An in-situ testing method for electrolyte particulate matter

Abstract

The present invention discloses an in-situ testing method for electrolyte particulate matter, which belongs to the field of lithium battery technology; the method comprises the following steps: connecting a sample detection pool Flow Cell to an electrolyte production or use device, and providing a fixed detection window; in the process of the electrolyte sample to be tested flowing through the sample detection pool Flow Cell, a high-frequency imaging detector dynamically and continuously detects the particulate matter in the electrolyte sample to be tested at the detection window, captures optical pixels and takes a microscopic image; by processing, classifying and counting the image information of the particulate matter, a particle size distribution and morphology photo of the electrolyte particulate matter is formed, thereby guiding the production and use process of the electrolyte. The present invention is used on an electrolyte production or use device through an in-situ connection method, and does not require the electrolyte to be separately extracted or dried for testing, thereby achieving rapid and accurate testing, and the electrolyte after testing will not be contaminated or damaged by the equipment and can be recycled.

Description

Technical Field

[0001] The present invention relates to a method for detecting insoluble particles in electrolyte, and in particular to an in-situ testing method for electrolyte particles, belonging to the technical field of lithium batteries. Background Art

[0002] The quantity and morphology of particulate matter in electrolytes are currently key control indicators in electrolyte production and testing. They directly impact the level of impurity control in electrolytes and are a significant factor influencing their safety performance. Direct detection of particulate matter would provide a reliable basis for selecting electrolyte filters and monitor their lifecycle.

[0003] In the past, we used cleanliness testers or light resistance methods to observe the morphology of particles in liquids and measure their relative counts, i.e., concentration tests. The hardware used in cleanliness testers is a metallographic microscope with a small field of view, so the number of particles seen is extremely limited. Figure 1 As shown, the cleanliness tester can obtain particle images, but it cannot perform a large number of in-situ tests. The operation is complicated, and the sample preparation process will also produce trace amounts of hydrofluoric acid due to the encounter with water vapor in the air, which will damage and pollute the testing equipment and the environment. In addition, the particulate matter tested by the cleanliness meter needs to pass through a filter membrane to intercept large particles in the solution, and then be imaged as a whole after drying. This method has too many complicated pretreatments. The electrolyte produces a large amount of volatilization effect during the filtration operation, which is harmful to the human body. In addition, more impurities will be introduced in different operation links, resulting in inaccurate counting. The light obstruction method detects particles through the extinction method, which only provides an equivalent volume diameter. It cannot obtain particle morphology information, and cannot classify or trace the particles. The light obstruction method cannot directly observe the actual sample particles, so it cannot identify whether they are harmful substances.

[0004] Therefore, a faster in-situ detection method is needed during the production or injection process of the electrolyte to solve the problems of electrolyte contamination and damage during the detection process and inaccurate detection results in the above-mentioned electrolyte particle detection method. Summary of the Invention

[0005] The purpose of the present invention is to overcome the problems existing in the prior art and provide an in-situ testing method for electrolyte particulate matter, which is in-situ connected to the electrolyte production or use device to avoid complicated processes such as separate extraction and drying during the electrolyte detection process, thereby achieving rapid and accurate testing. The electrolyte after detection will not be contaminated or damaged by the equipment and can be recycled.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: an in-situ testing method for electrolyte particulate matter, comprising the following steps:

[0007] S1. Arrange in-situ detection equipment:

[0008] Connect a sample detection pool Flow Cell to the electrolyte production or use device, and set a fixed detection window. Use a high-precision micro-flow pump to control the flow of electrolyte into the detection window. Synchronously control the strobe frequency and dispersion width of the background light and the shutter time of the camera according to the set sampling rate.

[0009] S2. Collect particle images:

[0010] When the electrolyte sample to be tested flows through the sample detection pool Flow Cell, a high-frequency imaging detector dynamically and continuously detects particles in the electrolyte sample to be tested and captures optical pixels at the detection window. The tiny particles are magnified by a high-precision lens and captured into high-definition and high-resolution microscopic images by a high-sensitivity area array camera.

[0011] S3. Digitally analyze and count particle information:

[0012] 1) Preprocessing the microscopic image information captured in step S2 above, wherein the preprocessing process includes matrix convolution processing, VL transformation, Gaussian filtering, automatic white balance and automatic contrast, to remove background noise and obtain a high-quality microscopic image;

[0013] 2) Based on the discretization processing of the image by a high-sensitivity area array camera, each pixel on the imaging surface is obtained, which only represents the color of its surrounding area. A sub-pixel edge algorithm is used to perform grayscale recognition and extraction and boundary calculation on the pre-processed microscopic image. The grayscale value in the scattering area is set as the threshold. The scattered pixels are removed through automatic threshold analysis to extract the image contour that is closer to the actual particle.

[0014] 3) Using a high-precision syringe pump, the system accurately counts particles per unit volume of fluid. Dynamic analysis is performed on multiple microscopic images to determine whether background particles in the extracted image outlines are adhered to the liquid pool. Repeatedly captured particles are removed, and the number of particles is corrected.

[0015] 4) Perform particle size and shape analysis based on the grayscale and outline of the extracted particles, calculate the particle size and shape parameters of the particles, and normalize the data;

[0016] 5) Particle classification based on a combination of particle size and shape parameters, using methods including pattern recognition and model training to identify and analyze the source of particles;

[0017] S4. Analyze and process the digitized data to generate a particle size and shape report:

[0018] The particle size and shape report includes information from three dimensions: statistical layer, characteristic layer, and identity layer. The statistical layer includes a global particle overview and a portrait of a single particle, and provides a statistical view of the particle size distribution by volume and number. The characteristic layer includes a variety of particle size and shape indicators, and provides display and analysis of curve charts, bar charts, box plots, and scatter plots. With the support of the above characteristic parameters, the identity layer combines the calculated physical parameters to classify and identify the particles, forming a particle size distribution and morphology photo of the electrolyte particles.

[0019] In step S1 , the strobe frequency is set to 10-50 Hz, the strobe time is set to 1-500 μs, and the camera shutter time is set to 1-10 ms.

[0020] In step S1, the sample detection pool Flow Cell is provided with an optical surface using blue parallel light, the matching clip is 50-500 μm, the middle channel is 0.1-10 mm, and the size range of the tested particles is 0.3-1000 μm.

[0021] In step S2, the tiny particles are magnified by 0.75 to 9 times using a high-precision lens, and the resolution of the captured microscopic image is 12 to 18 million.

[0022] In step S2, the viscosity of the electrolyte is in the range of 0-500 cp; the particulate matter includes insoluble lithium salt aggregates, fibers, metal foreign matter and crystalline impurities in the electrolyte.

[0023] In step S3, the particle size and shape of the particulate matter include: 1) shape information includes spherical, elongated, translucent, opaque, and sharp; 2) particle size parameters include area, volume, convex hull perimeter, area equivalent diameter, volume equivalent diameter, perimeter equivalent diameter, Legendre ellipse major axis, Legendre ellipse minor axis, Feret maximum diameter, Feret minimum diameter, diameter geodesic length, and thickness; 3) particle shape parameters include ellipticity, aspect ratio, elongation, straightness, irregularity, compactness, expansion, filling rate, Waddell sphericity, roundness, firmness, convexity, average concavity, particle robustness, maximum concavity index, and roughness.

[0024] The particle size and shape calculation of the particulate matter includes the following parameters:

[0025] Particle size parameters: Calculate the normalized parameter volume equivalent diameter x from the volume V V , Calculate the normalized parameter area equivalent diameter x from the projected area A A , Calculate the surface area equivalent diameter x from the surface area S S , Calculate the equivalent diameter x from the circumference P P , Legendre ellipse major axis x Lmax and the minor axis x of the Legendre ellipse Lmin ; Feret's maximum diameter x Fmax and Feret's minimum diameter x Fmin ; diameter geodesic length x LG and thickness x E ;

[0026] Particle shape parameters: d imax is the maximum inscribed circle diameter, d imin is the minimum circumscribed circle diameter; A box =x Fmin .x LF , x LF is the Feret diameter perpendicular to the minimum Feret diameter; Waddell sphericity ψ, Roundness C, Robustness = A / A c , A c is the area of the convex hull of the particle boundary; convexity = P c / P,P c is the length of the convex shell of the particle boundary; the average concavity ψ FP , in is the angle-average Feret diameter, Particle robustness Ω1, Where ω1 is the number of erosions required to completely eliminate the contour; the maximum concavity index Ω2, Where ω2 is the angle between the convex polygon A and the c The number of erosions required for the relevant contour remnants to completely disappear; the concavity / robustness ratio Ω3, The roughness is described as the fractal dimension D F In the complex logarithmic coordinate diagram, the boundary perimeter P(λ) is linear with the step length λ, and the upper limit of the step length is λ = 0.3x Fmax , the equation of the line is log 10 P(λ)=(1-DF)log 10 λ+log 10 b, where b is the intercept of the fractal dimension graph.

[0027] The beneficial effects of the present invention are:

[0028] 1) In the method of the present invention, the electrolyte flows directly through the detection equipment, and there is no need to extract or dry the electrolyte separately for testing, thereby achieving rapid and accurate testing. Moreover, the electrolyte after testing will not be contaminated or damaged by the equipment and can be recycled.

[0029] 2) The method of the present invention uses dynamic images for particle size analysis. Image analysis provides a clear numerical distribution of particle shape parameters by quantifying the distribution values of various shapes rather than qualitatively describing various shapes. Therefore, it can accurately reflect the actual particle size and morphology of the sample, and can count, classify and trace the concentration of particles, thereby achieving precise particle control in different production links and locating the entrance of pollutants; and can monitor impurities and visible foreign matter in each link of the entire electrolyte production process. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 is a statistical diagram of the number of electrolyte particles in an embodiment of the present invention;

[0031] Figure 2 This is an optical structure diagram of the present invention;

[0032] Figure 3 This is a morphology image of electrolyte particles taken in an embodiment of the present invention;

[0033] Figure 4 are morphological descriptors of several different particles listed in the embodiments of the present invention;

[0034] Figure 5 This is the electrolyte particulate matter report generated in an embodiment of the present invention. DETAILED DESCRIPTION

[0035] The present invention will be further explained below with reference to the accompanying drawings and specific embodiments.

[0036] Example: Figure 1-5 As shown, the present invention provides an in-situ testing method for electrolyte particulate matter, comprising the following steps:

[0037] S1. Arrange in-situ detection equipment:

[0038] like Figure 2 The optical structure shown in the figure is that a sample detection cell FlowCell is connected to the electrolyte production or use device and is provided with a fixed detection window. A high-precision micro-flow pump is used to control the flow of electrolyte into the detection window. The background light strobe frequency is synchronously controlled to be 10-50Hz according to the set sampling rate, the strobe time is 1-500μs, and the dispersion width and the shutter time of the camera are 1-10ms.

[0039] Among them, the sample detection pool Flow Cell is equipped with an optical surface that uses blue parallel light. Its matching clip is 50-500μm, the middle channel is 0.1-10mm, and the size range of the tested particles is 0.3-1000μm.

[0040] S2. Collect particle images:

[0041] As the electrolyte sample to be tested flows through the sample detection pool of fixed thickness, a high-frequency pulsed short-wave light source continuously illuminates the particles in the liquid pool at the detection window, dynamically and continuously detecting the particles in the electrolyte sample to be tested by capturing optical pixels. A high-precision telecentric lens magnifies the tiny particles by 0.75 to 9 times, and a high-sensitivity area array camera captures all the tiny particles into high-definition and high-resolution microscopic images and stores them. The resolution of the microscopic images is 12-18 million.

[0042] The viscosity of the electrolyte ranges from 0 to 500 cp; the particulate matter includes impurities such as insoluble lithium salt aggregates, fibers, metal foreign matter and crystals in the electrolyte.

[0043] S3. Digitally analyze and count particle information:

[0044] 1) Preprocessing the microscopic image information captured in step S2 above. The preprocessing process includes matrix convolution processing, VL transformation, Gaussian filtering, automatic white balance and automatic contrast to remove background noise in order to obtain high-quality microscopic images.

[0045] 2) Based on the discretization processing of the image by a high-sensitivity area array camera, each pixel on the imaging surface is obtained, which only represents the color of its surrounding area. A sub-pixel edge algorithm is used to perform grayscale recognition and extraction and boundary calculation on the pre-processed microscopic image. The grayscale value in the scattering area is set as the threshold. The scattered pixels are removed through automatic threshold analysis to extract the image contour that is closer to the actual particle.

[0046] The image grayscale varies depending on the size or shape of the particles, so the threshold settings are different. The threshold setting needs to be sufficient to remove scattered pixels and obtain a clearer and more realistic particle image outline.

[0047] 3) Using a high-precision syringe pump, the system accurately counts particles per unit volume of fluid. Dynamic analysis is performed on multiple microscopic images to determine whether background particles in the extracted image outlines are adhered to the liquid pool. Repeatedly captured particles are removed, and the number of particles is corrected.

[0048] like Figure 1 As shown, it is a statistical chart of the number of electrolyte particles based on camera capture and storage, combined with a high-precision injection pump to accurately count the particles in a unit volume of fluid. The sampling number is 4992 and the measurement range is 0.3 to 1000 μm.

[0049] 4) Perform particle size and shape analysis based on the grayscale and outline of the extracted particles, calculate the particle size and shape parameters of the particles, and normalize the data;

[0050] like Figure 3 The following is the particle size and shape information obtained by computer processing of the particle grayscale and outline. The particle size and shape of the particle include:

[0051] (1) Shape information includes spherical, long, translucent, opaque, and sharp;

[0052] (2) Particle size parameters include area, volume, convex hull perimeter, area equivalent diameter, volume equivalent diameter, perimeter equivalent diameter, Legendre ellipse major axis, Legendre ellipse minor axis, Feret's maximum diameter, Feret's minimum diameter, diameter geodesic length, and thickness;

[0053] (3) Particle shape parameters include ellipticity, aspect ratio, elongation, straightness, irregularity, compactness, expansion, filling fraction, Waddell sphericity, roundness, solidity, convexity, average concavity, particle robustness, maximum concavity index, and roughness.

[0054] like Figure 4 As shown in Figure 2, there are several morphological descriptors for different particles. The particle size and shape calculation of the particles includes the following parameters:

[0055] Particle size parameters: Calculate the normalized parameter volume equivalent diameter x from the volume V V , Calculate the normalized parameter area equivalent diameter x from the projected area A A , Calculate the surface area equivalent diameter x from the surface area S S , Calculate the equivalent diameter x from the circumference P P , Legendre ellipse major axis x Lmax and the minor axis x of the Legendre ellipse Lmin ; Feret's maximum diameter x Fmax and Feret's minimum diameter x Fmin ; diameter geodesic length x LG and thickness x E ;

[0056] Particle shape parameters: For particles that are not very long, For very long and thin particles (such as fibers), For very elongated particles (the inverse of the curvature), d imax is the maximum inscribed circle diameter, d imin is the minimum circumscribed circle diameter; in terms of the overall shape of the particle, it indicates the degree to which the particle (or its projection surface) is close to a circle. The ratio of the Feret cell area to the projected area, A box =x Fmin .x LF , x LF is the Feret diameter perpendicular to the minimum Feret diameter; Waddell sphericity ψ, Roundness C, Robustness = A / A c , A c is the area of the convex hull of the particle boundary; convexity = P c / P,P c is the length of the convex shell of the particle boundary; the average concavity ψ FP , in is the angle-average Feret diameter, Particle robustness Ω1, Where ω1 is the number of erosions required to completely eliminate the contour; the maximum concavity index Ω2, Where ω2 is the angle between the convex polygon A and the c The number of erosions required for the relevant contour remnants to completely disappear; the concavity / robustness ratio Ω3, The roughness is described as the fractal dimension D F In the complex logarithmic coordinate diagram, the boundary perimeter P(λ) is linear with the step length λ, and the upper limit of the step length is λ = 0.3x Fmax , the equation of the line is log 10 P(λ)=(1-DF)log 10 λ+log 10 b, where b is the intercept of the fractal dimension graph.

[0057] 5) Particle classification based on a combination of particle size and shape parameters, using methods including pattern recognition and model training to identify and analyze the source of particles;

[0058] Based on the calculated particle size and shape indicators, the results of different indicator combinations are analyzed. This analysis can classify the particles based on the existing model training method.

[0059] The classification method is as follows: particles with a roundness of 0.95 or above are round, and particles with a roundness of 0.87 are square.

[0060] S4. Analyze and process the digitized data to generate a particle size and shape report:

[0061] The particle size and shape report includes information from three dimensions: statistical layer, characteristic layer, and identity layer. The statistical layer includes a global particle overview and a portrait of a single particle, and provides a statistical view of the particle size distribution by volume and number. The characteristic layer includes a variety of particle size and shape indicators, and provides display and analysis of curve charts, bar charts, box plots, and scatter plots. With the support of the above characteristic parameters, the identity layer combines the calculated physical parameters to classify and identify the particles, forming a particle size distribution and morphology photo of the electrolyte particles.

[0062] like Figure 5 As shown, a large amount of particulate matter data is finally counted and processed to form a particle size and shape report. The calculated particle size and shape data are classified and counted, and the source of the particulate matter is determined based on the total number and morphology of the particles reflected in the report. This is used to guide the source of particulate impurities or product defects in the production process, so as to help production companies improve existing process flows.

[0063] The present invention provides an in-situ testing method for electrolyte particulate matter, in which a testing device is connected in situ to an electrolyte production or use device, avoiding complicated processes such as separate extraction and drying during electrolyte testing, thereby achieving rapid and accurate testing. The electrolyte after testing will not be contaminated or damaged by the equipment, can be recycled, and the judgment of impurities in the electrolyte is more accurate.

[0064] The above description is only used to illustrate the technical solution of the present invention and is not intended to limit it. Other modifications or equivalent substitutions made to the technical solution of the present invention by ordinary technicians in this field should be included in the scope of the claims of the present invention as long as they do not depart from the spirit and scope of the technical solution of the present invention.

Claims

1. An in-situ testing method for electrolyte particulate matter, characterized by: The following steps are involved: S1. Arrange in-situ detection equipment: Connect a sample detection pool Flow Cell to the electrolyte production or use device, and set a fixed detection window. Use a high-precision micro-flow pump to control the flow of electrolyte into the detection window. Synchronously control the strobe frequency and dispersion width of the background light and the shutter time of the camera according to the set sampling rate. S2. Collect particle images: When the electrolyte sample to be tested flows through the sample detection pool Flow Cell, a high-frequency imaging detector dynamically and continuously detects particles in the electrolyte sample to be tested and captures optical pixels at the detection window. The tiny particles are magnified by a high-precision lens and captured into high-definition and high-resolution microscopic images by a high-sensitivity area array camera. S3. Digitally analyze and count particle information: 1) Preprocessing the microscopic image information captured in step S2 above, wherein the preprocessing process includes matrix convolution processing, VL transformation, Gaussian filtering, automatic white balance and automatic contrast, to remove background noise and obtain a high-quality microscopic image; 2) Based on the discretization processing of the image by a high-sensitivity area array camera, each pixel on the imaging surface is obtained, which only represents the color of its surrounding area. A sub-pixel edge algorithm is used to perform grayscale recognition and extraction and boundary calculation on the pre-processed microscopic image. The grayscale value in the scattering area is set as the threshold. The scattered pixels are removed through automatic threshold analysis to extract the image contour that is closer to the actual particle. 3) Using a high-precision syringe pump, the system accurately counts particles per unit volume of fluid. Dynamic analysis is performed on multiple microscopic images to determine whether background particles in the extracted image outlines are adhered to the liquid pool. Repeatedly captured particles are removed, and the number of particles is corrected. 4) Perform particle size and shape analysis based on the grayscale and outline of the extracted particles, calculate the particle size and shape parameters of the particles, and normalize the data; 5) Particle classification based on a combination of particle size and shape parameters, using methods including pattern recognition and model training to identify and analyze the source of particles; S4. Analyze and process the digitized data to generate a particle size and shape report: The particle size and shape report includes information from three dimensions: statistical layer, characteristic layer, and identity layer. The statistical layer includes a global particle overview and a portrait of a single particle, and provides a statistical view of the particle size distribution by volume and number. The characteristic layer includes a variety of particle size and shape indicators, and provides display and analysis of curve charts, bar charts, box plots, and scatter plots. With the support of the above characteristic parameters, the identity layer combines the calculated physical parameters to classify and identify the particles, forming a particle size distribution and morphology photo of the electrolyte particles.

2. The in-situ testing method for electrolyte particulate matter according to claim 1, characterized in that: In step S1 , the strobe frequency is set to 10-50 Hz, the strobe time is set to 1-500 μs, and the camera shutter time is set to 1-10 ms.

3. The in-situ testing method for electrolyte particulate matter according to claim 1, characterized in that: In step S1, the sample detection pool Flow Cell is provided with an optical surface using blue parallel light, the matching clip is 50-500 μm, the middle channel is 0.1-10 mm, and the size range of the tested particles is 0.3-1000 μm.

4. The in-situ testing method for electrolyte particulate matter according to claim 1, characterized in that: In step S2, the tiny particles are magnified by 0.75 to 9 times using a high-precision lens, and the resolution of the captured microscopic image is 12 to 18 million.

5. The in-situ testing method for electrolyte particulate matter according to claim 1, characterized in that: In step S2, the viscosity of the electrolyte is in the range of 0-500 cp; the particulate matter includes insoluble lithium salt aggregates, fibers, metal foreign matter and crystalline impurities in the electrolyte.

6. The in-situ testing method for electrolyte particulate matter according to claim 1, characterized in that: In step S3, the particle size and shape of the particulate matter include: 1) shape information includes spherical, elongated, translucent, opaque, and sharp; 2) particle size parameters include area, volume, convex hull perimeter, area equivalent diameter, volume equivalent diameter, perimeter equivalent diameter, Legendre ellipse major axis, Legendre ellipse minor axis, Feret maximum diameter, Feret minimum diameter, diameter geodesic length, and thickness; 3) particle shape parameters include ellipticity, aspect ratio, elongation, straightness, irregularity, compactness, expansion, filling rate, Waddell sphericity, roundness, firmness, convexity, average concavity, particle robustness, maximum concavity index, and roughness.

7. The in-situ testing method for electrolyte particulate matter according to claim 6, characterized in that: The particle size and shape calculation of the particulate matter includes the following parameters: Particle size parameters: Calculate the normalized parameter volume equivalent diameter x from the volume V V , Calculate the normalized parameter area equivalent diameter x from the projected area A A , Calculate the surface area equivalent diameter x from the surface area S S , Calculate the equivalent diameter x from the circumference P P , Legendre ellipse major axis x Lmax and the minor axis x of the Legendre ellipse Lmin ; Feret's maximum diameter x Fmax and Feret's minimum diameter x Fmin ; diameter geodesic length x LG and thickness x E ; Particle shape parameters: d imax is the maximum inscribed circle diameter, d imin is the minimum circumscribed circle diameter; A box =x Fmin .x LF , x LF is the Feret diameter perpendicular to the minimum Feret diameter; Waddell sphericity ψ, Roundness C, Robustness = A / A c , A c is the area of the convex hull of the particle boundary; convexity = P c / P,P c is the length of the convex shell of the particle boundary; the average concavity ψ FP , in is the angle-average Feret diameter, Particle robustness Ω1, Where ω1 is the number of erosions required to completely eliminate the contour; the maximum concavity index Ω2, Where ω2 is the angle between the convex polygon A and the c The number of erosions required for the relevant contour remnants to completely disappear; the concavity / robustness ratio Ω3, The roughness is described as the fractal dimension D F In the complex logarithmic coordinate diagram, the boundary perimeter P(λ) is linear with the step length λ, and the upper limit of the step length is λ = 0.3x Fmax , the equation of the line is log 10 P(λ)=(1-D F )log 10 λ+log 10 b, where b is the intercept of the fractal dimension graph.

Images