Oil-based extraction as an efficient method for the quantification of microplastics in environmental samples

Background Wastewater treatment plant outlets are a major source of microplastics, with more than 90% retained in sewage sludge. No standardised method for the extraction, quantification, and characterisation of microplastics in sewage sludge or soil exists, and direct comparison of studies is often impossible. Our aim was to validate oil extraction efficiency with and without pre-treatment with Fenton’s reagent of selected microplastics in various types of environmental samples (sewage sludge and organic-rich substrates). Results Oxidation with Fenton’s reagent removed up to 90% of organic material, which improves the recovery rate and made quantification and characterisation easier and more reliable, regardless of type, shape, size, or density of the selected microplastic particles used in this study. Pre-treatment, as a pre-step of the oil extraction method, was shown to be important in reducing organic matter in all environmental samples, including sewage sludge and organic-rich substrates. It also improved the reliability of the selected method, shortened its duration, and, by reducing organic matter, made extracted microplastics more visible. The recovery rate was better for particles 1–5 mm and lower for particles 0.1 < 1 mm. Conclusions By achieving up to a 100% recovery rate for certain types of microplastics (polypropylene and poly-styrene), the selected method proved to be a promising extraction method. It was also shown to be efficient in the organic-rich substrates, for which the characterisation of microplastic particles was done by Fourier transform infrared spectroscopy. The most commonly detected types of microplastics in organic-rich substrates were polyethylene, polypropylene, polystyrene and polyester


Background
In recent years, interest in microplastics (MPs) research has been growing, mostly because the concentration of plastics in the environment is increasing [1,2].MPs, with high specific surface area, can adsorb contaminants of emerging concerns (CEC), such as potentially toxic elements, pesticides, and bisphenols, among others, and at the same time also contain various pollutants added during production, mostly to improve the flexibility and other characteristics of plastic materials [3,4].Once in the environment, MPs can release CEC [5] into different matrixes, causing potential environmental and health risks.MPs impact the growth of plants [6] and animals [7], are ingested by animals, and reach humans through the food chain or inhalation, raising serious concerns since the possible impact on human health remains unknown.
Currently, there is no uniform definition of MP sizes, although most of the literature defines MPs as particles of plastic material with a diameter less than 5 mm, as proposed by the National Oceanic and Atmospheric Administration (NOAA) [8].New definitions of size characterisation have been proposed by several researchers.For example, Hartmann et al. [9] suggested that plastic particles from 1 to < 1000 nm should be categorised as 'nanoplastics' , from 1 to < 1000 µm as 'microplastics' , from 1 to < 10 mm as 'mesoplastics' , and 1 cm and larger as 'macroplastics' .Because no uniform size definition exists, the extraction and identification methods are sometimes questionable and non-comparable.
Many studies have concluded that communal wastewater treatment plants (WWTP) outlets are a major source of MPs released into the environment [10][11][12].The most common shapes of MPs found in WWTPs are pellets, microbeads, fragments, films, and foams [10,11].Via sewage, MPs enter WWTPs and, after treatment, can end up in the environment [12,13].However, municipal WWTPs have been found to be efficient in removing MPs, with the removal efficiency of MPs increasing after each treatment phase [12,[14][15][16].Most bigger MP particles are removed in the pre-treatment units using screens and meshes, while smaller MPs are settled down during primary or secondary sedimentation.Further removal efficiency can also be increased in the tertiary treatment.The average efficiency of WWTPs is 90% but can reach up to 99.9% with tertiary treatment [12][13][14][15][16][17].Studies have also shown that more than 90% of MPs are retained in sewage sludge (SS) [18].The removal efficiency of MPs in WWTPs depends on the type of WWTPs, the level of treatment (primary, secondary, or tertiary), season, rainfall, and characteristics of communal wastewater [12,19,20].Different types of MPs were detected in SS and varied from 1000 to 24,000 particles/kg [21,22], even up to 113,000 particles/kg dw [23].The most commonly detected MPs types in SS were polyethylene (PE), polypropylene (PP) and polyamide (PA), followed by polyethylene terephthalate (PET), polyester (PES) and polystyrene (PS) in various shapes (films, fragments, films, lines, microbeads) [24].
The new European Union (EU) Wastewater Treatment Directive [25] and European Green Deal [26] aims for energy neutrality and circular economy by improving the existing SS management [27] and thus its reuse based on quality defined with new standards for micropollutants, including MPs and by quaternary treatment for potential micropollutant removal from wastewater.
MP extractions made in rich organic material, including SS, remain challenging and, due to people's constantly changing lifestyles, are also poorly controlled.Therefore, an efficient, fast, and reliable method for identifying MPs at individual process stages in the SS treatment process is needed to facilitate a decision support system for efficient SS management.
The lack of standard sampling methods for detecting MPs requires attention since there is still no standardised method for the extraction, quantification, and characterisation of MPs in SS or soil, and direct comparison of studies is often impossible or highly difficult [28].Methods of MP quantification can be summarised in the following steps: (1) sampling; (2) pre-treatment; (3) extraction; (4) quantification; and (5) identification [29].According to research results [30,31], different sample volumes, pre-treatment methods, extraction, quantification, and characterisation methods were used [32]; the size of extracted MPs remains a limitation of extraction methods.The standard MP quantification and identification protocol includes all types of MPs, regardless of their size or density.Organic material in environmental samples, microorganisms and extracellular polymeric substances may interfere with MP extraction and identification [30].This is why the pre-treatment of those samples is necessary before further analysis.Most researchers focused on the pre-treatment of samples to remove organic material.Most methods take days or even weeks to digest organic material, which may also cause the degradation of polymers.Fenton's oxidation seems to be the most efficient time-and cost-effective pre-treatment method, without visual and chemical impact on MP properties [30,[33][34][35].
After removing organic material, an extraction method is followed.Numerous different extraction methods were performed by several researchers, using density separation, electrostatic separation [36], froth flotation [37], magnetic extraction [38] and others.Density separation is the most common method; a salt solution of a specific density is mixed with an SS sample to be settled down.Particles with lower densities tend to float to the surface and are later recovered by filtration.Oil extraction proved to be a cost-effective, easy, and efficient method due to the lipophilic characteristics of plastics [39].Plastics, which are lipophilic, become hydrophobic and bound together in a water matrix.Scopetani et al. [40] used Fenton's reagent to remove organic material in soil and compost samples with an oil extraction method with custom-made polytetrafluoroethylene (PTFE) cylinders with removable caps and a piston.Cylinders, in which the mixture of sample, MPs and oil, after settling, were frozen at -40 °C.After freezing, ice columns with oil layers were defrosted and filtered through glass microfibre filters.Crichton et al. [39] compared oil extraction and density separation using NaI and CaCl 2 to determine the recovery rate of MPs in aquatic sediments, for which oil extraction reached the highest recovery rate, up to 96.2% ± 2.2%.To our knowledge, the combination of Fenton's reagent and oil extraction method has not yet been used for SS samples.
We hypothesised that the combination of digestion with Fenton's reagent and oil extraction is an efficient method regardless of the shape and type of MPs in different SS samples (anaerobic sludge and sludge cake) and organic-rich substrates for agricultural application (peat substrate, treated sludge and mixture of both).Our aim was to validate the oil extraction efficiency of MPs with and without possible pre-treatment with Fenton's reagent in various types of environmental samples (anaerobic sludge and sludge cake).After validation to extract MPs with a previously validated method (oil extraction and Fenton's pre-treatment) in environmental samples (organic-rich substrates), the extracted MPs are characterised using the Fourier transform infrared (FTIR) method.

Validation of pre-treatment and oil extraction recovery rate
The efficiency of the oil extraction procedure in combination with pre-treatment of SS samples was validated by comparing the recovery rate of the oil extraction procedure with and without pre-treatment of the samples spiked with MPs.

Spiking the sewage sludge with microplastics
For each experiment, MP particles, which differed in size, shape, type, and density (Additional file 1: Table S2, Fig. S1), were used to identify the impact on the extraction efficiency based on their characteristics.MPs were cut on smaller particles by hand and sieved on particles 0.1 < 1 mm and particles 1-5 mm by steel sieve.The densities of used MPs were determined using a certified glass pycnometer based on the principle of displaced liquid (by measuring the weight of an empty pycnometer, pycnometer filled with particles and pycnometer with added liquid of known density with and without particles).
The selected MPs were transferred into a 300 mL glass beaker with an SS sample (40 g of sludge cake and 150 mL of anaerobic sludge), 150 mL of deionised water was added to the sludge cake, and both samples were stirred at 600 rpm for 1 h until homogenisation with a magnetic stirrer (Velp Scientifica).Mixtures of larger and smaller MP particles with higher and lower densities were used, with average weights from 0.05 mg to 0.70 mg for particles from 0.1 mm to 1 mm and from 0.11 mg to 18.19 mg for particles sized from 1 to 5 mm (Additional file 1: Table S2); together 20 particles of each type and size of MP were used, from which 10 particles' sizes were 0.1 < 1 mm and 10 particles 1-5 mm.The MPs used were easily recognisable due to specific colour and shape and easily separated from other environmental particles already present in the SS samples (Additional file 1: Fig. S1).

Characteristics of sewage sludge (SS) samples
For each SS sample, we determined the moisture content and the organic material removal before and after digestion with Fenton's reagent according to APHA (2017) [41].Three trials were tested for each sample, and average values of moisture content and the percentage of organic material removal were calculated (Additional file 1: Table S1).

Method's description for microplastics' extraction
Two different SS samples were used to determine the efficiency of the extraction method: (i) anaerobic sludge and (ii) sludge cake.MPs were spiked into SS samples to determine the recovery rate with and without pre-treatment.
Three repetitions were done for each type of SS sample in three trials.
Some researchers suggested drying samples before performing the extraction method [39,42] to extract MPs from the soil easily.However, as investigated by Vermeiren et al. [43] and Lekše et al. [35], the drying of SS samples turned them into a solid agglomerate, which made it harder to extract MPs.Therefore, the SS samples were not dried before the extraction procedure.
To avoid contamination of the samples, nitrile gloves and 100% cotton clothes were used while treating the samples.Only glass and metal spoons were used, and the mixture was rinsed with deionised water before starting the experiment.Laboratory surfaces were wiped with cellulose tissue and ethanol each time after and prior to starting a new experiment.
Pre-treatment of sewage sludge (SS) samples SS samples contain a high amount of organic matter, microorganisms, and extracellular polymeric substances, which makes the extraction of MPs difficult due to high density and visibility [30].To reduce high levels of organic material (Fig. 1) and to increase extraction efficiency, Fenton's reagent as pre-treatment of samples was used prior to the extraction procedure; 30 mL of Fenton's reagent (0.05 M) and 30 mL of 30% H 2 O 2 was added to both SS samples of the same masses (see "Organic-rich substrates samples" section) and with spiked MPs.To avoid thermal reactions after adding Fenton's reagent, the temperature was regulated with an ice bath.An Fe(ll) solution was made of 3.6 g of iron (ll) sulphate heptahydrate (FeSO 4 •7H 2 O), 250 mL of deionised water and 1 mL of sulfuric acid (H 2 SO 4 ).
Oil extraction method Olive oil with a density of 912 kg m −3 was selected for the extraction of MPs according to the research results regarding the most appropriate oil for oil extraction (Fig. 2) [40].
After spiking of MPs with SS samples with a control addition of MPs, homogenisation, and digestion in the case of pre-treatment, samples were transferred from a glass beaker into 1 L separation glass funnel.Glass beaker was rinsed three times with deionized water to ensure that all content with MPs particles was transferred into glass funnel; 500 mL of deionised water was added and shaken by hand for 30 s.After mixing, 10 mL of olive oil was added, and to ensure homogenisation, the sample was again mixed with an aeration pump for 60 s.The funnel was sealed and shaken again by hand for 30 s to ensure contact between MPs and olive oil.The walls and lid of the funnel were rinsed with 200 mL of deionised water, and then the mixture was let to settle for 15-30 min.After settling, a lower layer with settled solids and supernatant was released from the funnel into another 1 L separation funnel.The upper oil layer was left in the funnel, and the procedure was repeated for the second and third times.All three oil layers, with captured MPs, were filtered through vacuum filtration with Whatmann filters GF/C, 47 mm.The lid and walls were rinsed with 100 mL of deionised water and 100 mL of ethanol (EtOH 96%) and filtered through Whatmann GF/C filters to remove oil layers from MPs particles.After adding ethanol, oil layer was easily filtered and only MPs particles remained on the filter.After filtration, the filters were carefully transferred to Petri dishes, half covered and left to dry at 37 ± 2 °C in an incubator.For each parallel, the oil extraction process was repeated three times to ensure a reliable recovery rate.
The effect of Fenton's reagent and ethanol used in vacuum filtration was investigated after the procedure, as described in Lekše et al. [35].FTIR and micrographic examination of spiked MPs (0.1 < 1 mm) was performed to determine any changes on the surface and their impact on particle identification before and after pre-treatment of samples with Fenton's reagent and oil extraction.

Quantification and recovery rate determination
MP particles obtained by the oil extraction protocol were examined under an optical microscope (Olympus CX43 Biological Binocular LED Upright Microscope) under 4 × objectives (40 × magnification).The recovery rate was determined after counting the MP particles (mean ± SD).

Organic-rich substrates samples
Four different environmental samples were used for the experiment: (i) peat substrate (raw sample, collected from the bag), (ii) peat substrate-control (peat fertilised with mineral fertilisers), and (iii) peat substrate-fertilised with treated sludge and (iv) treated sludge (activated sludge).For each organic-rich substrate sample, we determined the concentration of moisture content and the percentage of organic material removed before and after digestion with Fenton's reagent (Additional file 1: Table S1).

The description of the method for extraction of microplastics
Pre-treatment of organic-rich substrate samples (peat substrate, treated sludge and mixture of both) was done with Fenton's reagent (as described in "Pre-treatment of sewage sludge (SS) samples" section), followed by oil extraction method (described in "Oil extraction method" section).

Microplastic quantification
Whatmann filters GF/C, 47 mm were placed in Petri dishes and dried at 37 °C overnight in a Thermolyne (Type I42300) incubator and examined the following day under an Olympus CX43 Biological Binocular LED Upright microscope, which enabled identifying particles of size range between 0.1 < 1 mm.

Characterisation of environmental particles by Fourier transform infrared spectroscopy
After drying the MP particles and examination under a microscope, the extracted MP particles were saved in a desiccator until identification.The extracted particles were analysed using FTIR spectroscopy.Infrared spectra of the samples were recorded using a Perkin Elmer Spectrum Two FTIR Spectrometer equipped with Spectrum Two Universal ATR (Single Reflection Diamond).The wavenumber ranged from 4000 cm -1 to 400 cm -1 (resolution 2 cm -1 , 10 scans).

Recovery rate of oil extraction procedure (samples with spiked addition of MPs)
The recovery rate was determined to validate the oil extraction procedure with and without pre-treatment with Fenton's reagent.Any other extracted particles not added by spiking and determined in the SS samples (anaerobic or sludge cake) were not included in the recovery rate calculation.
The anaerobic sludge When comparing spiked MP recovery rate with different spiked sizes (0.1 < 1 mm and 1-5 mm), densities, types, and shapes, with and without pre-treatment, it appears that the recovery rate was, on average, higher for MP particles with the size of 1-5 mm in comparison with MP particles with lower densities and with MP shapes of fragments.
The recovery rate of pre-treated samples with Fenton's reagent was up to 100.0% ± 0.0% (size 1-5 mm) for PP fragments, PS fragments and PA lines, while for other types of MPs, the recovery rates reached up to 95.0% ± 4.7% for PP pellets and up to 98.9% ± 3.3% for PET, PE, and HDPE fragments.It seems that the oil extraction method is not effective for polymers with higher densities, such as PVC and PA in the shape of pellets and microbeads, since their recovery rate was lower and reached up to 40.0% ± 11.2% for PA and 62.8% ± 4.3% for PVC.As shown in Fig. 3, the method's reliability increases after pre-treatment.
The recovery rate of particles 0.1 < 1 mm decreased compared to the recovery rate of particles of 1-5 mm.Pre-treatment of sludge samples increased the recovery rate and reached up to 100.0% ± 0.0% for fragments of PP and PS, while for other types reached up to 80.0% ± 7.4% (PP pellets) and up to 97.8% ± 4.3% for HDPE fragments.Particles of PA, with the highest density, shaped as microbeads with sizes 0.1 < 1 mm, reached a lower recovery rate, up to 50.6% ± 17.3% after pre-treatment.As shown in Fig. 4, the method's reliability also increases after the pre-treatment of sludge samples, except for PA microbeads, where reliability decreased by 4%.
The reliability of the oil extraction method generally increases for particles of 1-5 mm after pre-treatment and reaches up to 98.9% ± 3.3% for PP and PS fragments without and 100% ± 0.0% with pre-treatment.However, reliability is lower for particles 0.1 < 1 mm.It appears that the method is suitable for the determination of bigger particles of 1-5 mm, shaped as fragments, lines and PES fibres, while for particles 0.1 < 1 mm, the reliability of the method should be improved for polymers with higher densities and the shapes of microbeads and pellets.
The sludge cake The recovery rate significantly increased after pre-treatment with Fenton's reagent for particles of 1-5 mm for sludge cake (Fig. 5) for all types, densities, and shapes of spiked MPs.The recovery rate with included pre-treatment reached 92.3% ± 2.1% (for PET fragment) and up to 100.0% ± 0.0% (for PA lines) but reached low efficiency for polymers with higher densities of PVC (52.4% ± 6.4%) pellets and PA microbeads (55.2% ± 8.0%).The reliability of the combined method, with pre-treatment included, increased for all types of MPs.
As shown in Fig. 6, the efficiency of the combined procedure increases with pre-treatment included for all MP particle types, densities, and shapes of polymers with sizes 0.1 < 1 mm, although the recovery rate is lower than for particles 1-5 mm.The highest recovery rate was reached for all types of fragments, lines, and fibres, where the highest recovery rate was for PE (94.4% ± 6.8%), PA lines (93.7% ± 2.5%), and PES fibres (91.3% ± 5.3%).The reliability of the combined method with pre-treatment increased for all spiked MPs.

Recovery rate after each extraction phase
Since oil extraction was repeated three times (from the phase of shaking/aeration to the phase of vacuum filtration), calculations regarding each extraction phase were made (Tables 1 and 2).The extraction efficiency after each phase increased, and it can be seen that extraction efficiency was higher for each phase at pre-treated samples, except for PA microbeads, for which the second and third phases of the extraction reached higher recovery rate without pre-treated samples for anaerobic sludge.The highest significant increase in the recovery rate of pre-treated samples in comparison to SS samples without pre-treatment was seen for PP pellets (up to 26% for particles 0.1 < 1 mm and 31% for particles of 1-5 mm for anaerobic sludge and 18% for particles 0.1 < 1 mm and 14% for particles of 1-5 mm), PVC pellets sizes of 0.1 < 1 mm (for up to 31% for anaerobic sludge and 21% for sludge cake) and PET fragments (for up to 39% for particles of 0.1 < 1 mm and 52% for particles of 1-5 mm for sludge cake).After a comparison of oil extraction efficiency with and without pre-treatment, it seems that the recovery rate was mostly higher (up to 15% for PP pellets) for particles of 1-5 mm and particles spiked in SS samples that were pre-treated with Fenton's reagent prior the oil extraction.The recovery rate was lower after pre-treatment for PA microbeads in anaerobic sludge (lower by as much as 16%), with higher densities and average weight of PA particles.It can be assumed that pre-treatment reduces the density of SS and MP particles, which (due to their weight and density) sunk to the bottom of the mixture before coming in contact with oil, thus negatively impacting the recovery rate.MP particles were quantified after the extraction, and the recovery rate was determined in anaerobic sludge and in sludge cake.The highest recovery rate was determined for types of MPs with densities from 825 kg m −3 to 1337 kg m −3 , shaped as fragments (PP, PS, PE, HDPE and PET), PES fibres with a density of 1332 kg m −3 and PA lines with a density of 1243 kg m −3 .
In addition to spiked MPs, 12 environmental MP particles were identified during the oil extraction procedure in the anaerobic sludge samples and 14 in the sludge cake samples, which originated from SS at WWTP and were not included in the determination of recovery rate.

Quantification of microplastic particles in the organic-rich substrates
The collected and obtained MPs were quantified after pre-treatment, and the oil extraction procedure, where the highest number (25 particles) were quantified in treated sludge and 17 in peat substrate fertilised with the same treated sludge.The lowest number (2 particles) was obtained in peat substrate, while in peat substrate-control, fertilised with mineral fertilisers, 11 particles were collected (Table 3).Particles that were too small to be characterised by FTIR were destroyed (see "Characterisation of microplastic particles in the organic-rich substrates" section) at the beginning of the analysis, and we continued the characterisation with particles of > 0.1 mm, while particles < 0.1 mm were collected and saved for further analysis.

Characterisation of microplastic particles in the organic-rich substrates
The research was carried out on only MP particles of > 0.1 mm because particles < 0.1 mm were destroyed during the characterisation with FTIR analysis and transformed into dust due to pressure.The most commonly detected types of MPs in organics-rich substrates were PE, PP, PS and PES in shapes of fragments, films, and fibres, where PE was detected and most abundant in almost all four samples, followed by PP and PS (Table 4).
The treated sludge contained EVA type of MPs, which was also found in personal care products [44].The most common shape, detected in organic-rich samples, were shapes of fragments and films.The sizes of extracted MPs ranged from 93.64 in peat substrate-control and up to 532.95 µm in treated sludge.Micrograph images (40× magnification) of some of the selected characterised types of MPs, determined in environmental samples, are shown in Fig. 7a-r.

Discussion
Based on the results, we can conclude that pre-treatment with Fenton's reagent generally improves the recovery rate of the MP particles used in this study, regardless of their type, shape, size, or density.Pretreatment, as a pre-step of the oil extraction method, was shown to be important in reducing organic matter in all environmental samples, such as SS and organicrich substrates.Reducing organic material in environmental samples is vital for the easier quantification and later characterisation of collected particles, for example, as a basis for improved SS management in the context of the new EU Wastewater Treatment Directive [25], promoting nutrient recovery while managing health and environmental risks.It also improved the reliability of the selected method and made it quicker (lasting approximately 4 h) since it reduced organic matter, making extracted MPs more visible and reducing the probability of clogging the funnels with solids, thus slowing down the filtration of environmental samples.By achieving up to a 100% recovery rate for certain types of MPs (PP and PS), the method seems highly promising for particles bigger than 1 mm.The results of recovery rates are comparable with the results of other studies; Lares et al. [45] reached a total 88.3% ± 5.5% recovery rate with oil extraction procedure from sludge samples without pre-treatment for PS beads, PE fragments, PVC fragments and SBR  (styrene-butadiene rubber) fragments with densities ranging 900-1600 kg m −3 and sizes 300-3400 µm.Mani et al. [46] achieved for 0.3-0.5 mm and 0.5-1 mm sized particles the highest recovery rate for PP and PS fragments (99% ± 3%), followed by PET-G (glycol modified PET) particles with a recovery rate of 98% ± 5% in environmental samples when using oil extraction procedure and H 2 O 2 oxidation.Crichton et al. [39] used an oil extraction protocol in sediment beach samples, with up to 7.02% of organic matter (without removal of organic matter) and reached 92.7% ± 4.3% extraction efficiency for fibres (PES, nylon and stain-resistant polyester) and 98.0% ± 4.5% for PVC particles.In our study, SS samples contained up to 70% organic matter, reaching a recovery rate of up to 100% for PP and PS and PA lines.
Despite high recovery rates for PP, PS, PE, HDPE, PET fragments, PA lines, PP pellets and PES fibres, the method is unsuitable for extracting particles with higher densities and weights, such as PA and PVC, shaped as microbeads or pellets, since particles remained in SS and avoided contact with added oil.Rodina et al. [47] also attained lower efficiency of PA-6 MPs extraction efficiency, possibly due to the higher density of used polymers.Wang et al. [48] used PS beads to test extraction from biosolids and soil samples and discovered similar findings, specifically that pre-treatment reduced extraction efficiency for smaller types of PS beads, while pre-treatment improved extraction efficiency for bigger particles of PS beads.Their study also showed that smaller particles were partially affected by H 2 O 2 , which could impact extraction Fig. 7 Micrograph images at 40 × magnification of selected particles after pre-treatment and oil extraction protocol efficiency, while larger particles were not negatively affected by H 2 O 2 digestion; the influence of pre-treatment was suspected and not proven.The results of our study, and those of Rodina et al. [47] and Wang et al. [48] indicate that the size, shape, density and weight of specific polymers can influence extraction efficiency; e.g. in the study particles of PA, with the highest density, shaped as microbeads with sizes < 1 mm, reached a lower recovery rate, up to 50.6% ± 17.3% after pre-treatment.It can be assumed that as the size decreases, the particles could also be attached to any residue (organic or inorganic) material in the SS samples, avoiding oil contact and being due to gravity sedimented while density separation.After pre-treatment the particles in our study showed no visual changes on the surface under microscopic examination with 40 × magnification (Fig. S2).No changes in absorption spectra of studied MPs samples were also observed at ATR-FTIR analysis, except that in almost all IR spectra of MPparticles after oil extraction, despite the extensive washing with EtOH, vibrational bands characteristic of oil are also observed (Fig. S3).Hurley et al. [30] and Li et al. [34] also reported no impact of chemical pre-treatments by Fenton's reagent, while Maw et al. [49] discovered no changes in the surface area for PET, LDPE, PVC, PP and PS; however, minimal changes in the weight for LDPE and PP were observed, while FTIR spectra were not affected for observed polymers.The method was also shown to be efficient in the organic-rich substrates, for which characterisation of MP particles was done using FTIR.The types of extracted MPs are consistent with other studies, in which PE, PP, PS, and PES were the most commonly detected types of MPs [18,50].In the organic-rich substrates, particles shaped as fibres, fragments and films, with sizes from 93.64-532.95µm, were extracted.Characteristics of extracted MPs coincide with the results of Liew et al. [51], who discovered different shapes and sizes of MPs particles in SS samples (from primary clarifier and activated sludge) and also observed that MP size reduces when going through different stages of wastewater treatment.Van der Berg et al. [51] confirmed PP and PVC as the most abundant types of MP particles with sizes between 150 and 250 µm, with shapes of fragments, fibres and films in olive and cereal farmland.Zhu et al. [52] discovered PE, PP and PS, with sizes 0.02-1 mm and fibres as the largest proportion of discovered particles, followed by films, fragments, and foams in farmland, while Liu et al. [53] found PP, PE, and PES with shapes of fibres, fragments, films and pellets as the most common types of MPs, sized between 0.03 and 5 mm in vegetable farmland.Discoveries of extracted MP particles in organic-rich samples confirm the validation of the oil extraction method with included pre-treatment.The highest recovery rate (up to 100%) was also reached for PP, PE, and PS fragments, as well as for PES fibres in SS.The results of extraction efficiency coincide with extracted MPs particles obtained from organic-rich substrates, where the most abundant types of MPs were PP, PE, PS, PES with shapes of fragments, films and fibers.

Benefits of the oil extraction method
The presented method is quick, reliable, and efficient, as well as being cost-effective and environmentally friendly.The oil extraction protocol took, on average, 4 h per sample, including pre-treatment, extraction, and characterisation of determined particles, making the method time-efficient compared to other pre-treatment and extraction methods that can take days [30,45].Pre-treatment with Fenton's reagent generally improves the recovery rate of the selected MP particles used in this study, regardless of their type, shape, size, or density.
To characterise obtained particles, the Fourier transform infrared (FTIR) spectroscopy method was identified as an effective method for characterising particles > 0.1 mm.

Limitations of the oil extraction method
The oil extraction method is not suitable for all types of fibres, as proven in the preliminary research with spiked polyacrylic and PA microfiber; during the oil extraction method, fibres degraded into many thinner fibres that were impossible to count.Another limitation of the proposed method is the size of the funnel outlet, which limits the size of MPs and could clog the funnel.To avoid or minimise clogging, the funnel outlet was physically enlarged.Further studies could be conducted using different glass devices to separate heterogeneous phases.Nakajima et al. used a small portable device provided by the Japan Agency for Marine-Earth Science and Technology (JAMSTEC), a microplastic-sediment separator (JAMSS) unit that separates sediment and supernatant by sliding two plates against each other [54].
MP particles with higher densities and masses (PA with a density of 1150 kg m −3 and an average weight of particles 0.1 < 1 mm 0.69 mg and PVC with a density of 1326 kg m −3 and average weight of particles 0.1 < 1 mm 0.36 mg) shaped as microbeads and pellets had lower efficiency since those particles remained in the SS and were not transferred to the upper oil phase.Some particles were also left on the glass surface of the separation funnel and could easily be missed if the separation funnel was not appropriately rinsed with deionized water and ethanol after each extraction.
While FTIR analysis, particles of < 0.1 mm were destroyed due to attachment pressure.Smaller MP particles (not destroyed during characterisation) that cannot be detected with FTIR can be characterised with the micro-FTIR as planned for future experiments.

Conclusions
The combination of digestion of waste sludge using Fenton's reagent and oil extraction method is a cost-effective and reliable method for the quantification and characterisation of particles of > 0.1 mm due to efficient removal of organic material and non-destructibility for degradation of polymers.The oleophilic interaction between oil and polymers is strong enough to extract denser polymers to the surface of the oil layer, but it has lower recovery rates and reliability for certain types and shapes (PA microbeads and PVC pellets).The reliability of the oil extraction method is higher with pre-treatment of environmental samples of sludge for particles of 1-5 mm, regarding the type and shape of MP.
The most frequently detected type of MPs in the analysed organic-rich substrates were PE, PP, PS, and PES in the shape of films, fragments, and fibres, sized from 93.64 to 532.95 µm.For the characterisation of detected particles (> 0.1 mm), the Fourier transform infrared (FTIR) method was identified as effective.The time efficiency and reliability of the oil extraction protocol with pretreatment included for organic-rich substrates enable the method to improve the decision support system for efficient SS management.Future experiments should focus on the oil extraction efficiency of particles smaller than 0.1 mm, investigate density and shape influence and possible mechanisms to improve oil extraction efficiency.

Table 1
Recovery rate [%] of MPs in anaerobic sludge after each extraction phase for particles 0.1 < 1 mm and 1-5 mm in anaerobic sludge

Table 2
Recovery rate [%] of MPs in anaerobic sludge after each extraction phase for particles 0.1 < 1 mm and 1−5 mm in sludge cake

Table 3
Number obtained, characterised, and destroyed particles a for an explanation, see "Organic-rich substrates samples" section

Table 4
Characteristics of microplastic particles in the organic-rich substrates