Method and system for measuring three-dimensional sphericity and components of micro-plastic particles

Abstract

The invention discloses a method and a system for measuring three-dimensional sphericity and components of micro-plastic particles. The method comprises the following steps: scanning a filter membrane by utilizing microscopic Raman or confocal Raman, screening and positioning suspected micro-plastic particles based on Raman fingerprint peaks, and determining polymer components; then, on the same filter membrane or slide glass, transferring the sample to a scanning electron microscope equipped with a variable inclination angle sample table, performing high-resolution imaging on the same particle at a plurality of inclination angles to obtain a single particle three-dimensional shape model, and calculating parameters such as volume, surface area and sphericity; one-to-one correspondence of Raman data and SEM data is realized through registration of grid coordinates and feature points of the filter membrane, and a single-particle database of spatial position-polymer type-three-dimensional morphology-sphericity is established. According to the method, a pretreatment process of enzyme digestion, mild oxidation and density separation is designed, micro-plastic particles are enriched on a filter membrane or a slide glass which is provided with a two-dimensional coordinate mark, can be used for Raman detection and is provided with a conductive coating on the surface, and the detection precision is improved.

Description

Technical Field

[0001] This invention pertains to microplastic detection and material characterization techniques, specifically involving a comprehensive measurement method and system for the three-dimensional sphericity and polymer composition of microplastic particles in human blood or saliva, combining scanning electron microscopy and micro Raman imaging. Background Technology

[0002] In recent years, numerous studies have shown that microplastics have entered the human body from the external environment, and have been detected in samples such as feces, placenta, lung tissue, and blood. Microplastics not only pose exposure risks in terms of quantity and quality, but their particle size, morphology, and sphericity are also considered to be closely related to cellular uptake, inflammatory responses, and toxicological effects: sharp or flaky particles are more likely to mechanically damage cell membranes, fibrous particles may trigger foreign body reactions, and the kinetic behavior of near-spherical particles in blood circulation is highly correlated with sphericity.

[0003] Existing microplastic detection techniques mainly include μ-FTIR, micro Raman spectroscopy, optical microscopy, SEM, and pyrolysis-GC / MS. While each method has its advantages, they all have significant limitations in terms of three-dimensional morphology and quantitative sphericity. (1) μ-FTIR and micro Raman spectroscopy can provide polymer composition information, but they are mostly two-dimensional projection images; (2) SEM has advantages in morphology and surface structure resolution, but it usually only provides two-dimensional images from a single angle and lacks composition information; (3) Traditional three-dimensional reconstruction technology is mostly used for large particles such as metal powder and medical implants, and is less used for microplastic particles with a size of 1-20μm; (4) In the context of biological fluids such as blood and saliva, improper pretreatment and slide selection can easily introduce background interference and positional errors, affecting subsequent related analyses.

[0004] Furthermore, although Raman and SEM analyses can be performed on the same sample separately, there is still a lack of standardized and reliable technical methods to ensure that the two instruments observe the same particles and to accurately reconstruct the three-dimensional morphology and calculate the sphericity.

[0005] In summary, existing technologies cannot adequately meet the following requirements: (1) In complex matrices such as blood / saliva, achieve single-particle-level component identification and joint measurement of three-dimensional morphology sphericity of microplastics; (2) Provide a set of repeatable and quantifiable SEM-Raman correlation analysis procedures and data processing methods. Summary of the Invention

[0006] Purpose of the invention: To solve the above-mentioned technical problems, the present invention provides a method for measuring the three-dimensional sphericity and composition of microplastic particles in human blood or saliva based on correlation SEM-Raman imaging.

[0007] This method targets complex biofluids such as blood and saliva, and designs a pretreatment process of enzymatic digestion-mild oxidation-density separation to enrich microplastic particles on filter membranes or slides with two-dimensional coordinate markers, Raman spectroscopy capability, and conductive coatings on their surfaces. This invention solves the technical problem in the prior art of simultaneously obtaining accurate chemical composition and three-dimensional morphology sphericity information in the detection of microplastics in biofluids, and provides a new quantitative means for studying the relationship between microplastic morphology and health effects.

[0008] Technical solution: The method for measuring the three-dimensional sphericity and composition of microplastic particles provided in the first aspect of this invention includes the following steps: (1) Sample pretreatment and microplastic enrichment Human blood or saliva samples are subjected to enzymatic digestion, mild oxidation, and density separation to remove proteins, cells, and organic residues, and to enrich microplastic particles. The resulting microplastic particles are then fixed onto a filter membrane or slide with two-dimensional coordinate markers and a conductive coating on its surface by vacuum filtration or drop coating to form the sample to be tested. (2) Raman imaging and component identification The filter membrane or slide was scanned at low magnification using a micro Raman or confocal Raman microscope. Suspected microplastic particles were screened based on the characteristic peaks of plastics in the Raman spectrum. The high signal-to-noise ratio Raman spectrum was obtained and its polymer composition type was determined by matching it with a standard polymer spectral library. (3) SEM multi-angle imaging Without changing the spatial distribution of particles on the filter membrane or slide, the particles are transferred to a scanning electron microscope. Using the tilting function of the sample stage, the target particles identified in step (2) are imaged at multiple different tilt angles to obtain a sequence of two-dimensional projection images of each particle at different observation angles. (4) Three-dimensional reconstruction and sphericity calculation Based on the multi-angle two-dimensional projection images obtained in step (3), a three-dimensional morphology model of a single particle is constructed using a multi-view three-dimensional reconstruction algorithm, and its volume is calculated. and surface area And according to the sphericity formula (3.1) Determine the sphericity of the particles ; (5) Raman-SEM related registration and data output: By using coordinate markers and feature point matching algorithms on the filter membrane or slide, spatial registration between the Raman imaging coordinate system and the SEM imaging coordinate system is achieved. The polymer composition information identified by Raman is correlated with the three-dimensional morphology and sphericity information reconstructed by SEM at the single-particle level, so as to obtain the three-dimensional sphericity distribution and statistical results of microplastic particles of different polymer types in the sample.

[0009] Furthermore, in the sample pretreatment step, enzyme digestion is performed using one or more of proteinase K, trypsin, and collagenase, incubated at 40-60°C for 2-24 hours, and mild oxidation is performed using hydrogen peroxide or persulfate, with the oxidation temperature not exceeding 60°C.

[0010] Furthermore, the density separation utilizes a high-density salt solution formed from one or more of zinc chloride, sodium iodide, and potassium iodide, with a density of 1.3-1.7 g / cm³. 3 The preferred value is 1.6 g / cm³. 3 .

[0011] Furthermore, the filter membrane or carrier material is a substrate with low autofluorescence and weak Raman background; the substrate surface is formed with a gold, platinum or carbon conductive coating with a thickness of 5-20 nm by sputtering or evaporation, and two-dimensional grid coordinates are etched or printed.

[0012] Furthermore, the Raman imaging employs a confocal microRaman system with an excitation wavelength of 488 nm, 532 nm, or 785 nm, achieving a spectral resolution better than 5 cm⁻¹. -1 The lateral spatial resolution is better than 1μm.

[0013] Furthermore, the scanning electron microscope is equipped with an adjustable tilt stage with a tilt angle range of -30° to +70° and a tilt step of 5°-15°, achieving a speed superior to [previous value] at each tilt angle. Acquire particle images at spatial resolution.

[0014] Furthermore, the multi-view 3D reconstruction algorithm is one of the following: contour-based shape reconstruction from silhouette, voxel drawing, photometric stereo, or deep learning-based 3D shape prediction model.

[0015] Furthermore, in the spatial registration step, by identifying the pre-made grid intersections, micro-engraved fiducial marks, or several feature particles on the filter membrane or slide as common reference points, an affine or perspective transformation is used to fit and establish the transformation matrix between the Raman coordinate system and the SEM coordinate system, thereby achieving the correspondence of the single particle position.

[0016] Furthermore, the sample is one of anticoagulated whole blood, plasma, serum, resting saliva, or irritated saliva, and the microplastic particles have a particle size range of 1-50 μm.

[0017] A measurement system provided in a second aspect of the present invention includes: The sample pretreatment module is used to perform enzymatic digestion, mild oxidation, density separation, and enrichment of microplastic particles in blood or saliva samples. Conductive filter membranes or carriers with coordinate markings are used to immobilize enriched microplastic particles. Raman imaging module, used to scan particles on filter membranes or slides using Raman and identify polymer components; The scanning electron microscope module is equipped with a variable tilt sample stage for multi-angle high-resolution imaging of particles on filter membranes or slides. The data processing module is used for spatial registration of Raman and SEM images, 3D morphology reconstruction and sphericity calculation, and outputs the 3D sphericity distribution of microplastic particles classified by polymer type.

[0018] Compared with the prior art, the present invention has the following beneficial effects: (1) Joint characterization of single-particle composition and three-dimensional morphology: Raman and SEM tests were performed sequentially on the same slide, and precise spatial registration was carried out to achieve joint characterization of the same particle polymer composition and three-dimensional morphology.

[0019] (2) Quantitative calculation of sphericity: The three-dimensional morphology of particles is reconstructed by multi-angle SEM images, and the volume and surface area are calculated to realize a quantitative description of sphericity, which is different from the traditional method that only has two-dimensional roundness index.

[0020] (3) Suitable for complex biological fluids: The specially designed enzyme digestion-mild oxidation-density separation process and the selection of low background filter membranes ensure that relatively clean and imageable microplastic particles are obtained in complex samples such as blood and saliva.

[0021] (4) The method has good reproducibility and scalability: the use of slides with grid coordinates and algorithmic registration-reconstruction process improves the reproducibility of experiments; the method can also be extended to the three-dimensional characterization of microplastics for other biological fluids or environmental samples. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the overall process of the present invention for measuring the three-dimensional sphericity and composition of microplastic particles in human blood or saliva based on relevant scanning electron microscopy and Raman imaging.

[0023] In the figure: 1 represents a blood / saliva sample; 2 represents the sample pretreatment module (including enzymatic digestion and gentle oxidation); 3 represents the density separation and microplastic enrichment module; 4 represents a filter membrane or slide with two-dimensional coordinate markers; 5 represents the Raman microscopy module; 6 represents the scanning electron microscope (SEM) multi-angle imaging module; and 7 represents the data processing, three-dimensional reconstruction, and sphericity calculation module. After being processed by the sample pretreatment module 2, the blood or saliva sample 1 is enriched with microplastic particles by the density separation module 3 and fixed onto the coordinate-marked filter membrane or slide 4. Subsequently, measurements are taken sequentially on the Raman microscopy module 5 and the SEM imaging module 6. Finally, the data processing module 7 completes the three-dimensional morphology reconstruction and sphericity calculation.

[0024] Figure 2 This is a schematic diagram of a conductive filter membrane or carrier structure with two-dimensional grid coordinate markings in this invention.

[0025] In the figure: 11 represents the overall structure of the filter membrane or slide; 12 represents the grid coordinate lines on the surface of the filter membrane or slide; 13 represents the marking area used to indicate coordinate scales or numbers; 14 represents the filtration and effective detection area; 15 represents the microplastic particles fixed on the surface of the filter membrane or slide; and 16 represents the outer frame or support ring structure of the filter membrane or slide. Through the grid coordinate lines 12 and the coordinate marking area 13, position tracking and spatial registration of the same particle can be achieved during Raman and SEM imaging.

[0026] Figure 3 This is a schematic diagram illustrating Raman imaging and localization of suspected microplastic particles on a filter membrane or slide according to the present invention.

[0027] In the figure: 21 is a filter membrane or slide containing microplastic particles; 22 is the selected Raman scanning area; 23 is the suspected microplastic particles identified within the scanning area 22; 24 is the Raman microscope objective; 25 is the laser incident beam guided from the objective 24 to the sample surface; and 26 is the Raman spectrometer or detector. The Raman microscopy imaging module 5 optically couples with the filter membrane 21 via the objective 24 to acquire spectra in the scanning area 22, identifies suspected microplastic particles 23 based on spectral characteristics, and records their positions in the filter membrane coordinate system.

[0028] Figure 4 This is a schematic diagram illustrating the multi-angle imaging of the same microplastic particle using a scanning electron microscope according to the present invention.

[0029] In the figure: 31 represents the SEM sample stage and the filter membrane or slide on it; 32 represents the position marker of the target microplastic particle on the sample stage 31; 33 represents the SEM image of the particle obtained at an inclination angle of 0°; 34 represents the SEM image of the particle obtained at a first inclination angle (e.g., +30°); 35 represents the SEM image of the particle obtained at a second inclination angle (e.g., -30° or +60°); and 36 represents arrows or angle markers indicating the tilt direction and angle of the sample stage. By changing the inclination angle of the sample stage and acquiring images 33, 34, and 35, two-dimensional projections of the target particle under multiple observation directions can be obtained, providing input data for subsequent three-dimensional morphology reconstruction.

[0030] Figure 5 This is a schematic diagram of 3D topography reconstruction and sphericity calculation based on multi-view SEM images.

[0031] In the figure: 41 represents the input set of multi-angle SEM images (e.g., Figure 4 (See images 33, 34, 35, etc.). 42 is the 3D reconstruction processing module, 43 is the reconstructed 3D microplastic particle model, 44 is the geometric parameter calculation module (used to calculate volume, surface area, etc.), and 45 is the sphericity calculation result output unit. The multi-angle image set 41 is processed by the reconstruction module 42 to obtain the 3D model 43. The geometric parameter calculation module 44 calculates the particle volume and surface area based on this model, and outputs the sphericity result in the sphericity calculation unit 45 according to a predetermined formula. Detailed Implementation

[0032] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be further described below.

[0033] Example 1: Measurement of three-dimensional sphericity and composition of microplastics in EDTA-anticoagulated whole blood (1) Sample pretreatment and microplastic enrichment a. Sample collection 10 mL of EDTA-anticoagulated whole blood was collected from healthy volunteers and placed in pre-cleaned polypropylene centrifuge tubes covered with aluminum foil.

[0034] b. Enzymatic digestion Add 50 mM Tris-HCl buffer (pH 8.0) at a 1:1 volume ratio. Add proteinase K to a final concentration of approximately 1 mg / mL, and simultaneously add SDS to a mass fraction of approximately 1%. Incubate with shaking at 50°C for 12 hours to degrade plasma proteins and blood cells.

[0035] c. Mild oxidation Add 30% H2O2 to the digestion solution to make the final volume fraction approximately 5%; Continue the reaction at 50°C for 4 hours to promote the decomposition of organic residues.

[0036] d. Density separation After digestion and oxidation, cool to room temperature and remove residue by low-speed centrifugation. Collect the supernatant and adjust the density to 1.6 g / cm³ by adding saturated zinc chloride solution. 3 about; After standing or centrifuging at 3000g for 30 minutes, collect the floating phase, which is enriched with most of the microplastics such as polyethylene (PE), polypropylene (PP), and polystyrene (PS).

[0037] e. Filtration fixed to the filter membrane Choose an alumina filter membrane or a polycarbonate filter membrane with a pore size of about 0.4-1.0 μm, and pre-etch two-dimensional grid coordinates with a certain spacing on the filter membrane; The floating phase is filtered onto the filter membrane by vacuum filtration, and the filtration volume is controlled to ensure that the particles are dispersed and do not overlap. The filter membrane is dried slowly in a vacuum drying oven to prevent particle migration.

[0038] f. Conductive coating treatment While ensuring compatibility with Raman detection (e.g., only locally on the back or side), a gold or platinum coating of about 5-10 nm thickness is deposited on the filter membrane surface by sputtering to improve the conductivity of SEM imaging.

[0039] If there are concerns that the metal layer may interfere with the Raman signal, Raman detection can be performed first, followed by sputtering.

[0040] (2) Raman imaging and polymer component identification a. Low-magnification panoramic scan Use a confocal Raman microscope with an excitation wavelength of 532 nm or 785 nm. The entire filter membrane was stitched and scanned under a 10× objective lens with a step size of approximately 5-10 μm. Extract characteristic peaks (e.g., 1000-1800 cm⁻¹) from the Raman spectrum of each pixel. -1 (Range), using an automatic peak recognition algorithm to filter out pixels with typical plastic characteristic peaks.

[0041] b. Location of suspected microplastic particles A "suspected microplastic distribution map" is generated based on the spectral intensity image; Combine the filter membrane grid coordinates to record the location coordinates (grid number + pixel coordinates) for each suspected particle.

[0042] (3) High-magnification fine spectral acquisition Perform point scanning or small area surface scanning on selected particles under a 50× or 100× objective lens; Baseline correction, fluorescence background subtraction, and normalization were performed on the spectral data; The processed spectra are compared with standard polymer libraries (PE, PP, PS, PET, etc.) by correlation coefficient matching or principal component analysis to determine the polymer type of each particle.

[0043] (4) SEM multi-angle imaging and 3D reconstruction a. Sample transfer and positioning Carefully remove the filter membrane after Raman detection and fix it on the SEM sample stage, ensuring that its orientation is consistent with the direction during Raman detection or a recordable rotation angle. Quickly browse the filter membrane in SEM low-magnification mode (e.g., 100×) to identify grid markings and several feature particles in order to establish a rough correspondence with the Raman image.

[0044] b. Multi-tilt SEM imaging Adjust the sample stage tilt angle, starting from -30°, and image every 10° until +60°; Imaging at each tilt angle with magnification of 5000×–20000×, achieving a resolution better than 10 nm; For each target particle, clear two-dimensional projected profile images are obtained at all tilt angles.

[0045] c. Three-dimensional topography reconstruction The contours of the same particles at different tilt angles are extracted to obtain the contour curves from each viewpoint; A shape-from-silhouette or voxel reconstruction algorithm is used to fuse multi-view contours to obtain a three-dimensional voxel model; Use surface reconstruction algorithms (such as marching cubes) to generate smooth 3D triangular mesh surfaces from voxel models.

[0046] d. Calculation of geometric parameters and sphericity Volume integration and surface area calculations are performed on the 3D mesh model to obtain the particle volume. and surface area ; Calculate sphericity: (4.2) in, ∈[0,1], The closer the value is to 1, the closer the particle is to an ideal sphere.

[0047] (4) Raman-SEM related registration and data output a. Coordinate system registration Mark several commonly visible grid intersections or feature particle coordinates in the Raman and SEM images respectively; Using at least 3-5 pairs of corresponding points, solve the coordinate transformation matrix through affine or perspective transformation to uniformly map the Raman coordinates to the SEM coordinate system.

[0048] b. Single-particle data fusion For each particle that has undergone 3D reconstruction, its corresponding position is located in the Raman data to obtain the polymer type label; The information such as "sample number, particle number, location, polymer type, volume, surface area, and sphericity" is stored in the database.

[0049] c. Statistical Analysis Statistical analysis of particle sphericity and volume distribution for different polymer types; By comparing the morphological differences of different types of microplastics in blood samples, a basis for health risk assessment can be provided.

[0050] Example 2: Application of saliva samples A preprocessing procedure similar to that in Example 1 can be used, with only appropriate adjustments to specific parameters, for example: Enzyme digestion time is shortened to 4-8 hours; The oxidation time is shortened to 2-3 hours; Because saliva has a low solids content, the centrifugation time for the density separation step can be appropriately reduced.

[0051] The subsequent Raman scanning, SEM imaging, and 3D reconstruction steps are the same as in Example 1, and will not be repeated here.

[0052] The above are merely preferred embodiments of the present invention and do not constitute any limitation on the present invention. Any equivalent substitutions or modifications made by those skilled in the art to the technical solutions and content disclosed in the present invention without departing from the scope of the present invention shall be deemed to have remained within the protection scope of the present invention.

Claims

1. A method for measuring the three-dimensional sphericity and composition of microplastic particles, characterized in that, First, the entire filter membrane is scanned at low magnification using micro Raman or confocal Raman spectroscopy. Suspected microplastic particles are screened and located based on Raman fingerprint peaks, and their high-resolution Raman spectra are obtained to determine the polymer composition. Subsequently, the sample is transferred to a scanning electron microscope equipped with a variable tilt stage on the same filter membrane or slide, and the same particle is imaged at high resolution at multiple tilt angles. A multi-view 3D reconstruction algorithm is used to obtain a single particle 3D morphology model, and various parameters are calculated. A one-to-one correspondence between Raman and SEM data is achieved by registering the filter membrane grid coordinates and feature points, and a single particle database of spatial location, polymer type, 3D morphology, and sphericity is established.

2. The method according to claim 1, characterized in that, Includes the following steps: (1) Sample pretreatment and microplastic enrichment The sample is subjected to enzymatic digestion, mild oxidation and density separation to remove proteins, cells and organic residues and enrich microplastic particles; the obtained microplastic particles are fixed on a filter membrane or slide with two-dimensional coordinate markers and a conductive coating to form the sample to be tested. (2) Raman imaging and component identification The filter membrane or slide was scanned using a micro Raman or confocal Raman microscope to screen for suspected microplastic particles based on the characteristic peaks of plastics in the Raman spectrum; its high signal-to-noise ratio Raman spectrum was obtained and its polymer composition type was determined by matching it with a polymer standard spectral library. (3) SEM multi-angle imaging Without changing the spatial distribution of particles on the filter membrane or slide, the particles are transferred to a scanning electron microscope. Using the tilting function of the sample stage, the target particles identified in step (2) are imaged at multiple different tilt angles to obtain a sequence of two-dimensional projection images of each particle at different observation angles. (4) Three-dimensional reconstruction and sphericity calculation Based on the multi-angle two-dimensional projection images obtained in step (3), a three-dimensional morphology model of a single particle is constructed using a multi-view three-dimensional reconstruction algorithm, and its volume is calculated. and surface area And according to the sphericity formula (3.1) Determine the sphericity of the particles ; (5) Raman-SEM related registration and data output: By using coordinate markers and feature point matching algorithms on the filter membrane or slide, spatial registration between the Raman imaging coordinate system and the SEM imaging coordinate system is achieved. The polymer composition information identified by Raman is correlated with the three-dimensional morphology and sphericity information reconstructed by SEM at the single-particle level, so as to obtain the three-dimensional sphericity distribution and statistical results of microplastic particles of different polymer types in the sample.

3. The method according to claim 2, characterized in that, In the sample pretreatment step, enzyme digestion uses one or more of proteinase K, trypsin, and collagenase, incubated at 40-60°C for 2-24 hours; mild oxidation uses hydrogen peroxide or persulfate, with the oxidation temperature not exceeding 60°C; density separation uses a high-density salt solution formed from one or more of zinc chloride, sodium iodide, and potassium iodide, with a density of 1.3-1.7 g / cm³. 3 The preferred value is 1.6 g / cm³. 3 .

4. The method according to claim 3, characterized in that, The filter membrane or carrier material is a substrate with low autofluorescence and weak Raman background; the substrate surface is formed with a gold, platinum or carbon conductive coating with a thickness of 5-20 nm by sputtering or evaporation, and two-dimensional grid coordinates are etched or printed.

5. The method according to claim 3, characterized in that, The Raman imaging employs a confocal micro-Raman system with an excitation wavelength of 488 nm, 532 nm, or 785 nm, achieving a spectral resolution better than 5 cm⁻¹. -1 The lateral spatial resolution is better than 1μm.

6. The method according to claim 3, characterized in that, The scanning electron microscope is equipped with an adjustable tilt stage, with a tilt angle range of -30° to +70° and a tilt step of 5°-15°, providing better performance at each tilt angle than... Acquire particle images at spatial resolution.

7. The method according to claim 3, characterized in that, The multi-view 3D reconstruction algorithm is one of the following: contour-based shape reconstruction from silhouette, voxel drawing, photometric stereo, or deep learning-based 3D shape prediction model.

8. The method according to claim 3, characterized in that, In the spatial registration step, by identifying the pre-made grid intersections, micro-engraved fiducial marks, or several feature particles on the filter membrane or slide as common reference points, an affine or perspective transformation is used to fit and establish the transformation matrix between the Raman coordinate system and the SEM coordinate system, thereby achieving the correspondence of the single particle position.

9. The method according to claim 3, characterized in that, The sample is one of anticoagulated whole blood, plasma, serum, resting saliva, or irritated saliva, and the microplastic particles have a particle size range of 1-50 μm.

10. A measurement system for implementing the method according to any one of claims 1-9, characterized in that, include: The sample pretreatment module is used to perform enzymatic digestion, mild oxidation, density separation, and enrichment of microplastic particles in blood or saliva samples. Conductive filter membranes or carriers with coordinate markings are used to immobilize enriched microplastic particles. Raman imaging module, used to scan particles on filter membranes or slides using Raman and identify polymer components; The scanning electron microscope module is equipped with a variable tilt sample stage for multi-angle high-resolution imaging of particles on filter membranes or slides. The data processing module is used for spatial registration of Raman and SEM images, 3D morphology reconstruction and sphericity calculation, and outputs the 3D sphericity distribution of microplastic particles classified by polymer type.

Images