visual

298 detection relies upon morphological criteria to decide whether a given particle is made

299 of plastic, an approach that works reasonably well for bigger MPs (>100 μm) but

300 progressively fails as MP size diminishes and morphological features become less

301 apparent. Thus, among those MPs <50 μm which were selected under the microscope,

302 only 63% could be confirmed as plastics after Raman analysis (Lenz et al., 2015). As

303 the authors point out, running a FTIR or Raman analysis following visual sorting

304 eradicates the problem of false positives but perpetuates the issue of false negatives.

305 This concern is well expressed in the study of Song et al. (Song et al., 2015) which

306 compares the effectiveness of visual pre-sorting under a microscope followed by FTIR

307 analysis with that of analyzing the spectra of all particles present in the sample. While

308 full FTIR analysis always leads to a larger estimate of MP abundance, the contrast is

309 most staggering for particles smaller than 50 μm for which the visual sorting procedure

310 failed to account for roughly half of the MPs detected using FTIR. One may argue that

311 if Raman had been used instead, the difference would be even more striking. It is then

14
 ACCEPTED MANUSCRIPT

312 clear that visual pre-sorting, besides taking up far too much operator time, likely leads

313 to less than representative results. An obvious solution to increase representativeness is

314 to analyze every single particle in a given sample, however such approach requires even

315 more operator time than visual pre-sorting. A more viable solution is automated analysis

316 and there are various ways of doing so. One possibility is full point mapping, where a

317 whole filter section is analyzed by collecting spectra at various points along a grid with

318 the aid of a motorized stage which moves the sample under the laser in small

319 increments. While thorough, this approach is rather slow, as illustrated in the work of

320 Käppler et al. (Kaeppler et al., 2016), whose point-by-point μ-Raman mapping without

321 pre-selection took a whopping 38 h to scan 1 mm2 of the sample filter using a point

322 distance of 10 μm and total integration time of 10 sec per point. By reducing the

323 integration time to 0.5 sec per point and using a topographical imaging tool to enhance

324 focus, the total measurement time was reduced to 90 min but the increase in speed came

325 at the expense of detection success rate, which was equivalent to that of the FTIR

326 imaging technique.

327 The major drawback in point by point mapping for MP analysis is that it does

328 not detect whether there is an actual particle under the laser – the whole grid is scanned,

329 so a good proportion of laser time will be wasted measuring the spectrum of the filter

330 itself. Time can be saved by guiding the laser directly to a small area where a particle is

331 located. Elert et al. (Elert et al., 2017) used this approach in their comparison of four

332 techniques (FTIR imaging, Raman mapping, mass spectrometry and chemical

333 extraction) for the analysis of soil spiked with reference MPs. Particles ranging from

334 150 to 200 μm were visually selected under the microscope and the scanning area

335 adjusted to fit each analyzed particle – an undesirable practice for environmental studies

336 but appropriate for comparison purposes. A pre-scan was run with a short integration

15
 ACCEPTED MANUSCRIPT

337 time of 20 ms per point in order to create a spectral image of each particle. Then, higher

338 resolution spectra were collected only at specific points of interest (i.e. a clear spot vs a

339 fluorescing spot). This procedure was more time-efficient than full mapping, taking

340 circa 15 hours per mm2. Besides, by focusing the analysis on a smaller area, a shorter

341 size step could be used (1 μm) so that different chemical compositions at the particle

342 surface could be distinguished. While this approach is valuable for in-depth studies,

343 routine monitoring protocols must be swifter.

344 An interesting strategy is the use of image analysis software which, by

345 enhancing contrast between particles and filter, creates a map of all particles in a given

346 area and Raman spectra are collected only at those points. Schymanski et al.

347 (Schymanski et al., 2018) used the Single Particle Explorer (rapID) software to

348 automatically detect particles larger than 5 μm up to a maximum of 5000 particles per

349 scanned area. Chemical identification was achieved by library matching and all MPs

350 with a ranking score between 550 and 700 were then individually analyzed using longer

351 integration times to improve signal to noise ratio. Despite the time consuming

352 verification step, the analysis was performed at a rate of 56 min/mm2, a significant

353 improvement relative to point mapping approaches. Frère et al. (Frere et al., 2016)

354 achieved an even shorter measurement time of 20s/mm2 by performing an automated

355 particle search using the ParticleFinder (Horiba) software and collecting the spectrum of

356 each detected particle. The higher speed efficiency of this measurement relative to that

357 of Schymanski’s study has to do with the size of the MPs sampled (> 300 μm) and the

358 absence of a thorough verification step, a comparison which nicely illustrates the trade-

359 off between meticulousness and celerity. An optimized software for the automated

360 detection of particles and fibres in the 1 μm – 500 μm range is currently being

361 developed by Käppler and colleagues, within the framework of the project

16
 ACCEPTED MANUSCRIPT

362 MICROPOLL - EU BONUS program (Oberbeckmann, 2017), which might allow a

363 more thorough analysis in a shorter time scale.

364 A truly innovative approach is the method for MP detection using a fluorescent

365 dye, initially suggested by Andrady (Andrady, 2011) and demonstrated in recent works

366 (Erni-Cassola et al., 2017; Maes et al., 2017; Shim et al., 2016), which currently yields

367 the best compromise between speed and accuracy . Nile Red, a solvatochromic

368 fluorescent dye, was found to selectively stain synthetic polymers – but not tire rubber -

369 enabling their fast detection under a UV light, as depicted in Figure 5.

370

371 [ INSERT FIGURE 5 HERE ]

372

373 Figure 5 - Microscope images of a) polyethylene, b) nylon 6 and c) rubber stained with Nile red. Left

374 panel: brightfield images, center and right panels: fluorescent image at excitation/emission 460/525 nm

375 (green) and 565/630 nm (red), respectively. Scale bar is 125 μm. Adapted with permission from

376 Environmental Science & Technology, 51, G. Erni-Cassola, M.I. Gibson, R.C. Thompson, et al., Lost, but

377 Found with Nile Red: A Novel Method for Detecting and Quantifying Small Microplastics (1 mm to 20

378 μm) in Environmental Samples, 13641−13648, Copyright 2017 American Chemical Society.

379 This procedure might seem counterproductive, as the fluorescence induced by

380 the dye may mask the Raman signal. However, one author (Maes et al., 2017) claims

381 that “The very small amounts of Nile Red adsorbed on the particles did not interfere

382 with IR or Raman spectroscopy” while another (Erni-Cassola et al., 2017) reports an

383 undesirable level of interference. The authors minimize the deleterious effects of

384 fluorescence by photo-bleaching the sample for 5 min prior to Raman analysis. The

385 great advantage of this method lies in the particle selection stage, where an automated

386 particle search is applied using imaging software and detecting only fluorescing

17
 ACCEPTED MANUSCRIPT

387 particles, so that only a fraction of the particles initially present (under 10%) are

388 selected for further analysis. Not all of the stained particles are necessarily plastics –

389 non-plastic fragments contaminated with hydrophobic biological material are stained as

390 well. However their detection can be minimized by combining an effective cleaning

391 step (prior to Nile Red staining) and the use of a UV wavelength that induces green

392 instead of red fluorescence – plastic particles fluoresce better in green while organic

393 material does not (Erni-Cassola et al., 2017).

394 Following this strategy, Erni-Cassola and co-workers located MPs as small as 20

395 μm and determined their chemical composition using μ-Raman totaling a measurement

396 time of almost 2 minutes per mm2. If the photo-bleaching step had been omitted, the

397 total analysis time would be roughly 40 seconds per mm2. A more time-efficient

398 procedure would be to: 1) run a first test without photo-bleaching MPs; 2) run the raw

399 spectra through an automated baseline correction script (Ghosal et al., 2018); 3) attempt

400 automatic identification through library matching and 4) photo-bleach and re-analyze

401 only those particles which could not be successfully identified in the first run.

402 The aforementioned studies of MP identification using μ-Raman have all relied

403 on a mapping – or “point and shoot” – strategy which collects spectra one point at a

404 time. However, it is already possible to image a whole area in a single scan using

405 widefield techniques, as discussed in section 6. Such strategy is often implemented for

406 FTIR analysis of MPs using a focal plane array (FPA) detector. Imaging a whole area in

407 a single measurement promises gains in expediency and representativeness by

408 dispensing with the step of particle selection and the use of a motorized stage. Yet, at

409 present, FTIR imaging with FPA is not necessarily more time-efficient than Raman

410 mapping. The measurement time/mm2 of a few selected works employing either Raman

411 mapping or FPA to detect MPs are presented in Table 2. Both categories showcase

18
 ACCEPTED MANUSCRIPT

412 examples in which 1 mm2 of sample area was analyzed within a few minutes, and even

413 seconds. On top of being comparable to FTIR imaging in terms of analysis time, Raman

414 mapping offers the possibility of identifying MPs in the 1-20 μm range. This

415 competitive “edge” of Raman mapping is likely to fade in coming years, as

416 advancements in instrumentation deliver affordable and time-efficient devices for

417 proper Raman imaging.

418

419

420 Table 2 – Comparison of studies employing automated routines for MP identification using μ-Raman and

421 μ-FTIR. The total analysis time per mm2 was calculated based on the total measurement time reported by

422 the authors divided by the measured area.

Min.

particle Int. time Total analysis

Ref. Technique Protocol
size /point time per mm2

detected

μ-Raman - No pre-selection

(Kaeppler -Automated scanning of the

alpha 300R
et al., Raman whole filter area in 10 μm steps 5 μm 10 s 38 h/mm2

2016) microscope -ID by automated library

(WITec) matching

-Visual pre-selection of potential

μ-Raman MPs

-Automated scanning of small 15 h/mm2 +

((Elert et alpha 300R

Raman filter sub-sections (max: 0.2 x 150 μm 0.02 s pre-selection

al., 2017)
microscope 0.19 mm) in 1 μm steps time
(WITec) -ID by comparison with ref

spectra

(Schymans μ-Raman -Automated particle search by 5 μm 5s 56 min/mm2

19
 ACCEPTED MANUSCRIPT

ki et al., Single the Single Particle Explorer

Particle -Raman spectra collected at
2018)
Explorer
each particle location
(rapID)
-ID by automated library

matching followed by individual

verification of low confidence

matches

-Automatic particle search by

μ-Raman the ParticleFinder software

(Frere et -Raman spectra collected at

LabRAM HR800
300 μm 20 s 20 s/mm2
al., 2016) (Horiba each particle location

Scientific) -ID by automated library

matching

-Staining with fluorescent dye

μ-Raman followed by automatic particle

(Erni-
search by the ImageJ software
inVia Raman
Cassola et 20 μm 200 s ~2 min/mm2*
microscope -Raman spectra collected at
al., 2017)
(Renishaw) each particle location

-ID method not specified

μ-FTIR
Use of a focal plane array for
(Elert et
Vertex 70
direct imaging of a whole filter 100 μm - 111 min/mm2
al., 2017) spectrometer
section
(Bruker)

(Kaeppler μ-FTIR
Use of a focal plane array for
Tensor 27
et al., direct imaging of a whole filter 15 μm - 20 min/mm2
spectrometer
2016) section
(Bruker)

μ-FTIR
Use of a focal plane array for
(Loeder et
PerkinElmer
direct imaging of a whole filter 20 μm - 6 min/mm2
al., 2015) Spotlight micro-
section
FT-IR

20
 ACCEPTED MANUSCRIPT

19 s/mm2
μ-FTIR
Use of a focal plane array for
(Tagg et (1 scan)
Tensor 27
direct imaging of a whole filter 150 μm -
al., 2015) spectrometer 33 s/mm2
section
(Bruker) (2 scans)

423 *The authors only disclosed the time necessary for particle detection (2.45 s/mm2) so the total

424 analysis time was estimated based on the time necessary to measure all the analyzed particles (199x200s)

425 plus the pre-measurement bleaching step (199x5min) divided by the total scanned area (23 sections of

426 23x1.8 mm each)

427

428 5. Spectral Libraries

429

430 Automated μ-Raman routines employ library matching software to compare the

431 spectrum of the sample with that of custom made and/or commercial libraries. The

432 likelihood of successful matching greatly depends upon the comprehensiveness of the

433 spectral library. However, custom made libraries usually rely on spectra acquired from

434 pristine polymer pellets and may differ significantly from those of MPs collected from

435 environmental compartments. Environmental MPs mostly result from the fragmentation

436 of commercial polymers, available in different morphologies (foam, sheet, fibre) and

437 include in their composition a variety of additives, fillers and colouring agents. These

438 components, in some instances, overlay or even mask the underlying polymer spectrum.

439 The use of more comprehensive commercial libraries is a step in the right direction to

440 enhancing matching scores, especially if the matching algorithm uses multicomponent

441 correlations (Lenz et al., 2015). Among 31 studies using automated matching software

442 for MP identification, 6 rely solely on custom made reference libraries, 13 used solely

443 commercial libraries (although most of them use a collection of libraries from different

21
 ACCEPTED MANUSCRIPT

444 vendors) and 12 use a combination of commercial and custom made libraries. The need

445 for different commercial libraries, often complemented with own reference spectra is a

446 good indicator that no single commercial library is currently broad enough to assure

447 extensive success rates. Perhaps part of this problem is due to the fact that both

448 commercial and custom made libraries only include spectra from plastics that were not

449 exposed to environmental degradation.

450 MPs in the environment are continuously exposed to a variety of environmental

451 stressors (UV light, heat, biodegradation) that result in plastic weathering (Andrady,

452 2017; Lenz et al., 2015). For example, polyvinylchloride (PVC) is prone to photo-

453 degradation in aquatic media, as the leakage of additives, such as photo-stabilizers,

454 accelerates under high humidity conditions. UV degradation significantly alters the

455 spectral fingerprint of PVC showing a simultaneous intensity decrease of the

456 neighboring peaks at 693 cm−1 and 637 cm−1, which stem from the characteristic C-Cl

457 bonds of the polymer (Lenz et al., 2015). For the spectrum corresponding to the highest

458 UV exposure, a complete absence of these double peaks and the appearance of two

459 strong peaks at 1139 cm−1 and 1540 cm−1, assigned to carbon double bonds (C=C), are

460 observed. A successful match through library search was impossible since the only

461 reference for PVC was that of a pristine (non-degraded) sample. In the light of this

462 drawback, it is of utmost importance that spectra of degraded polymers at different

463 degradation stages be included in the reference library, thus increasing the chance of the

464 matching software correctly identifying the polymeric composition. It should however

465 be noted that photo-oxidation is not as problematic for MP identification using Raman

466 as it is when using FTIR techniques. Photo-oxidation of the most common polymers

467 (i.e. PE, PP) results in the formation of oxygenated moieties, chiefly C=O and –OH,

468 which display strong intensities in the infrared spectrum (Andrady, 2017; Cai et al.,

22
 ACCEPTED MANUSCRIPT

469 2018; Costa et al., 2018; Rodrigues et al., 2018), contrasting with their typically weak

470 Raman signals. For example Cai et al. (2018) found obvious differences among the

471 infrared spectra of pristine and degraded PE, PP and PS while the corresponding Raman

472 spectra only registered slight differences in band intensities. This feature of Raman may

473 be considered as an advantage – when the purpose is to simply identify plastic identity –

474 or a disadvantage – for those studies focused on assessing the extent of polymer

475 degradation.

476 Moreover, as Lenz and colleagues have done, it is of value to include the spectra

477 of non-plastic materials which are often confused with MPs, such as cellulose, keratin,

478 inorganic particles and, most important of all, synthetic fibres (i.e. viscose), whose

479 presence is ubiquitous among MP samples. The obvious hurdle with this approach is

480 that it is time consuming for each individual laboratory to build a comprehensive

481 spectral library. An open source, curated spectral database would be of great utility: the

482 effort and material resources spent by each individual researcher would be minimized

483 by avoiding repetition (i.e. each group having to build their own library) while

484 achieving a degree of complexity and completeness currently unavailable in a single

485 commercial library. The inclusion of spectra from real environmental samples, whose

486 identity, when in doubt, could be confirmed through other identification techniques,

487 would greatly increase matching scores. Besides, having access to a free library would

488 encourage more MP identification studies for research groups with narrow funding,

489 thereby expanding our current perception of global MP pollution. Such an enterprise

490 might seem far-fetched, however an open source database for mass spectra – Curatr -

491 has been developed at the European Molecular Biology Laboratory, Germany, (Palmer

492 et al., 2017) and is available at http://curatr.mcf.embl.de/about/curatr/. The spectra

493 uploaded by authorized curators are converted into a format compatible with library

23
 ACCEPTED MANUSCRIPT

494 matching software so that any user can easily download and use them in their own

495 studies.

496 Finally, an interesting way of optimizing automated identification of MPs with

497 complex composition might be using the Raman barcode strategy outlined by Lawson

498 and Rodriguez ((Lawson and Rodriguez, 2016) for the fast identification of counterfeit

499 drugs. Briefly, both reference (APIs) and sample (counterfeit drug) spectra are

500 converted into a barcode where each line represents a spectral peak, as illustrated in

501 Figure 6. The barcode of the sample is then identified through binary comparison with

502 the reference barcodes, thus speeding up matching and dispensing with spectral

503 intensity normalization. The barcode strategy has not yet been applied for the

504 identification of microplastics although it could easily be adapted for that purpose: the

505 reference barcodes would be those of polymers, colouring agents and common plastic

506 additives. The advantage of using the barcode for Raman spectral matching is that

507 ignoring spectral intensity would improve the identification rates for those MPs whose

508 spectra are overshadowed by the peaks of the colouring agent and/or additives.

509

510 [ INSERT FIGURE 6 HERE ]

511

512 Figure 6 – Illustration of the Raman barcode generation: peaks marked with a circle are included as

513 vertical stripes in the barcode spectrum. Reprinted with permission from Analytical Chemistry, 88, L.S.

514 Lawson, J.D. Rodriguez, Raman Barcode for Counterfeit Drug Product Detection, 4706−4713. Copyright

515 2016 American Chemical Society.

516

517

518

24
 ACCEPTED MANUSCRIPT

519

520 6. Future prospects – emerging non-conventional Raman

521 techniques

522

523 This section discusses non-conventional Raman techniques. These are not, at

524 present, commonly used for the identification of environmental MPs yet might grow in

525 importance for MP research as the necessary instrumentation becomes more easily

526 accessible. A theoretical description of unconventional Raman techniques falls outside

527 the scope of this review, having been comprehensively discussed elsewhere (Stewart et

528 al., 2012; Opilik et al., 2013). Furthermore, the aim here is not to delineate novel

529 detection protocols but merely bringing the reader’s attention to a wider set of possible

530 Raman tools using concrete examples from the available literature.

531 As mentioned before, the majority of works employing μ-Raman for MP

532 detection follows a mapping, point by point strategy. The spectra collected at each

533 spatial location may then be put together to create a Raman image. However, this cannot

534 be considered a true imaging technique. Proper Raman imaging may rely on either

535 spontaneous or nonlinear processes. In spontaneous Raman imaging – wide-field

536 Raman - a whole sample section is illuminated by defocusing the laser and collecting an

537 image in a single measurement, a strategy not yet applied for environmental MP

538 analysis. In contrast, proper FTIR imaging using FPAs for MP detection has been

539 demonstrated before. An interesting example of wide-field Raman imaging is the proof

540 of concept demonstration of Schmälzlin et al. (Schmaelzlin et al., 2014), where a fiber-

541 coupled high-performance astronomy spectrograph (MUSE) was used to image

542 polystyrene (PS) (50 μm) and PMMA (100 μm) microshperes scattered on a metal plate.

25
 ACCEPTED MANUSCRIPT

543 The exposure time was 2 minutes and, after background subtraction and automated

544 fluorescence correction, the area of non-overlapping characteristic peaks was integrated

545 over the sample area to produce the false color images shown in Figure 7. Although the

546 image presents poor spatial resolution the location and size of the PS and PMMA

547 microspheres are clearly defined. The authors themselves warn that this preliminary test

548 is intended to illustrate the lower limit of the technique’s capabilities, since the MUSE

549 spectrograph (designed for astrophysical exploration of distant galaxies) had not been

550 optimized for this specific purpose.

551

552 [ INSERT FIGURE 7 HERE ]

553

554 Figure 7 –Left panel: optical image of plastic microbeads. Center panel: Raman image of PMMA (600

555 cm−1); Right panel: Raman image of PS (1035 cm−1). Scale bar is 50 μm. Reprinted from Sensors, 14, E.

556 Schmälzlin, B. Moralejo, M. Rutowska, et al., Raman Imaging with a Fiber-Coupled Multichannel

557 Spectrograph, 21968-21980, Copyright 2014, under the terms of the Creative Commons Attribution

558 license 3.0.

559

560 The next frontier in MP characterization is real-time analysis of flowing

561 particles. Conventional Raman techniques, based on spontaneous Raman scattering, are

562 unfit for this purpose. The inherently weak scattering intensity results in the necessity of

563 long dwelling times – milliseconds to seconds - to achieve a good signal. However, the

564 improved signal intensity achieved by nonlinear Raman techniques such as coherent

565 anti-Stokes Raman scattering (CARS) and stimulated Raman scattering (SRS) opens up

566 new possibilities for real-time MP analysis. In CARS and SRS, a strong signal is

567 elicited only from the molecular vibrational modes of interest. Thus, the signal is free

568 from fluorescence, even if strongly fluorescing contaminants are present - as long as
26
 ACCEPTED MANUSCRIPT

569 these are inactive at the frequency of interest, the sample signal will not be influenced.

570 Therefore, a sufficient signal can be acquired quickly and sample preparation becomes

571 less important – an ideal scenario for real time detection of MPs in environmental

572 compartments.

573 The use of CARS for locating MPs within biological organisms has been

574 demonstrated before for zooplankton (Cole et al., 2013) and crabs (Watts et al., 2016,

575 2014), as briefly discussed elsewhere (Ribeiro-Claro et al., 2017), and more recently,

576 fairy shrimp and zebrafish embryos (Galloway et al., 2017). The latter study, by

577 Galloway and colleagues, detected nanoparticles of acrylic co-polymers as small as 80

578 nm, thereby pushing the limit of plastic particle detection from the micro- to the

579 nanoscale.

580 The potential of SRS for fast analysis of MPs did not go unnoticed by the

581 microplastics research community, as demonstrated in a recent proof of principle study

582 by Zada and co-workers (2018). SRS microscopy was used, in a time-efficient manner,

583 to detect and distinguish MPs of five commonly encountered polymers. The analysis

584 was calibrated at six different wavenumbers (Figure 8a), selected by statistical methods

585 to tell apart the MPs from the background while maximizing polymer type

586 discrimination. Figure 8b displays the SRS identification images of model MPs (upper

587 panel) and its respective color maps identifying polymeric composition (lower panel).

588 The technique was successfully applied in the detection of MPs present in glitter nail

589 polish and sediments from the Rhine estuary. Zada et al. report a total SRS analysis time

590 of 2.7 minutes per mm2, an impressive speed especially when taking into account that

591 there was no particle pre-selection: the whole filter area was scanned for Raman signals.

592 A comparable full mapping analysis using conventional Raman took 38 hours per mm2,

593 as reported by the same authors. With conventional Raman, short measurement times

27
 ACCEPTED MANUSCRIPT

594 ranging from a few seconds to a few minutes per mm2 are achievable, but only by

595 drastically reducing the number of data collection points through automated particle

596 pre-selection (see Table 2). The advantage of the latter strategy is greater flexibility,

597 since it can detect a broad range of polymer types, while the SRS scheme suggested by

598 Zada and co-workers only identifies the polymer classes included in the calibration step.

599

600 [INSERT FIGURE 8 HERE]

601

602 Figure 8 – a) Spontaneous Raman spectra of Nylon, PET, PS, PE and PP with circles indicating

603 the wavenumbers used in the SRS analysis. b) Upper panel: SRS identification images of the

604 aforementioned polymers, with unequivocal identification points shown as white. Lower panel: Color

605 coded version of the identification image above. Adapted from (Zada et al., 2018) under a Creative

606 Commons license (https://creativecommons.org/licenses/by/4.0/legalcode)

607 Zhang et al. (C. Zhang et al., 2017) demonstrated the capabilities of SRS

608 coupled with flow cytometry in the fast identification of plastic microspheres. Beads of

609 PMMA and PS, 10 μm in size, were suspended in water and run through a flow

610 cytometry device coupled to a multichannel SRS system (FC-SRS). The FC-SRS

611 prototype includes a custom-built signal amplifier, allowing the collection of a Raman

612 spectrum in merely 5 μs. With the flow speed set to 400 mm/s, ~11 000 microspheres

613 per second were successfully identified, an astonishing detection rate compared with the

614 timescales previously discussed in this review. Figure 9 (left) shows a time-stack SRS

615 image corresponding to 1.8 ms, a short time window during which 8 PMMA and 5 PS

616 beads were identified through their characteristic peaks in the CH stretching region,

617 shown in Figure 9 (right), aided by multivariate curve resolution.

618

28
 ACCEPTED MANUSCRIPT

619

620 [ INSERT FIGURE 9 HERE ]

621

622 Figure 9 – Left panel: Time stacked window recorded in 1.8 ms, 8 PMMA beads (peak centered at 2955

623 cm−1) and 5 PS beads (peak centered at 3060 cm−1) were detected. Right panel: SRS (dashed) and

624 spontaneous Raman Spectra (bold) of PMMA and PS. Reprinted from Optica, 4, C. Zhang, K. Huang, B.

625 Rajwa et al., Stimulated Raman scattering flow cytometry for label-free single-particle analysis, 103-109,

626 Copyright 2017, under OSA Open Access License

627

628 Combining real time Raman analysis and Raman imaging, Liao et al. (2017)

629 recently proposed a handheld SRS microscope with optical fibre laser delivery that

630 permits in situ and in vivo chemical identification. The device, depicted in Figure 10,

631 permits imaging at a rate of 8 frames per second, thereby reducing distortion generated

632 by hand movement. The Raman image shown in Figure 10, displaying the spatial

633 locations of PS and PMMA microbeads (5 μm), as well cellulose fibers (from the filter

634 paper underneath) was generated in a mere 3 seconds. Fast, steady and accurate, the

635 handheld SRS device might be an interesting option for real-time monitoring of MPs in

636 food, cosmetics and other consumer products. The handheld device could be suspended

637 above the production line treadmill, attached to an automated arm which scans the

638 surface of the product and rapidly detects the presence of MPs.

639 [ INSERT FIGURE 10 HERE ]

640

29
 ACCEPTED MANUSCRIPT

641 Figure 10 –Handheld SRS microscope. Inset: SRS image at 2890 cm-1 of PS/PMMA microspheres on

642 paper (bottom) and output spectra (top). Reprinted with permission from ACS Photonics, C. Liao, P.

643 Wang, C.Y. Huang et al., In Vivo and in Situ Spectroscopic Imaging by a Handheld Stimulated Raman

644 Scattering Microscope, Article ASAP, DOI: 10.1021/acsphotonics.7b01214. Copyright 2017 American

645 Chemical Society

646

647 While, in the examples discussed throughout this review, the sample is placed

648 within close reach to the Raman detector, it is possible to collect Raman spectra from

649 afar, at distances up to hundreds of meters. This approach, deemed Standoff Raman

650 spectroscopy, has been developed for detecting samples which are hard to reach and/or

651 dangerous to handle, such as explosives (Gares et al., 2016; Stewart et al., 2012).

652 Östmark et al. (Ostmark et al., 2011) employed hyperspectral spontaneous Raman

653 imaging for distinguishing explosives from inorganic particles, measuring circa 5 mm,

654 at a distance of 10 m. The method was then successfully used to detect sulfur

655 microparticles (20 μm) dispersed over a brick surface.

656 Bremer et al. (Bremer et al., 2011) used standoff CARS to identify explosives

657 present in trace quantities in a complex chemical background – a polymer solution of

658 PS and PMMA dispersed in toluene. The Raman images, recorded from a distance of 1

659 m, were not subject to any background subtraction or other processing procedures, yet

660 regions rich in PMMA, PS and DNT (explosive) were clearly distinguishable, as shown

661 in Figure 11.

662

663 [ INSERT FIGURE 11 HERE ]

664

665 Figure 11 – Stand-off Raman images created from a distance of 1 m showing trace detection in a complex

666 environment. Title on each chemical image refers to the resonance monitored: PS = 1200 cm_1,

30
 ACCEPTED MANUSCRIPT

667 DNT=1350 cm_1, PMMA=1750 cm_1, and 2,6-DNT=1090 cm_1. Adapted from Applied Physics

668 Letters, 99:10, M.T. Bremer, P.J. Wrzesinski, N. Butcher et al., Highly selective standoff detection and

669 imaging of trace chemicals in a complex background using single-beam coherent anti-Stokes Raman,

670 101109, Copyright 2011, with the permission of AIP Publishing.

671 Standoff Raman techniques may be particularly interesting for monitoring MPs

672 in the ocean surface by passing ships or, indeed, for monitoring any surface with the aid

673 of a drone or a helicopter, as currently done in the search for explosives in enemy

674 territory using the LIDAR technology.

675

676

677 7. Conclusions

678

679 This review outlined the major advantages and limitations of Raman

680 spectroscopy for the identification of environmental microplastics. Recently suggested

681 good practices for overcoming the well-known drawbacks of µ-Raman were discussed

682 and some possible future trends for MP analysis briefly introduced. In sum, µ-Raman

683 stands out as the method of choice for the non-destructive identification of very small

684 microplastics (<20 µm). Commonly cited drawbacks of Raman techniques, such as

685 weak signal and fluorescence interference, can be overcome by following appropriate

686 cleaning protocols, applying baseline removal algorithms and using more efficient

687 detectors. The development of automated Raman mapping routines enables fast

688 detection of plastic particles with minimal operator time. By streamlining particle pre-

689 selection, through particle detection software and, recently, fluorescent tagging of

690 microplastics, the number of data points to be collected is greatly

31
 ACCEPTED MANUSCRIPT

691 reduced without compromising representativeness. The current great challenge in

692 microplastic identification is real-time monitoring, a goal achievable through the use of

693 ultrafast, nonlinear Raman techniques such as stimulated Raman scattering which, when

694 coupled with flow cytometry, proved feasible for high-throughput screening of plastic

695 microbeads. The next decade promises to be a bustling period of activity for µ-Raman

696 MP analysis. The authors of this review sincerely hope their humble contribution will

697 further the completion of such project.

698 Acknowledgments

699

700 This work was developed within the scope of the project CICECO-Aveiro

701 Institute of Materials, POCI-01-0145-FEDER-007679 (FCT Ref. UID /CTM

702 /50011/2013), financed by portuguese funds through the FCT/MEC and when

703 appropriate co-financed by FEDER under the PT2020 Partnership Agreement. FCT is

704 gratefully acknowledged for funding a PhD grant to CFA (SFRH/BD/129040/2017) and

705 a researcher contract under the program IF 2015 to MMN (IF/01468/2015).

706

707

708 References

709

710 Alexander, J., Barregard, L., Bignami, M., Ceccatelli, S., Cottrill, B., Dinovi, M., Edler, L.,
711 Grasl-Kraupp, B., Hogstrand, C., Hoogenboom, L., Knutsen, H.K., Nebbia, C.S., Oswald, I.,
712 Petersen, A., Rogiers, V.M., Rose, M., Roudot, A.-C., Schwerdtle, T., Vleminckx, C., Vollmer,
713 G., Wallace, H., 2016. Presence of microplastics and nanoplastics in food, with particular focus
714 on seafood. EFSA J. 14, 6, 4501, 1-30. https://doi.org/10.2903/j.efsa.2016.4501
715 Andor Technology, 2007. Optimizing an EMCCD for Spectroscopy [WWW Document].
716 Photonics Media. URL https://www.photonics.com/a31367 (accessed 5.21.18).

32
 ACCEPTED MANUSCRIPT

717 Andrady, A.L., 2017. The plastic in microplastics: A review. Mar. Pollut. Bull. 119, 12–22.
718 https://doi.org/10.1016/j.marpolbul.2017.01.082
719 Andrady, A.L., 2011. Microplastics in the marine environment. Mar. Pollut. Bull. 62, 1596–
720 1605. https://doi.org/10.1016/j.marpolbul.2011.05.030
721 Avio, C.G., Gorbi, S., Regoli, F., 2017. Plastics and microplastics in the oceans: From emerging
722 pollutants to emerged threat. Mar. Environ. Res. 128, 2–11.
723 https://doi.org/10.1016/j.marenvres.2016.05.012
724 Ballent, A., Corcoran, P.L., Madden, O., Helm, P.A., Longstaffe, F.J., 2016. Sources and sinks
725 of microplastics in Canadian Lake Ontario nearshore, tributary and beach sediments. Mar.
726 Pollut. Bull. 110, 383–395. https://doi.org/10.1016/j.marpolbul.2016.06.037
727 Beer, S., Garm, A., Huwer, B., Dierking, J., Nielsen, T.G., 2018. No increase in marine
728 microplastic concentration over the last three decades – A case study from the Baltic Sea. Sci.
729 Total Environ. 621, 1272–1279. https://doi.org/10.1016/j.scitotenv.2017.10.101
730 Blaesing, M., Amelung, W., 2018. Plastics in soil: Analytical methods and possible sources.
731 Sci. Total Environ. 612, 422–435. https://doi.org/10.1016/j.scitotenv.2017.08.086
732 Borman, S.A., 1982. Nonlinear Raman spectroscopy. Anal. Chem. 54, 1021A-1026A.
733 https://doi.org/10.1021/ac00246a002
734 Bremer, M.T., Wrzesinski, P.J., Butcher, N., Lozovoy, V.V., Dantus, M., 2011. Highly selective
735 standoff detection and imaging of trace chemicals in a complex background using single-beam
736 coherent anti-Stokes Raman scattering. Appl. Phys. Lett. 99, 101109.
737 https://doi.org/10.1063/1.3636436
738 Cai, L., Wang, J., Peng, J., Wu, Z., Tan, X., 2018. Observation of the degradation of three types
739 of plastic pellets exposed to UV irradiation in three different environments. Sci. Total Environ.
740 628–629, 740–747. https://doi.org/10.1016/j.scitotenv.2018.02.079
741 Clunies-Ross, P.J., Smith, G.P.S., Gordon, K.C., Gaw, S., 2016. Synthetic shorelines in New
742 Zealand? Quantification and characterisation of microplastic pollution on Canterbury’s
743 coastlines. N. Z. J. Mar. Freshw. Res. 50, 317–325.
744 https://doi.org/10.1080/00288330.2015.1132747
745 Cole, M., Lindeque, P., Fileman, E., Halsband, C., Goodhead, R., Moger, J., Galloway, T.S.,
746 2013. Microplastic Ingestion by Zooplankton. Environ. Sci. Technol. 47, 6646–6655.
747 https://doi.org/10.1021/es400663f
748 Collard, F., Gilbert, B., Compere, P., Eppe, G., Das, K., Jauniaux, T., Parmentier, E., 2017a.
749 Microplastics in livers of European anchovies (Engraulis encrasicolus, L.). Environ. Pollut. 229,
750 1000–1005. https://doi.org/10.1016/j.envpol.2017.07.089
751 Collard, F., Gilbert, B., Eppe, G., Parmentier, E., Das, K., 2015. Detection of Anthropogenic
752 Particles in Fish Stomachs: An Isolation Method Adapted to Identification by Raman
753 Spectroscopy. Arch. Environ. Contam. Toxicol. 69, 331–339. https://doi.org/10.1007/s00244-
754 015-0221-0
755 Collard, F., Gilbert, B., Eppe, G., Roos, L., Compère, P., Das, K., Parmentier, E., 2017b.
756 Morphology of the filtration apparatus of three planktivorous fishes and relation with ingested
757 anthropogenic particles. Mar. Pollut. Bull. 116, 182–191.
758 https://doi.org/10.1016/j.marpolbul.2016.12.067

33
 ACCEPTED MANUSCRIPT

759 Conkle, J.L., Del Valle, C.D.B., Turner, J.W., 2018. Are We Underestimating Microplastic
760 Contamination in Aquatic Environments? Environ. Manage. 61, 1–8.
761 https://doi.org/10.1007/s00267-017-0947-8
762 Costa, J.P.D., Nunes, A.R., Santos, P.S.M., Girão, A.V., Duarte, A.C., Rocha-Santos, T., 2018.
763 Degradation of polyethylene microplastics in seawater: Insights into the environmental
764 degradation of polymers. J. Environ. Sci. Health Part A 0, 1–10.
765 https://doi.org/10.1080/10934529.2018.1455381
766 Cozar, A., Echevarria, F., Ignacio Gonzalez-Gordillo, J., Irigoien, X., Ubeda, B., Hernandez-
767 Leon, S., Palma, A.T., Navarro, S., Garcia-de-Lomas, J., Ruiz, A., Fernandez-de-Puelles, M.L.,
768 Duarte, C.M., 2014. Plastic debris in the open ocean. Proc. Natl. Acad. Sci. U. S. A. 111,
769 10239–10244. https://doi.org/10.1073/pnas.1314705111
770 Crawford, C.B., Quinn, B., 2017. 10 - Microplastic identification techniques, in: Microplastic
771 Pollutants. Elsevier, pp. 219–267. https://doi.org/10.1016/B978-0-12-809406-8.00010-4
772 da Costa, J.P., Santos, P.S.M., Duarte, A.C., Rocha-Santos, T., 2016. (Nano)plastics in the
773 environment - Sources, fates and effects. Sci. Total Environ. 566, 15–26.
774 https://doi.org/10.1016/j.scitotenv.2016.05.041
775 De Tender, C.A., Devriese, L.I., Haegeman, A., Maes, S., Ruttink, T., Dawyndt, P., 2015.
776 Bacterial Community Profiling of Plastic Litter in the Belgian Part of the North Sea. Environ.
777 Sci. Technol. 49, 9629–9638. https://doi.org/10.1021/acs.est.5b01093
778 Dehaut, A., Cassone, A.-L., Frere, L., Hermabessiere, L., Himber, C., Rinnert, E., Riviere, G.,
779 Lambert, C., Soudant, P., Huvet, A., Duflos, G., Paul-Pont, I., 2016. Microplastics in seafood:
780 Benchmark protocol for their extraction and characterization. Environ. Pollut. 215, 223–233.
781 https://doi.org/10.1016/j.envpol.2016.05.018
782 Di, M., Wang, J., 2018. Microplastics in surface waters and sediments of the Three Gorges
783 Reservoir, China. Sci. Total Environ. 616, 1620–1627.
784 https://doi.org/10.1016/j.scitotenv.2017.10.150
785 Dieing, T., Hollricher, O., 2008. High-resolution, high-speed confocal Raman imaging. Vib.
786 Spectrosc. 48, 22–27. https://doi.org/10.1016/j.vibspec.2008.03.004
787 Duis, K., Coors, A., 2016. Microplastics in the aquatic and terrestrial environment: sources
788 (with a specific focus on personal care products), fate and effects. Environ. Sci. Eur. 28:2.
789 https://doi.org/10.1186/s12302-015-0069-y
790 Elert, A.M., Becker, R., Duemichen, E., Eisentraut, P., Falkenhagen, J., Sturm, H., Braun, U.,
791 2017. Comparison of different methods for MP detection: What can we learn from them, and
792 why asking the right question before measurements matters? Environ. Pollut. 231, 1256–1264.
793 https://doi.org/10.1016/j.envpol.2017.08.074
794 Enders, K., Lenz, R., Beer, S., Stedmon, C.A., 2017. Extraction of microplastic from biota:
795 recommended acidic digestion destroys common plastic polymers. Ices J. Mar. Sci. 74, 326–
796 331. https://doi.org/10.1093/icesjms/fsw173
797 Enders, K., Lenz, R., Stedmon, C.A., Nielsen, T.G., 2015. Abundance, size and polymer
798 composition of marine microplastics >= 10 mu m in the Atlantic Ocean and their modelled
799 vertical distribution. Mar. Pollut. Bull. 100, 70–81.
800 https://doi.org/10.1016/j.marpolbul.2015.09.027

34
 ACCEPTED MANUSCRIPT

801 Eriksen, M., Lebreton, L.C.M., Carson, H.S., Thiel, M., Moore, C.J., Borerro, J.C., Galgani, F.,
802 Ryan, P.G., Reisser, J., 2014. Plastic Pollution in the World’s Oceans: More than 5 Trillion
803 Plastic Pieces Weighing over 250,000 Tons Afloat at Sea. PLOS ONE 9, e111913.
804 https://doi.org/10.1371/journal.pone.0111913
805 Erni-Cassola, G., Gibson, M.I., Thompson, R.C., Christie-Oleza, J.A., 2017. Lost, but Found
806 with Nile Red: A Novel Method for Detecting and Quantifying Small Microplastics (1 mm to
807 20 mu m) in Environmental Samples. Environ. Sci. Technol. 51, 13641–13648.
808 https://doi.org/10.1021/acs.est.7b04512
809 Fischer, D., Kaeppler, A., Eichhorn, K.-J., 2015. Identification of Microplastics in the Marine
810 Environment by Raman Microspectroscopy and Imaging. Am. Lab. 47, 32–34.
811 Frere, L., Paul-Pont, I., Moreau, J., Soudant, P., Lambert, C., Huvet, A., Rinnert, E., 2016. A
812 semi-automated Raman micro-spectroscopy method for morphological and chemical
813 characterizations of microplastic litter. Mar. Pollut. Bull. 113, 461–468.
814 https://doi.org/10.1016/j.marpolbul.2016.10.051
815 Frere, L., Paul-Pont, I., Rinnert, E., Petton, S., Jaffre, J., Bihannic, I., Soudant, P., Lambert, C.,
816 Huvet, A., 2017. Influence of environmental and anthropogenic factors on the composition,
817 concentration and spatial distribution of microplastics: A case study of the Bay of Brest
818 (Brittany, France). Environ. Pollut. 225, 211–222. https://doi.org/10.1016/j.envpol.2017.03.023
819 Gago, J., Galgani, F., Maes, T., Thompson, R.C., 2016. Microplastics in Seawater:
820 Recommendations from the Marine Strategy Framework Directive Implementation Process.
821 Front. Mar. Sci. 3:219. https://doi.org/10.3389/fmars.2016.00219
822 Galloway, T.S., Dogra, Y., Garrett, N., Rowe, D., Tyler, C.R., Moger, J., Lammer, E.,
823 Landsiedel, R., Sauer, U.G., Scherer, G., Wohlleben, W., Wiench, K., 2017. Ecotoxicological
824 assessment of nanoparticle-containing acrylic copolymer dispersions in fairy shrimp and
825 zebrafish embryos. Environ. Sci.-Nano 4, 1981–1997. https://doi.org/10.1039/c7en00385d
826 Gares, K.L., Hufziger, K.T., Bykov, S.V., Asher, S.A., 2016. Review of explosive detection
827 methodologies and the emergence of standoff deep UV resonance Raman. J. Raman Spectrosc.
828 47, 124–141. https://doi.org/10.1002/jrs.4868
829 Ghosal, S., Chen, M., Wagner, J., Wang, Z.-M., Wall, S., 2018. Molecular identification of
830 polymers and anthropogenic particles extracted from oceanic water and fish stomach - A Raman
831 micro-spectroscopy study. Environ. Pollut. 233, 1113–1124.
832 https://doi.org/10.1016/j.envpol.2017.10.014
833 Goldstein, M.C., Goodwin, D.S., 2013. Gooseneck barnacles (Lepas spp.) ingest microplastic
834 debris in the North Pacific Subtropical Gyre. Peerj 1, e184. https://doi.org/10.7717/peerj.184
835 Griffiths, P.R., Miseo, E.V., 2014. Infrared and Raman Instrumentation for Mapping and
836 Imaging, in: Infrared and Raman Spectroscopic Imaging. Wiley-Blackwell, pp. 1–56.
837 https://doi.org/10.1002/9783527678136.ch1
838 Gündoğdu, S., 2018. Contamination of table salts from Turkey with microplastics. Food Addit.
839 Contam. Part A 0, 1–9. https://doi.org/10.1080/19440049.2018.1447694
840 Halstead, J.E., Smith, J.A., Carter, E.A., Lay, P.A., Johnston, E.L., 2018. Assessment tools for
841 microplastics and natural fibres ingested by fish in an urbanised estuary. Environ. Pollut., 234,
842 552–561. https://doi.org/10.1016/j.envpol.2017.11.085

35
 ACCEPTED MANUSCRIPT

843 Hanvey, J.S., Lewis, P.J., Lavers, J.L., Crosbie, N.D., Pozo, K., Clarke, B.O., 2017. A review of
844 analytical techniques for quantifying microplastics in sediments. Anal. Methods 9, 1369–1383.
845 https://doi.org/10.1039/c6ay02707e
846 Hidalgo-Ruz, V., Gutow, L., Thompson, R.C., Thiel, M., 2012. Microplastics in the Marine
847 Environment: A Review of the Methods Used for Identification and Quantification. Environ.
848 Sci. Technol. 46, 3060–3075. https://doi.org/10.1021/es2031505
849 Horton, A.A., Jürgens, M.D., Lahive, E., Bodegom, P.M. van, Vijver, M.G., 2018. The
850 influence of exposure and physiology on microplastic ingestion by the freshwater fish Rutilus
851 rutilus (roach) in the River Thames, UK. Environ. Pollut. 236, 188–194.
852 https://doi.org/10.1016/j.envpol.2018.01.044
853 Horton, A.A., Svendsen, C., Williams, R.J., Spurgeon, D.J., Lahive, E., 2017a. Large
854 microplastic particles in sediments of tributaries of the River Thames, UK - Abundance, sources
855 and methods for effective quantification. Mar. Pollut. Bull. 114, 218–226.
856 https://doi.org/10.1016/j.marpolbul.2016.09.004
857 Horton, A.A., Walton, A., Spurgeon, D.J., Lahive, E., Svendsen, C., 2017b. Microplastics in
858 freshwater and terrestrial environments: Evaluating the current understanding to identify the
859 knowledge gaps and future research priorities. Sci. Total Environ. 586, 127–141.
860 https://doi.org/10.1016/j.scitotenv.2017.01.190
861 Imhof, H.K., Ivleva, N.P., Schmid, J., Niessner, R., Laforsch, C., 2013. Contamination of beach
862 sediments of a subalpine lake with microplastic particles. Curr. Biol. 23, R867–R868.
863 https://doi.org/10.1016/j.cub.2013.09.001
864 Imhof, H.K., Laforsch, C., Wiesheu, A.C., Schmid, J., Anger, P.M., Niessner, R., Ivleva, N.P.,
865 2016. Pigments and plastic in limnetic ecosystems: A qualitative and quantitative study on
866 microparticles of different size classes. Water Res. 98, 64–74.
867 https://doi.org/10.1016/j.watres.2016.03.015
868 Imhof, H.K., Schmid, J., Niessner, R., Ivleva, N.P., Laforsch, C., 2012. A novel, highly efficient
869 method for the separation and quantification of plastic particles in sediments of aquatic
870 environments. Limnol. Oceanogr.-Methods 10, 524–537.
871 https://doi.org/10.4319/lom.2012.10.524
872 Imhof, H.K., Wiesheu, A.C., Anger, P.M., Niessner, R., Ivleva, N.P., Laforsch, C., 2018.
873 Variation in plastic abundance at different lake beach zones - A case study. Sci. Total Environ.
874 613, 530–537. https://doi.org/10.1016/j.scitotenv.2017.08.300
875 Ivleva, N.P., Wiesheu, A.C., Niessner, R., 2017. Microplastic in Aquatic Ecosystems. Angew.
876 Chem.-Int. Ed. 56, 1720–1739. https://doi.org/10.1002/anie.201606957
877 Jahnke, A., Arp, H.P.H., Escher, B.I., Gewert, B., Gorokhova, E., Kühnel, D., Ogonowski, M.,
878 Potthoff, A., Rummel, C., Schmitt-Jansen, M., Toorman, E., MacLeod, M., 2017. Reducing
879 Uncertainty and Confronting Ignorance about the Possible Impacts of Weathering Plastic in the
880 Marine Environment. Environ. Sci. Technol. Lett. 4, 85–90.
881 https://doi.org/10.1021/acs.estlett.7b00008
882 Jiang, J.-Q., 2018. Occurrence of microplastics and its pollution in the environment: A review.
883 Sustain. Prod. Consum. 13, 16–23. https://doi.org/10.1016/j.spc.2017.11.003
884 Kaeppler, A., Fischer, D., Oberbeckmann, S., Schernewski, G., Labrenz, M., Eichhorn, K.-J.,
885 Voit, B., 2016. Analysis of environmental microplastics by vibrational microspectroscopy:

36
 ACCEPTED MANUSCRIPT

886 FTIR, Raman or both? Anal. Bioanal. Chem. 408, 8377–8391. https://doi.org/10.1007/s00216-
887 016-9956-3
888 Kaeppler, A., Windrich, F., Loeder, M.G.J., Malanin, M., Fischer, D., Labrenz, M., Eichhorn,
889 K.-J., Voit, B., 2015. Identification of microplastics by FTIR and Raman microscopy: a novel
890 silicon filter substrate opens the important spectral range below 1300 cm(-1) for FTIR
891 transmission measurements. Anal. Bioanal. Chem. 407, 6791–6801.
892 https://doi.org/10.1007/s00216-015-8850-8
893 Karami, A., Golieskardi, A., Choo, C.K., Larat, V., Galloway, T.S., Salamatinia, B., 2017a. The
894 presence of microplastics in commercial salts from different countries. Sci. Rep. 7:46173.
895 https://doi.org/10.1038/srep46173
896 Karami, A., Golieskardi, A., Choo, C.K., Larat, V., Karbalaei, S., Salamatinia, B., 2018.
897 Microplastic and mesoplastic contamination in canned sardines and sprats. Sci. Total Environ.
898 612, 1380–1386. https://doi.org/10.1016/j.scitotenv.2017.09.005
899 Karami, A., Golieskardi, A., Choo, C.K., Romano, N., Bin Ho, Y., Salamatinia, B., 2017b. A
900 high-performance protocol for extraction of microplastics in fish. Sci. Total Environ. 578, 485–
901 494. https://doi.org/10.1016/j.scitotenv.2016.10.213
902 Karami, A., Golieskardi, A., Ho, Y.B., Larat, V., Salamatinia, B., 2017c. Microplastics in
903 eviscerated flesh and excised organs of dried fish. Sci. Rep. 7:5473.
904 https://doi.org/10.1038/s41598-017-05828-6
905 Karami, A., Romano, N., Galloway, T., Hamzah, H., 2016. Virgin microplastics cause toxicity
906 and modulate the impacts of phenanthrene on biomarker responses in African catfish (Clarias
907 gariepinus). Environ. Res. 151, 58–70. https://doi.org/10.1016/j.envres.2016.07.024
908 Karlsson, T.M., Vethaak, A.D., Almroth, B.C., Ariese, F., van Velzen, M., Hassellov, M.,
909 Leslie, H.A., 2017. Screening for microplastics in sediment, water, marine invertebrates and
910 fish: Method development and microplastic accumulation. Mar. Pollut. Bull. 122, 403–408.
911 https://doi.org/10.1016/j.marpolbul.2017.06.081
912 Klein, S., Dimzon, I.K., Eubeler, J., Knepper, T.P., 2018. Analysis, Occurrence, and
913 Degradation of Microplastics in the Aqueous Environment, in: Wagner, M., Lambert, S. (Eds.),
914 Freshwater Microplastics : Emerging Environmental Contaminants? Springer International
915 Publishing, Cham, pp. 51–67. https://doi.org/10.1007/978-3-319-61615-5_3
916 Lares, M., Ncibi, M.C., Sillanpää, Markus, Sillanpää, Mika, 2018. Occurrence, identification
917 and removal of microplastic particles and fibers in conventional activated sludge process and
918 advanced MBR technology. Water Res. 133, 236–246.
919 https://doi.org/10.1016/j.watres.2018.01.049
920 Lawson, L.S., Rodriguez, J.D., 2016. Raman Barcode for Counterfeit Drug Product Detection.
921 Anal. Chem. 88, 4706–4713. https://doi.org/10.1021/acs.analchem.5b04636
922 Lei, K., Qiao, F., Liu, Q., Wei, Z., Qi, H., Cui, S., Yue, X., Deng, Y., An, L., 2017.
923 Microplastics releasing from personal care and cosmetic products in China. Mar. Pollut. Bull.
924 123, 122–126. https://doi.org/10.1016/j.marpolbul.2017.09.016
925 Lenz, R., Enders, K., Stedmon, C.A., Mackenzie, D.M.A., Nielsen, T.G., 2015. A critical
926 assessment of visual identification of marine microplastic using Raman spectroscopy for
927 analysis improvement. Mar. Pollut. Bull. 100, 82–91.
928 https://doi.org/10.1016/j.marpolbul.2015.09.026

37
 ACCEPTED MANUSCRIPT

929 Li, J., Liu, H., Chen, J.P., 2017. Microplastics in freshwater systems: A review on occurrence,
930 environmental effects, and methods for microplastics detection. Water Res. 137, 362-374
931 https://doi.org/10.1016/j.watres.2017.12.056
932 Liao, C.-S., Wang, P., Huang, C.Y., Lin, P., Eakins, G., Bentley, R.T., Liang, R., Cheng, J.-X.,
933 2017. In Vivo and in Situ Spectroscopic Imaging by a Handheld Stimulated Raman Scattering
934 Microscope. ACS Photonics 5, 947–954. https://doi.org/10.1021/acsphotonics.7b01214
935 Liboiron, F., Ammendolia, J., Saturno, J., Melvin, J., Zahara, A., Richárd, N., Liboiron, M.,
936 2018. A zero percent plastic ingestion rate by silver hake (Merluccius bilinearis) from the south
937 coast of Newfoundland, Canada. Mar. Pollut. Bull. 131, 267–275.
938 https://doi.org/10.1016/j.marpolbul.2018.04.007
939 Loeder, M.G.J., Gerdts, G., 2015. Methodology Used for the Detection and Identification of
940 Microplastics—A Critical Appraisal, in: Bergmann, M., Gutow, L., Klages, M. (Eds.), Marine
941 Anthropogenic Litter. Springer International Publishing, Cham, pp. 201–227.
942 https://doi.org/10.1007/978-3-319-16510-3_8
943 Loeder, M.G.J., Kuczera, M., Mintenig, S., Lorenz, C., Gerdts, G., 2015. Focal plane array
944 detector-based micro-Fourier-transform infrared imaging for the analysis of microplastics in
945 environmental samples. Environ. Chem. 12, 563–581. https://doi.org/10.1071/EN14205
946 Lots, F.A.E., Behrens, P., Vijver, M.G., Horton, A.A., Bosker, T., 2017. A large-scale
947 investigation of microplastic contamination: Abundance and characteristics of microplastics in
948 European beach sediment. Mar. Pollut. Bull. 123, 219–226.
949 https://doi.org/10.1016/j.marpolbul.2017.08.057
950 Lusher, A.L., Burke, A., O’Connor, I., Officer, R., 2014. Microplastic pollution in the Northeast
951 Atlantic Ocean: Validated and opportunistic sampling. Mar. Pollut. Bull. 88, 325–333.
952 https://doi.org/10.1016/j.marpolbul.2014.08.023
953 Lusher, A.L., Welden, N.A., Sobral, P., Cole, M., 2017. Sampling, isolating and identifying
954 microplastics ingested by fish and invertebrates. Anal. Methods 9, 1346–1360.
955 https://doi.org/10.1039/c6ay02415g
956 Maes, T., Jessop, R., Wellner, N., Haupt, K., Mayes, A.G., 2017. A rapid-screening approach to
957 detect and quantify microplastics based on fluorescent tagging with Nile Red. Sci. Rep. 7:
958 44501. https://doi.org/10.1038/srep44501
959 Mai, L., Bao, L.-J., Shi, L., Wong, C.S., Zeng, E.Y., 2018. A review of methods for measuring
960 microplastics in aquatic environments. Environ. Sci. Pollut. Res. 25, 11319–11332.
961 https://doi.org/10.1007/s11356-018-1692-0
962 Miller, M.E., Kroon, F.J., Motti, C.A., 2017. Recovering microplastics from marine samples: A
963 review of current practices. Mar. Pollut. Bull. 123, 6–18.
964 https://doi.org/10.1016/j.marpolbul.2017.08.058
965 Mintenig, S.M., Int-Veen, I., Loeder, M.G.J., Primpke, S., Gerdts, G., 2017. Identification of
966 microplastic in effluents of waste water treatment plants using focal plane array-based micro-
967 Fourier-transform infrared imaging. Water Res. 108, 365–372.
968 https://doi.org/10.1016/j.watres.2016.11.015
969 Muhlschlegel, P., Hauk, A., Walter, U., Sieber, R., 2017. Lack of evidence for microplastic
970 contamination in honey. Food Addit. Contam. Part -A 34, 1982–1989.
971 https://doi.org/10.1080/19440049.2017.1347281

38
 ACCEPTED MANUSCRIPT

972 Murray, F., Cowie, P.R., 2011. Plastic contamination in the decapod crustacean Nephrops
973 norvegicus (Linnaeus, 1758). Mar. Pollut. Bull. 62, 1207–1217.
974 https://doi.org/10.1016/j.marpolbul.2011.03.032
975 Naidu, S.A., Ranga Rao, V., Ramu, K., 2018. Microplastics in the benthic invertebrates from
976 the coastal waters of Kochi, Southeastern Arabian Sea. Environ. Geochem. Health.
977 https://doi.org/10.1007/s10653-017-0062-z
978 Oberbeckmann, S., 2017. BONUS MICROPOLL - Multilevel assessment of microplastics and
979 associated pollutants in the Baltic Sea [WWW Document]. URL https://www.io-
980 warnemuende.de/micropoll-home.html (accessed 5.17.18).
981 Ogonowski, M., Gerdes, Z., Gorokhova, E., 2018. What we know and what we think we know
982 about microplastic effects – A critical perspective. Curr. Opin. Environ. Sci. Health 1, 41–46.
983 https://doi.org/10.1016/j.coesh.2017.09.001
984 Opilik, L., Schmid, T., Zenobi, R., 2013. Modern Raman Imaging: Vibrational Spectroscopy on
985 the Micrometer and Nanometer Scales. Annu. Rev. Anal. Chem. 6, 379–398.
986 https://doi.org/10.1146/annurev-anchem-062012-092646
987 Ossmann, B.E., Sarau, G., Schmitt, S.W., Holtmannspoetter, H., Christiansen, S.H., Dicke, W.,
988 2017. Development of an optimal filter substrate for the identification of small microplastic
989 particles in food by micro-Raman spectroscopy. Anal. Bioanal. Chem. 409, 4099–4109.
990 https://doi.org/10.1007/s00216-017-0358-y
991 Ostmark, H., Nordberg, M., Carlsson, T.E., 2011. Stand-off detection of explosives particles by
992 multispectral imaging Raman spectroscopy. Appl. Opt. 50, 5592–5599.
993 https://doi.org/10.1364/AO.50.005592
994 Palmer, A., Phapale, P., Fay, D., Alexandrov, T., 2017. Curatr: a web application for creating,
995 curating and sharing a mass spectral library. Bioinformatics 34, 1436-1438.
996 https://doi.org/10.1093/bioinformatics/btx786
997 Qiu, Q., Tan, Z., Wang, J., Peng, J., Li, M., Zhan, Z., 2016. Extraction, enumeration and
998 identification methods for monitoring microplastics in the environment. Estuar. Coast. Shelf
999 Sci. 176, 102–109. https://doi.org/10.1016/j.ecss.2016.04.012
1000 Remy, F., Collard, F., Gilbert, B., Compere, P., Eppe, G., Lepoint, G., 2015. When Microplastic
1001 Is Not Plastic: The Ingestion of Artificial Cellulose Fibers by Macrofauna Living in Seagrass
1002 Macrophytodetritus. Environ. Sci. Technol. 49, 11158–11166.
1003 https://doi.org/10.1021/acs.est.5b02005
1004 Renner, G., Schmidt, T.C., Schram, J., 2018. Analytical methodologies for monitoring
1005 micro(nano)plastics: Which are fit for purpose? Curr. Op. Env. Sci. Hea1th 1, 55–61.
1006 https://doi.org/10.1016/j.coesh.2017.11.001
1007 Revel, M., Châtel, A., Mouneyrac, C., 2018. Micro(nano)plastics: A threat to human health?
1008 Curr. Op. Env. Sci. Hea1th 1, 17–23. https://doi.org/10.1016/j.coesh.2017.10.003
1009 Ribeiro-Claro, P., Nolasco, M.M., Araújo, C., 2017. Chapter 5 - Characterization of
1010 Microplastics by Raman Spectroscopy, in: Rocha-Santos, T.A.P., Duarte, A.C. (Eds.),
1011 Comprehensive Analytical Chemistry, Characterization and Analysis of Microplastics. Elsevier,
1012 pp. 119–151. https://doi.org/10.1016/bs.coac.2016.10.001

39
 ACCEPTED MANUSCRIPT

1013 Rocha-Santos, T., Duarte, A.C., 2015. A critical overview of the analytical approaches to the
1014 occurrence, the fate and the behavior of microplastics in the environment. Trac-Trends Anal.
1015 Chem. 65, 47–53. https://doi.org/10.1016/j.trac.2014.10.011
1016 Rodrigues, M.O., Abrantes, N., Gonçalves, F.J.M., Nogueira, H., Marques, J.C., Gonçalves,
1017 A.M.M., 2018. Spatial and temporal distribution of microplastics in water and sediments of a
1018 freshwater system (Antuã River, Portugal). Sci. Total Environ. 633, 1549–1559.
1019 https://doi.org/10.1016/j.scitotenv.2018.03.233
1020 Rodríguez-Seijo, A., Pereira, R., 2017. Chapter 3 - Morphological and Physical
1021 Characterization of Microplastics, in: Rocha-Santos, T.A.P., Duarte, A.C. (Eds.),
1022 Characterization and Analysis of Microplastics, Comprehensive Analytical Chemistry. Elsevier,
1023 pp. 49–66. https://doi.org/10.1016/bs.coac.2016.10.007
1024 Scheurer, M., Bigalke, M., 2018. Microplastics in Swiss Floodplain Soils. Environ. Sci.
1025 Technol. 52, 3591–3598. https://doi.org/10.1021/acs.est.7b06003
1026 Schmaelzlin, E., Moralejo, B., Rutowska, M., Monreal-Ibero, A., Sandin, C., Tarcea, N., Popp,
1027 J., Roth, M.M., 2014. Raman Imaging with a Fiber-Coupled Multichannel Spectrograph.
1028 Sensors 14, 21968–21980. https://doi.org/10.3390/s141121968
1029 Schymanski, D., Goldbeck, C., Humpf, H.-U., Fürst, P., 2018. Analysis of microplastics in
1030 water by micro-Raman spectroscopy: Release of plastic particles from different packaging into
1031 mineral water. Water Res. 129, 154–162. https://doi.org/10.1016/j.watres.2017.11.011
1032 Shan, J., Zhao, J., Liu, L., Zhang, Y., Wang, X., Wu, F., 2018. A novel way to rapidly monitor
1033 microplastics in soil by hyperspectral imaging technology and chemometrics. Environ. Pollut.
1034 238, 121–129. https://doi.org/10.1016/j.envpol.2018.03.026
1035 Shim, W.J., Hong, S.H., Eo, S.E., 2017. Identification methods in microplastic analysis: a
1036 review. Anal. Methods 9, 1384–1391. https://doi.org/10.1039/c6ay02558g
1037 Shim, W.J., Song, Y.K., Hong, S.H., Jang, M., 2016. Identification and quantification of
1038 microplastics using Nile Red staining. Mar. Pollut. Bull. 113, 469–476.
1039 https://doi.org/10.1016/j.marpolbul.2016.10.049
1040 Silva, A.B., Bastos, A.S., Justino, C.I.L., da Costa, J.P., Duarte, A.C., Rocha-Santos, T.A.P.,
1041 2018. Microplastics in the environment: Challenges in analytical chemistry - A review. Anal.
1042 Chim. Acta 1017, 1-19. https://doi.org/10.1016/j.aca.2018.02.043
1043 Smith, W.E., Dent, G., 2005. Modern Raman Spectroscopy: A Practical Approach. John Wiley
1044 & Sons, Ltd, West Sussex, England.
1045 Song, Y.K., Hong, S.H., Jang, M., Han, G.M., Rani, M., Lee, J., Shim, W.J., 2015. A
1046 comparison of microscopic and spectroscopic identification methods for analysis of
1047 microplastics in environmental samples. Mar. Pollut. Bull. 93, 202–209.
1048 https://doi.org/10.1016/j.marpolbul.2015.01.015
1049 Sruthy, S., Ramasamy, E.V., 2017. Microplastic pollution in Vembanad Lake, Kerala, India:
1050 The first report of microplastics in lake and estuarine sediments in India. Environ. Pollut. 222,
1051 315–322. https://doi.org/10.1016/j.envpol.2016.12.038
1052 Stewart, S., Priore, R.J., Nelson, M.P., Treado, P.J., 2012. Raman Imaging. Annu. Rev. Anal.
1053 Chem. 5, 337–360. https://doi.org/10.1146/annurev-anchem-062011-143152
1054 Sujathan, S., Kniggendorf, A.-K., Kumar, A., Roth, B., Rosenwinkel, K.-H., Nogueira, R.,
1055 2017. Heat and Bleach: A Cost-Efficient Method for Extracting Microplastics from Return

40
 ACCEPTED MANUSCRIPT

1056 Activated Sludge. Arch. Environ. Contam. Toxicol. 73, 641–648.

1057 https://doi.org/10.1007/s00244-017-0415-8
1058 Syakti, A.D., 2017. Microplastics Monitoring in Marine Environment. Omni-Akuatika 13, 1-6.
1059 https://doi.org/10.20884/1.oa.2017.13.2.430
1060 Tagg, A.S., Sapp, M., Harrison, J.P., Ojeda, J.J., 2015. Identification and Quantification of
1061 Microplastics in Wastewater Using Focal Plane Array-Based Reflectance Micro-FT-IR
1062 Imaging. Anal. Chem. 87, 6032–6040. https://doi.org/10.1021/acs.analchem.5b00495
1063 Van Cauwenberghe, L., Claessens, M., Vandegehuchte, M.B., Janssen, C.R., 2015.
1064 Microplastics are taken up by mussels (Mytilus edulis) and lugworms (Arenicola marina) living
1065 in natural habitats. Environ. Pollut. 199, 10–17. https://doi.org/10.1016/j.envpol.2015.01.008
1066 Van Cauwenberghe, L., Janssen, C.R., 2014. Microplastics in bivalves cultured for human
1067 consumption. Environ. Pollut. 193, 65–70. https://doi.org/10.1016/j.envpol.2014.06.010
1068 Van Cauwenberghe, L., Vanreusel, A., Mees, J., Janssen, C.R., 2013. Microplastic pollution in
1069 deep-sea sediments. Environ. Pollut. 182, 495–499.
1070 https://doi.org/10.1016/j.envpol.2013.08.013
1071 Wagner, J., Wang, Z.-M., Ghosal, S., Rochman, C., Gassel, M., Wall, S., 2017. Novel method
1072 for the extraction and identification of microplastics in ocean trawl and fish gut matrices. Anal.
1073 Methods 9, 1479–1490. https://doi.org/10.1039/c6ay02396g
1074 Watts, A.J.R., Lewis, C., Goodhead, R.M., Beckett, S.J., Moger, J., Tyler, C.R., Galloway, T.S.,
1075 2014. Uptake and Retention of Microplastics by the Shore Crab Carcinus maenas. Environ. Sci.
1076 Technol. 48, 8823–8830. https://doi.org/10.1021/es501090e
1077 Watts, A.J.R., Urbina, M.A., Goodhead, R., Moger, J., Lewis, C., Galloway, T.S., 2016. Effect
1078 of Microplastic on the Gills of the Shore Crab Carcinus maenas. Environ. Sci. Technol. 50,
1079 5364–5369. https://doi.org/10.1021/acs.est.6b01187
1080 Wiesheu, A.C., Anger, P.M., Baumann, T., Niessner, R., Ivleva, N.P., 2016. Raman
1081 microspectroscopic analysis of fibers in beverages. Anal. Methods 8, 5722–5725.
1082 https://doi.org/10.1039/c6ay01184e
1083 Wright, S.L., Kelly, F.J., 2017. Plastic and Human Health: A Micro Issue? Environ. Sci.
1084 Technol. 51, 6634–6647. https://doi.org/10.1021/acs.est.7b00423
1085 Wu, C., Zhang, K., Xiong, X., 2018. Microplastic Pollution in Inland Waters Focusing on Asia,
1086 in: Wagner, M., Lambert, S. (Eds.), Freshwater Microplastics : Emerging Environmental
1087 Contaminants? Springer International Publishing, Cham, pp. 85–99.
1088 https://doi.org/10.1007/978-3-319-61615-5_5
1089 Xiong, X., Zhang, K., Chen, X., Shi, H., Luo, Z., Wu, C., 2018. Sources and distribution of
1090 microplastics in China’s largest inland lake – Qinghai Lake. Environ. Pollut. 235, 899–906.
1091 https://doi.org/10.1016/j.envpol.2017.12.081
1092 Yonkos, L.T., Friedel, E.A., Perez-Reyes, A.C., Ghosal, S., Arthur, C.D., 2014. Microplastics in
1093 Four Estuarine Rivers in the Chesapeake Bay, USA. Environ. Sci. Technol. 48, 14195–14202.
1094 https://doi.org/10.1021/es5036317
1095 Young, A.M., Elliott, J.A., 2016. Characterization of microplastic and mesoplastic debris in
1096 sediments from Kamilo Beach and Kahuku Beach, Hawai’i. Mar. Pollut. Bull. 113, 477–482.
1097 https://doi.org/10.1016/j.marpolbul.2016.11.009

41
 ACCEPTED MANUSCRIPT

1098 Yu, Y., Zhou, D., Li, Z., Zhu, C., 2018. Advancement and Challenges of Microplastic Pollution
1099 in the Aquatic Environment: a Review. Water. Air. Soil Pollut. 229, 140.
1100 https://doi.org/10.1007/s11270-018-3788-z
1101 Zada, L., Leslie, H.A., Vethaak, A.D., Tinnevelt, G., Janssen, J., Boer, J.F. de, Ariese, F., 2018.
1102 Fast microplastics identification with stimulated Raman scattering microscopy. J. Raman
1103 Spectrosc. Special Issue, 1-9 https://doi.org/10.1002/jrs.5367
1104 Zettler, E.R., Mincer, T.J., Amaral-Zettler, L.A., 2013. Life in the “Plastisphere”: Microbial
1105 Communities on Plastic Marine Debris. Environ. Sci. Technol. 47, 7137–7146.
1106 https://doi.org/10.1021/es401288x
1107 Zhang, C., Huang, K.-C., Rajwa, B., Li, J., Yang, S., Lin, H., Liao, C., Eakins, G., Kuang, S.,
1108 Patsekin, V., Robinson, J.P., Cheng, J.-X., 2017. Stimulated Raman scattering flow cytometry
1109 for label-free single-particle analysis. Optica 4, 103–109.
1110 https://doi.org/10.1364/OPTICA.4.000103
1111 Zhang, K., Su, J., Xiong, X., Wu, X., Wu, C., Liu, J., 2016. Microplastic pollution of lakeshore
1112 sediments from remote lakes in Tibet plateau, China. Environ. Pollut. 219, 450–455.
1113 https://doi.org/10.1016/j.envpol.2016.05.048
1114 Zhang, K., Xiong, X., Hu, H., Wu, C., Bi, Y., Wu, Y., Zhou, B., Lam, P.K.S., Liu, J., 2017.
1115 Occurrence and Characteristics of Microplastic Pollution in Xiangxi Bay of Three Gorges
1116 Reservoir, China. Environ. Sci. Technol. 51, 3794–3801.
1117 https://doi.org/10.1021/acs.est.7b00369
1118 Zhao, S., Danley, M., Ward, J.E., Li, D., Mincer, T.J., 2017. An approach for extraction,
1119 characterization and quantitation of microplastic in natural marine snow using Raman
1120 microscopy. Anal. Methods 9, 1470–1478. https://doi.org/10.1039/c6ay02302a
1121 Zhao, S., Zhu, L., Li, D., 2015. Microplastic in three urban estuaries, China. Environ. Pollut.
1122 206, 597–604. https://doi.org/10.1016/j.envpol.2015.08.027

1123

42
 ACCEPTED MANUSCRIPT

Highlights

Raman is the method of choice for identifying small microplastics (<20 µm)

Automated mapping routines and library matching allow fast microplastic detection

Nonlinear Raman techniques enable real-time monitoring of microplastics