Microscopic Fourier infrared spectrum rapid detection method for soil micro-plastics

Abstract

The invention discloses a soil micro-plastic microscopic Fourier infrared spectrum rapid detection method, and relates to the technical field of environmental pollutant analysis and detection.The method comprises the steps that a soil sample of a target area is collected; performing micro-plastic separation and extraction on the soil sample to obtain a micro-plastic sample; performing vacuum filtration on the micro-plastic sample and transferring the micro-plastic sample to a to-be-detected filter membrane; performing Fourier infrared microscopic scanning on the filter membrane to be detected to obtain a Mosaic image; carrying out particle point positioning on the Mosaic image by adopting a Fourier infrared detection method, and collecting spectral information of each particle point; the spectrum information of each particle point is matched with a micro-plastic standard spectrum library, micro-plastic type information of each particle point is determined, and a detection report is generated and comprises the point position matching rate, the area ratio, the particle length, the particle width and the content of different types of micro-plastics. According to the invention, rapid and accurate detection of the soil micro-plastic can be realized, and the detection precision and efficiency of the soil micro-plastic can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of environmental pollutant analysis and detection, in particular to a soil microplastic microscopic Fourier infrared spectrum rapid detection method. BACKGROUND

[0002] With the wide application of plastic products in agriculture, industry and daily life, the problem of soil microplastic (plastic particles with a size of less than 5 mm) pollution is becoming increasingly serious. Microplastics are difficult to degrade in soil, and long-term accumulation can change the physical and chemical properties of soil, affect soil microbial activity, and thus threaten the safety of the ecological system and human health. Therefore, it is of great significance to establish an efficient and accurate soil microplastic detection technology for soil pollution monitoring and management.

[0003] Currently, soil microplastic detection mainly relies on spectroscopy. Although pyrolysis gas chromatography-mass spectrometry (Py-GC-MS) can obtain the chemical composition and mass concentration of microplastics, it requires destructive treatment of the sample, which cannot preserve the morphological and abundance information of microplastics, and the final detection accuracy is low. Raman spectroscopy is susceptible to fluorescence interference, which limits the detection accuracy. Existing Fourier transform infrared spectroscopy methods mostly rely on manual focusing to collect spectra at a single point, or use machine learning algorithms but have insufficient model generalization ability, resulting in low detection efficiency, long time consumption (the traditional detection process takes more than 24 hours), and poor adaptability to complex soil samples. For example, related technologies rely on standard samples to train machine learning models, which are difficult to cope with complex organic matter interference in actual soil, and the detection precision and efficiency are difficult to balance.

[0004] In summary, how to realize the rapid and accurate detection of soil microplastics and improve the detection precision and efficiency of soil microplastics has become a technical problem to be solved in the field. SUMMARY

[0005] The purpose of the present application is to provide a soil microplastic microscopic Fourier infrared spectrum rapid detection method, which can realize the rapid and accurate detection of soil microplastics and improve the detection precision and efficiency of soil microplastics.

[0006] To achieve the above purpose, the present application provides the following solutions.

[0007] The present application provides a soil microplastic microscopic Fourier infrared spectrum rapid detection method, which specifically comprises the following steps.

[0008] Collecting soil samples in the target area.

[0009] Separating and extracting microplastics from the soil samples to obtain a microplastic sample.

[0010] Vacuum filtration is used to transfer the microplastic sample to the filter membrane to be detected.

[0011] Fourier infrared micro-scan is performed on the filter membrane to be detected to obtain a Mosaic image.

[0012] The Fourier infrared detection method is used to locate the particle points of the Mosaic image and collect the spectral information of each particle point.

[0013] The spectral information of each particle point is matched with a microplastic standard spectrum library to determine the microplastic type information of each particle point and generate a detection report; the microplastic standard spectrum library is a database including standard spectrum information of microplastic standard samples and corresponding microplastic type information; the detection report includes point matching rate, area ratio, particle length, particle width, and content of different types of microplastics.

[0014] Optionally, the soil sample is subjected to microplastic separation and extraction to obtain a microplastic sample, specifically including the following steps.

[0015] The soil sample is mixed with Fenton reagent at a predetermined ratio to obtain a mixed solution.

[0016] The mixed solution is subjected to water bath heating treatment to obtain a sample after water bath heating.

[0017] The sample after water bath heating is subjected to cleaning treatment using a screen mesh to obtain a cleaned sample.

[0018] Based on the cleaned sample, ZnCl2 solution is used for filter residue transfer and flotation to collect an upper layer microplastic and organic matter mixture.

[0019] The upper layer microplastic and organic matter mixture is subjected to digestion treatment, and L-ascorbic acid is added as a reducing agent for reduction reaction to obtain a sample after reduction reaction.

[0020] Polytetrafluoroethylene filter membrane is used to filter the sample after reduction reaction to obtain a microplastic sample.

[0021] Optionally, the Fe 2+ The volume ratio of the Fe 2+ The preparation process of the FeSO4 solution is as follows: 1.39 g of ferrous sulfate solid is added to 50 mL of RO water to obtain 50 mL of FeSO4 solution with a mass fraction of 2.78%; the mass fraction of the H2O2 solution is 30%.

[0022] The density of the ZnCl2 solution is 1.5 g / cm 3 , and the flotation time is 1 h.

[0023] The screen uses a 600-mesh stainless steel screen.

[0024] The temperature condition of the digestion treatment is 60 DEG C, the shaking speed is 115 rmp, and the digestion duration is 8 hours.

[0025] Optionally, the microplastic sample is vacuum-filtered and transferred to a filter membrane to be detected, specifically including the following steps.

[0026] The microplastic sample is transferred to a beaker for mixing, stirring and ultrasonic treatment, to obtain an ultrasonically treated microplastic sample.

[0027] The ultrasonically treated microplastic sample is filtered by a vacuum filtration device to a filter membrane to be detected.

[0028] Optionally, the filter membrane to be detected is a polyester gold-coated membrane with a pore size of 5 microns, a thickness of 19 microns, a diameter of 13 mm, and an effective filtration diameter of 10 mm.

[0029] Optionally, the filter membrane to be detected is subjected to Fourier infrared microscopic scanning to obtain a Mosaic image, specifically including the following steps.

[0030] The gold mirror surface is cleaned and wetted to obtain a cleaned and wetted gold mirror.

[0031] The filter membrane to be detected is tightly and flatly pasted on the surface of the cleaned and wetted gold mirror.

[0032] The electrically driven object platform of the Fourier infrared microscope is used to scan the filter membrane to be detected row by row according to a preset resolution, to splice a Mosaic image; the splicing process of the Mosaic image includes: extracting image features based on the SIFT algorithm, matching the features using the Euclidean distance formula, calculating a homography matrix to realize image registration, and completing image splicing using an image fusion formula with feathering processing.

[0033] Optionally, the gradient amplitude formula is used for image feature extraction based on the SIFT algorithm, and the gradient amplitude formula is as follows.

[0034] .

[0035] .

[0036] .

[0037] .

[0038] wherein, represents the gradient amplitude of the image at coordinates , and respectively represent the image in and the gradient of the direction, represent the original image, represent the image after Gaussian smoothing, is a Gaussian function, is the standard deviation of the Gaussian kernel.

[0039] The Euclidean distance formula is as follows.

[0040] .

[0041] wherein, represent the Euclidean distance between the feature points and the feature points , represent the coordinates of the feature points in the image, represent the coordinates of the feature points in the image.

[0042] The expression of the homography matrix is as follows.

[0043] .

[0044] wherein, H is a 3*3 homography matrix, represent the coordinates of a pixel in the original image, represent the corresponding coordinates of the pixel in the target registration image.

[0045] The image fusion formula of the feathering processing is as follows.

[0046] .

[0047] wherein, is the fused image, and are two images to be spliced, , is the weight of the corresponding image, and .

[0048] Optionally, Fourier infrared detection method is adopted to locate the particle points of the Mosaic image, and the spectral information of each particle point is collected, specifically including the following steps.

[0049] Image recognition technology is adopted to segment the particle points of the Mosaic image, and the coordinate information of each particle point is determined.

[0050] The Fourier infrared microscope is used as a spectrum acquisition device, is set to a reflection mode, and the spectrum acquisition range and acquisition time are set.

[0051] Based on the coordinate information of each particle point, the spectrum information of each particle point in the Mosaic image is collected by using the guide analysis function of the Fourier infrared microscope.

[0052] Based on the spectrum information of each particle point, water vapor absorption peak and carbon dioxide peak interference correction processing is performed to obtain corrected spectrum information; the corrected spectrum information is used as the final spectrum information of each particle point, and is matched with a microplastic standard spectrum library to determine the microplastic type information of each particle point and generate a detection report.

[0053] Optionally, the spectrum information of each particle point is matched with the microplastic standard spectrum library to determine the microplastic type information of each particle point, and specifically includes the following steps.

[0054] The spectrum information of each particle point is matched with the standard spectrum information of a microplastic standard in the microplastic standard spectrum library to determine the spectrum matching degree of the spectrum information of each particle point and the standard spectrum information.

[0055] According to the spectrum matching degree of the spectrum information of each particle point and the standard spectrum information, the microplastic type information corresponding to the particle point is determined, including the following steps.

[0056] When the spectrum matching degree of the spectrum information of the particle point and the standard spectrum information is greater than or equal to 75%, the particle point is determined to be microplastic.

[0057] When the spectrum matching degree of the spectrum information of the particle point and the standard spectrum information is greater than or equal to 45% and less than 75%, the particle point is determined to be suspected microplastic.

[0058] According to the specific embodiments provided in the present application, the present application has the following technical effects.

[0059] The application provides a soil microplastic microscopic Fourier infrared spectrum rapid detection method. The method first collects a soil sample in a target area; then microplastics are separated and extracted from the soil sample to obtain a microplastic sample; the microplastic sample is vacuum filtered and transferred to a filter membrane to be detected; then Fourier infrared microscopic scanning is performed on the filter membrane to be detected to obtain a Mosaic image; then the Fourier infrared detection method is used to locate particle points of the Mosaic image and collect spectral information of each particle point; finally, the spectral information of each particle point is matched with a microplastic standard spectrum library to determine the microplastic type information of each particle point and generate a detection report. The detection process of the application, i.e., "sample collection, separation and extraction, vacuum filtration and transfer, infrared scanning, spectrum matching, and report generation", does not need to rely on the experience judgment of an operator, reduces the manual intervention error, can enable a novice to quickly master the operation, and can ensure that the detection processes of different laboratories and different batches of samples are unified, greatly improves the accuracy and repeatability of the detection results, and solves the problems of the traditional method, i.e., the operation varies from person to person and the results are difficult to reproduce. Moreover, the Fourier infrared microscopic scanning is used to automatically obtain the Mosaic image, which replaces the traditional manual single-point focusing imaging, can cover the whole area of the filter membrane to be detected at one time, and shortens the imaging time. At the same time, with the automatic process of "particle point positioning, spectrum automatic collection, and standard spectrum library matching", the microplastic type and key parameters can be quickly determined without manual spectrum analysis point by point. Compared with the traditional method relying on manual analysis, the overall detection period is significantly shortened, the detection efficiency is improved, and the timely monitoring demand of soil microplastics can be met. In addition, the detection report not only contains the microplastic type information, but also synchronously outputs the point matching rate (reflecting the identification reliability), the area ratio, the particle length and width (reflecting the microplastic morphology), and the content of different types (reflecting the pollution degree), which presents the soil microplastic pollution condition in multiple dimensions. These information can be directly used to evaluate the soil pollution grade, trace the pollution source, and provide detailed data support for subsequent pollution control scheme formulation, and has practical value. BRIEF DESCRIPTION OF DRAWINGS

[0060] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the following will briefly introduce the drawings needed in the embodiments. Obviously, the drawings in the following description only constitute some embodiments of the application, and for those skilled in the art, other drawings can be obtained from these drawings without creative labor.

[0061] Figure 1 A flowchart of a soil microplastic microscopic Fourier infrared spectrum rapid detection method provided by an embodiment of the application.

[0062] Figure 2 A farmland soil microplastic gold film full-scan Mosaic image provided by an embodiment of the application.

[0063] Figure 3 For Figure 2 The corresponding particle segmentation image.

[0064] Figure 4 The Fourier infrared detection image splicing flowchart provided by an embodiment of the present application.

[0065] Figure 5 The farmland soil microplastic detection report schematic diagram provided by an embodiment of the present application.

[0066] Figure 6 The absorbance change with wave number schematic diagram of gold film, nylon film, silver film and aluminum film provided by an embodiment of the present application.

[0067] Figure 7 The farmland soil microplastic extraction, separation and detection flowchart schematic diagram provided by an embodiment of the present application.

[0068] Figure 8 The farmland 0-30cm soil microplastic content schematic diagram provided by an embodiment of the present application.

[0069] Figure 9 The soil microplastic vacuum filtration device schematic diagram provided by an embodiment of the present application. DETAILED DESCRIPTION

[0070] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application.

[0071] The purpose of the present application is to provide a soil microplastic microscopic Fourier infrared spectrum rapid detection method, which is suitable for qualitative and quantitative detection of microplastics in various soils such as agricultural land and urban green land. Through optimization of sample pretreatment process, Mosaic image splicing technology and automatic spectrum acquisition technology, rapid and accurate detection of soil microplastics can be realized, and the detection accuracy and efficiency of soil microplastics can be improved.

[0072] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application will be further described in detail below with reference to the drawings and specific embodiments.

[0073] As Figure 1 shown, the present embodiment provides a soil microplastic microscopic Fourier infrared spectrum rapid detection method, which specifically includes the following steps.

[0074] S1: Collecting soil samples in a target area.

[0075] S2: Separating and extracting microplastics from the soil samples to obtain microplastic samples.

[0076] S3: Vacuum-filtering and transferring the microplastic samples to a filter membrane to be detected.

[0077] S4: Fourier infrared microscopic scanning of the filter membrane to be detected to obtain a Mosaic image.

[0078] S5: Particle point positioning of the Mosaic image by Fourier infrared detection, and collecting spectral information of each particle point.

[0079] S6: Matching the spectral information of each particle point with a microplastic standard spectral library to determine the microplastic type information of each particle point and generate a detection report. The microplastic standard spectral library is a database including standard spectral information of microplastic standard samples and corresponding microplastic type information. The detection report includes point matching rate, area ratio, particle length, particle width, and content of different types of microplastics.

[0080] As an implementable way, step S2 of separating and extracting microplastics from the soil samples to obtain microplastic samples specifically includes the following steps.

[0081] S21: Mixing and soaking the soil samples with Fenton reagent at a predetermined ratio to obtain a mixed solution.

[0082] S22: Water bath heating treatment of the mixed solution to obtain a sample after water bath heating.

[0083] S23: Cleaning the sample after water bath heating by using a screen to obtain a cleaned sample.

[0084] S24: Based on the cleaned sample, using ZnCl2 solution for residue transfer and flotation to collect the upper layer of microplastic and organic matter mixture.

[0085] S25: Digestion treatment of the upper layer of microplastic and organic matter mixture, and adding L-ascorbic acid as a reducing agent for reduction reaction to obtain a sample after reduction reaction.

[0086] S26: Filtering the sample after reduction reaction by using a polytetrafluoroethylene filter to obtain a microplastic sample.

[0087] As an implementable way, in step S2, the Fe 2+The volume ratio of the ferrous ion solution to the H2O2 (hydrogen peroxide) solution is 1:50; the Fe 2+ The preparation process of the solution is as follows: 1.39 g of ferrous sulfate (FeSO4) solid is added into 50 mL of RO (Reverse Osmosis) water to obtain 50 mL of FeSO4 solution with a mass fraction of 2.78%; the mass fraction of the H2O2 solution is 30%. The density of the ZnCl2 solution is 1.5 g / cm 3 The floating time is 1 h. The screen mesh is a 600-mesh stainless steel screen mesh. The temperature condition of the digestion treatment is 60℃, the shaking bed rotation speed is 115 rmp, and the digestion duration is 8 h.

[0088] As an implementable manner, step S3 transfers the microplastic sample to the filter membrane to be detected by vacuum filtration, specifically including the following steps.

[0089] S31: The microplastic sample is transferred to a beaker for mixing, stirring treatment and ultrasonic treatment to obtain the microplastic sample after ultrasonic treatment.

[0090] S32: The microplastic sample after ultrasonic treatment is filtered to the filter membrane to be detected by using a vacuum filtration device.

[0091] As an implementable manner, in step S3, the filter membrane to be detected is a gold-coated polyester (PETG) film (referred to as a gold film) with a pore size of 5 μm, a thickness of 19 μm, a diameter of 13 mm and an effective filtering diameter of 10 mm.

[0092] As an implementable manner, step S4 performs Fourier infrared microscopic scanning on the filter membrane to be detected to obtain a Mosaic image, specifically including the following steps.

[0093] S41: The surface of the gold mirror is cleaned and wetted to obtain the cleaned and wetted gold mirror.

[0094] S42: The filter membrane to be detected is closely and flatly pasted on the surface of the cleaned and wetted gold mirror.

[0095] S43: The electrically driven object platform of the Fourier infrared microscopic spectrometer is used to scan the filter membrane to be detected row by row according to a preset resolution to splice a Mosaic image; the splicing process of the Mosaic image includes: extracting image features based on a SIFT (Scale-Invariant Feature Transform) algorithm, performing feature matching by using a Euclidean distance formula, calculating a homography matrix to realize image registration, and completing image splicing by using a feathering processing image fusion formula.

[0096] As an implementable manner, step S5 adopts Fourier infrared detection method to locate the particle points of the Mosaic image and collect the spectral information of each particle point, specifically including the following steps.

[0097] S51: adopt image recognition technology to segment the particle points of the Mosaic image and determine the coordinate information of each particle point.

[0098] S52: adopt Fourier infrared microscope spectrometer as a spectral collection device, set the Fourier infrared microscope spectrometer to a reflection mode, and set the spectral collection range and collection time.

[0099] S53: based on the coordinate information of each particle point, use the guide analysis function of the Fourier infrared microscope spectrometer to directionally collect the spectral information of each particle point in the Mosaic image.

[0100] S54: based on the spectral information of each particle point, perform water vapor absorption peak and carbon dioxide peak interference correction processing to obtain corrected spectral information; the corrected spectral information is used as the final spectral information of each particle point, which is matched with a microplastic standard spectral library to determine the microplastic type information of each particle point and generate a detection report.

[0101] As an implementable manner, step S6 matches the spectral information of each particle point with a microplastic standard spectral library to determine the microplastic type information of each particle point, specifically including the following steps.

[0102] S61: match the spectral information of each particle point with the standard spectral information of a microplastic standard in the microplastic standard spectral library to determine the spectral matching degree of the spectral information of each particle point with the standard spectral information.

[0103] S62: according to the spectral matching degree of the spectral information of each particle point with the standard spectral information, determine the microplastic type information corresponding to the particle point, including the following two cases.

[0104] (1) when the spectral matching degree of the spectral information of the particle point with the standard spectral information is greater than or equal to 75%, it is determined that the particle point is microplastic.

[0105] (2) when the spectral matching degree of the spectral information of the particle point with the standard spectral information is greater than or equal to 45% and less than 75%, it is determined that the particle point is suspected microplastic.

[0106] As an implementable manner, the microplastic species include, but are not limited to, PP (polypropylene), PE (polyethylene), PET (polyethylene terephthalate), PP-PE (polypropylene-polyethylene), PVC (polyvinyl chloride), PS (polystyrene), and the like.

[0107] As an implementable manner, the point matching rate in the detection report is obtained based on the spectral matching result. The point matching rate is a core index for measuring the recognition credibility of microplastics. First, the effective detection point is determined: all particle points with a particle size of 50-1300 μm are segmented out through Mosaic image recognition, each particle point corresponds to one detection point, and the total number of points (denoted as Ntotal) is counted. Then, the matching successful point is counted: the spectral information of each point is matched with the microplastic standard spectral library, and the point with a spectral matching degree of ≥45% (including microplastic points and suspected microplastic points, denoted as Nmatching) is screened out. Finally, the matching rate is calculated: point matching rate=(Nmatching / Ntotal)×100%, reflecting the proportion of recognizable particles in the detection area. The higher the matching rate, the stronger the coverage or recognition effectiveness of microplastics (including suspected ones) in the sample.

[0108] As an implementable manner, the area ratio in the detection report is obtained based on the proportion of the particle area and the detection area of the filter membrane. The area ratio reflects the distribution density of a single microplastic in the detection area. First, the single particle area is calculated: based on the pixel calibration coefficient of the Mosaic image, the total number of pixels in each particle contour is counted, and then converted into the actual area (unit: μm 2 ). Then, the total area of a single species of microplastic is calculated: the total area of a single species of microplastic (such as PE, PP) is obtained by summing up the areas of all particles of the same species. Finally, the area ratio is calculated: area ratio=(Ssingle species / Sdetection area)×100%, wherein Sdetection area is the actual coverage area of the Mosaic image, reflecting the proportion of a single species of microplastic in the detection range.

[0109] As an implementable way, the particle length and width in the detection report are measured based on Mosaic image pixel calibration. The Mosaic image obtained by the Fourier infrared microscope has self-pixel size calibration information (determined by the instrument resolution parameter). First, image pixel calibration is performed: according to the scanning resolution set by the instrument, the calibration coefficient of 1 pixel corresponding to the actual length (for example, 1 pixel = 1 μm) in the Mosaic image is determined. Then, particle contour extraction is performed: the edge contour of each particle point is automatically outlined by an image recognition algorithm to form a closed particle shape area. Finally, size measurement is performed: the length (the straight line distance between the two farthest points of the particle) is measured along the longest axis of the particle contour, and the width (the straight line distance between the two farthest points of the particle) is measured along the widest axis perpendicular to the longest axis, and then the pixel distance is converted into actual length (unit: μm) by combining the pixel calibration coefficient.

[0110] As an implementable way, the content of different kinds of microplastics in the detection report is obtained based on particle counting and mass conversion (or relative content statistics). The content statistics includes relative content and absolute content. For the relative content (particle number ratio), the total number of particles of each kind of microplastic is first counted (for example, the number of PE particles is NPE, and the number of PP particles is NPP); then the ratio of single microplastic particles = (Nsingle / Nall microplastics) x 100% is calculated, reflecting the number distribution relationship of different kinds of microplastics. For the absolute content (mass / unit soil), the mass of single particle is first calculated based on the microplastic density (known standard value, for example, the density of PE is 0.92 g / cm 3 ) and the volume of single particle (calculated from the length, width and approximate shape of the particle, such as being regarded as a cylinder or a sphere); then the total mass of this kind of microplastic (Msingle kind) is obtained by summation; finally, the content of microplastic per unit soil is converted (for example, Msingle kind / 10 g soil, unit: mg / kg) by combining the mass of the soil sample used for detection.

[0111] In order to make the technical scheme of the present application more clear, the specific implementation process of the technical scheme of the present application will be described in the form of examples as follows.

[0112] The main steps of the soil microplastic micro-Fourier infrared spectrum rapid detection method proposed in this embodiment are as follows.

[0113] Step 1: The soil sample is mixed with Fenton reagent and soaked, heated in a water bath, and the temperature is controlled at 60℃; the reaction between Fenton reagent and organic matter in the soil is an exothermic reaction, the sample is pretreated by using a stainless steel screen, the filter residue on the filter membrane and screen is transferred with ZnCl2 solution and further flotation, the mixture of upper layer microplastics and organic matter is collected, the Fenton reagent is fully digested to remove the interfering organic matter in the microplastics, and finally L-ascorbic acid is added as a reducing agent to reduce for 0.5h, and Fe 3+ is reduced to Fe 2+ , which is convenient for filter membrane filtration and easy observation; further filtration is performed to obtain clean microplastics on a new filter membrane, and the soil microplastics extracted on the filter membrane are transferred to a beaker.

[0114] As an optional implementation, the mesh number of the stainless steel screen in step 1 is 600 meshes, the density of ZnCl2 is 1.5g / cm 3 , the flotation time is 1h, the volume ratio of Fe 2+ solution to H2O2 solution in the Fenton reagent is 1:50; the Fe 2+ solution is prepared as follows: 1.39g of ferrous sulfate solid is added to 50mL of RO water to obtain 50mL of FeSO4 solution with a mass fraction of 2.78%; the mass fraction of H2O2 solution is 30%. Preferably, the pH of RO water is 3.

[0115] As an optional implementation, the digestion condition in step 1 is 115rpm, 60℃, 8h.

[0116] As an optional implementation, the filter membrane in step 1 is made of polytetrafluoroethylene material with a pore size of 20μm.

[0117] Step 2: The soil microplastics extracted on the filter membrane are transferred to a beaker, mixed and stirred with a reducing agent, vacuum filtered to a filter membrane for Fourier infrared detection, the vacuum filtration device mainly includes a filter cup, a sand core filter head, a silica gel conical plug, a fixing clamp and a filtration bottle, the diaphragm vacuum pump pressure is 0.08Mpa, and the air suction speed is 120L / min to make the microplastics uniformly distributed on the membrane.

[0118] As an optional implementation, the ultrasonic time in step 2 is 30min, the filter membrane is a polyester gold coating membrane with a pore size of 5μm, a diameter of 13mm and an effective filtration diameter of 10mm.

[0119] As an optional implementation, the type of reducing agent in step 2 is L-ascorbic acid, and the weight is 15g.

[0120] Step 3: Clean and wet the surface of the gold mirror, and tightly paste the filter membrane to be detected on the gold mirror. Use the electrically driven carrier platform of the Fourier infrared microscope spectrometer to automatically obtain the Mosaic image of the filter membrane. The electrically driven carrier platform scans the entire gold membrane according to the resolution of 8 cm -1 Scan the entire gold membrane row by row, enable the surface scanning tool to splice the Mosaic image, realize the full scanning of the gold membrane, and identify and segment the image to obtain the coordinates of each particle point. Figure 2 and Figure 3 are respectively the Mosaic image and the particle segmentation image of the farmland soil microplastic gold membrane full scan, and it can be seen that the distribution of different microplastics on the gold membrane.

[0121] Among them, as shown in Figure 4 , the main steps of splicing the Mosaic image are as follows.

[0122] S1: Feature extraction. In the SIFT algorithm, the gradient amplitude formula is the core step of direction assignment and feature descriptor generation, which is defined as follows.

[0123] (1).

[0124] (2).

[0125] (3).

[0126] (4).

[0127] Among them, and respectively represent the gradients of the image in and directions. First, the original image is Gaussian smoothed to de-noise, and then the horizontal and vertical gradients are calculated through , and finally the square root of the sum is obtained . The greater the value, the more significant the texture change. This result is used to determine the main direction of the feature point and generate the descriptor, ensuring the invariance of the feature to scale and rotation.

[0128] S2: Feature matching. Match the extracted feature points using the Euclidean distance formula. Euclidean distance is the most intuitive and most commonly used distance measurement method. Its essence is the straight-line distance between two feature points in n-dimensional space, which is used to measure the similarity of two features in feature matching. The similarity of matching is usually calculated by Euclidean distance, and the formula is as follows.

[0129] (5).

[0130] Among them, representing a feature point between a feature point , representing a feature point in an image, representing a feature point in an image.

[0131] S3: Image registration. Compute homography matrix H for aligning images. Given corresponding points and in two images, the homography matrix can be represented as follows.

[0132] (6).

[0133] wherein, wherein, H is a 3x3 homography matrix containing 8 free parameters (the last parameter is usually set to 1), represents the coordinates of a pixel in the original image, represents the corresponding coordinates of the pixel in the target registration image.

[0134] S4: Image fusion. Fuse the aligned images to generate a mosaic image. The fusion process includes feathering processing, which is a common method for natural transition in image fusion. The core is to achieve smooth transition through "weight distribution" to avoid harsh edges in the fusion. To reduce the visibility of the seam. The fusion formula is as follows.

[0135] (7).

[0136] wherein, is the fused image, and are two images to be spliced, , are the weights of the corresponding images, and .

[0137] Step 4: Set the Fourier infrared microscope spectrometer to reflection mode, set the step size, scan times, and spectral range, and use the Fourier infrared spectrometer guide analysis function to collect the spectrum of each particle point. Match with the microplastic standard spectrum library to determine the type of particle microplastic.

[0138] As an optional implementation, the step size of the Mosaic image collected in step 4 is 100 μm.

[0139] As an optional implementation, the area of the Mosaic image collected in step 4 is 13.6 13.6mm 2 .

[0140] As an optional implementation, the particle size segmentation range of the Mosaic image in step 4 is 50-1300 μm.

[0141] Step 5: Generate a detection report to determine the point matching rate, area ratio, length, width, and other information, and obtain specific information of different types of microplastics. The form of the detection report is as shown in Figure 5 .

[0142] As an optional implementation, the spectral collection range in step 5 is set to 3700-2800 cm -1 , 1800-800 cm -1 , and the collection time is 16 s.

[0143] As an optional implementation, the spectral data preprocessing step in step 5 mainly includes correcting the interference of water vapor absorption peaks and carbon dioxide peaks in the infrared spectrum.

[0144] As an optional implementation, in step 5, the microplastics are: the spectral matching degree of the particle spectrum with the microplastic standard spectrum library is ≥75%. Suspected microplastics: the spectral matching rate is 45%-75%.

[0145] In this embodiment, the main instruments and reagents used are as follows.

[0146] (1) Reagents.

[0147] Ferrous sulfate (FeSO4·5H2O): purchased from Tianjin Zhiyuan Chemical Reagent Co., Ltd.

[0148] Zinc chloride (ZnCl2): purchased from Tianjin Yongda Chemical Reagent Co., Ltd.

[0149] Hydrogen peroxide (H2O2): mass fraction of 30%, purchased from Tianjin Chemical Reagent Co., Ltd.

[0150] (2) Experimental instruments.

[0151] Polyester (PETG) metal-coated membrane filter, manufacturer: Sterlitech, USA.

[0152] Glass mercury thermometer, manufacturer: Hebei Wuyuan County Red Star Instrument Factory, China.

[0153] GZX-9140 MBE drying box, manufacturer: Shanghai Boxun Company, China.

[0154] HZ-9612K high-temperature vertical oscillator, manufacturer: Taicang City Science and Technology Instrument Factory, China.

[0155] SHA-BA water bath constant temperature oscillator: manufacturer: Changzhou Guowang Instrument Manufacturing Co., Ltd., China.

[0156] JL-UPR-Ⅲ-10T ultrapure water device: manufacturer: Sichuan Jinlan Technology Co., Ltd., China.

[0157] GM-1.0A vacuum filtration device, manufacturer: Tianjin Teng Company, China.

[0158] Meilen MTB 200 electronic balance, manufacturer: Shenzhen Meifeng Electronics Co., Ltd., China.

[0159] Nicolet TM iN10 Fourier transform infrared microscope, manufacturer: Thermo Fisher Scientific, USA.

[0160] In this embodiment, the soil microplastic Fourier infrared spectrum rapid detection mainly includes the following work.

[0161] 1. Soil sample acquisition: collect deep soil samples to minimize microplastic pollution; crush the soil sample with a soil crusher and use a 5mm sieve for preliminary particle size screening; observe the soil sample under a stereomicroscope to determine that there is no microplastic pollution, and weigh 10g of soil sample for standby.

[0162] 2. The detection of microplastic particles on gold film (PETG), silver film (Silver Membrance Filter), nylon film (Nylon), and aluminum film (Alumina) can be accurately detected: In this application, 500μm irregular standard PE, PP, and PET particles are selected as research objects, and different microplastic samples are added to verify the experiment. Figure 6 The absorbance of gold film, nylon film, silver film, and aluminum film (i.e. alumina film, referred to as aluminum film) changes with the wave number. As can be seen, the absorbance of gold film is always constant, and it hardly absorbs infrared spectrum. Therefore, in this embodiment, gold film is selected as the experimental substrate to avoid interference.

[0163] 3. Microplastic detection of environmental soil samples: In this application, 30-year farmland soil is selected. In order to ensure the universality of the environmental sample, five-point sampling method is used to dig 1 1m soil pit, and 0-10cm, 10-20cm, and 20-30cm soil samples are uniformly mixed and taken back to the laboratory. The soil sample is ground with a soil crusher and screened with a 5mm sieve for preliminary particle size screening as an environmental sample.

[0164] The technical solutions of the present application will be further described in conjunction with specific embodiments, and the purposes and advantages of the present application will be clarified.

[0165] Example 1.

[0166] The present embodiment uses a standard microplastic mixture as a sample to be detected, and the soil microplastic rapid detection method by Fourier infrared spectroscopy comprises the following steps.

[0167] (1) 30 irregular standard PE, PP, and PET particles with a size of 500 μm were mixed and added to 10 g of clean soil sample, respectively, to make four groups of identical samples.

[0168] (2) The four groups of soil samples added with standard microplastics were mixed and soaked with Fenton reagent, and heated in a water bath with a temperature controlled at 60°C. The reaction between Fenton reagent and organic matter in the soil is an exothermic reaction. The sample was pretreated by washing with a 600-mesh stainless steel screen. The residue on the filter membrane and screen was transferred with ZnCl2 solution and further floated. The mixture of upper microplastics and organic matter was collected. The Fenton reagent was used to fully digest and remove the interfering organic matter in the microplastics. The volume ratio of Fe 2+ to H2O2 solution was 1:50. The Fe 2+ solution was prepared by adding 1.39 g of ferrous sulfate solid to 50 mL of RO water, adjusting the pH of the RO water to 3, and obtaining 50 mL of 2.78% FeSO4 solution. The mass fraction of H2O2 solution was 30%, the digestion temperature was 60°C, the shaking speed was 115 rmp, and the digestion time was 8 h.

[0169] (3) 15 g of L-ascorbic acid reducing agent was added to each group of samples for ultrasonic reaction. The soil microplastics extracted on the filter membrane were transferred to a beaker, and the four groups of samples were fully mixed and stirred, and then vacuum filtered onto the gold, silver, nylon, and aluminum membranes for Fourier infrared detection, so that the microplastics were evenly distributed on the membranes.

[0170] (4) The surface of the gold mirror was cleaned and moistened with alcohol, and the filter membrane to be detected was tightly pasted on the gold mirror. No bubbles were allowed to appear between the gold, silver, nylon, and aluminum membranes and the gold mirror to ensure the relative flatness of the filter membrane, which had a great influence on the accuracy of Mosaic imaging and infrared spectrum acquisition. The Mosaic image of the filter membrane was automatically obtained by using the electric sample loading platform of the Fourier infrared microscope.

[0171] (5) The face scanning tool of the full-automatic sample loading platform was started to splice the 13.6 13.6 mm 2 Mosaic image, and the image was segmented to obtain the coordinates of each particle point. The particle size segmentation range was 50 μm-1300 μm.

[0172] (6) The Fourier infrared acquisition parameters were set, the reflection mode was set, the MCT detector was set, and the spectral acquisition range was set to 4000-675 cm -1, the collection time is 16s, the Fourier infrared spectrometer guide analysis function is used to collect the spectrum of each particle point, the interference of water vapor absorption peak and carbon dioxide peak in the infrared spectrum is corrected, and the standard puku is matched to determine the type of particle microplastic, and the spectrum matching rate ≥75% is determined as microplastic; the spectrum matching rate is 45%-75%, which is determined as suspected microplastic, and the spectrum matching rate is other range, which is determined as non-microplastic and non-suspected microplastic. From Figure 6 It can be seen that the absorbance of the gold film is always constant, it hardly absorbs infrared signal, and it has the least influence on the detection result, so as to improve the spectrum matching rate of microplastic type, and therefore the gold film is determined as the substrate for subsequent experimental detection.

[0173] Example 2.

[0174] In this embodiment, 30-year farmland soil samples are used as detection samples, and the soil microplastic microscopic Fourier infrared spectrum rapid detection method includes the following steps.

[0175] (1) Select 30-year farmland soil, in order to ensure the universality of the environmental sample, adopt five-point sampling method to dig 1 m soil pit, mix 0-10cm, 10-20cm, 20-30cm soil samples uniformly and take back to laboratory, grind the soil sample by extrusion type soil crusher and use 5mm screen for preliminary particle size screening as environmental sample.

[0176] (2) As shown in Figure 7 , 10g of soil sample is mixed with Fenton reagent and soaked, and heated to 60℃ in water bath. The reaction of Fenton reagent and organic matter in soil is exothermic reaction. The sample is washed and pretreated by using 600 mesh stainless steel screen, the residue on the filter membrane and screen is transferred by ZnCl2 solution for further flotation, and the upper layer of microplastic and organic matter mixture is collected. The interference organic matter in microplastic is removed by sufficient digestion of Fenton reagent, and the volume ratio of Fe 2+ and H2O2 solution in Fenton reagent is 1:50. The preparation method of Fe 2+ solution is to add 1.39g of ferrous sulfate solid into 50ml of RO water, and adjust the pH value of RO water to 3, to obtain 50ml of 2.78% FeSO4 solution. The mass fraction of H2O2 solution is 30%, the digestion temperature is 60℃, the shaking bed rotation speed is set to 115rpm, and the digestion time is 8 hours. The mixed solution is filtered onto the gold film by using 15g of L-ascorbic acid reducing agent combined with ultrasonic reaction, the soil microplastic extracted onto the gold film is transferred to the beaker, and after being mixed and stirred, it is vacuum filtered onto the Fourier infrared detection gold film, so that the microplastic is uniformly distributed on the film.

[0177] Figure 8The soil microplastic vacuum filtration device is shown, which includes a filtration bottle, a filter cup, a sand core filter head, a fixing clamp and a silica gel conical plug. The vacuum filtration process of soil microplastics can be realized through the soil microplastic vacuum filtration device.

[0178] (3) After cleaning and wetting the gold mirror surface, the filter membrane to be detected is tightly attached to the gold mirror, ensuring that there is no air bubble between the gold membrane and the gold mirror, because it significantly affects the flatness of the gold membrane and the accuracy of Mosaic imaging and infrared spectrum acquisition. The Mosaic image of the gold membrane is automatically obtained by using the electric sample platform of the Fourier transform infrared microscope.

[0179] (4) The full-automatic sample platform enables the face scanning tool to splice 13.6 13.6mm 2 Mosaic image, segment the image and obtain the coordinates of each particle point, and the particle size segmentation range is 50-1300μm.

[0180] (5) Set the collection parameters of the Fourier infrared spectrometer, in reflection mode, MCT detector, and the spectral collection range is 3700-2800cm -1 , 1800-800cm -1 , and the collection time is 16s. Use the Fourier infrared spectrometer's guided analysis function to identify and analyze, collect the spectrum of each particle point, and correct the water vapor absorption peak and carbon dioxide peak interference in the infrared spectrum. Match with the microplastic standard spectrum library to determine the type of microplastic particles, and those with a spectrum matching rate of ≥75% are identified as microplastics; those with a spectrum matching rate between 45%-75% are identified as suspected microplastics. Finally, after preprocessing the spectrum data, the key information of microplastics such as class, particle size, circumference, area and microscopic morphology can be obtained, and a detection report can be generated. Figure 9 The soil microplastic content of 0-30cm farmland is shown, from which the content corresponding to different types of soil microplastics can be seen.

[0181] The application provides a soil micro-plastic microscopic Fourier infrared spectrum rapid detection method, and key steps of the method include: soil micro-plastic sample separation and extraction; automatically acquiring a Mosaic image of a to-be-detected area by using an electric sample loading platform; collecting a particle point spectrum by using a Fourier infrared guide plane scanning analysis function, and realizing qualitative and quantitative detection of micro-plastics in the to-be-detected area by analyzing and processing the spectrum data. The soil organic matter is rapidly digested by using Fenton reagent, imaging interference is reduced, the Mosaic image of the to-be-detected area is automatically acquired by using the electric sample loading platform, and the particle point spectrum is collected by using the Fourier infrared guide plane scanning analysis function, so that the chemical composition, abundance, size and morphological data of the micro-plastics are accurately acquired without damaging the sample, the detection time is shortened, the detection efficiency is improved, an efficient, rapid and accurate soil micro-plastic identification method is formed, the soil micro-plastics can be rapidly detected, the detection precision is high, the method is accurate and reliable, the operation is simple, the intelligent level of soil micro-plastic detection is improved, and a reliable detection means is provided for soil micro-plastic research and monitoring.

[0182] The application is based on rapid digestion of soil organic matter by using Fenton reagent, the digestion time is greatly shortened compared with traditional H2O2 oxidation, the digestion time is shortened to 8 hours, the sample is clean and free of excess impurities, the sample is convenient for on-machine detection, the Mosaic image segmentation is accurate, the on-machine detection time of the Fourier infrared microscopic spectrometer is shortened to 6 hours. The Fourier infrared detection method acquires particle point coordinates by segmenting a static image, and the full-automatic sample loading platform is used for rapid directional collection of target particle infrared spectrum of the to-be-detected object, so that the Fourier infrared spectrum detection efficiency is improved. The Fourier infrared spectrum can acquire key information such as infrared spectrum, microscopic morphology and particle size category of the micro-plastics, accurately identifies the micro-plastics, and meets the rapid qualitative and quantitative detection requirements of the soil micro-plastics without damaging the sample.

[0183] The technical features of the above embodiments can be combined in any manner. To make the description concise, all possible combinations of the technical features in the above embodiments are not described, but as long as the combinations of the technical features do not exist, they should be considered as the scope of the present application.

[0184] The principles and implementation modes of the application are described by using specific examples, and the above embodiments are only used to help understand the method and core idea of the application; meanwhile, according to the idea of the application, the specific implementation modes and application ranges can be changed by those skilled in the art. In conclusion, the content of the specification should not be understood as a limitation of the application.

Claims

1. A method for rapid detection of soil microplastics by microscopic Fourier infrared spectroscopy, characterized in that, The soil micro-plastic microscopic Fourier infrared spectrum rapid detection method comprises the following steps: Collecting soil samples in a target area; Separating and extracting micro-plastics from the soil samples to obtain a micro-plastic sample; Vacuum-filtering and transferring the micro-plastic sample to a filter membrane to be detected; Performing Fourier infrared microscopic scanning on the filter membrane to be detected to obtain a Mosaic image; Using the Fourier infrared detection method, locating particle points of the Mosaic image, and collecting spectral information of each particle point; Matching the spectral information of each particle point with a micro-plastic standard spectrum library to determine the micro-plastic type information of each particle point and generate a detection report; the micro-plastic standard spectrum library is a database comprising standard spectrum information of micro-plastic standard samples and corresponding micro-plastic type information; the detection report comprises a point matching rate, an area ratio, a particle length, a particle width, and contents of different types of micro-plastics.

2. The method according to claim 1, wherein, Separating and extracting micro-plastics from the soil samples to obtain a micro-plastic sample, specifically comprising the following steps: Mixing and soaking the soil samples with Fenton reagent at a preset ratio to obtain a mixed solution; Performing water bath heating treatment on the mixed solution to obtain a sample after water bath heating; Cleaning the sample after water bath heating using a screen to obtain a cleaned sample; Based on the cleaned sample, using ZnCl2 solution to transfer and float residues, and collecting an upper-layer micro-plastic and organic matter mixture; Performing digestion treatment on the upper-layer micro-plastic and organic matter mixture, and adding L-ascorbic acid as a reducing agent to perform a reduction reaction to obtain a sample after reduction reaction; Using a polytetrafluoroethylene filter membrane to filter the sample after reduction reaction to obtain a micro-plastic sample.

3. The method according to claim 2, wherein, Fe in the Fenton reagent 2+ The volume ratio of the solution to the H2O2 solution is 1:50; the Fe 2+ The preparation process of the solution is as follows: 1.39 g of ferrous sulfate solid is added into 50 mL of RO water to obtain 50 mL of FeSO4 solution with a mass fraction of 2.78%; the mass fraction of the H2O2 solution is 30%; The density of the ZnCl2 solution is 1.5 g / cm3 3 , the floatation time is 1 h; The screen is a 600-mesh stainless steel screen; The temperature condition of the digestion treatment is 60 DEG C, the shaking bed rotation speed is 115 rmp, and the digestion duration is 8 h.

4. The method according to claim 1, wherein, Vacuum-filtering and transferring the micro-plastic sample to a filter membrane to be detected, specifically comprising the following steps: Transferring the micro-plastic sample to a beaker to perform mixing and stirring treatment and ultrasonic treatment to obtain a micro-plastic sample after ultrasonic treatment; Using a vacuum filtration device to filter the micro-plastic sample after ultrasonic treatment to the filter membrane to be detected.

5. The method according to claim 4, wherein, The filter membrane to be detected is a polyester gold-coated membrane with a pore size of 5 mu m, a thickness of 19 mu m, a diameter of 13 mm, and an effective filtration diameter of 10 mm.

6. The method according to claim 1, wherein, Performing Fourier infrared microscopic scanning on the filter membrane to be detected to obtain a Mosaic image, specifically comprising the following steps: Cleaning and wetting the surface of a gold mirror to obtain a cleaned and wetted gold mirror; Closely pasting the filter membrane to be detected on the surface of the cleaned and wetted gold mirror; Using an electric sample loading platform of a Fourier infrared microscope to scan the filter membrane to be detected row by row at a preset resolution to splice a Mosaic image; the splicing process of the Mosaic image comprises the following steps: extracting image features based on a SIFT algorithm, matching features using a Euclidean distance formula, calculating a homography matrix to realize image registration, and completing image splicing using an image fusion formula of feathering processing.

7. The method according to claim 6, wherein, The gradient amplitude formula is used for image feature extraction based on the SIFT algorithm, and the gradient amplitude formula is: ； ； ； ； wherein, denotes the magnitude of the gradient of the image at the coordinates , and denotes the gradient of the image in and direction, denotes the original image, denotes the image smoothed by a Gaussian filter, is a Gaussian function, is the standard deviation of the Gaussian kernel. The Euclidean distance formula is: ； wherein, denotes a feature point and a Euclidean distance between the feature point and the feature point denotes a coordinate of the feature point in the image, denotes a coordinate of the feature point in the image; The expression of the homography matrix is: ； wherein, H is a 3x3 homography matrix, denotes the coordinates of a pixel in the original image, denotes the corresponding coordinates of this pixel in the target registration image; The image fusion formula of the feathering processing is: ； wherein, is the fused image, and are two images to be stitched, , are the weights of the corresponding images, and .

8. The method according to claim 1, wherein, The Fourier infrared detection method is used to locate the particle points of the Mosaic image, and the spectral information of each particle point is collected, specifically including: An image recognition technology is used to segment the particle points of the Mosaic image, and the coordinate information of each particle point is determined. A Fourier infrared microscope spectrometer is used as a spectral acquisition device, the Fourier infrared microscope spectrometer is set to a reflection mode, and the spectral acquisition range and acquisition time are set. Based on the coordinate information of each particle point, the guide analysis function of the Fourier infrared microscope spectrometer is used to directionally collect the spectral information of each particle point in the Mosaic image. Based on the spectral information of each particle point, water vapor absorption peak and carbon dioxide peak interference correction processing is performed to obtain corrected spectral information; the corrected spectral information is used as the final spectral information of each particle point, which is used to match with a microplastic standard spectral library to determine the microplastic type information of each particle point, and a detection report is generated.

9. The method according to claim 1, wherein, The spectral information of each particle point is matched with a microplastic standard spectral library to determine the microplastic type information of each particle point, specifically including: The spectral information of each particle point is matched with the standard spectral information of the microplastic standard in the microplastic standard spectral library to determine the spectral matching degree of the spectral information of each particle point and the standard spectral information. According to the spectral matching degree of the spectral information of each particle point and the standard spectral information, the microplastic type information corresponding to the particle point is determined, including: When the spectral matching degree of the spectral information of the particle point and the standard spectral information is greater than or equal to 75%, the particle point is determined to be microplastic. When the spectral matching degree of the spectral information of the particle point and the standard spectral information is greater than or equal to 45% and less than 75%, the particle point is determined to be suspected microplastic.

Images