Chemosphere 308 (2022) 136407

Contents lists available at ScienceDirect

Chemosphere
journal homepage: www.elsevier.com/locate/chemosphere

The Raman spectroscopy and SEM/EDS investigation of the primary

sources of microplastics from cosmetics available in Poland
Agnieszka Dąbrowska a, b, *, Maria Mielańczuk a, Marcin Syczewski c
a
University of Warsaw, Faculty of Chemistry, Laboratory of Spectroscopy and Intermolecular Interactions, Pasteura 1 Str., 02-093, Warsaw, Poland
b
University of Warsaw Biological and Chemical Research Centre, Żwirki i Wigury 101 Str., 02-089, Warsaw, Poland
c
Helmholtz Centre Potsdam, GFZ German Research Centre for Geoscience, Telegrafenberg, 14473, Potsdam, Germany

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

• Raman spectroscopy of microbeads in

cosmetics currently available on the
Polish market.
• The underlined crucial role of
morphology in sorptive properties.
• SEM/EDX evidence of diversified
morphology and composition of
microbeads.
• Numerical description and classification
of primary MPs.
• PE quantitative characterization.

A R T I C L E I N F O A B S T R A C T

Handling Editor, Govarthanan Muthusamy The increasing amount of marine microplastics and nanoplastics is due to primary (P-MPs) and secondary sources
(S-MPs). Among small P-MPs and fibres from textiles, nurdles or industrial dust, the microbeads from cosmetics
Keywords: (peelings, scrubs) are dominant. The polyethylene and polypropylene debris, already <5 mm, enlarge Plasti­
Cosmetics microplastic sphere and have a complex influence on environmental processes. Fortunately, the situation has changed with
Nanoplastics
the limitation of synthetic materials in scrubs in recent years. This paper summarizes the six years of studies
Surface characterization
(2015–2021) on microplastics in cosmetics available on the Polish market. It focuses on those still available and
Plastisphere
Primary microplastics presents the potential of Raman spectroscopy for the efficient qualitative and quantitative characterization of
Raman spectroscopy those materials. Finally, surface morphology’s vital role in microplastics’ general behaviour is underlined. The
SEM/EDX and numerical modelling enables a more detailed description of particles.

1. Introduction synthetic materials and the primary sources of polymers already <5 mm
entering the freshwater systems (Triebskorn et al., 2019) and finally
The number of marine microplastics (MPs, < 5 mm) (Silva et al., reaching the global ocean. Even the remote and pristine polar regions
2018) and nanoplastics (NPs, < 100 nm) (Gigault et al., 2018) (Jiang et al., 2020) or seabed are contaminated. The total surface of MPs
constantly increases due to the fragmentation of macro pieces of and NPs (Barría et al., 2020) available in the environment, called

* Corresponding author. University of Warsaw, Faculty of Chemistry, Laboratory of Spectroscopy and Intermolecular Interactions, Pasteura 1 Str., 02-093, Warsaw,
Poland.
E-mail address: [email protected] (A. Dąbrowska).

https://doi.org/10.1016/j.chemosphere.2022.136407
Received 17 May 2022; Received in revised form 1 September 2022; Accepted 7 September 2022
Available online 12 September 2022
0045-6535/© 2022 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).
A. Dąbrowska et al. Chemosphere 308 (2022) 136407

Plastisphere (Amaral-Zettler et al., 2021), is already new, rapidly Methodology, target and general conclusions were comparable,
expanding and extensively studied (Tarafdar et al., 2021; Zhang et al., although, in Malaysia the size of MPs in tested cosmetic (3–178 μm)
2021) ecological niche (Delacuvellerie et al., 2019). Although one can were one order of magnitude lower than in China or Spain (particles ~
restrict this term only to the biofilm presence, it is advisable to use it in hundreds of μm). In China primary MPs were detected in 7.1% of facial
this broader sense concerning new findings. The ecotoxicological impact cleaners (~25.04 mg MP per g with an average size ~313 μm) and 2.2%
of NPs is already extensively discussed in the literature (Shen et al., of tested shower gels (~17.80 mg MP per g with an average size ~422
2019) and the natural ageing processes (Kalčíková et al., 2020) play an μm). One can conclude that MPs ~420 μm are the most efficient exfo­
important role in their fate and behaviour. Secondary sources of plastics liating agents. In Spain the higher content of PE among ingredients had
are forming the MPs and NPs (El Hadri et al., 2020) due to abrasion, foot (>80% of tested products) and body scrubs (>60% of tested prod­
degradation and gradual fragmentation processes. As primary micro­ ucts). Sizes of MPs in personal products on a Spanish market varies in
plastics (Boucher and Friot, 2017) have a high surface-to-volume ratio range from 8 μm to 2 mm and some of the products had >3000 particles
(which is increasing with the decreasing linear dimension), their role in per g. Each geographic region has its particular characteristic in terms of
ecological impact is crucial, especially as vectors (Caruso, 2019) for dominant types of MPs containing cosmetics, patterns of their use and
pollutants and viruses or substrates for biofilm (Harrison et al., 2018). consumption, and content in sewage. This works aims to contribute to
One can classify them further as those originating from industrial dust, this field by local data and basic studies, but more importantly, presents
washed-out from synthetic textiles, transferred to waters as a spillage of a broad perspective on morphology description, physical and chemical
nurdles, or being an ingredient added to cause surface abrasion (in classification of debris. The Raman spectroscopy (Lv et al., 2020) itself is
cosmetics or detergents). Particles originating from those sources are an efficient and powerful tool for qualitative and quantitative charac­
persistent and ubiquitous, freely passing the wastewater treatment terization of MPs (Xu et al., 2019), especially if combined with FTIR
plants (WWTP). One can estimate the number of microbeads added to (Fourier-transform Infrared Spectroscopy) data (Cowger et al., 2020).
domestic sewage in a single use is > 200 000. Primary MPs (Wang et al., However, the typical in infrared ATR (Attenuated Total Reflectance) mode
2019) dominate environmental samples filtered from waters, soils, or has lower (~2 cm− 1) than Raman special resolution. Moreover, Raman
air-born. The microplastics from cosmetics, > 90% polyethylene (PE), spectroscopy is suitable for the nanoplastics detection, providing that
but also polypropylene (PP), polyethylene terephthalate (PET), poly­ other challenges are overcome (such as the samples preparation, pur­
methyl methacrylate (PMMA), and nylon, were already studied locally ofocation, and the adequate control of blanks). The most common
in many research with some standards proposed (V. V Godoy et al., polyethylene (PE) exhibits four typical bands between 1000 and 1500
2019; Lei et al., 2017; Napper et al., 2015; Praveena et al., 2018). Ac­ cm− 1 in Raman spectra (1439 cm− 1, 1295 cm− 1, 1129 cm− 1, 1062
cording to the European Cosmetic Ingredient Database, currently added cm− 1), whereas characteristic FTIR peaks (for instance due to the
abrasive agents in cosmetics are, among others, the following: poly­ asymmetric and symmetric stretching of –CH3 group) can be observed
ethylene, polylactic acid, hydrogenated poly (c6-20olefin), ethyl­ at: ~2915-2918 cm− 1, ~2845-2850 cm− 1, ~1463-1469 cm− 1,
ene/propylene copolymer, ammonium acryloyldimethyltaurate/ ~713-720 cm− 1. Other used methods include gas chromatography, mass
laureth-7 methacrylate copolymer, 1,4-butanodiol/succinic acid/­ spectrometry, and fluorescent staining (Liu et al., 2022).
adipic acid copolymer. Research conducted in the UK (Napper et al., In addition, scanning electron microscopy with the energy dispersive
2015) estimated the number of microbeads released in a single use to be X-ray spectroscopy (SEM/EDX) analysis of particle surfaces is also pro­
between 4594 and 94 500 particles. This considerable amount of par­ vided in this research. One may underline the crucial role of its
ticles vary in size, shape, roughness and sorptive properties comparable morphology (Burrows et al., 2020) in modelling the MPs’ behaviour in
with the transport potential. Primary MPs are potential vectors for hy­ the environment (Song et al., 2019). That also includes the importance
drophobic pollutants with a considerable affinity to of functional groups changing material affinity, thus determining their
dichloro-diphenyl-trichloroethane (DDT). The transfer of hydrophobic ecotoxicological effect (Gonçalves and Bebianno, 2021). Due to that,
substances (PAHs, PCBs, PBDEs) from plastic to biota was already within this work, the numerical description of the structure is provided
proved (Rochman et al., 2013). Fortunately, the increasing awareness of from the obtained microscopic pictures using Python scripts. In 2016
the problem caused the systematic reduction in primary MPs on the many European countries observed modified cosmetics formulas with
market (Xanthos and Walker, 2017). UNEP passed the resolution on decreasing content of PE. However, even of a natural origin, all
microplastics and marine litter in 2017, and at the beginning of 2018, an microbeads can serve as transport agents for pollutants and exhibit
action plan for managing plastics accordingly to the Circular Economy sorption properties directly correlated with their morphology and sur­
was approved. The EU has reduced primary MPs in cosmetics from 4360 face structure, less to the type of material itself. Studies on the sorption
tonnes in 2012 to 793 tonnes in 2015. From 2020 MPs as exfoliating of a phenanthrene and DDT mixture proved that roughness is crucial,
agents in cosmetics are entirely forbidden. US and UK bans prohibit MPs not synthetic or natural origin (Napper et al., 2015). Smooth microbeads
microbeads in rinse-off cosmetics. However, there is still a need to adsorb lower concentrations of persistent organic pollutants, known as
extend this strategy and policies. Worldwide contributions to environ­ POPs (Rodrigues et al., 2019), compared to rough ones. Thus, MPs’
mental impute of MPs from cosmetics is estimated to be: 18.3% South detailed qualitative and quantitative description seems necessary to
Asia, 17.2% North America, 15.9% Europe and Central Asia, 15.8% understand the Plastisphere phenomena. This research addresses those
China, 15% East Asia and Oceania, 9.1% South America and Africa, issues.
8.7% Middle East (Guerranti et al., 2019). Experts underline that so far Finally, it is essential to consider the long-term behaviour of MPs and
only entirely banned were MPs from rinse off cosmetics, what does not their changes in time due to the natural ageing or gradual biofouling and
include glitters, mascara, etc., which impact is not negligible. Regula­ sedimentation of initially floating debris. Studies (Jemec et al., 2019)
tions and consumers choice are both important for MPs reduction. The revealed that physical (morphology, roughness, size) and chemical
local evidences of increasing MPs amount and impact were the direct (hydrophobicity) properties strongly influence the biofilm type and
cause for those first bans and limitations. For instance, in Spain (V V. mechanisms of ecocorona formation. For polyethylene (PE), the stan­
Godoy et al., 2019), China (Lei et al., 2017) and Malaysia (Praveena dard crystallinity parameters could be calculated from Raman spectra.
et al., 2018) the first attempts to classify the MPs from cosmetics and Although it has many limitations discussed elsewhere, the crystallinity
estimations of their environmental impact were carried out. Mainly the parameter might be the first step to better model quantitatively the
toothpastes, scrubs, shower gels and facial cleaners were analysed. density and ageing of the material.

2
A. Dąbrowska et al. Chemosphere 308 (2022) 136407

Table 1
Samples chosen as representative for a set of similar ones and further referred to
by numbers.
Number of Product (producer/ Microbeads Macroscopic
the sample name/type) observations and
Additional Comments

1 GA/“cloudberry”/ PE + natural Visible abrasive

body peeling macroscopic particles; MPs
debris confirmed
2 GA/" cedrat"/body PE + natural Visible abrasive
peeling macroscopic particles; MPs
debris confirmed
3 TF/“coco”/shower natural Visible spherical
gel microbeads
4 CI/shower gel Mix White spheres; Green or
orange matrix
5 TF/" yellow"/ natural Visible spherical
peeling microbeads
6 TF/“kiwi”/peeling natural Visible spherical
microbeads
7 TF/“watermelon”/ natural Visible spherical
Fig. 1. An overview of product types and their number analysed in 2020/21
body sugar peeling microbeads
(@M.Mielańczuk).
8 TF/" cranberry"/ natural Visible spherical
peeling microbeads; MPs
confirmed Natural and synthetic abrasive and exfoliating agents exhibited
9 GA/" red"/face PE Visible abrasive
various shapes and sizes (super spherical, irregular, ellipsoid, elon­
peeling particles; MPs of two
types confirmed
gated). They were purified from the matrix on chromium-nickel sieves
10 GA/" white"/face PE Visible white abrasive (18–20 μm), rinsed in distilled water, and documented by optical mi­
peeling particles; MPs croscopy (raw and in the product). In addition, two face peelings and
confirmed one shower gel were tested and chosen from products available in 2021 -
11 GA/" honey"/body PE + natural Visible abrasive
all with PE microbeads. For one of those products, the total amount of
peeling macroscopic particles; MPs
debris confirmed MPs was estimated.
12 ZI/"olive"/shower PE, white Small debris in a green Finally, we only give a short statistic overview from 2020/21 years.
gel matrix; MPs confirmed Thirty-seven samples were selected from a pool of scrubs, hygiene
13 G/peeling Mix Different abrasive
products, skin care products and make-up (Fig. 1). One can compare it
particles
14 EV/cream Mix Homogenous matrix
with Fig. S1 microscopic overview and pictures provided with spectral
15 LIR/" Mocca caf."/ Natural Visible abrasive data. All chosen products included microplastics in their ingredient lists,
peeling particles both used in their matrixes and in the form of microbeads. Ingredient
16 SPA Mix Homogenous matrix; lists mainly were included on the packaging (thirty products). Other­
regular spheres and
wise, they were found on websites of chain stores selling them (seven
irregular shapes
products).
All microscopic pictures have the following scale – the longer edge of
2. Materials and methods the picture is equal to 800 μm.

2.1. Cosmetic samples 2.2. Raman spectroscopy measurements

The measured samples were collected within the time span of six All spectra were registered on the Thermo Fisher Scientific DXR
years (2015–21). In 2015 materials from dozens of producers and Raman Microscope with four laser lines: 455 nm, 532 nm, 633 nm, and
hundreds of products, including scrubs, peelings, toothpaste, cleaners, 780 nm. If it is not marked as different, the standard green line, aperture
shower gels, shampoo etc., exhibited huge diversity (Supplementary of 50 μm, focal length of 10 mm, and laser power of 10 mW were used as
materials, Fig. S1) and contained microplastics that were tested as MPs’ standard parameters. The IR laser radiation was used in the case of the
presence was proved by Raman spectroscopy. The majority, however dominant self-luminescence quenching the signal. The laser power was
not all, were adequately labelled, and polyethylene (PE) or poly­ reduced when destroying the sample or degrading the particles. Before,
propylene (PP) was indicated in the ingredient list. A substantial spectral the visible debris was identified in situ in cosmetics or cleaned in
library was created after those studies. However, last year changed the distilled water from the matrix. The exposition time and the number of
situation as the European Union market significantly recalled plastic repetitions were adjusted to ensure the proper noise-to-signal ratio. All
scrubs and peelings. Thus, the currently available specimens containing spectra had their background cut off using the standard OMNIC
plastic MPs in Poland and other EU countries come mainly from cos­ approach and were identified by OMNIC libraries and those constructed
metics imported from, for instance, Russia (directly or from producers in by authors. Signals were further scrutinized in ORIGIN and using dedi­
Latvia, Estonia, and other countries) or Asia. To update this work, the cated Python scripts for the quantitative analyses (e.g. crystallinity).
products from the brand named here as GA were chosen for detailed
studies being currently the dominant source of primary MPs as the most 2.3. SEM/EDX analyses
common Russian cosmetics imported to Poland. One should note that,
even in those products, the MPs’ presence is limited and constantly Before the scanning electron microscopy (SEM) analysis, the samples
decreases as natural microbeads gradually substitute them. The polish were prepared as follows one micro spatula (~150 mg) of the cosmetic
peelings (named here as TF) with natural microbeads were taken for sample was dissolved in 10 mL of ultrapure water. The following sample
comparison on the local market. All reported samples are chosen as was filtered on glass filters. Next, they were rinsed three times with
representative of a set of products analysed in past years and summa­ ultrapure water. The samples were washed twice with ultrapure hexane
rized in Table 1. to remove the non-polar components of the cosmetics. The SEM analyses

3
A. Dąbrowska et al. Chemosphere 308 (2022) 136407

Fig. 2. The PE microbeads a) mimetics perfectly the cedarwood in sample 2; however, b) under the microscope and in spectra, the MPs differ.

provided information about microplastic particles’ morphology and surface of the MP particles. First, the selected particles were placed on
surface roughness. A scanning electron microscope FE-SIGMA VP (Carl the glass microscope slide with conductive carbon tape. Then, the
Zeiss Microscopy GmbH) coupled with two energy-dispersive X-ray samples were coated with a 25 nm carbon layer with a vacuum coater
spectrometers (Quantax XFlash 6|10, Bruker Nano GmbH) was used to (Quorum 150T ES). Furthermore, carbon tape bridges were made for
characterize the type of accumulated metals and minerals covering the each microscope slide to avoid excessive charge accumulation. Analyses

Fig. 3. Although being from one product, the PE microbeads (four tested) in sample 12 differs significantly in their spectra, thus density and crystallinity.

4
A. Dąbrowska et al. Chemosphere 308 (2022) 136407

Fig. 4. Raman spectra of a) glitters in two similar creams and b) one for feet, and c) a particle in a feet cream.

were performed with a 120 μm aperture and 20 keV acceleration 3. Results and discussion
voltage. The beam intensity was 2.5 nA. The working distance depended
on the sample height, which was chosen to achieve approximately 7.8 3.1. Raman spectroscopy identification of microbeads
mm.
Raman spectroscopy proved the presence of synthetic microbeads in
2.4. Numerical modelling of surfaces and crystallinity tested samples. Although all products tested in the last years contained
detailed and correct information about their contents, some misleading
Qualitative and quantitative analyses were performed in OMNIC and practices were observed, which was not always the case previously. For
ORIGIN, and detailed numerical modelling was done using Python. instance, the PE coloured looks intentionally like the parts of cloud­
Python scripts, open-source, available from authors and dedicated to berries or cedarwood. Debris that biomimetic perfectly natural in­
this research, were used to describe the morphology of scrub particles gredients (samples 2 and 9) were revealed to be plastics (Fig. 2).
registered by SEM quantitatively. The following libraries were used: Interestingly, the synthetic microbeads differ from each other, both in
NumPy, matplotlib, cv2, math, mahotas, skimage, mpl_toolkits, and size and shape.
mplot3d.axes3d, sys. One can quantify the roughness by the total length In sample 9, particles were different at first glance due to their size
of edges, determine the grain size distribution, estimate surface homo­ and colour and, under optical microscopy, were confirmed to be MPs
geneity by histograms analysis, and reconstruct the 3D profiles of (Fig. S2). Sometimes, the strong self-luminescence quench part of bands
abrasive particles. Finally, the preliminary picture preparation was done that might complicate the material’s identification or even make it
in ACDSee software and IrfanView. impossible (Fig. S3). However, this considerable signal around ~1500

5
A. Dąbrowska et al. Chemosphere 308 (2022) 136407

Fig. 5. Two face scrubs with PE (a) cucumber-honey, b) cedar-ginseng) and one c) shower gel with a smaller peeling fraction.

cm− 1 is a typical fluorescence often encountered in PE of lower density presented in Fig. S5. One can briefly conclude the current situation
or in the case of the presence of polyacrylic compounds. Bands in a –CH2 considering the products collected in season 2020/21. MPs in the sam­
stretching zone confirmed the polymeric origin of those microbeads. PE ples were identified using Raman spectroscopy, with 532 nm and 633
grains were also detected in one cosmetic from a different producer, in nm laser wavelengths. Out of thirty-seven samples tested, only twenty-
sample 12 of olive fragrance (green). Interestingly, the morphology of one produced Raman spectra indicating the presence of microplastics.
microbeads differs significantly within one shower gel (Fig. 3), forming While including MPs in their ingredient lists, the remaining samples had
a set of different MPs rather than a homogenous set. Also, the grain size unclear or irrelevant results.
distribution was considerable. Some polymers found on ingredient lists that could not be identified
Some spheres marked as natural in some tested products were through Raman spectroscopy include polyethylene terephthalate,
degraded under the laser beam (as shown in Fig. S4). Also, added poloxamer 407, polyvinyl alcohol, sodium polyacrylate and methacry­
without any restrictions, creams and glitter can be identified as synthetic late crosspolymer. Their spectra were either covered by other com­
materials (Fig. 4). Moreover, creams themselves can contain MPs, and pounds (ex. titanium oxide, used in cosmetics as a colouring agent), too
synthetic polymers are frequently added. weak, or too warped due to self-luminescence to identify correctly.
Most products contain complex matrices that interfere with the However, the complementary measurements by FTIR resolve the iden­
spectra (even in purified samples). Raman spectroscopy can be efficient tification problem. The MPs found were acrylate copolymer (eleven
in monitoring their content. Different fractions within one product are samples), polyethylene (seven samples), carbomer (two samples) and

6
A. Dąbrowska et al. Chemosphere 308 (2022) 136407

expected to develop rapidly, should be sustainable and deeply consid­

ered at this stage. The European Chemical Agency (ECHA) published
2019 the restriction proposal for MPs added to products. Obviously,
their constant presence in the market is possible due to the lack of a total
ban. One can point out that, apart from abrasive particles, cosmetics
contain many polymers in their matrices. For instance, the popular
polyacrylate crosspolymer-6 thickens formulas and provides a gel
texture that leaves a velvet feel on the skin. It is a popular viscosity
controlling agent.
Other chemical additives, such as silicones or oxides, metals, and
rubbers, are better visible with EDS, as shown in the 3.3 paragraph.
Moreover, the SEM/EDX analyses enable detailed surface visualization,
and the results are discussed in paragraphs 3.3 and 3.4.

3.2. An approximative quantitative description of PE density

The “cryst_1′′ parameter is determined as the ratio between peaks’

intensity (I) of CH2 symmetric and CH2 asymmetric bands: I at ~2848
cm− 1 to I at ~2882 cm− 1, and the “cryst_2′′ parameter is calculated as
the intensity of the CH2 bending bands at ~1416 cm− 1 (crystal.) and
~1440 cm− 1 (amorph.). Although both have their limitations discussed
further in (Rytelewska and Dąbrowska, 2022), one can use them for
preliminary analyses of the crystallinity and density of PE particles.
Those parameters would also be sensitive to any natural ageing. Fig. 6
presents the data for PE confirmed in analysed cosmetics. One can note
Fig. 6. The “cryst 1′′ and “cryst 2′′ parameters were calculated for different
the substantial differences even within one product. Accordingly, in the
debris from “Agafia” samples (2 and 9) and sample 12.
authors’ opinion, it is worth monitoring the structural parameters of
MPs and NPs as they may have a crucial impact on their properties. In
polybutene (one sample). MPs could be found in microbeads and as part other words, it is already more than just a promising hypothesis to claim
of the cosmetics matrix. Acrylate copolymer was found in the matrix of that the morphology determines the environmental impact far more
shower gels, shampoos and scrubs. It has most widely included in hair than the type of polymer chain. Thus, within this study, the evidence of
products as a smoothing and conditioning agent. The name is used for the variety of P-MPs even within one product is provided.
various copolymers, including acrylic acid monomers. Polyethylene can Evaluating microbeads’ final impact and interaction with biota, their
be used as both a scrubbing agent and a viscosity enhancer. The samples, fate in the environment, and transport patterns is essential. Different
including PE, were found in the form of microbeads. All but one were starting structures in pristine material will result in different ageing
recently imported cosmetics from producers beyond EU borders. Poly­ paths and alter the final toxicological effect. However, more studies on
acrylic acid, commonly known as carbomer, was found in two scrubs reference materials are needed to properly correlate the debris type with
produced by the same company. It has been used as a viscosity enhancer its biological impact.
and a stabilizer in creams and scrubs. Polybutene was found in the
matrix of a lip gloss sample. It is most commonly used as a waterproofing
3.3. SEM/EDX results and findings
agent and a viscosity enhancer in lip and eye products.
Several products were labelled as “natural”, with imagery leading
The morphology of the tested samples is shown in Fig. 7. Particles
the consumer towards assuming they included plant-based alternatives
used in the cosmetic were present in agglomerates with several hundred
to synthetic agents. They also included natural and artificial MPs in their
micrometres. They consist of smaller MP particles with a lamellar or
matrix and as scrubbing agents. The presence of microplastic abrasive
spherical shape (Fig. 7 D, G and H). In addition, some types of products
particles (PE) was confirmed in three additional products currently on
also had rounded silicon oxide particles. The chemical composition of
the market (October 2021), two face peelings and one shower gel. The
MP is shown in Fig. S6. The EDS results showed enrichment of MPs in Fe,
final product contained a smaller fraction than body scrubs (typically
Ti. Contrary to the popular hypothesis, it was not confirmed that the
dozens of μm instead of hundreds). Representative P-MPs are presented
synthetic particles would have a more regular shape.
in Fig. 5. They reflect the most up-to-date situation on the market.
From the product marked as sample 1 in Table 1, all PE P-MPs were
3.4. Preliminary numerical characterization of microbeads
isolated and counted. In 300 ml of sample 1 (one box of the product),
one can detect 6.83 g of plastic peeling particles collected and rinsed on
Relatively simple and basic numerical processing of SEM pictures
the chromium-nickel sieve. That gives 22.8 mg of P-MPs per 1 ml of a
enables the first attempts to characterize morphology quantitatively. For
product. The PE content is equal to 2.28 wt %. The natural particles were
example, one can estimate the size of particles (Fig. S7) directly from the
at least one order of magnitude more abundant. The same is observed for
scale and their roughness by the total length of edges (Fig. S8) and
the face scrubs in which the 3.49 wt % contents of PE are gradually
discuss the shape of histograms (Fig. 8), or reconstruct the 3D profiles
replaced by natural equivalents.
(Fig. S9). The sizes of spherical particles were the following: 7.4–10 μm
Fortunately, the number of synthetic particles in peelings and scrubs
(sample 3), 0.9–1.3 μm (sample 4), 580–750 μm (sample 7), 433–613 μm
was drastically limited in EU products during the last years. They are
(sample 8). The average diameter value for irregular abrasive
substituted mainly by natural abrasive agents such as carbon, silica,
microbeads equals 464–567 μm (sample 1), 414–482 μm (sample 2), and
pumice, oatmeal, apricot or walnut husks, perlite, sodium chloride,
~274 μm (sample 12). The mean value for a set of grains was: width
sugar crystals, cellulose (also efficiently confirmed by Raman spectros­
~199.4 μm, length ~297,5 μm (sample 9), and grains ~150 μm (sample
copy), seeds of fruits, milled shells, and cacao. Even being “natural” can
16). Thus, microbead sizes vary from a few up to hundreds of μm. In
exhibit sorptive or vectors behaviour not neglected from the point of
peeling available, the optimal size was estimated to be ~400 μm,
view of environmental balance. Therefore, their growing production,
ensuring the best abrasive properties for the human skin. Notably, one

7
A. Dąbrowska et al. Chemosphere 308 (2022) 136407

Fig. 7. The morphology of investigated particles. A – sample 1, B – sample 2, C – sample 3, D – sample 4, E − sample 5, F – sample 6, G – sample 7, H– sample 8, I –
sample 9, J – sample 10, K – sample 11, L – sample 12, M − sample 13, N – sample 15, O – sample 16.

can focus on the crucial role of shape, which can be quantified by the to mitigate the growing MPs problem. It seems crucial to evaluate the
total length of edges (Fig. S8). In sample 7, the microsphere is covered possible impact of MPs according to their morphology. The luminosity
with adsorbed particles and cracks leading to the 74.07 nm of edge per histograms are sometimes precise enough (in the case of the uniform
area of μm2. In sample 1, on the rough surface, one can detect 149.5 nm signal) to determine the grain size distribution (Fig. 8). For the quanti­
of edge per area of μm2. Comparing samples 3 and 6 with 341.6 nm and tative comparison, the magnification value should be maintained. From
107.6 nm of edge per area of μm2, respectively, one concludes that this the narrow histogram signal, one can conclude the smaller grain size
first one will be more efficient. distribution and higher surface homogeneity. This simple test can pre­
In 2017 (Xanthos and Walker, 2017), local strategies were developed liminarily quantify the features of P-MPs as biofilm substrates or vectors

8
A. Dąbrowska et al. Chemosphere 308 (2022) 136407

Fig. 8. Histograms quantify the uniformity of samples.

9
A. Dąbrowska et al. Chemosphere 308 (2022) 136407

Fig. 8. (continued).

10
A. Dąbrowska et al. Chemosphere 308 (2022) 136407

of contamination. Author credits

Thus, this numerical approach provides information on the material
structure, which is crucial to determining its toxicity, behaviour, and Agnieszka Dąbrowska (85%): Conceptualization, Methodology,
fate, sorptive properties in the environment. For accurate data recovery Software, Validation, Formal analysis, Investigation, Resources, Data
from the pictures, the even particular intensity in greyscale is crucial, Curation, Writing – Original Draft, Writing – Review & Editing, Visu­
which was one of the significant limitations in this case. Here, for better alization, Supervision, Project administration, Funding acquisition,
visualization of morphology and the detector configuration, some parts Maria Mielańczuk (5%): Investigation (double-check – subsequent
were better “exposed” than others. Due to that, the data are not quan­ measurements of the same materials and report), Visualization, Marcin
titative at this stage. Natural microbeads can be rough and irregular in Syczewski (10%): Investigation (SEM/EDX), Resources (SEM/EDX),
shape, forming the developed and large surface for biofilm and being Writing – Original Draft (part of SEM/EDX), Visualization (part of SEM/
vectors for pathogens or pollutants. From that point of view, the clas­ EDX).
sification based on the morphology and hydrophobicity of particular
debris is more important than labelling “natural” or “synthetic”. How­
ever, the “natural” particles are less persistent and potentially less Declaration of competing interest
harmful but remain not smooth (Fig. S9), which is one of their main
disadvantages from the ecotoxicological point of view. The authors declare that they have no known competing financial
Along with personal care products, the primary MPs can be found in interests or personal relationships that could have appeared to influence
industrial dust, textiles, filters (also for water), and e-cigarettes. Prob­ the work reported in this paper.
ably, the natural abrasive particles could also behave as vectors for
pollutants. Further studies are needed to compare the sorptive proper­ Data availability
ties of plastics and other debris. A numerical description of MPs’ surface
is essential to understand better and control these processes. A lot in­ Data will be made available on request.
dicates that roughness and size will be the most valid parameters (Xu
et al., 2021). Finally, microbeads are an exciting model of general MPs’ Acknowledgements
behaviour in the environment: natural ageing, biofouling, transport, and
biological impact. The toxicological impact is directly related to surface This research did not receive any specific grant from funding
charge, particle size and hydrophobicity, especially in the case of agencies in the public, commercial, or not-for-profit sectors. It was
nanoplastics (Gonçalves and Bebianno, 2021). Marine microplastics partially supported at its final, but crucial, stage by the IDUB POB I
(MPs) are constantly increasing (Bonanno and Orlando-Bonaca, 2018), project (BOB-IDUB-622-168/2021/PSP: 501-D112-20-1004310) “Mik­
and their phenomena are within the interest of expanding research on roplastik morski i Plastisfera – holistyczne badania podstawowe
the Plastisphere. (jakościowe i ilościowe)” (University of Warsaw). The authors are
grateful to Barbara Pałys for the access to Raman spectroscopy and Piotr
4. Conclusions and final remarks Machowski for the substantial help in the programming and the design
of Python scripts.
Raman spectroscopy combined with SEM/EDX can provide the
detailed qualitative and quantitative characterization of microbeads. Appendix A. Supplementary data
The physical and chemical description is beneficial in case of misleading
labelling when coloured PE particles resemble natural abrasives (such as Supplementary data to this article can be found online at https://doi.
parts of cloudberries or cedar). That was observed in the case of two org/10.1016/j.chemosphere.2022.136407.
products. Furthermore, one can provide the preliminary numerical data
of surfaces’ structures and PE crystallinity. Those parameters were
References
successfully tested within this study. Modelling morphology is crucial
for understanding the biological impact of microbeads, both synthetic Amaral-Zettler, L.A., Ballerini, T., Zettler, E.R., Asbun, A.A., Adame, A., Casotti, R.,
and natural. In scrubs and peelings, PE, green polymers and natural Dumontet, B., Donnarumma, V., Engelmann, J.C., Frère, L., Mansui, J.,
agents of various origins were confirmed. Moreover, their matrices Philippon, M., Pietrelli, L., Sighicelli, M., 2021. Diversity and predicted inter- and
intra-domain interactions in the Mediterranean Plastisphere. Environ. Pollut. 286,
included a couple of synthetic polymers (mainly of acrylic origin). Ac­ 117439 https://doi.org/10.1016/j.envpol.2021.117439.
cording to the obtained results, one can also conclude that exfoliating Barría, C., Brandts, I., Tort, L., Oliveira, M., Teles, M., 2020. Effect of nanoplastics on fish
and abrasive agents in personal care products vary significantly in type, health and performance: a review. Mar. Pollut. Bull. 151 https://doi.org/10.1016/j.
marpolbul.2019.110791.
size, grain size distribution, and roughness. SEM/EDX confirmed the Bonanno, G., Orlando-Bonaca, M., 2018. Ten inconvenient questions about plastics in the
presence of metal and some traces of byproducts (organic and inorganic sea. Environ. Sci. Pol. 85, 146–154. https://doi.org/10.1016/j.envsci.2018.04.005.
compounds) even in the purified material. PE is dominant among syn­ Boucher, J., Friot, D., 2017. Primary Microplastics in the Oceans. Marine Environmental
Research.
thetic microbeads in all types of products, and polypropylene (PP) Burrows, S.D., Frustaci, S., Thomas, K.V., Galloway, T., 2020. Expanding exploration of
abrasive particles, once popular, are currently nearly absent in the dynamic microplastic surface characteristics and interactions. TrAC, Trends Anal.
market products. The mixture of different microbeds is frequently Chem. 130, 115993 https://doi.org/10.1016/j.trac.2020.115993.
Caruso, G., 2019. Microplastics as vectors of contaminants. Mar. Pollut. Bull. 146,
observed within one product, some of which can be regular spheres.
921–924. https://doi.org/10.1016/j.marpolbul.2019.07.052.
Carbon black is an increasingly used abrasive agent. Products imported Cowger, W., Gray, A., Christiansen, S.H., DeFrond, H., Deshpande, A.D.,
to the EU still can contain synthetic and natural particles in their scrubs Hemabessiere, L., Lee, E., Mill, L., Munno, K., Ossmann, B.E., Pittroff, M.,
Rochman, C., Sarau, G., Tarby, S., Primpke, S., 2020. Critical review of processing
and peelings. It is worth underlining that the amount of MPs is
and classification techniques for images and spectra in microplastic research. Appl.
constantly decreasing (even if previously used, like in the analysed Spectrosc. 74, 989–1010. https://doi.org/10.1177/0003702820929064.
products with cloudberries and cedar). Green polymers, in many as­ Delacuvellerie, A., Cyriaque, V., Gobert, S., Benali, S., Wattiez, R., 2019. The plastisphere
pects, cause the same threat as PE and PP to ocean fauna, so it is not a in marine ecosystem hosts potential specific microbial degraders including
Alcanivorax borkumensis as a key player for the low-density polyethylene
solution to change them for polycaprolactone (PCL) or polylactic acid degradation. J. Hazard Mater. 380, 120899 https://doi.org/10.1016/j.
(PLA) (Wang et al., 2021). jhazmat.2019.120899.
El Hadri, H., Gigault, J., Maxit, B., Grassl, B., Reynaud, S., 2020. Nanoplastic from
mechanically degraded primary and secondary microplastics for environmental
assessments. NanoImpact 17, 100206. https://doi.org/10.1016/j.
impact.2019.100206.

11
A. Dąbrowska et al. Chemosphere 308 (2022) 136407

Gigault, J., Baudrimont, M., Pascal, P., 2018. Current opinion : what is a nanoplastic. * Rodrigues, J.P., Duarte, A.C., Santos-Echeandía, J., Rocha-Santos, T., 2019. Significance
235, 1030–1034. https://doi.org/10.1016/j.envpol.2018.01.024. of interactions between microplastics and POPs in the marine environment: a critical
Godoy, V., Calero, M., Blázquez, G., 2019. Physical-chemical characterization of overview. TrAC, Trends Anal. Chem. 111, 252–260. https://doi.org/10.1016/j.
microplastics present in some exfoliating products from Spain. Mar. Pollut. Bull. 139, trac.2018.11.038.
91–99. https://doi.org/10.1016/j.marpolbul.2018.12.026. Rytelewska, S., Dąbrowska, A., 2022. The Raman spectroscopy approach to different
Godoy, V., Martín-Lara, M.A., Calero, M., Blázquez, G., 2019. Physical-chemical freshwater microplastics and quantitative characterization of polyethylene aged in
characterization of microplastics present in some exfoliating products from Spain. the environment. Microplastics 1, 263–282. https://doi.org/10.3390/
Mar. Pollut. Bull. 139, 91–99. https://doi.org/10.1016/j.marpolbul.2018.12.026. microplastics1020019.
Gonçalves, J.M., Bebianno, M.J., 2021. Nanoplastics impact on marine biota: a review. Shen, M., Zhang, Y., Zhu, Y., Song, B., Zeng, G., Hu, D., Wen, X., Ren, X., 2019. Recent
Environ. Pollut. 273 https://doi.org/10.1016/j.envpol.2021.116426. advances in toxicological research of nanoplastics in the environment: a review.
Guerranti, C., Martellini, T., Perra, G., Scopetani, C., Cincinelli, A., 2019. Microplastics in Environ. Pollut. 252, 511–521. https://doi.org/10.1016/j.envpol.2019.05.102.
cosmetics : environmental issues and needs for global bans. Environ. Toxicol. Silva, A.B., Bastos, A.S., Justino, C.I.L., Duarte, A.C., Rocha-santos, T.A.P., 2018.
Pharmacol. 68, 75–79. https://doi.org/10.1016/j.etap.2019.03.007. Analytica Chimica Acta Microplastics in the Environment : Challenges in Analytical
Harrison, J.P., Hoellein, T.J., Sapp, M., Tagg, A.S., Ju-Nam, Y., Ojeda, J.J., 2018. Chemistry - A Review 1017. https://doi.org/10.1016/j.aca.2018.02.043.
Microplastic-associated biofilms: a comparison of freshwater and marine Song, Z., Yang, X., Chen, F., Zhao, F., Zhao, Y., Ruan, L., Wang, Y., Yang, Y., 2019. Fate
environments. Handb. Environ. Chem. https://doi.org/10.1007/978-3-319-61615- and transport of nanoplastics in complex natural aquifer media: effect of particle size
5_9. and surface functionalization. Sci. Total Environ. 669, 120–128. https://doi.org/
Jemec, A., Kuehnel, D., Puntar, B., Zgajnar, A., Kal, G., 2019. An Exploratory Ecotoxicity 10.1016/j.scitotenv.2019.03.102.
Study of Primary Microplastics versus Aged in Natural Waters and Wastewaters * Tarafdar, A., Lee, J.U., Jeong, J.E., Lee, H., Jung, Y., Oh, H., Bin Woo, H.Y., Kwon, J.H.,
254. https://doi.org/10.1016/j.envpol.2019.112980. 2021. Biofilm development of Bacillus siamensis ATKU1 on pristine short chain low-
Jiang, Y., Yang, F., Zhao, Y., Wang, J., 2020. Science of the total environment Greenland density polyethylene: a case study on microbe-microplastics interaction. J. Hazard
sea gyre increases microplastic pollution in the surface waters of the nordic seas. Sci. Mater. 409, 124516 https://doi.org/10.1016/j.jhazmat.2020.124516.
Total Environ. 712, 136484 https://doi.org/10.1016/j.scitotenv.2019.136484. Triebskorn, R., Braunbeck, T., Grummt, T., Hanslik, L., Huppertsberg, S., Jekel, M.,
Kalčíková, G., Skalar, T., Marolt, G., Jemec Kokalj, A., 2020. An environmental Knepper, T.P., Krais, S., Müller, Y.K., Pittroff, M., Ruhl, A.S., Schmieg, H., Schür, C.,
concentration of aged microplastics with adsorbed silver significantly affects aquatic Strobel, C., Wagner, M., Zumbülte, N., Heinz-r, K., 2019. Trends in Analytical
organisms. Water Res. 175 https://doi.org/10.1016/j.watres.2020.115644. Chemistry Relevance of Nano- and Microplastics for Freshwater Ecosystems : A
Lei, K., Qiao, F., Liu, Q., Wei, Z., Qi, H., Cui, S., Yue, X., Deng, Y., An, L., 2017. Critical Review 110. https://doi.org/10.1016/j.trac.2018.11.023.
Microplastics releasing from personal care and cosmetic products in China. Mar. Wang, C., Yu, J., Lu, Y., Hua, D., Wang, X., Zou, X., 2021. Biodegradable microplastics
Pollut. Bull. 123, 122–126. https://doi.org/10.1016/j.marpolbul.2017.09.016. (BMPs): a new cause for concern? Environ. Sci. Pollut. Res. https://doi.org/10.1007/
Liu, S., Shang, E., Liu, J., Wang, Y., Bolan, N., Kirkham, M.B., Li, Y., 2022. What have we s11356-021-16435-4.
known so far for fluorescence staining and quantification of microplastics: a tutorial Wang, T., Li, B., Zou, X., Wang, Y., Li, Y., Xu, Y., Mao, L., Zhang, C., Yu, W., 2019.
review. Front. Environ. Sci. Eng. 16, 1–14. https://doi.org/10.1007/s11783-021- Emission of primary microplastics in mainland China: invisible but not negligible.
1442-2. Water Res. 162, 214–224. https://doi.org/10.1016/j.watres.2019.06.042.
Lv, L., He, L., Jiang, S., Chen, J., Zhou, C., Qu, J., Lu, Y., Hong, P., Sun, S., Li, C., 2020. Xanthos, D., Walker, T.R., 2017. International policies to reduce plastic marine pollution
Science of the Total Environment in situ surface-enhanced Raman spectroscopy for from single-use plastics (plastic bags and microbeads): a review. Mar. Pollut. Bull.
detecting microplastics and nanoplastics in aquatic environments. Sci. Total Environ. 118, 17–26. https://doi.org/10.1016/j.marpolbul.2017.02.048.
728, 138449 https://doi.org/10.1016/j.scitotenv.2020.138449. Xu, G., Yang, Y., Yu, Y., 2021. Size effects of polystyrene microplastics on the
Napper, I.E., Bakir, A., Rowland, S.J., Thompson, R.C., 2015. Characterisation , quantity accumulation and toxicity of (semi-)metals in earthworms. Environ. Pollut. 291,
and sorptive properties of microplastics extracted from cosmetics. Mar. Pollut. Bull. 118194 https://doi.org/10.1016/j.envpol.2021.118194.
99, 178–185. https://doi.org/10.1016/j.marpolbul.2015.07.029. Xu, J.L., Thomas, K.V., Luo, Z., Gowen, A.A., 2019. FTIR and Raman imaging for
Praveena, S.M., Shaifuddin, S.N.M., Akizuki, S., 2018. Exploration of microplastics from microplastics analysis: state of the art, challenges and prospects. TrAC, Trends Anal.
personal care and cosmetic products and its estimated emissions to marine Chem. 119, 115629 https://doi.org/10.1016/j.trac.2019.115629.
environment: an evidence from Malaysia. Mar. Pollut. Bull. 136, 135–140. https:// Zhang, X., Zhang, Y., Wu, N., Li, W., Song, X., Ma, Y., Niu, Z., 2021. Colonization
doi.org/10.1016/j.marpolbul.2018.09.012. characteristics of bacterial communities on plastic debris: the localization of
Rochman, C.M., Hoh, E., Kurobe, T., Teh, S.J., 2013. Ingested plastic transfers hazardous immigrant bacterial communities. Water Res. 193, 116883 https://doi.org/10.1016/
chemicals to fish and induces hepatic stress. Sci. Rep. 3, 1–7. https://doi.org/ j.watres.2021.116883.
10.1038/srep03263.

12