中文   |   English   |   Email Alert
Fig/Tab
Advanced Search
image

Raman Spectroscopy in the Detection of Environmental Micro- and Nanoplastics: Applications and Challenges

Kefu Ye , Minjie Xie , Xingqi Chen , Zhiyu Zhu , Shixiang Gao

Progress in Chemistry ›› 2025, Vol. 37 ›› Issue (1) : 2 -15.

PDF
Progress in Chemistry ›› 2025, Vol. 37 ›› Issue (1) : 2-15 DOI: 10.7536/PC240710
research-article
Scan to read & Share
Raman Spectroscopy in the Detection of Environmental Micro- and Nanoplastics: Applications and Challenges
Author information +
History +
PDF

Abstract

This review highlights the advantages and research advancements of Raman spectroscopy in detecting micro- and nanoplastics in the environment. With the worsening issue of microplastic pollution, particularly its widespread presence in aquatic and terrestrial ecosystems, Raman spectroscopy has emerged as a non-destructive, high-resolution analytical technique widely employed for identifying and quantitatively analyzing micro- and nanoplastics. This is attributed to its unique spectral characteristics and reduced susceptibility to water interference compared to infrared spectroscopy. The strengths of Raman spectroscopy in detecting micro- and nanoplastics lie in its high spatial resolution, broad spectral range, and exceptional sensitivity. However, challenges such as fluorescence interference and low signal-to-noise ratios persist in the detection process. To enhance Raman signals, researchers have introduced various approaches, including sample pretreatment, surface-enhanced Raman spectroscopy (SERS), and nonlinear Raman spectroscopy techniques. Furthermore, this paper underscores the necessity of building a comprehensive Raman spectroscopy database to boost detection accuracy and efficiency. Future research directions include developing more effective preprocessing methods, dynamically monitoring the behavior of micro- and nanoplastics, and integrating intelligent detection systems.

Contents

1 Introduction

2 Raman spectroscopy methods for micro-and nanoplastics

2.1 Basic principles and conventional Raman spectroscopy

2.2 Surface-enhanced Raman spectroscopy (SERS)

2.3 Coherent Raman spectroscopy (CRS)

2.4 Raman imaging

3 Identification in environmental samples with Raman spectroscopy

3.1 Fluorescence interference and its elimination

3.2 Machine learning applications with Raman spectral databases

4 Quantitative Analysis

4.1 In situ concentration and mass concentration

4.2 Number concentration via µ-Raman and imaging

5 Conclusion and outlook

Graphical abstract

Keywords

Raman spectroscopy / micro- and nanoplastics / qualitative identification / quantitative detection / signal enhancement and optimization / machine learning

Cite this article

Download citation ▾
Kefu Ye, Minjie Xie, Xingqi Chen, Zhiyu Zhu, Shixiang Gao. Raman Spectroscopy in the Detection of Environmental Micro- and Nanoplastics: Applications and Challenges. Progress in Chemistry, 2025, 37(1): 2-15 DOI:10.7536/PC240710

登录浏览全文

4963

注册一个新账户 忘记密码

1 Introduction

Plastic, as a type of polymer material, is typically formed by monomer molecules through addition and condensation reactions[1]. With the continuous development of the chemical industry, humans can customize various plastic products according to their usage. However, during this process, the phenomenon of micro- and nano-plastic pollution has also become increasingly severe[1⇓-3]. Microplastics are defined as plastic particles with a particle size (or one-dimensional length) less than 5 mm, while nanoplastics, borrowing from the definition of engineered nanoparticles, are defined as plastic particles with a particle size less than 100 nm[4]; some scientists define them as plastic particles smaller than 1 μm. These plastic particles are widely present in aquatic and terrestrial environments[5-6] and primarily enter the human body intentionally or unintentionally through ingestion, inhalation, and dermal contact[7⇓-9]. According to statistics, there are already more than 5 trillion plastic particles floating in the oceans, with a total weight of about 270,000 tons[10]. As plastics weather and degrade, each particle breaks down into smaller and smaller fragments, forming microplastics and even nanoplastics, with their numbers increasing exponentially, thereby increasing the risk to the safety and health of animals and humans[11]. Like other impacts of human activities on natural systems, although the problem of plastic pollution has been widely recognized, the pollution situation continues to deteriorate, and even if it is immediately controlled, the adverse effects will persist for hundreds of years[12].
The assessment of the safety risks associated with micro- and nanoplastics requires an analysis and identification of their abundance, size distribution, and chemical composition[13]. In the process of analyzing and measuring micro- and nanoplastics, a series of identification methods have been developed, including vibrational spectroscopy, density measurement, differential scanning calorimetry, gas chromatography-mass spectrometry-based methods, and hyperspectral imaging techniques[3,14⇓-16]. Compared to other identification methods, vibrational spectroscopy can rapidly obtain high-throughput geometric morphology and chemical characteristics of particulate samples with only a small amount of sample, simple pretreatment steps, and without damaging the sample, thus minimizing the possibility of misjudgment[17-18].
In vibrational spectroscopy, Raman spectroscopy and Fourier Transform Infrared (FTIR) spectroscopy are commonly used vibrational spectroscopic techniques for the identification of microplastics and have been recommended as complementary techniques by the European Union's Marine Litter Expert Group. They advocate that all suspected microplastics within the size range of 1 to 100 μm should be confirmed for their polymer composition through spectral analysis[19]. Compared to FTIR spectroscopy, Raman spectroscopy is a vibrational spectroscopic technique based on inelastic scattering of light, featuring higher spatial resolution (down to 1 μm, while FTIR spectroscopy is 10-20 μm), wider spectral coverage, higher sensitivity to nonpolar functional groups, no interference from water, and narrower spectral bands[20]. The drawbacks are that Raman spectroscopy is susceptible to fluorescence interference, has a lower signal-to-noise ratio, and due to the use of lasers as the light source, can cause sample heating, occasional background emission, and even degradation of polymers[21].
This paper will focus on the application of Raman spectroscopy in the analysis of micro- and nanoplastics in environmental samples, providing researchers with a critical and novel perspective to promote research on Raman spectroscopy in the field of micro- and nanoplastics. The discussion will be conducted from three aspects: (1) theoretical analysis of methods for detecting micro- and nanoplastics using Raman spectroscopy; (2) overcoming the interference of other impurities in environmental samples on the qualitative detection of micro- and nanoplastics by Raman spectroscopy; (3) quantitative detection of micro- and nanoplastics through Raman spectroscopy.

2 Common Methods For Detecting Micro And Nanoplastics Using Raman Spectroscopy

Through various microscopy techniques, such as optical microscopes, electron microscopes, and scanning probe microscopes, the geometric shapes and surface characteristics of particulate samples can be directly obtained. On the other hand, the chemical identification of particulates is equally important, as it not only confirms the presence of micro- and nanoplastics in the samples but also provides additional information on chemical composition, such as the presence of additives or the degree of aging. In the analysis of micro- and nanoplastics, Raman spectroscopy is often combined with optical microscopy to provide the capability for imaging and analyzing individual particles. However, there are still drawbacks such as insufficient spatial resolution and low signal intensity. This section mainly introduces the commonly used methods of Raman spectroscopy for the detection of micro- and nanoplastics, including micro-Raman spectroscopy, surface-enhanced Raman spectroscopy, coherent Raman spectroscopy, and Raman spectroscopic imaging (Figure 1).

2.1 Principles of Raman Spectroscopy and Common Raman Spectroscopy Detection Methods

Raman spectroscopy is essentially inelastic scattering of light at the molecular level. The vibrational energy levels of molecules are excited by incident light (with a frequency of ω0), transitioning from the ground state to an excited state (which is highly unstable), and then returning to a lower energy state through the emission of scattered photons (with a frequency of ωR). The difference between ω0 and ωR corresponds to the vibrational energy levels of the molecule. The frequency shift of the scattered light relative to the incident light (ω0 ± ωR) is represented by the frequencies of the vibrational peaks displayed in the Raman spectrum. Therefore, Raman scattering has significant advantages in studying molecular vibrations and rotations[22].
Raman spectroscopy, as a non-destructive, rapid, and highly sensitive detection method for nonpolar functional groups, can perform precise qualitative and quantitative analysis on micro- and nano-plastics primarily composed of nonpolar covalent bonds. Raman spectroscopy can identify the chemical composition of micro- and nano-plastics, thereby understanding their sources and degradation processes. Different types of plastics, such as polyethylene (PE), polypropylene (PP), and polyethylene terephthalate (PET), have unique spectral characteristics in Raman spectra[23]. By comparing and analyzing these characteristic spectra, we can trace the pollution pathways of micro- and nano-plastics and assess their environmental impact.
In addition, Raman spectroscopy technology can detect the interactions between micro- and nanoplastics and other pollutants. For example, the surfaces of micro- and nanoplastics easily adsorb organic pollutants and heavy metals, which may enter organisms through the food chain, further exacerbating ecotoxicity[24-25]. Through Raman spectroscopy, it is possible to study in detail the forms and concentrations of these complex pollutants, thereby providing a scientific basis for environmental risk assessment.
Plastic samples exposed to the environment undergo long-term aging and fragmentation, leading to a significant increase in the number of small-sized microplastics[26-27]. However, in many studies, small-sized microplastics are often overlooked due to the overly large mesh size of the nets used during sampling. As demonstrated in a study by Enders et al.[28], when collecting microplastics from the Atlantic Ocean and identifying them using micro-Raman (µ-Raman) technology, 64% were found to be smaller than 40 µm, with the distribution of microplastic particles in the size range of 10-100 µm following an exponential distribution, and a scaling exponent λ of 1.96. Erni-Cassola et al.[29] used µ-Raman technology to identify microplastics smaller than 40 µm in surface waters of Plymouth Bay (UK). Schymanski et al.[8] identified, using µ-Raman technology, that 80% of plastic particles in a study of 38 brands of bottled water had a particle size between 5-20 µm.
The applicability of classical Raman scattering techniques (such as µ-Raman) for analyzing environmental micro- and nanoplastic samples is quite limited. This is due to the low environmental concentration of micro- and nanoplastics, coupled with the low efficiency of Raman scattering (even under favorable conditions, only 10-8 of the incident photons are converted into Raman photons)[22], which makes the signal of the target substance easily overwhelmed by noise. A solution can be to add a concentration and enrichment step in the sample pretreatment process, increasing the concentration of micro- and nanoplastics in the detection samples. Typically, the operational measures on the instrument side include extending the laser exposure time and increasing the laser intensity, but this will prolong the testing time and may even cause damage to the plastic samples from the laser. Therefore, in actual operation, the experimenter needs to make a trade-off between exposure time and laser intensity based on the actual situation of the sample. Another effective approach is to replace the conventional charge-coupled device (CCD) with an electron-multiplying charge-coupled device (EM-CCD). Compared to the conventional CCD, the EM-CCD is equipped with a multiplication register that can amplify the gain by a factor of 1,000[30]. This can greatly save sampling time and suppress the intrinsic noise of the instrument. Dieing et al.[31] used a conventional CCD to obtain Raman images of polymethyl methacrylate (PMMA) on glass, with a collection time of 36 ms per spectrum. The collected images had a very low signal-to-noise ratio, making it almost impossible to distinguish between glass and PMMA. Using an EM-CCD detector, they were able to obtain images with the same signal-to-noise ratio in one-tenth of the time, and with the same acquisition time, it was very easy to distinguish between glass and PMMA. With the expansion of Raman theory, the emergence of surface-enhanced Raman spectroscopy and coherent Raman spectroscopy has effectively improved the Raman signals of the target substances[32-33], and there has been considerable research on their application in the analysis of environmental samples of micro- and nanoplastics.

2.2 Surface-Enhanced Raman Spectroscopy

Typically, the content of micro- and nanoplastics in environmental media is relatively low. To facilitate detection, the sample pretreatment process by scientists often includes concentration and enrichment, and the use of enhanced Raman techniques to increase the Raman signal response of micro- and nanoplastics[32,34]. Surface-enhanced Raman spectroscopy (SERS) is a commonly used method that focuses electromagnetic energy through metal nanostructures, thereby enhancing the Raman spectral signals. The enhancement principles are divided into electromagnetic enhancement and chemical enhancement. First, incident light drives the conduction band electrons on the metal surface. When the frequency of the light matches the oscillation frequency of the electrons, local surface plasmon resonance (LSPR) occurs, leading to the formation of a photoelectric field at the interface of the metal nanostructure. This photoelectric field is much stronger than the incident light's electric field, providing an enhancement factor (EF) of approximately 106 ∼ 108. Second, the spatial position of the target molecule directly affects the intensity of the Raman scattering signal. The gaps where plasmonic coupling occurs on the surface of nanoparticles are typically referred to as "hot spots," where adsorbed molecules exhibit significantly different SERS enhancement effects compared to other areas. As the distance from the hot spot increases, the enhancement effect rapidly decays. Chemical enhancement refers to the interaction between the target substance and the metal substrate, where the adsorption of molecules on the substrate surface undergoes chemical complexation or charge transfer, leading to changes in the Raman polarizability of the molecules, with an EF of 10 ∼ 100. The definition of the enhancement factor (EF) for SERS technology when detecting organic molecules is given by equation (1).
\[EF=\frac{{{{I}_{\mathrm{SERS}}}}/{{{N}_{\mathrm{SERS}}}}\;}{{{{I}_{\mathrm{NRS}}}}/{{{N}_{NRS}}}\;}\]
in the equation, ISERS and INRS are the peak intensities of a specific peak in SERS and normal Raman spectra, respectively, while NSERS and NNRS are the average numbers of signal molecules in the focal area.
For the determination of single plastic particles, NSERS and NNRS are typically both fixed at N = 1. It is worth noting that although this simplifies the calculation of the EF value, the Raman spectra collected must use the same sample and the same testing conditions, especially the distance from the objective lens to the sample, to ensure that the polymer is exposed to laser spots of the same intensity[35].
SERS can be achieved through two forms: metal nanoarray substrates and colloidal metal nanoparticles (as shown in Table 1 and Table 2). Metal nanoarray substrates can further be divided into filter-type and non-filter-type. Filter-type substrates have the characteristic of filtration and concentration, which simplifies the sample pretreatment steps, allowing for in-situ detection of micro- and nanoplastics on SERS filters. The Klarite enhancement substrate is a commonly used commercial non-filter-type metal nanoarray substrate, often serving as a reference for developing other enhancement substrates. Xu et al.[35] used gold to prepare an inverted pyramidal Klarite enhancement substrate, which was able to identify standard PS and PMMA microspheres down to 360 nm, a diffraction size that ordinary Raman spectroscopy struggles to reach. The primary reason for this is the electric field enhancement caused by particles entering the bottom of the pyramidal pits, rather than the electric field enhancement between particles and gold nanoparticles. However, environmental plastic particles vary in size, not all of which can perfectly enter the bottom of the pyramidal pits, leading to different enhancement effects for different particles. Additionally, the manufacturing process and cost of such substrates are relatively complex and expensive. Repeated use after multiple cleanings can cause the gold film to peel off, damaging the structure, making it less cost-effective. Park et al.[52] utilized the low melting point of plastics, heating an Ag nanoparticle-constructed SERS substrate to melt the plastic particles and allow them to enter the SERS "hot spots," overcoming the difficulty of weak Raman signals due to various shaped micro- and nanoplastic particles failing to enter the "hot spots." To address the issue of varying enhancement effects for different sizes of nanoplastics on metal nanoarray substrates, Chang et al.[36] designed a SiO2@Ag substrate, using the coffee ring effect and the gravity of the nanoplastics to trap individual nanoplastics in nanowells formed by three adjacent SiO2@Ag nanoparticles, achieving higher spatial resolution and better consistency. Using a 100 µmol/L 4-aminothiophenol ethanol solution to evaluate the uniformity of the SiO2@Ag substrate, the coefficient of variation (CV) was 23.4%, better than the 39.05% for the Klarite substrate. The high consistency of the signal provides favorable conditions for quantitative analysis of microplastics, with the Raman response intensity of PS nanoplastics showing a highly linear relationship with concentration in the range of 0.00005% to 0.1% (R2 > 0.97). Yang et al.[46] proposed using one-dimensional Ag nanowires to self-assemble into a filtration membrane for pre-concentrating nanoplastics. The high SERS activity of the Ag nanostructures can enhance the Raman signals of nanoplastics in situ. Combining these two features, the detection limit for 50 nm diameter PS plastic microspheres was as low as 0.10 µg/L.
The metal nanoparticle colloids used for micro-nanoplastic analysis are mainly composed of metals such as Ag, Au, and Cu, which can be easily photoexcited to generate surface plasmons. In the process of using metal nanoparticle colloids, the metal nanoparticles first need to be mixed with micro-nanoplastics. The effect of signal enhancement is mainly related to the shape and size of the metal nanoparticles and the degree of binding between the metal particles and the micro-nanoplastics. Lee et al.[51] significantly enhanced the Raman signal of 600 nm PS plastic using approximately 50 nm Au nanourchin particles. Kleinman et al.[53] found that Au nanoparticles with more vertices produce a richer number of "hotspots" (number of hotspots: spherical < triangular < star-shaped), further improving the Raman response. Lv et al.[34] and Hu et al.[48] used Ag nanoparticles to detect micro-nanoplastics in water samples. They used the aggregating agents NaCl or KI to disrupt the double electron layer of Ag nanoparticles, allowing the Ag particles to aggregate and envelop the surface of PS nanoparticles. The detection limit for 100 nm PS microspheres could reach 6.25 µg/mL. Although some studies have shown that competitive adsorption from the aggregating agent ions can significantly reduce the SERS sensitivity of the target substance[54⇓-56], researchers still recommend using KI as an aggregating agent to increase the aggregation of metal nanoparticles with micro-nanoplastics. This not only removes impurities on the surface of the metal nanoparticles[57] but also increases the number of "hotspots"[58].
Overall, metal nanoarray substrates have more efficient and stable "hotspot" structures and good reproducibility, which are conducive to achieving single-particle imaging and quantitative research on micro- and nanoplastics. However, their manufacturing process is complex and requires more expensive processing equipment. In the future, it will be possible to seek more convenient manufacturing processes and cheaper preparation materials to achieve the goal of efficient, rapid, and convenient identification of micro- and nanoplastics in the environment.

2.3 Coherent Raman Spectroscopy

Coherent Raman spectroscopy is a nonlinear optical method that primarily uses two synchronized pulsed lasers (pump light and Stokes light). When the frequency difference between the two beams corresponds to the vibration frequency of the target chemical bond, coherent anti-Stokes Raman scattering (CARS) and stimulated Raman scattering (SRS) occur, achieving a stronger signal at the expense of some spectral information[33]. Rhee et al.[59] used the CARS method for direct and rapid detection and identification of PE, PS, PMMA, and PA-12 micro- and nanoplastics in soil samples. Laptenok et al.[60] utilized SRS technology to detect plastic fibers in drinking water, surface seawater, coastal sediments, deep-sea sediments, and fish gastrointestinal tracts, distinguishing them from natural fibers. Qian et al.[61] directly detected micro- and nanoplastics in bottled water using SRS technology, quantifying that there are more than 105 plastic particles per liter, most of which are nanoplastic particles. Additionally, as nonlinear optical techniques, CARS and SRS also have characteristics such as real-time rapid imaging, fluorescence suppression, minimal sample damage, and high spatial resolution. Huber et al.[62] rapidly detected PS, PE, and PMMA plastic particles with diameters ranging from 100 to 5000 nm in a flow field using SRS technology. Thanks to the high temporal resolution of 60.5 µs, this technique can directly detect the single signal of each particle and infer the particle diameter from the average peak width and intensity of the signals. CARS and SRS technologies can also perform label-free detection in biological samples[63⇓⇓⇓-67]. For example, Choi et al.[66] used CARS to track the movement of 2 µm PS microspheres in live human cells in real time, successfully obtaining evidence of their internalization into cells from the dynamic characteristics of the PS microspheres. Fueser et al.[68] used CARS technology to detect the ingestion of 1 µm PS microspheres by Caenorhabditis elegans, finding that the lipid distribution area in worms exposed to PS microspheres was 79% higher than in unexposed worms. Xin et al.[69] used SRS technology to quantitatively monitor the bioaccumulation and metabolic toxicity caused by micro- and nanoplastic pollution during the early developmental stages of zebrafish, discovering that the gut and liver are the main target organs for micro- and nanoplastics, revealing that these plastics disrupt the development and lipid metabolism of zebrafish larvae, leading to developmental delays. It is worth noting that plastics are composed of elements such as C, H, and O, and in cells, molecules like proteins, nucleic acids, and polysaccharides are also polymers made up of C, H, and O. This similarity makes their coherent Raman signals extremely similar and difficult to distinguish. One solution is to use isotope labeling, for example, using deuterium-labeled molecules, which produce a C—D bond with a different vibration frequency from the C—H bond, thereby distinguishing the background signal[70].

2.4 Raman Spectroscopy Imaging

Currently, the concentration levels of micro- and nanoplastics in most environmental sample test reports are primarily determined by using microscopes to ascertain the total number and morphology of plastic particles in the samples. However, some studies have pointed out that this method is prone to false positives and false negatives[71]. An effective approach would be to incorporate chemical composition detectors, such as Raman spectroscopy, in tandem with microscopy. Nevertheless, these methods rely on manual pre-selection, which significantly wastes human and time resources. Especially, nanoplastics beyond the diffraction limit of visible light are often overlooked, leading to a considerable discrepancy between statistical results and actual concentrations[72]. Park et al.[52] used dark-field microscopy, which can clearly distinguish particles from the substrate. A more feasible method is the identification and visualization of micro- and nanoplastics through Raman spectral imaging.
Compared to single-point analysis, Raman spectroscopy imaging scans the surface of the sample by changing the position of the laser, collecting signals from a preset area in matrix form, and generating a hyperspectral matrix. Subsequent analysis can convert this matrix into a mapping image, visualizing the scanned area from a spectral perspective or through chemical information channels. The spatial resolution of Raman imaging typically reaches the sub-micron level, allowing for the analysis of complex details such as the shape, distribution, and behavior of micro- and nanoplastics in complex environmental samples[73]. Qian et al.[61] used SRS technology to rapidly identify standard PS particles within tens of microseconds. Although the optimal spatial resolution was 365 nm, it was still easy to identify individual PS particles as small as 100 nm based on the diffraction limit model and intensity distribution. Sobhani et al.[74] successfully mapped microparticles and nanoparticles down to 100 nm in diameter and demonstrated that during the polishing of automotive paint, trillions of 200 nm to 7 µm microparticles and nanoparticles are generated. Using conventional microscopic Raman techniques, scanning an area of 4 µm2 requires 400 seconds, and the signal-to-noise ratio is low, making it difficult to qualitatively analyze some particles. Ruan et al.[43] used SRS technology to image and analyze 1248 µm2 areas of multiple edible sea salt samples on SERS substrates, reducing the imaging time to about 2 minutes. They assumed that if an adult consumes 5 g of sea salt daily, in some regions, an adult could ingest up to 6×106 nanoplastics annually through salt. Notably, this number may be significantly underestimated because the method did not account for nanoparticles with diameters below 200 nm. The interconnection between particle morphology and chemical composition has significant implications for toxicology. Studies have shown that the toxicity induced by micro- and nanoparticles is not only related to the dose but also to the physicochemical properties of the particles and their interactions with cells and effects on uptake[75-76]. Therefore, whether in water samples or biological tissues, the visualization and analysis of individual micro- and nanoplastic particles help to understand their aggregation behavior, transport mechanisms, and potential interactions with organisms and ecosystems. Real-time data acquisition is another attractive feature of Raman imaging. This capability facilitates dynamic studies of the behavior of micro- and nanoplastics, tracking changes in their properties over time[77]. This is particularly important for observing how micro- and nanoplastics transform in response to environmental conditions, chemical processes, or biological interactions, providing deeper insights into their persistence and toxicity.
To efficiently convert hyperspectral matrix signals into images, different algorithms (as shown in Table 3) need to be employed. Advanced algorithms that integrate machine learning techniques, pattern recognition, image fusion, and signal deconvolution methods have significantly improved the accuracy and efficiency of plastic identification[78-79]. For example, Su et al.[80] used machine learning algorithms trained on large datasets containing various types of plastics and environmental conditions, thereby discerning subtle differences in spectra and enhancing their ability to distinguish between plastic and background signals. Moreover, emerging signal processing techniques are capable of extracting meaningful information from complex Raman spectra even in the presence of interfering substances[81]. These advancements in algorithms not only improve the precision of micro- and nanoplastic identification but also facilitate the automation of the analysis process, thus accelerating research on the distribution, transformation, and potential environmental impacts of micro- and nanoplastics. With continuous technological development, the synergistic effect of Raman imaging and spectral analysis algorithms provides crucial support for studying the behavior of micro- and nanoplastics in a variety of environmental and biological contexts.

3 Raman Spectroscopy Is Used For The Identification Of Micro-Nanoplastics In Environmental Samples

The sources of micro- and nanoplastics in the environment are diverse, which can be categorized into primary and secondary sources. Primary sources refer to micro- and nanoplastics that are directly produced and eventually released into the environment, while secondary sources are formed by the fragmentation of larger plastic pieces under physical, chemical, and biological effects[87-88]. Additives inherent in primary and secondary plastics or contamination in environmental media can lead to fluorescence and interference peaks in Raman spectra, thus increasing the difficulty of qualitative identification of micro- and nanoplastics using Raman spectroscopy. This section mainly introduces several techniques for reducing fluorescence interference and methods to enhance the recognition of micro- and nanoplastics in complex samples through machine learning (Figure 2).

3.1 Interference And Elimination Of Fluorescence On Raman Spectroscopy Signals Of Micro-And Nanoplastics

In the Raman spectroscopy analysis of micro- and nanoplastics, fluorescence signals originate from a wide range of sources, including the inherent fluorescence of plastics, colorants, stabilizers, degradation products, and various impurities in the environment. The presence of fluorescence leads to an elevated baseline and can even completely obscure the Raman signal. There are many methods to suppress fluorescence signals, such as pre-cleaning the materials to remove fluorescent contaminants, photobleaching, and changing the excitation wavelength, which are alternative approaches.
Currently, a common method to reduce the fluorescence effect caused by organic matter is to use a cleaning step to remove contaminants[89-90]. However, chemical cleaning can damage micro- and nanoplastics, making it difficult to accurately identify plastics. Typically, collected environmental samples are pretreated with acids, bases, oxidants, and enzymes to reduce the interference of surface adsorbed contaminants, facilitating accurate analysis. Acids (such as sulfuric acid, nitric acid, or hydrochloric acid) and bases (such as sodium hydroxide and potassium hydroxide) can rapidly digest organic matter in the samples, with an organic digestion efficiency of over 90%[91]. However, strong acids and bases, in addition to their function of digesting organic matter, can also cause oxidation on the surface of plastics[92-93]. The oxidation process can also lead to discoloration and slight degradation of the plastics[94]. Since the Raman spectra of degraded plastics do not match those of standard samples, this digestion or oxidation-induced plastic degradation may affect the accuracy of Raman spectroscopy identification of microplastics. Enzymatic digestion is an alternative to chemical digestion and can minimize plastic degradation. However, the efficiency of enzymatic digestion is relatively low and is very sensitive to the matrix and digestion conditions[95]. Therefore, appropriate digestion methods should be selected and optimized by combining several digestion processes.
Despite a thorough dissolution process, some samples may still exhibit fluorescence from impurities, as colorants in plastics are often difficult to remove. Two commonly used methods are switching the excitation light to different wavelengths and photobleaching the sample[96-97]. As shown in Table 1 and Table 2, the most widely used excitation wavelength is the 785 nm near-infrared laser. This laser wavelength achieves a high level of balance in terms of signal intensity, fluorescence suppression, cost, and overall performance. Compared to expensive laser sources, photobleaching is more widely applied, which involves placing the sample under a laser to continuously degrade the fluorescent substances. However, apart from increasing the detection time, this method can also lead to the photodegradation or thermal degradation of the micro- and nanoplastics themselves, making photobleaching not always effective[96-97]. A more aggressive approach is the use of nonlinear Raman techniques[22], which can provide a high signal-to-noise ratio without interference from fluorescence but require much more expensive equipment.
Ghosal et al.[98] proposed a rapid detection scheme that can accurately identify fluorescent samples. They used an automatic polynomial fitting algorithm to eliminate the broad background, thereby removing the fluorescence background and enhancing the polymer spectra. When analyzing plastic samples covered by biofilms, the directly obtained spectra were saturated with fluorescence signals, almost masking the characteristic peaks of the plastics. After processing the spectra with the automatic algorithm, the characteristic peaks of the polymers became clearly visible, allowing for the identification of the chemical composition of the plastics through Raman database software. Xie et al.[99] used a random forest algorithm to overcome the interference of fluorescence generated by complex environmental media in the identification of plastics, achieving an identification accuracy of over 97% in spiked tap water. Validation experiments showed successful detection of PS and PVC nanoplastics in rainwater.
Another approach to eliminate fluorescence interference is the use of coherent Raman techniques. Since the signal originates solely from molecular vibrations, this technique can effectively avoid fluorescence interference and obtain Raman spectra with a high signal-to-noise ratio. Rhee et al.[46] used the CARS method to identify PE, PS, PMMA, and PA-12 micro- and nanoplastics in undigested soil samples, significantly reducing the preprocessing steps. C—H bond-specific CARS imaging and spectral analysis facilitate rapid searching for microplastic particles, enabling quick chemical identification even when interfered by residual particles and highly fluorescent substances in the soil. Zada et al.[100] efficiently measured 1 cm2 of the filter membrane surface within less than 5 hours after density separation of sediment samples from the Rhine River estuary, successfully identifying 88 types of micro- and nanoplastics, including nylon, PET, PS, PP, and PE. Wang et al.[67] successfully detected PE, PP, PVC, PET, PS, and PMMA particles in protozoan cells using SRS technology, effectively reducing the impact of other fluorescent substances in water samples and cells on the Raman signals of plastics.

3.2 Identification Of Micro-Nanoplastics Based On Machine Learning Using A Raman Spectroscopy Database

An automated Raman spectroscopy program uses spectral library matching software to compare sample spectra with customized or commercial spectral libraries and automatically identify the specific chemical composition of microplastics. The probability of successful matching largely depends on the comprehensiveness of the spectral library. However, customized spectral libraries are typically based on spectra obtained from original polymer pellets, which may differ from the spectra of micro- and nanoplastics collected from the environment. Micro- and nanoplastics in the environment are mainly composed of fragments of commercial plastics, with morphologies including foams, sheets, and fibers, and often contain various additives, fillers, and colorants. In some cases, these components may obscure the polymer spectra, so using a more comprehensive commercial spectral library would undoubtedly improve the accuracy of the match[71].
Under the influence of various environmental factors (such as ultraviolet light, heat, and biodegradation), plastics undergo aging[26,71], for example, polyvinyl chloride (PVC) is prone to photodegradation in aqueous media. Under high humidity conditions, additives (such as light stabilizers) may accelerate leakage. After UV exposure, the spectral fingerprint of PVC changes significantly, with a simultaneous weakening of the double peaks at 693 cm-1 and 637 cm-1 representing C—Cl bonds[71]. In the spectra after strong UV irradiation, these double peaks are completely absent, and two strong peaks appear at 1139 cm-1 and 1540 cm-1 due to the vibrations of carbon-carbon double bonds (C=C). Since the spectral library only contains spectra of undegraded PVC, it is impossible to successfully match the spectra through a simple search in the spectral library. Considering this issue, it is crucial to include the spectra of polymers at different stages of aging and degradation in the spectral library, which can increase the chances of the matching software correctly identifying the polymer components. It is worth noting that compared to FTIR technology, Raman spectroscopy can better identify photo-oxidized micro- and nanoplastics by excluding fluorescence interference. For the most common micro- and nanoplastics (such as PE, PP), their photo-oxidation leads to the formation of oxygen-containing groups, mainly C=O and —OH, which exhibit strong intensity in infrared spectra[26,101⇓-103], in stark contrast to their weak Raman signals. For instance, Cai et al.[101] found significant differences between the infrared spectra of pristine and degraded PE, PP, and PS, while the corresponding Raman spectral peak intensities showed only minor differences, facilitating automatic matching of standard Raman spectra by software. Therefore, when it comes to simply identifying the polymer types of aged plastics, Raman spectroscopy has a significant advantage.
In addition, it has been suggested by scholars that incorporating the spectra of some non-plastic materials, which are often confused with micro- and nanoplastics, into the spectral library is also valuable. These non-plastic materials include cellulose, keratin, inorganic particles, and most importantly, synthetic fibers (i.e., viscose), which are commonly found in micro- and nanoplastic samples[71,104]. The main limitation to the widespread adoption of this approach is that each laboratory needs to establish a comprehensive spectral library, and building such a library is a time-consuming task. The field of micro- and nanoplastics is highly interdisciplinary, with researchers having diverse academic backgrounds and experiences. Therefore, establishing an open-source, well-designed spectral database would be extremely beneficial. This would avoid the duplication of efforts in building spectral libraries and minimize the effort and material resources required by each researcher, while achieving a level of complexity and comprehensiveness not possible with individual commercial spectral libraries. Such a spectral database could also include spectra from real environmental samples, allowing for confirmation of their composition through other identification techniques when in doubt, thereby greatly improving the accuracy of matches. Furthermore, a free spectral library would encourage research groups with limited funding to conduct more studies on microplastic identification, thus increasing awareness of global microplastic pollution.
After having a complete Raman spectral library, the challenge is to quickly compare the sample Raman spectra with the spectral library. Machine learning has received widespread attention due to its ability to automatically analyze complex spectral data. The combination of machine learning and Raman spectroscopy can improve the efficiency and accuracy of micro- and nanoplastic identification[105⇓-107]. Before machine learning, it is necessary to properly process the spectra. The first step is the smoothing of the spectral curves. A commonly used smoothing method is Savitzky-Golay smoothing, which works by dynamically fitting polynomials to a continuous window of data points (local least-squares polynomial approximation) to follow the shape of the spectrum, thereby reducing the impact of random noise signals. However, in some cases, this smoothing method may have a negative effect on spectral features. When applied to highly noisy spectra that require a high degree of smoothing, it may severely affect sharp local features, leading to errors in subsequent machine learning. Barton et al.[108] used maximum likelihood estimation to enhance Savitzky-Golay smoothing, effectively preventing significant deviations from the true Raman signal while retaining the powerful smoothing properties of the Savitzky-Golay method. This approach increased the signal-to-noise ratio by at least 50%, providing good Raman spectra for subsequent machine learning.
Machine learning can dynamically and rapidly process complex Raman spectral data, using algorithms to automatically extract features from the spectra and train classification models to explore potential connections between multiple points of interest for researchers, ultimately achieving accurate classification of micro- and nanoplastic types[109]. Xie et al.[99] used a random forest model to identify five common types of micro- and nanoplastics (PE, PTFE, PS, PMMA, and PVC), with average accuracy and sensitivity of 98.8% and 98.5%, respectively. In rainwater testing, they identified nanoscale PS and PVC, demonstrating the potential of their model for analyzing real environmental samples. Lim et al.[110] utilized a convolutional neural network to achieve rapid identification of individual microplastics in the 1-10 µm range, with an exposure time of just 0.4 s and a confidence level of around 85.47%, significantly accelerating qualitative analysis capabilities for microplastics in different environmental scenarios. Feng et al.[106] argued that single machine learning models are not efficient in identifying micro- and nanoplastics exposed to the environment, thus proposing a multi-model algorithm with three layers of discrimination. For unknown spectra, after applying the Savitzky-Golay smoothing method, principal component analysis-linear discriminant analysis (PCA-LDA) was first used to identify PP, PS, PET, and PVC plastics. Then, principal component analysis-k-nearest neighbors (PCA-KNN) was employed to identify chlorinated polyethylene (CPE). Finally, a multilayer perceptron (MLP) model was used on the Raman spectral segment of 980-1385 cm-1 to classify HDPE and LDPE, with an accuracy exceeding 97%, successfully identifying real microplastic samples from daily life.
The rapid identification of microplastics with complex compositions through the Raman barcode method is an intriguing approach[111]. This method converts the Raman spectra of various plastics into barcodes, where each line represents a Raman spectral peak. Identification is then achieved by binary comparison of the sample barcode with reference barcodes, which accelerates matching and eliminates issues related to the normalization of spectral intensity. Although this barcode method has not yet been applied to the identification of micro- and nanoplastics in actual environmental samples, its reference barcodes can include those of polymers, colorants, and common plastic additives. By ignoring spectral intensity, it can enhance the recognition rate of micro- and nanoplastic spectra that are obscured by signals from colorants or additives.

4 Raman Spectroscopy Is Used For The Quantitative Analysis Of Micro And Nanoplastics

On the basis of accurately qualitative analysis of micro- and nanoplastics in environmental samples, Raman spectroscopy can also be used for quantitative detection of micro- and nanoplastics. However, due to the limited spatial resolution of the instrument and the complexity of particle morphology, when calculating quantities, micro- and nanoplastics can be quantified using two units: mass concentration and particle number concentration. This section introduces the quantitative methods of micro- and nanoplastics in environmental samples using these two different concentration units, including the improvement of quantitative precision through the "coffee ring" effect and membrane filtration pre-concentration.

4.1 In-Situ Concentration And Mass Concentration Analysis Of Micro-Nanoplastics On Substrates

In the quantitative determination of micro- and nanoplastics, mass concentration is an important parameter for formulating pollution control strategies, and the differences in particle size and density of microplastics are significant factors affecting the accuracy of quantification. Chaisrikhwun et al.[112] proposed an interesting experiment where micro- and nanoplastics (PS particles) in water were dried and dissolved in an organic solvent to avoid the influence of particle size on the Raman signal response. The plastic-containing organic solution was then dropped onto a gold film substrate and dried, with plastic molecules concentrated at the edge of the droplet due to the "coffee ring" effect. There are three points worth noting: first, this method effectively avoids the dependence of the signal on particle size; second, based on the principle that "like dissolves like," the solvents used to dissolve the plastic molecules are generally nonpolar, so polar interfering substances will be separated during the dissolution process; third, the "coffee ring" effect not only concentrates the plastic molecules but also separates other interfering substances from them. Based on these three points, this experimental method can detect as low as 0.10 µg/L and has a quantification range between 10~40 µg/L, excluding interference from salts, sugars, detergents, and proteins.
The "coffee ring" effect refers to the phenomenon where, on a hydrophilic surface, the boundary of a droplet is fixed due to capillary action, and the liquid near the boundary that evaporates must be replenished by the internal liquid, thus forming a flow from the inside to the boundary. This flow can carry most of the solute to the boundary[113-114]. Although the "coffee ring" effect helps concentrate micro- and nanoplastics at the gas-liquid-solid interface, in most cases, these particles do not distribute completely within the ring, leading to significant differences in their Raman signal peaks. Using a hydrophobic detection substrate can effectively concentrate micro- and nanoplastics in the smaller contact area between the droplet and the hydrophobic substrate. Li et al.[115] immersed clean silicon wafers in a hexane solution of methyltrichlorosilane to create a hydrophobic substrate with a contact angle as high as 143.71°, effectively concentrating the target material uniformly and densely in a smaller area. Additionally, using Ag nanoparticles mixed with plastic particles further enhances the Raman signal. This method has a detection limit as low as 0.5 mg/L (500 nm PS) and 1 mg/L (100 nm PS), with a relative standard deviation (RSD) as low as 1.94%, and it offers the best reproducibility currently, providing a potential direction for quantitative detection of microplastics. However, the intensity of SERS depends on the distribution of "hotspots," which may be influenced by the preparation method of Ag nanoparticles, the binding mode between Ag particles and micro- and nanoplastics, and experimental operations, thus posing a significant challenge to the operator's skills.
Physical interception-based membrane filtration methods are simple to operate, have high enrichment efficiency, and can maintain the original morphology of particles while retaining particles of different sizes. They can effectively separate and enrich micro- and nano-plastics[116-117]. Due to the background signal from plastic filter membranes masking the signals of micro- and nano-plastic samples, glass fiber membranes, inorganic membranes (alumina), and mixed cellulose membranes are often used in the membrane filtration steps for micro- and nano-plastics. In addition to direct analysis of micro- and nano-plastics on the membrane using optical instruments, when membrane filtration is used as a pre-concentration method before other analytical techniques, sample transfer is also required. However, small-sized plastic particles tend to adhere to the membrane, leading to low transfer rates (only 53.1% for microplastics)[91]. Directly preparing nano-enhanced structures on the filtration membrane is an effective method to combine the filtration and detection of microplastics, improving the sensitivity of Raman spectroscopy. Qin et al.[45] synthesized spiky Au nanocrystals and uniformly deposited them onto glass fiber filtration membranes, aiming for in-situ enrichment and highly sensitive SERS detection of micro- and nano-plastics. Using this SERS filter membrane, PS microspheres ranging from 20 nm to 10 µm were successfully detected, with a detection limit of 0.1 mg/L. The lowest detection limit for 100 nm PS microspheres dispersed in tap water and rainwater was 0.1 mg/L. Yang et al.[35] proposed using one-dimensional Ag nanowires to self-assemble into a filtration membrane for pre-concentrating nano-plastics. The retention rate for 50 nm PS microspheres was as high as 86.7%, and the high SERS activity of the Ag nanostructures could enhance the Raman signals of the nano-plastics in situ, achieving a detection limit as low as 0.10 µg/L. When analyzing actual water samples from seafood markets, considering the low concentration of nano-plastics in the water samples, the amount of actual water sample filtered was increased within the maximum retention capacity of the membrane to increase the concentration of nano-plastics on the membrane. This method detected 500 nm PS plastics at the µg/L level.
Whether through the "coffee ring" effect or membrane filtration for concentration, we recommend first selecting the particle size of micro- and nanoplastics that needs to be focused on. Larger particles often have a weaker ability to form a "coffee ring"[118], and a mismatch between the particle size and the pore size of the filter membrane can severely affect the sample concentration efficiency[119], leading to lower quantitative accuracy.

4.2 Raman Spectroscopy Is Used For The Analysis Of Micro And Nanoplastic Number Concentration

In addition to the detection of micro- and nanoplastic mass concentrations, analyzing the size distribution and number concentration of micro- and nanoplastics in environmental samples is also crucial, as plastic particles of different sizes may enter organisms and interact with biological target organs in various ways. Zhang et al.[38] used electron beam evaporation technology to construct a gold triangular cavity array substrate. When particles fall into or approach the edges or vertices of the triangular cavities, significant electric field enhancement can be achieved, allowing for the detection of polystyrene (PS) standard microspheres with a concentration of 1.5 × 1011 particles/mL and a diameter of 50 nm. Using the gold triangular cavity array substrate and a nanoparticle tracking analyzer, they detected PET plastic particles with an average diameter of 88.2 nm and a concentration of 108 particles/mL in commercial bottled water. However, this was a qualitative analysis of particles using SERS technology, and the quantification of the number concentration of micro- and nanoplastics was done using a nanoparticle tracking analyzer, not a true quantitative study using a Raman spectrometer. Huber et al.[62] selected standard PS microspheres with diameters of 300 nm and 600 nm as representatives and explored the relationship between SRS signal values and number concentrations in a flow field using SRS technology. They found that this technique is only suitable for the quantitative study of plastic particles at high number concentrations, and the signal intensity depends on the particle diameter. This is because most particles cannot appear near the very small SRS focal point during the experiment, thus producing no signal, requiring further optimization of the flow cell to increase the probability of particles appearing at the focal point. Another approach is to use Raman spectral imaging within the instrument's spatial resolution to visualize individual micro- and nanoplastic particles, making it easy for researchers to count the number of particles within the scanning range. Qian et al.[61] used SRS technology to scan micro- and nanoplastics filtered onto 13 mm diameter alumina membranes, selecting five or more 0.2 mm × 0.2 mm imaging areas on each filter membrane to estimate the total number of micro- and nanoplastics on the entire membrane, ultimately finding over 105 particles per liter of bottled water, with most being plastic. Qian et al. conducted research transforming Raman spectral imaging from qualitative identification to quantitative number concentration studies, which is helpful for understanding the impact of micro- and nanoplastics on organisms from the perspective of number concentration. Raman spectral imaging also aids in analyzing the number concentration of micro- and nanoplastics of different shapes and sizes, especially nanoplastics that are easily overlooked in mass concentration tests. Although these nanoplastics make a negligible contribution to the mass concentration, they have the ability to cross biological barriers and play a dominant role in toxicity assessments[120-121]. Therefore, assessing the number concentration of nanoplastics remains profoundly significant.

5 Conclusion and Prospects

Raman spectroscopy has shown great potential and application value in the detection of micro- and nanoplastics in the environment. Although this technology has been widely used for the detection of micro- and nanoplastics, there are still some issues worth considering.
(1) Sample pretreatment remains a key challenge. Although Raman spectroscopy is insensitive to oxygen-containing polar groups resulting from acid, base, and enzymatic digestion, it can still affect the identification of micro- and nanoplastics. Different pre-concentration methods impact the quantitative measurement data for micro- and nanoplastics of different particle sizes. A unified and standardized sample pretreatment method is crucial for conducting research on the detection of micro- and nanoplastics using Raman spectroscopy. These procedures can establish benchmarks for obtaining consistent, reproducible, and comparable results.
(2) Micro- and nanoplastics, especially nanoplastics, often adsorb other substances in the environment, including their own additives, leading to significant differences between the collected Raman spectra and standard spectra. Although CARS and SRS can detect micro- and nanoplastics in real time with high efficiency, there is still a lack of a complete spectral database. In the future, a reference database should be established, containing as many types of micro- and nanoplastics as possible, aged micro- and nanoplastics, additives, biopolymers, and humic substances, to provide references for accurately identifying unknown micro- and nanoplastics in natural environments.
(3) The quantitative determination of micro- and nanoplastic particles using Raman spectroscopy remains a challenging issue to be addressed. Although there have been studies that accurately quantify small organic molecules using SERS, micro- and nanoplastics are often unevenly distributed on the detection substrate, making it difficult to obtain highly reproducible signals. Furthermore, whether to use particle concentration or mass concentration in research is still worth further discussion.
(4) Imaging analysis is a highly complex and time-consuming task. Although the combination of electron microscopy, atomic force microscopy, and Raman spectroscopy can accelerate imaging speed and improve spatial resolution, it also significantly increases the cost of detection instruments. Future work should focus on advancements in algorithms, particularly the integration with machine learning, which can simplify and enhance analytical efficiency, making it more widely applicable. Integrating machine learning into future research can overcome the challenges posed by analytical complexity. However, machine learning model methods are complex, and some have internal mechanisms that are difficult to understand. The relative importance of each variable is also hard to estimate, all of which increase the uncertainty of their application in the field of micro- and nanoplastic identification. Currently available machine learning algorithms cannot easily improve these characteristics. In the future, there will be a need to integrate multiple algorithms or develop more advanced algorithms to provide richer and more comprehensive sample information.

References

Publishing order | Descend order by publishing year | Descend order by cited within
[1]

MacLeod M, Arp H P H, Tekman M B, Jahnke A. Science, 2021, 373(6550): 61.
[2]

Ding J F, Ju P, Ran Q, Li J X, Jiang F H, Cao W, Zhang J, Sun C J. Sci. Adv., 2023, 9(27): eadf5897.
[3]

Hanvey J S, Lewis P J, Lavers J L, Crosbie N D, Pozo K, Clarke B O. Anal. Methods, 2017, 9(9): 1369.
[4]

Hernandez L M, Xu E G, Larsson H C E, Tahara R, Maisuria V B, Tufenkji N. Environ. Sci. Technol., 2019, 53(21): 12300.
[5]

Duis K, Coors A. Environ. Sci. Eur., 2016, 28(1): 2.
[6]

Horton A A, Walton A, Spurgeon D J, Lahive E, Svendsen C. Sci. Total Environ., 2017, 586: 127.
[7]

Karami A, Golieskardi A, Choo C K, Larat V, Karbalaei S, Salamatinia B. Sci. Total Environ., 2018, 612: 1380.
[8]

Schymanski D, Goldbeck C, Humpf H U, Fürst P. Water Res., 2018, 129: 154.
[9]

Chen Y L, Xu H Y, Luo Y L, Ding Y T, Huang J G, Wu H H, Han J N, Du L J, Kang A Q, Jia M Y, Xiong W P, Yang Z H. Water Res., 2023, 242: 120243.
[10]

Eriksen M, Lebreton L C M, Carson H S, Thiel M, Moore C J, Borerro J C, Galgani F, Ryan P G, Reisser J. PLoS One, 2014, 9(12): e111913.
[11]

Osman A I, Hosny M, Eltaweil A S, Omar S, Elgarahy A M, Farghali M, Yap P S, Wu Y S, Nagandran S, Batumalaie K, Gopinath S C B, John O D, Sekar M, Saikia T, Karunanithi P, Hatta M H M, Akinyede K A. Environ. Chem. Lett., 2023, 21: 1.
[12]

Barnes D K A, Galgani F, Thompson R C, Barlaz M. Phil. Trans. R. Soc. B, 2009, 364(1526): 1985.
[13]

Jahnke A, Arp H P H, Escher B I, Gewert B, Gorokhova E, Kühnel D, Ogonowski M, Potthoff A, Rummel C, Schmitt-Jansen M, Toorman E, MacLeod M. Environ. Sci. Technol. Lett., 2017, 4(3): 85.
[14]

Shim W J, Hong S H, Eo S E. Anal. Methods, 2017, 9(9): 1384.
[15]

Silva A B, Bastos A S, Justino C I L, da Costa J P, Duarte A C, Rocha-Santos T A P. Anal. Chim. Acta, 2018, 1017: 1.
[16]

Shan J, Zhao J, Liu L, Zhang Y, Wang X, Wu F. Environ. Pollut., 2018, 238:121.
[17]

Elert A M, Becker R, Duemichen E, Eisentraut P, Falkenhagen J, Sturm H, Braun U. Environ. Pollut., 2017, 231: 1256.
[18]

Ribeiro-Claro P, Nolasco M M, Araújo C. Characterization and Analysis of Microplastics. Amsterdam: Elsevier, 2017: 119.
[19]

Gago J, Galgani F, Maes T, Thompson R C. Front. Mar. Sci., 2016, 3: 219.
[20]

Araujo C F, Nolasco M M, Ribeiro A M P, Ribeiro-Claro P J A. Water Res., 2018, 142: 426.
[21]

Ivleva N P, Wiesheu A C, Niessner R. Angew. Chem. Int. Ed., 2017, 56(7): 1720.
[22]

Borman S A. Anal. Chem., 1982, 54(9): 1021A.
[23]

Käppler A, Fischer D, Oberbeckmann S, Schernewski G, Labrenz M, Eichhorn K J, Voit B. Anal. Bioanal. Chem., 2016, 408(29): 8377.
[24]

Lehner R, Weder C, Petri-Fink A, Rothen-Rutishauser B. Environ. Sci. Technol., 2019, 53(4): 1748.
[25]

Rillig M C, Kim S W, Kim T Y, Waldman W R. Environ. Sci. Technol., 2021, 55(5): 2717.
[26]

Andrady A L. Mar. Pollut. Bull., 2017, 119(1): 12.
[27]

Cózar A, Echevarría F, González-Gordillo J I, Irigoien X, Úbeda B, Hernández-León S, Palma Á T, Navarro S, García-de-Lomas J, Ruiz A, Fernández-de-Puelles M L, Duarte C M. Proc. Natl. Acad. Sci. U. S. A., 2014, 111(28): 10239.
[28]

Enders K, Lenz R, Stedmon C A, Nielsen T G. Mar. Pollut. Bull., 2015, 100(1): 70.
[29]

Erni-Cassola G, Gibson M I, Thompson R C, Christie-Oleza J A. Environ. Sci. Technol., 2017, 51(23): 13641.
[30]

Griffiths P R, Miseo E V. Infrared and Raman Spectroscopic Imaging: Second, Completely Revised and Updated Edition, 2014. 1-56.
[31]

Dieing T, Hollricher O. Vib. Spectrosc., 2008, 48(1): 22.
[32]

Caldwell J, Taladriz-Blanco P, Rothen-Rutishauser B, Petri-Fink A. Nanomaterials, 2021, 11(5): 1149.
[33]

Min W, Freudiger C W, Lu S J, Xie X S. Annu. Rev. Phys. Chem., 2011, 62: 507.
[34]

Lv L L, He L, Jiang S Q, Chen J J, Zhou C X, Qu J H, Lu Y Q, Hong P Z, Sun S L, Li C Y. Sci. Total Environ., 2020, 728: 138449.
[35]

Xu G J, Cheng H Y, Jones R, Feng Y Q, Gong K D, Li K J, Fang X Z, Ali Tahir M, Valev V K, Zhang L W. Environ. Sci. Technol., 2020, 54(24): 15594.
[36]

Chang L, Jiang S, Luo J, Zhang J F, Liu X H, Lee C Y, Zhang W. Environ. Sci.: Nano, 2022, 9(2): 542.
[37]

Li J, Liu H, Chen S Y, Liang X, Gao Y M, Zhao X F, Li Z, Zhang C, Lei F C, Yu J. J. Phys. Chem. Lett., 2022, 13(25): 5815.
[38]

Zhang J J, Peng M, Lian E K, Xia L, Asimakopoulos A G, Luo S H, Wang L. Environ. Sci. Technol., 2023, 57(22): 8365.
[39]

Zhu Z D, Han K H, Feng Y T, Li Z H, Zhang A X, Wang T, Zhang M F, Zhang W. ACS Appl. Mater. Interfaces, 2023, 15(30): 36988.
[40]

Liu J, Xu G J, Ruan X J, Li K J, Zhang L W. Front. Environ. Sci. Eng., 2022, 16(11): 143.
[41]

Lê Q T, Ly N H, Kim M K, Lim S H, Son S J, Zoh K D, Joo S W. J. Hazard. Mater., 2021, 402: 123499.
[42]

Jeon Y, Kim D, Kwon G, Lee K Y, Oh C S, Kim U J, You J. Carbohydr. Polym., 2021, 272: 118470.
[43]

Ruan X J, Ao J P, Ma M L, Jones R R, Liu J, Li K J, Ge Q Y, Xu G J, Liu Y Y, Wang T, Xie L F, Wang W, You W B, Wang L C, Valev V K, Ji M B, Zhang L W. Environ. Sci. Technol., 2024, 58(21): 9091.
[44]

Xu D W, Su W, Lu H W, Luo Y L, Yi T A, Wu J, Wu H, Yin C, Chen B Y. Phys. Chem. Chem. Phys., 2022, 24(19): 12036.
[45]

Qin Y Z, Qiu J X, Tang N, Wu Y Z, Yao W X, He Y S. Environ. Res., 2023, 228: 115926.
[46]

Yang Q, Zhang S Y, Su J, Li S, Lv X C, Chen J, Lai Y C, Zhan J H. Environ. Sci. Technol., 2022, 56(15): 10818.
[47]

Ruan X J, Xie L F, Liu J, Ge Q Y, Liu Y Y, Li K J, You W B, Huang T T, Zhang L W. J. Hazard. Mater., 2024, 462: 132702.
[48]

Hu R, Zhang K N, Wang W, Wei L, Lai Y C. J. Hazard. Mater., 2022, 429: 128388.
[49]

Kihara S, Chan A, In E, Taleb N, Tollemache C, Yick S, McGillivray D J. RSC Adv., 2022, 12(32): 20519.
[50]

Lin P Y, Wu I H, Tsai C Y, Kirankumar R, Hsieh S. Anal. Chim. Acta, 2022, 1198: 339516.
[51]

Lee C H, Fang J K. J. Environ. Sci., 2022, 121: 58.
[52]

Park C, Lim D, Kong S M, Won N I, Na Y H, Shin D. Water Res., 2023, 244: 120459.
[53]

Kleinman S L, Frontiera R R, Henry A I, Dieringer J A, Van Duyne R P. Phys. Chem. Chem. Phys., 2013, 15(1): 21.
[54]

Dong X, Gu H M, Kang J, Yuan X J, Wu J W. Colloids Surf. A Physicochem. Eng. Aspects, 2010, 368(1-3): 142.
[55]

Bell S E J, Sirimuthu N M S. J. Phys. Chem. A, 2005, 109(33): 7405.
[56]

Nickel U, Mansyreff K, Schneider S. J. Raman Spectrosc., 2004, 35(2): 101.
[57]

Xu L J, Lei Z C, Li J X, Zong C, Yang C J, Ren B. J. Am. Chem. Soc., 2015, 137(15): 5149.
[58]

Xu L J, Zong C, Zheng X S, Hu P, Feng J M, Ren B. Anal. Chem., 2014, 86(4): 2238.
[59]

Rhee H, Jeong S, Lee H, Cho M G, Choi D S. Environ. Pollut., 2024, 342: 123080.
[60]

Laptenok S P, Martin C, Genchi L C, Duarte C M, Liberale C. Environ. Pollut., 2020, 267: 115640.
[61]

Qian N X, Gao X, Lang X Q, Deng H P, Bratu T M, Chen Q X, Stapleton P, Yan B Z, Min W. Proc. Natl. Acad. Sci. U. S. A., 2024, 121(3): e2300582121.
[62]

Huber M J, Zada L, Ivleva N P, Ariese F. Anal. Chem., 2024, 96(22): 8949.
[63]

Watts A J R, Urbina M A, Goodhead R, Moger J, Lewis C, Galloway T S. Environ. Sci. Technol., 2016, 50(10): 5364.
[64]

Watts A J R, Lewis C, Goodhead R M, Beckett S J, Moger J, Tyler C R, Galloway T S. Environ. Sci. Technol., 2014, 48(15): 8823.
[65]

Cole M, Lindeque P, Fileman E, Halsband C, Goodhead R, Moger J, Galloway T S. Environ. Sci. Technol., 2013, 47(12): 6646.
[66]

Choi D S, Lim S, Park J S, Kim C H, Rhee H, Cho M. Environ. Sci. Technol., 2022, 56(5): 3045.
[67]

Wang M, Huang Z L, Wu C, Yan S, Fang H T, Pan W, Tan Q G, Pan K, Ji R, Yang L Y, Pan B C, Wang P, Miao A J. Environ. Sci. Technol., 2024, 58(6): 2922.
[68]

Fueser H, Pilger C, Kong C H, Huser T, Traunspurger W. Environ. Pollut., 2022, 294: 118662.
[69]

Xin L, Huang M Z, Huang Z W. Environ. Int., 2024, 187: 108679.
[70]

Shi L Y, Zheng C G, Shen Y H, Chen Z X, Silveira E S, Zhang L Y, Wei M, Liu C, de Sena-Tomas C, Targoff K, Min W. Nat. Commun., 2018, 9: 2995.
[71]

Lenz R, Enders K, Stedmon C A, MacKenzie D M A, Nielsen T G. Mar. Pollut. Bull., 2015, 100(1): 82.
[72]

Fang C, Sobhani Z, Zhang X, Gibson C T, Tang Y H, Naidu R. Water Res., 2020, 183: 116046.
[73]

Jakubowicz I, Enebro J, Yarahmadi N. Polym. Test., 2021, 93: 106953.
[74]

Sobhani Z, Zhang X, Gibson C, Naidu R, Megharaj M, Fang C. Water Res., 2020, 174: 115658.
[75]

Krug H F, Wick P. Angew. Chem. Int. Ed., 2011, 50(6): 1260.
[76]

Domingues C, Santos A, Alvarez-Lorenzo C, Concheiro A, Jarak I, Veiga F, Barbosa I, Dourado M, Figueiras A. ACS Nano, 2022, 16(7): 9994.
[77]

Yu E S, Jeong E T, Lee S, Kim I S, Chung S, Han S, Choi I, Ryu Y S. ACS Nano, 2023, 17(3): 2114.
[78]

Fang C, Luo Y L, Naidu R. Trac Trends Anal. Chem., 2023, 166: 117158.
[79]

Fang C, Sobhani Z, Zhang X, McCourt L, Routley B, Gibson C T, Naidu R. Water Res., 2021, 194: 116913.
[80]

Su J M, Zhang F P, Yu C X, Zhang Y S, Wang J C, Wang C Q, Wang H, Jiang H R. J. Environ. Manag., 2023, 344: 118756.
[81]

Luo Y L, Zhang X, Zhang Z X, Naidu R, Fang C. Anal. Chem., 2022, 94(7): 3150.
[82]

Anger P M, Prechtl L, Elsner M, Niessner R, Ivleva N P. Anal. Methods, 2019, 11(27): 3483.
[83]

Luo Y L, Chuah C, Al Amin M, Khoshyan A, Gibson C T, Tang Y H, Naidu R, Fang C. J. Hazard. Mater., 2022, 431: 128636.
[84]

Fang C, Luo Y L, Zhang X, Zhang H P, Nolan A, Naidu R. Chemosphere, 2022, 286: 131736.
[85]

Jin N F, Song Y Z, Ma R, Li J Y, Li G H, Zhang D Y. Anal. Chim. Acta, 2022, 1197: 339519.
[86]

Tian M P, Morais C L M, Shen H Q, Pang W Y, Xu L, Huang Q Y, Martin F L. J. Hazard. Mater., 2022, 422: 126892.
[87]

Peng L C, Fu D D, Qi H Y, Lan C Q, Yu H M, Ge C J. Sci. Total Environ., 2020, 698: 134254.
[88]

Auta H S, Emenike C U, Fauziah S H. Environ. Int., 2017, 102: 165.
[89]

Bläsing M, Amelung W. Sci. Total Environ., 2018, 612: 422.
[90]

Zhang S L, Yang X M, Gertsen H, Peters P, Salánki T, Geissen V. Sci. Total Environ., 2018, 616-617: 1056.
[91]

Li Q C, Lai Y J, Yu S J, Li P, Zhou X X, Dong L J, Liu X, Yao Z W, Liu J F. Anal. Chem., 2021, 93(10): 4559.
[92]

Enders K, Lenz R, Beer S, Stedmon C A. ICES J. Mar. Sci., 2017, 74(1): 326.
[93]

Hurley R R, Lusher A L, Olsen M, Nizzetto L. Environ. Sci. Technol., 2018, 52(13): 7409.
[94]

Nuelle M T, Dekiff J H, Remy D, Fries E. Environ. Pollut., 2014, 184: 161.
[95]

Kiran B R, Kopperi H, Venkata Mohan S. Rev. Environ. Sci. Bio/Technol., 2022, 21(1): 169.
[96]

Sullivan K D, Gugliada V. Mar. Pollut. Bull., 2018, 133: 622.
[97]

Halenkovič T, Kotrla M, Gutwirth J, Nazabal V, Němec P. Photon. Res., 2022, 10(9): 2261.
[98]

Ghosal S, Chen M, Wagner J, Wang Z M, Wall S. Environ. Pollut., 2018, 233: 1113.
[99]

Xie L F, Luo S H, Liu Y Y, Ruan X J, Gong K D, Ge Q Y, Li K J, Valev V K, Liu G K, Zhang L W. Environ. Sci. Technol., 2023, 57(46): 18203.
[100]

Zada L, Leslie H A, Vethaak A D, Tinnevelt G H, Jansen J J, de Boer J F, Ariese F. J. Raman Spectrosc., 2018, 49(7): 1136.
[101]

Cai L Q, Wang J D, Peng J P, Wu Z Q, Tan X L. Sci. Total Environ., 2018, 628-629: 740.
[102]

Da Costa J P, Nunes A R, Santos P S M, Girão A V, Duarte A C, Rocha-Santos T. J. Environ. Sci. Health Part A, 2018, 53(9): 866.
[103]

Rodrigues M O, Abrantes N, Gonçalves F J M, Nogueira H, Marques J C, Gonçalves A M M. Sci. Total Environ., 2018, 633: 1549.
[104]

Munno K, De Frond H, O’Donnell B, Rochman C M. Anal. Chem., 2020, 92(3): 2443.
[105]

Gong R Y, Wang J, Wang X, Liu Y, Shan J J. Food Packag. Shelf Life, 2023, 39: 101152.
[106]

Feng Z K, Zheng L N, Liu J. Chemosphere, 2023, 325: 138312.
[107]

Lei B, Bissonnette J R, Hogan Ú E, Bec A E, Feng X Y, Smith R D L. Anal. Chem., 2022, 94(49): 17011.
[108]

Barton S J, Ward T E, Hennelly B M. Anal. Methods, 2018, 10(30): 3759.
[109]

Weisser J, Pohl T, Heinzinger M, Ivleva N P, Hofmann T, Glas K. Trac Trends Anal. Chem., 2022, 148: 116535.
[110]

Lim J, Shin G, Shin D. Anal. Chem., 2024, 96(17): 6819.
[111]

Lawson L S, Rodriguez J D. Anal. Chem., 2016, 88(9): 4706.
[112]

Chaisrikhwun B, Ekgasit S, Pienpinijtham P. J. Hazard. Mater., 2023, 442: 130046.
[113]

Deegan R D, Bakajin O, Dupont T F, Huber G, Nagel S R, Witten T A. Nature, 1997, 389(6653): 827.
[114]

Hu H, Larson R G. J. Phys. Chem. B, 2002, 106(6): 1334.
[115]

Li D D, Tian X Y, Yang W Q, Wang X, Liu Y, Shan J J. Chemosphere, 2023, 339: 139775.
[116]

Schwaferts C, Niessner R, Elsner M, Ivleva N P. Trac Trends Anal. Chem., 2019, 112: 52.
[117]

Li P, Li Q C, Hao Z N, Yu S J, Liu J F. J. Environ. Sci., 2020, 94: 88.
[118]

Jung J Y, Kim Y W, Yoo J Y, Koo J, Kang Y T. Anal. Chem., 2010, 82(3): 784.
[119]

Covernton G A, Pearce C M, Gurney-Smith H J, Chastain S G, Ross P S, Dower J F, Dudas S E. Sci. Total Environ., 2019, 667: 124.
[120]

Fournier S B, D’Errico J N, Adler D S, Kollontzi S, Goedken M J, Fabris L, Yurkow E J, Stapleton P A. Part. Fibre Toxicol., 2020, 17(1): 55.
[121]

Banerjee A, Shelver W L. Sci. Total Environ., 2021, 755: 142518.

Funding

National Natural Science Foundation of China(22241601)

PDF (1618KB)

893

Accesses

0

Citation

Detail
Sections
Recommended

/

〈 〉

Address of Editorial Office: No. 33 Beisihuan Xilu, Zhongguancun , Beijing 100190, China
Tel: +86 10 8262 7757
Fax: +86 10 8262 7757
E-mail: progchem@mail.las.ac.cn
http://www.progchem.ac.cn
    • 　
    Copyright © Editorial Office of Progress in Chemistry