Microchimica Acta (2023) 190:285 https://doi.org/10.1007/s00604-023-05851-7 ORIGINAL PAPER Validated method for polystyrene nanoplastic separation in aqueous matrices by asymmetric‑flow field flow fraction coupled to MALS and UV–Vis detectors Iris H.Valido1 · Victor Fuentes‑Cebrian1 · Alba Hernández2 · Manuel Valiente1 · Montserrat López‑Mesas1 Received: 23 December 2022 / Accepted: 26 May 2023 / Published online: 7 July 2023 © The Author(s) 2023 Abstract Plastics with nanosize (nanoplastics, NPLs) must be characterized, since they can be toxic or act as carriers of organic and inorganic pollutants, but there is a lack of reference materials and validated methods in the nanosize range. Therefore, this study has focused on the development and validation of a separation and size characterization methodology of polystyrene latex nanospheres, by using an asymmetric-flow field flow fraction system coupled to multi-angle light scattering and ultra- violet–visible detectors (AF4-MALS-UV). Hence, this work presents a fully validated methodology in the particle size range 30 to 490 nm, with bias between 95 and 109%, precision between 1 and 18%, LOD and LOQ below 0.2 and 0.3 µg respec- tively, except for 30-nm standard, for both detectors, and showing stable results for 100 analyses. Keywords Nanoplastics · AF4-MALS-UV · Validation · Polystyrene · Spheres · Reference standards Introduction The large and growing use, together with the waste mis- management, of the plastic products has impacted the envi- Plastics are multi-functional materials used for a variety of ronment in many ways, as being disease vectors, marine, purposes, from food preservation to medicine supplies or soil, and air pollutants, even impacting global warming, clothing. Due to this large usage nowadays, its global pro- among others [3]. Plastic without any type of major degrada- duction has increased from 250 million metric tons (Mt) tion can cause physical damage to marine creatures mainly, in 2009 to 367 million Mt in 2020, being China the major due to its form by getting stuck between extremities, hin- producer (32%) and with Europe producing around 15% [1]. dering the movement, or in the mouth, making impossible Considering the rising use tendency and the lack of proper animals to eat [4]. Furthermore, when this waste suffers residue management, it has been estimated that 33,000 MT from physical and chemical weathering, their size reduces, of plastic waste being generated by 2050, from which only forming particles from 1 µm to 5 mm, called microplastics the 27% will be recycled, the 36% will be incinerated, and (MPLs) [5], which can enter the animal’s system, making the final 37% will be disposed into the environment [2]. this material a much more dangerous pollutant. Microplas- tics can act in two different ways in the animal’s body, the first one is to induce physical injuries in the digestive tract that can lead to major problems from digestive or endocrinal Iris H. Valido and Victor Fuentes-Cebrian contributed equally. harm to death [6]. Moreover, they can also act as carriers of organic such as polychlorinated biphenyls or bisphenol A * Montserrat López‑Mesas montserrat.lopez.mesas@uab.cat [7], which have been reported to have toxic and mutagenic activities in several marine creatures and being some of the 1 GTS Research Group, Department of Chemistry, Faculty most common organic pollutants in the marine ecosystem, of Science, Universitat Autònoma de Barcelona, Cerdanyola as well as metals [8, 9]. Nevertheless, plastics and micro- del Vallès, 08193 Barcelona, Spain 2 plastics can also endure more weathering until they reach Group of Mutagenesis, Department of Genetics smaller sizes being equal or less than 1000 nm, forming and Microbiology, Faculty of Biosciences, Universitat Autònoma de Barcelona, Cerdanyola del Vallès, what it is called nanoplastics (NPLs) [10]. Human exposure 08193 Barcelona, Spain to microplastics and nanoplastics (MNPLs) occurs directly 13 Vol.:(0123456789) 285 Page 2 of 9 Microchim Acta (2023) 190:285 through ingestion, inhalation, and dermal contact but there energy beam and scanning the backscattering electrons to is a lack of human biomonitoring studies [11]. Nanoplastics integrate this signal into an image [25]. Moreover, dynamic and their potential hazard come mainly from their size, since light scattering (DLS) or multi-angle dynamic light scattering they are in the nanoscale range, they are able to trespass the (MALS) is also used when determining the size of NPLs by blood–brain barrier [12, 13], which presents a new risk for detecting the scattered light by the sample into a certain angle, living beings as this pollutant can easily enter to the blood- or multiple angles, and transforming its intensity to a size [26]. stream and cellular medium [14]. In the latest years, several Both techniques work with static samples, which makes it investigations have reported the toxicity of the NPLs them- difficult to obtain a full-size characterization of the analyte, selves in a variety of animals, mostly, aquatic specimens mainly due to shielding effect of larger molecules or particles [15]. Like microplastics, nanoplastics can act as potential to smaller ones [27]. However, this problem can be solved carriers of several types of pollutants, such as plastic addi- by the application of asymmetric flow field flow fractionation tives, heavy metals [9, 16], or volatile organic compounds (AF4), which allows to separate nanoparticles by size inside [17], and due to their size, they will be able to enter the cells a channel. This technique bases its separation on the creation without any physical harm and, potentially, release those of a parabolic flow (Fig. 1A) by the combination of a parallel compounds into the intracellular medium [18]. flow (TIP Flow) and a vertical flow (X-Flow) to the separation This type of materials can be detected in several ways channel (Fig. 1B), which originates different velocities along depending on its size, being microplastics mostly analyzed by the height of the channel, being the fastest one in the middle the naked eye, dividing by color and size or by microscopy if [28]. Considering that smaller particles have a higher diffu- the size is smaller [19, 20]. Nevertheless, due to the different sion coefficient, these will diffuse to the center of the channel behaviors of NPLs, when compared to MPLs, methodologies acquiring larger velocities and eluting faster, and on the con- already described in the literature for MPLs cannot be directly trary, bigger particles will have lower velocities due to its low applied. Furthermore, a common technique used in the char- diffusion, which will result in larger elution times [29]. acterization of these pollutants’ composition is pyrolysis gas The AF4 technique is usually coupled to two types of detec- chromatography mass spectrometry (Pyr-GC–MS). This tech- tors, concentration detectors, and size detectors. An example nique fragments the plastic by applying high temperature and of concentration detector is the ultraviolet–visible spectro- analyzes the volatile molecules generated, which are charac- photometers (UV). On the other hand, size detectors such as teristic compounds of the plastic or other pollutants adsorbed MALS can directly detect the size of a sample by its irradia- on it [21, 22]. On the other hand, non-destructive techniques tion with a beam and the detection of the scattered light into are also used in this field, as attenuated total reflectance fourier the different detectors situated at known angles [30]. Gyration infrared spectroscopy (ATR FT-IR) which detects the charac- radius (Rg) can be obtained using a modified static light scat- teristic vibrations of specific regions of the analyte molecule tering equation, to be later transformed into geometric diam- [23]. On the other hand, due to the increasing interest in the eter (Dgeo) by the fitting and approximation depending on the characterization of NPL size, it has become a new path of geometry of the particle [31], as shown in Eq. 1, being 0.775 research in many areas. Recently, the European Commission the typical form factor of a sphere. has published a document showing the state of knowledge Rgeo about NPLs and the challenges to be faced in order to stablish Dgeo = ⋅2 (1) a new policy regarding NPLs [24]. This type of material is 0.775 usually characterized using tools such as scanning electron Although there are publications that explore the use of microscopy, which can approximately determine the size of the AF4 to obtain size information of NPLs [32, 33], there a nanoscale material by irradiating the sample with a high Fig. 1  A Illustration of the parabolic flow and the diffusion difference among particle size. B Illustration of the channel and the two main flows involved in the creation of the parabolic flow 13 Microchim Acta (2023) 190:285 Page 3 of 9 285 is still a lack of validated methodologies to apply in this field followed. The validation parameters optimized studied [11, 24], also for the validation of standards in aqueous solu- were accuracy (as precision and bias), linearity, limit tion. Reference materials are being developed and validated of detection (LOD), limit of quantification (LOQ), and methodologies; alternatives to SEM and TEM are needed resolution. for their measurement. The main objective of this research is to develop and validate an AF4-MALS-UV method for the determination of a wide range of nanoplastic sizes of Precision standards in aqueous solutions. Repeatability was studied by injecting five times a volume of 25 µL of the standard mix. The retention times (tr) for both, Materials and methods UV and MALS, detectors were compared between them, and the results were expressed in RSD (%). Reagents and standard preparation Intermediate precision was studied by injecting five times a volume of 25 µL of the standard mix, two different days by Polystyrene (latex) spheres SEM/TEM standards (Std.) (Ted the same analyst. The variance and means between days of Pella, Inc., Redding, CA), provided in an aqueous stabilized the retention times and Dgeo of each detector, UV and MALS, solution, for non-aggregation purpose, at a concentration of were studied by F-test and t-test: two-sample assuming equal 0.1% (w/v) were used to prepare a standard mix contain- variances, and the results were expressed in RSD (%). ing a combination of diameters: 40 ppm of 30 ± 9 nm and 80 ± 14 nm, 8 ppm of 170 ± 9 nm and 304 ± 9 nm, and 6 ppm of 490 ± 15 nm. To avoid the aggregation of the NPLs, the Bias Std. mix was prepared using NovaChem Surfactant 100 (IESMAT, Madrid, Spain) 0.2% (v/v). The bias was evaluated to assess trueness [32]. It was tested Polystyrene (latex) Spheres NIST standards (IESMAT, injecting five times a volume of 25 µL of the NIST stand- Madrid, Spain) of certified mean diameter (and nominal ard mix. The average geometric diameter (Av. Experimen- diameter): 62 ± 3 nm (60 nm), 122 ± 3 nm (125 nm), and tal Dgeo) calculated from MALS data were compared to the 345 ± 7 nm (350 nm), at a concentration of 1% (v/v) in an theoretical ones (Dgeo) and the results were expressed as aqueous stabilized solution were used for the bias test. apparent recovery (R, %), following Eq. 2. All solutions were prepared using “ultrapure type 1” Av.Experimental Dgeo water (Milli-Q, Millipore, 18.2 mS ­cm−1). R (%) = ⋅ 100 (2) Dgeo Equipment The NPL separation was performed using an AF4-MALS- UV. The equipment consists of an AF2000 Asymmetric Linearity, LOD, and LOQ Flow FFF (AF4) system with two PN1130 Isocratic Pumps, a Kloehn V6 Pump, a PN5300 Autosampler, a PEEK AF4 These parameters were studied by the injection of sev- Channel with a 10 kDa regenerated Cellulose membrane, eral volumes of the standard mix from 5 to 250 µL. The and a 350 µm height spacer. The mobile phase used for the results were expressed in injected mass (µg) vs. inten- separation methodology was the same used for the standard sity 90º MALS (V), injected mass (µg) vs. intensity UV preparation, NovaChem Surfactant 100 at 0.2% (v/v). The (V), and injected mass (µg) vs. Area UV. Linearity of spectral data was provided by a PN3212 UV–VIS detector at the results was considered where the overall studied 248 nm. Moreover, a PN3609 MALS detector with 9 angles points expressed a linear fitting with an R 2 parameter provided the Rg by applying the sphere fitting provided by higher to 0.99 [35]. Furthermore, the limit of detection the AF2000 Software. All the equipment mentioned above (LOD) was considered as the point of least injected mass was provided by Postnova Analytics GmbH, Germany. The where NPLs could be detected, considering three times Dgeo of the NPLs was obtained applying Eq. 1. the signal to noise ratio of the blank (σ Blank) of each of the detectors and substituting each of the results in the Method validation corresponding calibration curve to obtain the minimum injected mass able to be detected. The method validation performance and ranges of accept- The limit of quantification (LOQ) was considered as the ance from EURACHEM guidelines [34], Food and Drug point of least injected mass where the intensity was above Administration [35], and ISO/TS 21362:2018 [36] were ten times the signal to noise ratio. 13 285 Page 4 of 9 Microchim Acta (2023) 190:285 Resolution Results and discussion This parameter was calculated in both detectors at the same Optimized AF4 analysis method time taking into account the retention times of consecutive peaks using the peak full width at half maximum. After the optimization of the different parameters, the final developed method used for the separation of polystyrene Additional studied parameters nanospheres NPLs by AF4-MALS-UV is presented in Table 1. Two additional parameters were studied: membrane satura- An example of the results obtained with the optimized tion and injection volume. The membrane saturation was method for the analysis of NPLs Std. mix in aqueous tested by comparing five injections of 25 µL of standard mix matrix can be seen in Fig. 2. The MALS detector pre- in a 100-h-of-use cellulose membrane with five injections sented a linear signal with no affectation by the pressure of 25 µL of standard mix in a new cellulose membrane. To changes in the pumps (Fig. 2A). However, the UV detector do so, F-test and t-Student test were performed, and bias showed noticeable changes in its baseline signal due to and precision were compared. On the other hand, the mass the pressure changes in the system (Fig. 2B). The reten- injected precision was tested by comparing five injections tion times obtained from both detectors showed a clear of 25 µL of standard mix with five injections of 50 µL of the exponential fitting (r2 > 0.99) when plotted against stand- same standard mix. To do so, F-test and t-Student test were ard size (Fig. 2C), which results from the exponential of performed, and accuracy of the retention times of the UV the separation step programmed in the method described detector and the Dgeo of the MALS detector were compared. in Table 1. Table 1  Optimized parameters Tip flow (mL/min) Focus flow Cross-flow (mL/min) Time (min) Function type for the separation of polystyrene (mL/min) nanospheres by AF4- MALS-UV Focusing step 1.5 0.2 1.2 3 - Separation step 2.0 0 1.5 0.2 Constant 2.0 to 1.5 0 1.5 to 1.0 10 Linear 1.5 to 0.5 0 1.0 to 0 80 Exponential (0.7) Cleaning step 0.5 0 0 20 Constant Fig. 2  AF4 fractograms of the Standard mix separation with their respective retention times (from lower to higher NPL size) measured by A MALS detector at 90° and B UV detector; and C exponential regression of the retention time calibration 13 Microchim Acta (2023) 190:285 Page 5 of 9 285 Method validation for each one of the three standards: 99% for the 62 ± 3 nm Std. (calculated as 62 ± 2 nm), 100% for the 122 ± 3 nm Std. Precision (calculated as 122 ± 2 nm), and 111% for the 345 ± 7 nm Std. (calculated as 381 ± 1 nm). Repeatability was studied for the Dgeo and tr for MALS and UV detectors, respectively. The results (Table 2, top) con- firmed that both detectors showed a repeatability within Linearity, LOD, and LOQ range of acceptance (RSD = 1–20%) [33, 35]. A noticeable tendency is the increasing precision in the MALS detec- The study of the linearity, limit of detection (LOD), and tor, with the increase of the standard size, agreeing with limit of quantification (LOQ) was performed independently the increasing sensibility of the equipment for higher sizes. for each detector [34]. As it can be seen in Table 3, a gen- On the contrary, UV detector precision tends to decrease eral increasing tendency in the LOD can be observed when as the retention time increases, this can be due to the lower the standard size decreases, which correlates with the low mobility of higher size standards which increases the disper- sensitivity of MALS toward lower size particles and the low sion of the peaks, leading to a lower precision. These results of absorbance of small particles in the UV. For the LOQ, imply that there are no statistical differences between the the results shown in Table 3 presented a tendency similar to obtained using the optimized method for analyzing NPLs the LOD where an increase of sensitivity and linear range within a day. can be observed as the particle size increases. The lowest Moreover, intermediate precision was studied for the standard size, 30 nm, was well detected and shown a lin- same parameters as the repeatability. All of F-test and equal ear tendency along all volume injections, but it could only variance t-Student test showed that there were no signifi- be properly quantified with bias within range of acceptance cative differences between-days (Fexp < Ftab) variances and between 1- and 2-µg injection (25–50 µL standard mix) due between-days means (texp < ttab) from both tr from UV detec- to the closeness with the void peak and its interference at tor and Dgeo from MALS detector. On the other hand, both the integration of the 30-nm standard peak. Furthermore, detectors precision had RSDs within the range of accept- the UV detector showed the same LOQ than MALS detector ance, 80–120%, and presented the same increasing/decreas- (Table 3). However, when using the UV area (Vmin) instead ing tendency explained in the repeatability (Table 2, bot- of the MALS Intensity (V), the linearity range increased in tom), indicating that there are no significant differences in most of the standards. The linear range for the standards pre- between-days analysis. sented the same tendency for the UV and MALS detectors, where the higher size standards showed a significant larger Bias linear range than lower size standards. Overall, both detectors showed a LOD below the range of To obtain the bias, in order to assess trueness, the NIST study (< 0.2 µg, except for 30 nm) and the LOQ were also Std. mix was injected (n = 5), and the experimental Dgeo was lower when compared to the overall range. Moreover, com- calculated using the separation method developed, obtain- paring both detectors, UV detectors area resulted in a much ing a bias within the range of acceptance (80–120%) [35] more efficient and reliable method for quantifying NPLs in Table 2  Repeatability (top) and intermediate precision (bottom) results from MALS and UV detectors using the NPL standard mix Std. ­Dgeo MALS detector UV detector Average ­Dgeo Std. Dev. RSD (%) Average ­tr (min) Std. Dev. (min) RSD (%) (nm) (nm) Repeatability 30 33 4 13 12.9 0.1 1 80 82 3 4 26.8 0.2 1 170 169 5 3 48 1 3 304 323 4 1 69 3 4 490 527 4 1 95 3 3 Intermediate precision 30 30 5 18 13 0.1 1 80 84 4 5 27 0.3 1 170 167 4 3 49 2 4 304 323 5 2 75 6 8 490 531 5 1 96 3 3 13 285 Page 6 of 9 Microchim Acta (2023) 190:285 Table 3  Limit of detection (LOD), limit of quantification (LOQ), and linearity range results for each NPL standard size (Std. mix) and for each detector (MALS and UV) Std. ­Dgeo (nm) MALS UV LOD (µg) LOQ (µg) Linear range (µg) LOD (µg) LOQ (µg) Linear range (µg) 30 0.31 ± 0.07 1.03 ± 0.01 1.0–2.0 0.8 ± 0.1 2.7 ± 0.2 3.00–8.00 80 0.05 ± 0.03 0.17 ± 0.03 0.20–2.00 0.11 ± 0.04 0.4 ± 0.1 0.40–10.00 170 0.003 ± 0.001 0.009 ± 0.001 0.04–0.40 0.04 ± 0.02 0.20 ± 0.09 0.20–2.00 304 0.04 ± 0.01 0.14 ± 0.08 0.10–0.80 0.04 ± 0.01 0.13 ± 0.05 0.10–2.00 490 0.04 ± 0.01 0.13 ± 0.09 0.10–0.60 0.06 ± 0.03 0.21 ± 0.09 0.20–1.50 aqueous matrixes giving larger linear ranges and lower LOD Dgeo obtained from the analysis of both membranes were in most of the standards. within range of acceptance (Table 4), having the lowest pre- cision at 30 nm. The same results were obtained when the Resolution retention times of the UV detector were studied (Table 4), showing no significant differences between the results The resolution of each peak resulted higher than one unit and obtained with a new membrane and a 100-h used membrane. close to 1.5. Baseline resolution is achieved when R = 1.5, but it is not generally expected to achieve baseline separation Injection volume study with the AF4 systems, so the target should be to achieve a resolution of at least R = 1.0 [37]. As can be seen in Fig. 3, Volume injection test study between 25 and 50 µL showed the resolution decreases as the standard size increases, non-significant differences in the results concerning both, which is due to the method flow development. Since the retention times of the UV detector and geometric radius flow decreases with an exponential function, at longer times, obtained from the MALS detector, as F exp < F tab and the flow decreases faster, which leaves less space between texp < ttab. Furthermore, as shown in Table 5, diameters standards and, hence, less resolution. However, even having obtained with the two injections have accuracies within a lower resolution at longer times, it is still above the range ranges of acceptance, observing no differences in the results of acceptance, as well as the bias and the precision results. obtained between the two injections. The same tendency is observed when studying the retention times of the UV detec- Membrane saturation study tor, where the precision between retention time injections is less than 5% (Table 5). Overall, there are no significa- Results of the F and t-Student tests, assuming equal vari- tive differences between the two injection volumes, with the ances for membrane saturation, indicated that the membrane precision and bias of both detectors within ranges of accept- saturation did not affect significantly to either the retention ance, with the exception of the 30-nm standard due to its times studied in the UV detector nor the Dgeo from the peak closeness to the void peak of the fractogram, since as MALS detector, as Fexp < Ftab and texp < ttab. Moreover, pre- higher volume is injected, longer is the tail of the void peak, cision and bias results concerning the Average Experimental increasing the integration effect due to the peak overlapping. Fig. 3  Resolution for each standard size (Std. mix) 13 Table 4  Precision and bias results of the Std. Dgeo determined with the MALS and UV detectors from the membrane saturation study. New membrane is expressed as NM, while the 100 h used membrane is expressed as UM Std. ­Dgeo (nm) MALS detector UV detector Microchim Acta (2023) 190:285 Exp. ­Dgeo Exp. ­Dgeo Bias NM (%) Bias UM (%) Std. Dev RSD (%) Av. ­tr NM (min) Av. ­tr UM (min) Std. Dev (min) RSD (%) NM (nm) UM (nm) (nm) 30 30 33 101 109 4 12 12.6 12.6 0.3 3 80 82 84 102 105 4 4 26.2 26.2 0.3 1 170 169 165 99 97 4 2 47 46 3 7 304 323 319 106 105 4 1 65.4 66.6 0.3 0.4 490 527 533 108 109 4 1 82 83 1 2 Table 5  Precision and bias results of the Std. Dgeo determined with the MALS and UV detectors from the 25 and 50 µL volume injection study Std. ­Dgeo (nm) MALS detector UV detector Exp. ­Dgeo 25 Exp. ­Dgeo 50 Bias 25 µL (%) Bias 50 µL (%) Std. Dev (nm) RSD (%) Av. ­tr 25 µL Av. ­tr 50 µL Std. Dev (min) RSD (%) µL (nm) µL (nm) (min) (min) 30 29 32 97 105 2 6 12.4 12.2 0.2 2 80 82 83 102 103 1 1 26.2 26.2 0.3 1 170 167 169 98 99 1 1 44.4 44.1 0.3 1 304 320 324 105 107 3 1 65.8 66.1 0.6 1 490 500 519 102 106 13 3 83.5 83.5 0.3 0.3 13 Page 7 of 9 285 285 Page 8 of 9 Microchim Acta (2023) 190:285 Conclusions were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in The results of this research show the successful develop- the article's Creative Commons licence and your intended use is not ment and validation of an AF4-MALS-UV methodology to permitted by statutory regulation or exceeds the permitted use, you will determine NPL size in the range from 30 to 490 nm for need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. standards in aqueous matrices. All the parameters studied agree and are within the acceptance ranges of the referenced publications and standards. Furthermore, this method pro- vides accurate results within same day and between days References showing an increasing sensitivity toward higher sizes in both 1. Plastics Europe Market Research Group (PEMRG) / Conversio detectors with a good linearity in all the range of interest. Market & Strategy GmbH (2019) Estimated data In addition, for the studied conditions, neither the injection 2. John J, Nandhini AR, Velayudhaperumal Chellam P, Sillanpää M volume nor the membrane saturation affected the veracity of (2022) Microplastics in mangroves and coral reef ecosystems: a review. Environ Chem Lett 20:397–416. https://​doi.​org/​10.​1007/​ the results, except for the 30-nm standard due to its closeness s10311-​021-​01326-4 to the void peak and the interference generated from it when 3. Ferronato N, Torretta V (2019) Waste mismanagement in develop- performing its integration. Hence, the use of AF4-MALS- ing countries: a review of global issues. Int J Environ Res Public UV optimized method results in an efficient and accurate Health 16. https://​doi.​org/​10.​3390/​ijerp​h1606​1060 4. Wayman C, Niemann H (2021) The fate of plastic in the ocean way of determining the size of polydisperse NPL standards, environment-a minireview. Environ Sci Process Impacts 23:198– being complementary to other composition characterization 212. https://​doi.​org/​10.​1039/​d0em0​0446d techniques such as pyrolysis–gas chromatrography-mass 5. Derraik JGB (2002) The pollution of the marine environment by spectrometry or Fourier transformed infrared spectroscopy. plastic debris: a review. Mar Pollut Bull 44:842–852. https://​doi.​ org/​10.​1016/​S0025-​326X(02)​00220-5 However, if the described method wants to be extrapo- 6. Law KL (2017) Plastics in the marine environment. Ann lated to environmental samples, it must be further devel- Rev Mar Sci 9:205–229. https:// ​ d oi. ​ o rg/ ​ 1 0. ​ 1 146/ ​ a nnur​ oped. Since the method has been optimized and validated for ev-​marine-​010816-​060409 the analysis of polystyrene standards in aqueous solutions, 7. Fu L, Li J, Wang G et al (2021) Adsorption behavior of organic pollutants on microplastics. Ecotoxicol Environ Saf 217:112207. other shapes than spheres that could be present in real envi- https://​doi.​org/​10.​1016/j.​ecoenv.​2021.​112207 ronmental samples are not contemplated in the presented 8. Santana-Viera S, Montesdeoca-Esponda S, Guedes-Alonso R et al method. Moreover, due to the focus on the validation of a (2021) Organic pollutants adsorbed on microplastics: analytical method for the size determination on standard mixtures in methodologies and occurrence in oceans. Trends Environ Anal Chem 29:e00114 aqueous solutions, the possible matrix effect of real samples 9. Domenech J, Cortés C, Vela L et al (2021) Polystyrene nanoplas- has not been evaluated. tics as carriers of metals. Interactions of polystyrene nanoparticles with silver nanoparticles and silver nitrate, and their effects on Acknowledgements Marc Díaz is highly acknowledged for his support human intestinal Caco-2 cells. Biomolecules 11(6):859. https://​ in the laboratory. doi.​org/​10.​3390/​biom1​10608​59 10. Ekvall MT, Lundqvist M, Kelpsiene E et al (2019) Nanoplastics Author contribution Conceptualization: M.López-Mesas, I. H.Valido; formed during the mechanical breakdown of daily-use polystyrene methodology: V. Fuentes-Cebrian; formal analysis and investigation: products. Nanoscale Adv 1:1055–1061. https://​doi.​org/​10.​1039/​ V. Fuentes-Cebrian; writing — original draft preparation: V. Fuentes- c8na0​0210j Cebrian; writing — review and editing: I. H.Valido, M.López- 11 Domenech J, Marcos R (2021) Pathways of human exposure to Mesas; funding acquisition: A.Hernández; resources: A.Hernández, microplastics, and estimation of the total burden. Curr Opin Food M.Valiente; supervision: M.López-Mesas. Sci 39:144–151. https://​doi.​org/​10.​1016/j.​cofs.​2021.​01.​004 12. Guerrera MC, Aragona M, Porcino C, et al (2021) Micro and nano Funding Open Access Funding provided by Universitat Autonoma plastics distribution in fish as model organisms: histopathology, de Barcelona. This project has received funding from the European blood response and bioaccumulation in different organs. Appl Sci Union’s Horizon 2020 research and innovation program under grant 11. https://​doi.​org/​10.​3390/​app11​135768 agreement No 965196. Iris H.Valido acknowledges the funding support 13. Prüst M, Meijer J, Westerink RHS (2020) The plastic brain: neu- from “Ayudas Margarita Salas para la formación de jóvenes doctores,” rotoxicity of micro- and nanoplastics. Part Fibre Toxicol 17:1–16. Ministerio de Universidades (Spain). https://​doi.​org/​10.​1186/​s12989-​020-​00358-y 14. Domenech J, de Britto M, Velázquez A, et al (2021) Long-term Declarations effects of polystyrene nanoplastics in human intestinal Caco-2 cells. Biomolecules 11. https://​doi.​org/​10.​3390/​biom1​11014​42 Conflict of interest The authors declare no competing interests. 15. Paul MB, Fahrenson C, Givelet L et al (2022) Beyond microplas- tics - investigation on health impacts of submicron and nanoplas- Open Access This article is licensed under a Creative Commons Attri- tic particles after oral uptake in vitro. Microplastics Nanoplastics bution 4.0 International License, which permits use, sharing, adapta- 2:16. https://​doi.​org/​10.​1186/​s43591-​022-​00036-0 tion, distribution and reproduction in any medium or format, as long 16 Naqash N, Prakash S, Kapoor D, Singh R (2020) Interaction of as you give appropriate credit to the original author(s) and the source, freshwater microplastics with biota and heavy metals: a review. provide a link to the Creative Commons licence, and indicate if changes 13 Microchim Acta (2023) 190:285 Page 9 of 9 285 Environ Chem Lett 18:1813–1824. https://​d oi.​o rg/​1 0.​1 007/​ 28. Meier F, Heinzmann G (2017) Field-flow fractionation: a powerful s10311-​020-​01044-3 technology for the separation and advanced characterization of 17. Katsumiti A, Losada-Carrillo MP, Barros M, Cajaraville MP proteins, antibodies, viruses, polymers and nano-/microparticles. (2021) Polystyrene nanoplastics and microplastics can act as Tro- White Pap 1–34 jan horse carriers of benzo(a)pyrene to mussel hemocytes in vitro. 29 Messaud FA, Sanderson RD, Runyon JR et al (2009) An overview Sci Rep 11:22396. https://​doi.​org/​10.​1038/​s41598-​021-​01938-4 on field-flow fractionation techniques and their applications in 18. Abdolahpur Monikh F, Hansen SF, Vijver MG et al (2022) Can the separation and characterization of polymers. Prog Polym Sci current regulations account for intentionally produced nanoplas- 34:351–368. https://​doi.​org/​10.​1016/j.​progp​olyms​ci.​2008.​11.​001 tics? Environ Sci Technol 56:3836–3839. https://d​ oi.o​ rg/1​ 0.1​ 021/​ 30. Alasonati E, Caebergs T, Pétry J et al (2021) Size measure- acs.​est.​2c009​65 ment of silica nanoparticles by asymmetric flow field-flow frac- 19 Picó Y, Barceló D (2021) Analysis of microplastics and nano- tionation coupled to multi-angle light scattering: a comparison plastics: how green are the methodologies used? Curr Opin Green exercise between two metrological institutes. J Chromatogr A Sustain Chem 31:100503. https://​doi.​org/​10.​1016/j.​cogsc.​2021.​ 1638:461859. https://​doi.​org/​10.​1016/j.​chroma.​2020.​461859 100503 31. Schwafert C, Niessner R, Elsner M et al (2019) Nanoplastic 20. Schwaferts C, Niessner R, Elsner M, Ivleva NP (2019) Methods Analysis enabled by Centrifugal Field-Flow Fractionation cou- for the analysis of submicrometer- and nanoplastic particles in the pled with Multi Angle Light Scattering and Raman Microscopy. environment. TrAC - Trends Anal Chem 112:52–65. https://​doi.​ POSTNOVA analytics, Tech note, ID0011 org/​10.​1016/j.​trac.​2018.​12.​014 32. Correia M, Loeschner K (2018) Detection of nanoplastics in food 21. Reichel J, Graßmann J, Letzel T, Drewes JE (2020) Systematic by asymmetric flow field-flow fractionation coupled to multi- development of a simultaneous determination of plastic particle angle light scattering: possibilities, challenges and analytical identity and adsorbed organic compounds by thermodesorption- limitations. Anal Bioanal Chem 410:5603–5615. https://​doi.​org/​ pyrolysis GC/MS (TD-Pyr-GC/MS). Molecules 25. https://​doi.​ 10.​1007/​s00216-​018-​0919-8 org/​10.​3390/​molec​ules2​52149​85 33. Battistini B, Petrucci F, Bocca B (2021) In-house valida- 22. Castelvetro V, Corti A, La Nasa J et al (2021) Polymer identi- tion of AF4-MALS-UV for polystyrene nanoplastic analysis. fication and specific analysis (Pisa) of microplastic total mass Anal Bioanal Chem 413:3027–3039. https://​doi.​org/​10.​1007/​ in sediments of the protected marine area of the meloria shoals. s00216-​021-​03238-2 Polymers 13:1–18. https://​doi.​org/​10.​3390/​polym​13050​796 34. Magnusson B, Örnemark U (eds) (2014) Eurachem Guide: The 23. Käppler A, Fischer M, Scholz-Böttcher BM et al (2018) Com- fitness for purpose of analytical methods – a laboratory guide to parison of μ-ATR-FTIR spectroscopy and py-GCMS as identifica- method validation and related topics (2nd ed. 2014). Available tion tools for microplastic particles and fibers isolated from river from www.​eurac​hem.​org sediments. Anal Bioanal Chem 410:5313–5327. https://​doi.​org/​ 35. Food and Drug Administration Office of Regulatory Affairs, ORA 10.​1007/​s00216-​018-​1185-5 Laboratory Manual Volume II (2020). Methods, method verifica- 24. Science for Environment Policy (2023) Nanoplastics: state of tion and validation. ORA-LAB.5.4.5, pp 1–32 knowledge and environmental and human health impacts. Future 36. ISO TS 21362 (2018) Nanotechnologies — Analysis of nano- Brief 27. Brief produced for the European Commission DG objects usingasymmetrical-flow and centrifugal field-flow Environment by the Science Communication Unit, UWE Bristol. fractionation Available at: https://​ec.​europa.​eu/​scien​ce-​envir​onment-​policy 37. Gilliland D, Cascio C, Franchini F, Geiss O, Barrero-Moreno J 25. Inkson BJ (2016) 2 - Scanning electron microscopy (SEM) and (2017) Asymmetric flow field flow fractionation (AF4) and induc- transmission electron microscopy (TEM) for materials charac- tively coupled plasma- mass spectrometry (ICP-MS) methods as a terization. In: Hübschen G, Altpeter I, Tschuncky R, Herrmann means to determine nanoparticle number size distributions of sil- H-GBT-MCUNE (NDE) M (eds) Materials characterization using ver nanoparticle dispersions. EUR 28682 EN, Publication Office nondestructive evaluation (NDE) methods. Woodhead Publishing, of the European Union, Luxembourg. https://​doi.​org/​10.​2760/​ pp 17–43. https://​doi.​org/​10.​1016/​B978-0-​08-​100040-​3.​00002-X 737551 26. Stetefeld J, McKenna SA, Patel TR (2016) Dynamic light scatter- ing: a practical guide and applications in biomedical sciences. Bio- Publisher's note Springer Nature remains neutral with regard to phys Rev 8:409–427. https://​doi.​org/​10.​1007/​s12551-​016-​0218-6 jurisdictional claims in published maps and institutional affiliations. 27. Caputo F, Vogel R, Savage J et al (2021) Measuring particle size distribution and mass concentration of nanoplastics and micro- plastics: addressing some analytical challenges in the sub-micron size range. J Colloid Interface Sci 588:401–417. https://​doi.​org/​ 10.​1016/j.​jcis.​2020.​12.​039 13