Raman imaging method based on PLS and quantitative analysis method of polymer material

Abstract

The present application relates to a kind of PLS-based raman imaging method and the quantitative analysis method of polymer material, the imaging method includes (1) start raman testing instrument and calibration;(2) to the sample with single state change is carried out large spot raman test;(3) after using raman spectrum analysis technique to the raman spectrum data obtained in step (2) is preprocessed, using PLS is analyzed, and thus establish state-spectrum model;(4) using the wavelength range where characteristic peak is located carries out raman mapping scanning test;(5) using raman spectrum analysis technique to the mapping result obtained in step (4) is preprocessed, to make mapping scanning data and the model obtained in step (3) compatible;(6) according to the peak intensity of characteristic peak or the model to the mapping scanning data obtained in step (5) is plotted, to obtain distribution diagram.The present application patent broadens raman imaging application field, and it is important to obtain the sample topography and component distribution under microscale.

Description

Technical Field

[0001] This invention belongs to the field of Raman spectroscopy characterization, and relates to a Raman imaging method based on PLS (partial least squares) and a quantitative analysis method for polymer materials, particularly a Raman imaging characterization method for polymer materials based on PLS (partial least squares). Background Technology

[0002] Raman spectroscopy is a type of scattering spectroscopy based on the Raman scattering effect. It analyzes the scattered spectra with frequencies different from the incident light to obtain information about molecular vibrations and rotations, and is applied to molecular structure research. For polymers, Raman spectroscopy can effectively characterize their internal structural differences. Confocal Raman spectroscopy combines confocal microscopy with Raman spectroscopy, allowing scanning of small regions on the material surface. The large spot scanning function allows the spot to move within the region during the test, thus obtaining regional Raman spectra, which is of great significance for Raman testing of microscopically inhomogeneous samples.

[0003] Raman imaging divides the test area into arrays of equally spaced points, scans each point, and stores the data as spectral data. Patent CN201910178930.9 discloses a method for determining the order degree of graphite using Raman mapping. This method uses Raman spectroscopy to detect the characteristic G and D peaks of graphite and marks their ratios with different colors, thereby obtaining a graphite order degree distribution map.

[0004] Patent CN201810312121.8 describes a method for characterizing lithium iron phosphate-amorphous carbon composite materials using Raman imaging. This method uses the P0 peak of lithium iron phosphate as a characteristic peak and obtains a high-resolution distribution map of the composite structure using Raman mapping. This method is convenient, fast, and can achieve non-destructive testing.

[0005] Dow Chemical Company describes a method for olefin synthesis and real-time monitoring in patent CN201180068142.6. This monitoring method uses Raman spectroscopy to measure the concentration of carbon-carbon unsaturated states within the system in real time, thereby characterizing the degree of polymerization. This method is simple, rapid, and does not affect the polymerization process.

[0006] Researchers can extract the intensity or integrated area of ​​characteristic peaks from the spectrum and reconstruct the image according to the original test position to obtain a distribution image based on molecular structure. Raman spectroscopy itself has a strong ability to identify the carbon skeleton, so it can be used for imaging in fields such as polymers and carbon materials. However, Raman testing itself is easily affected by factors such as fluorescence and impurities, and the peaks in mixtures can mask each other, making it difficult to intuitively analyze the characteristic peaks containing information. Summary of the Invention

[0007] To overcome the shortcomings of existing technologies, this invention provides a PLS-based Raman imaging method and a quantitative analysis method for polymer materials. For complex mixture samples, a new processing method is proposed to enable local quantitative analysis of complex mixture samples.

[0008] To achieve the first aspect of the aforementioned objective, the technical solution adopted by the present invention is as follows:

[0009] A PLS-based Raman imaging method, the Raman imaging method comprising the following steps:

[0010] (1) Turn on the Raman test instrument and calibrate it using a standard sample to ensure that the test peak position is displayed accurately.

[0011] (2) Perform large-spot Raman testing on samples with a single state change;

[0012] (3) After preprocessing the Raman spectral data obtained in step (2) using Raman spectroscopy analysis technology, PLS is used to analyze it, and a state-spectral model is established accordingly.

[0013] (4) Raman mapping scan test is performed using the wavelength range of the characteristic peak;

[0014] (5) The mapping results obtained in step (4) are preprocessed using Raman spectroscopy to make the mapping scanning data compatible with the model obtained in step (3);

[0015] (6) Plot the mapping scan data obtained in step (5) based on the peak intensity of the characteristic peak or the model to obtain the distribution map.

[0016] In this invention, step (1) is to calibrate the Raman instrument, for example, by using a silicon standard sample to confirm the accuracy of the peak position. This is well known in the art and will not be described in detail here.

[0017] In step (2) of the present invention, a large-spot Raman test is performed on the sample exhibiting a single state change. In one embodiment, the state change includes changes in component ratio, temperature, pressure, atmosphere, crystallinity, and stress. Specifically, it can be a change in component ratio, that is, the state refers to the component ratio. For example, for polymer blends, the state can refer to the component composition ratio of the polymer blend, such as a blend of PGA (polyglycolic acid) and PBAT (polybutylene terephthalate), or for polymer copolymers, the state refers to the monomer ratio forming the copolymer, such as an ethylene-propylene copolymer.

[0018] In one embodiment, in step (2), the sample is a mixture of two or more components with Raman differences or an apparent homogeneous system composed of the same substance with Raman differences in different states; preferably, the sample is a polymer blend or polymer copolymer sample, more preferably a polymer blend composed of both polymer components or a polymer copolymer obtained by polymerizing two monomers.

[0019] According to the Raman imaging method of the present invention, preferably, in steps (3) and (5), the preprocessing method includes smoothing and denoising, baseline correction, and normalization. The above preprocessing methods are well known in the art. For example, the above preprocessing can be performed using LabSpec5 software or Origin software. For example, "smoothing and denoising" can be completed first by using the polynomial moving average filtering method, then "baseline correction" can be completed by using the polynomial fitting method, and finally "normalization" can be completed by selecting the spectral peak intensity as a reference value. In one embodiment, the processed spectrum is converted into data and PLS analysis is performed to obtain the fitting model and load diagram (for example, by using JMP software, or Minitab and MATLAB software can also be used). The characteristic peak positions are judged and analyzed based on the load diagram.

[0020] In one embodiment of the present invention, in step (2), the sample is a polymer blend with different component ratios, such as a PGA and PBAT blend with PGA content in the range of 10wt%-90wt%; in step (4), the sample in step (2) is subjected to Raman mapping scanning test using the wavelength range of the characteristic peaks; preferably, the sample is a PGA and PBAT blend sample, such as a PGA content of 20wt%, 50wt%, or 90wt%.

[0021] In one embodiment of the present invention, in step (2), the sample is a standard sample of ethylene-propylene random copolymer with different monomer content ratios, for example, the ethylene monomer content ratio varies in the range of 0wt%-20wt%, such as 0wt% (i.e., excluding), 4wt%, 9wt%, 11wt%, 16wt%, and 18wt%; in step (4), Raman mapping scan is performed on another ethylene-propylene copolymer sample to be tested using the wavelength range of the characteristic peak.

[0022] In this invention, in step (4), the wavelength range of the characteristic peak can be obtained by consulting literature; in one embodiment, preferably, in step (4), the wavelength range of the characteristic peak is obtained by a load diagram, which is derived from the PLS-based analysis model in step (3); wherein, based on the load diagram, the characteristic peaks that affect the spectrum due to the state change can be found, specifically, based on the load diagram, the wavelength range ab (cm) of the characteristic peak is obtained. －1 In the given information, (ba) should be greater than 100, and the wavelength range should encompass at least n effective characteristic peaks with the largest absolute Y-axis values, where n is not less than 3, for example, 5, 10, or 20. That is, the absolute Y-axis values ​​corresponding to all effective characteristic peaks are arranged in ascending order, and the effective characteristic peaks corresponding to the first n largest absolute Y-axis values ​​should be included. An effective characteristic peak refers to a characteristic peak with a peak width greater than 10 wavenumbers. "Included" means that the wavelength of the X-axis corresponding to the characteristic peak is within the wavelength range. In one embodiment, the wavelength range can also be a wavelength range encompassing all characteristic peaks.

[0023] According to the Raman imaging method of the present invention, in one embodiment, in step (2), the large spot Raman test range is set between 1μm*1μm and 100μm*100μm, such as 10μm*10μm, 20μm*20μm, 50μm*50μm or 80μm*80μm, which can be appropriately selected by those skilled in the art as needed.

[0024] According to the Raman imaging method of the present invention, in one embodiment, in step (4), the size of the scanning area is set between 1μm*1μm and 100μm*100μm, such as 10μm*10μm, 20μm*20μm, 50μm*50μm or 80μm*80μm, which can be appropriately selected by those skilled in the art as needed.

[0025] According to the Raman imaging method of the present invention, in one embodiment, in step (4), the single scan time can be set to 0.1s-50s, such as 1s, 2s, 5s, 10s or 20s, and the scan step size can be 0.1μm to 10μm, such as 0.2μm, 0.5μm, 0.8μm, 1μm, 5μm to 8μm, and multiple scans can be performed on a single point.

[0026] According to the Raman imaging method of the present invention, in one embodiment, in step (6), based on the characteristic peak intensity shown in the load map or by directly substituting the mapping result into the model, imaging is performed on each sample region to obtain a proportional distribution image.

[0027] In a preferred embodiment of the present invention, the Raman imaging method includes:

[0028] (1) First, obtain the sample, turn on the Raman tester, and use the silicon standard sample for calibration to confirm the accuracy of the peak position; first, use 532nm and 325nm lasers for scanning, preferably 532nm;

[0029] (2) Take samples with different component ratios and perform large spot Raman mode region scanning. The preferred scanning range is 10μm*10μm to 100μm*100μm. The obtained spectra are preprocessed using Raman spectroscopy analysis technology.

[0030] (3) Convert the processed spectrum into data and perform PLS analysis to obtain the fitting model and load diagram. Based on the load, the characteristic peak positions are judged and analyzed.

[0031] (4) Perform Raman mapping scans on each sample or the sample to be tested (preferably with a scanning area of ​​20μm*20μm to 50μm*50μm, a scanning step size of 0.2μm to 0.8μm, and a single scan time of 1-4s) to obtain mapping data;

[0032] (5) The mapping results were preprocessed using Raman spectroscopy to ensure compatibility with the model;

[0033] (6) Based on the characteristic peak intensity shown in the load diagram or by directly substituting the mapping result into the model, imaging and mapping are performed on each sample region to obtain its proportional distribution image.

[0034] To achieve the second aspect of the above-mentioned objective, the present invention also provides a method for quantitative analysis of polymer materials, wherein a distribution map is first obtained according to the Raman imaging method described above, and then the local quantitative composition distribution of the polymer material is obtained based on the distribution map.

[0035] In this invention, the term "Raman difference" refers to the difference in Raman spectral data obtained by Raman testing in step (2).

[0036] Compared with the prior art, the present invention has the following advantages:

[0037] This method images samples by detecting molecular vibrational modes, thereby obtaining the sample structure and morphology. The method involves obtaining regional Raman spectra of samples under different conditions using a large-spot scanning mode; using PLS (partial least squares) analysis to compare the impact of sample changes on the spectrum, identifying relevant characteristic peaks and establishing a model; and using a mapping mode for scanning and visually characterizing the intensity of characteristic peaks using imaging functions. This invention broadens the application field of Raman imaging and is of great significance for obtaining sample morphology and component distribution at the microscale; it also proposes a new processing method for the complex composition of polymers. Compared with previous characterization methods, this method offers more flexible testing methods and a wider range of applications.

[0038] This invention is not merely a simple combination of conventional PLS modeling and mapping analysis; its significance lies in the quantitative study of inhomogeneity at the microscale of materials. For polymer materials, inhomogeneity is often manifested at the micrometer scale, while uniformity can be achieved at the millimeter-scale macroscopic level. Conventional Raman scanning test spot diameter is 1-2 micrometers, and the obtained spectral data only represents the composition data within the 1-2 micrometer range (for liquid substances such as crude oil, the scanning scale can be ignored because it is also uniform at the micrometer scale, and therefore its micromorphology does not need to be studied). In real-world polymer materials, due to the prevalence of inhomogeneity, a mixture of 10% A and 90% B can exhibit local proportions that vary from 100% A and 0% B to 0% A and 100% B at the microscale (existing in the form of localized crystallization and agglomeration), but the macroscopic proportion remains 1:9. Therefore, by using a large spot scanning method, such as expanding the scanning range to 100 micrometers * 100 micrometers, the measured macroscopic spectrum can reflect the uniform composition ratio of the material, which can be used for PLS modeling. Mapping tests using microscopic light spots with micrometer diameters yield a large number of microscopic local spectra. These spectra are then compared with macroscopic spectral models, and combined with the location information of each point, the concentration distribution at the microscale can be calculated, and the size of each phase and the changes in the proportion within each phase region can be observed. This method allows for mapping analysis of any blended material lacking detailed theoretical research; simply using blends with different concentration ratios is sufficient to observe the size of the internal phase regions and quantitatively characterize material compatibility.

[0039] For polymer materials, it's impossible to conduct detailed studies on the blending of any two materials and explain the meaning of each peak. PLS (Profilometry-based modeling) can provide correlation evidence; for example, peak A is positively correlated with substance A, while peak B is negatively correlated, without needing to explain the specific reasons for the correlation. This provides the theoretical support needed for mapping and greatly expands its application scope. Even when combining conventional PLS modeling with mapping analysis, without discussing the macroscopic and microscopic differences in the measured spectral range, an effective model cannot be established, and therefore, quantitative analysis of local areas is impossible. Attached Figure Description

[0040] Figure 1 This is a PLS fitting load diagram of Embodiment 1 of the present invention;

[0041] Figure 2 This refers to the sample mapping data from Example 1 of the present invention.

[0042] Figure 3 This is a mapping imaging effect diagram of Embodiment 1 of the present invention;

[0043] Figure 4 This is the Raman spectrum of the homopolymer in Example 2 of the present invention;

[0044] Figure 5 This is a graph of the PLS fitting model coefficients for Embodiment 2 of the present invention;

[0045] Figure 6 This is a PLS fitting load diagram of Embodiment 2 of the present invention;

[0046] Figure 7 This refers to the sample mapping data for Example 2 of the present invention;

[0047] Figure 8 This is a mapping imaging effect diagram of Embodiment 2 of the present invention; wherein,

[0048] a is an image of the actual object under an optical microscope, b is a mapping image, and c is an image showing the two overlapping effects.

[0049] Figure 9 The standard Raman spectrum of PBAT is shown in Comparative Example 1.

[0050] Figure 10 The standard Raman spectrum of PGA in Comparative Example 1;

[0051] Figure 11 Raman spectroscopy was performed on the PGA-PBAT blend sample of Comparative Example 1.

[0052] Figure 12 Optical image of the polyethylene-polypropylene block copolymer in Comparative Example 2;

[0053] Figure 13 The standard Raman spectra of polyethylene and polypropylene in Comparative Example 2 are shown.

[0054] Figure 14 The mapping Raman spectrum of the polyethylene-polypropylene block copolymer in Comparative Example 2 is shown.

[0055] Figure 15 Mapping imaging of the polyethylene-polypropylene block copolymer in Comparative Example 2. Detailed Implementation

[0056] The present invention will be further described below with reference to the embodiments and accompanying drawings. However, the present invention is not limited to the listed embodiments, but should also include equivalent improvements and modifications to the technical solutions defined in the appended claims of the present invention.

[0057] Example 1

[0058] First, three PGA and PBAT blend samples were obtained, with PGA contents of 20 wt% (Ratio 1), 50 wt% (Ratio 2), and 80 wt% (Ratio 3), respectively. The Raman spectrometer was turned on (using a 532 nm laser for scanning, the same below), and the peak positions were confirmed to be accurate using silicon standard samples.

[0059] Each blend sample underwent large-spot Raman mode scanning with a scanning range of 20 μm * 20 μm. The resulting spectra were then preprocessed (smoothing, noise reduction, baseline correction, and normalization, hereinafter the same). The processed spectra were converted into data and analyzed using PLS (jmp software, hereinafter the same), yielding the component ratio-spectral model and loading plots. The loading plots are shown below. Figure 1 As shown.

[0060] Based on the load diagram, the correlation between each peak position and the component ratio can be obtained. A positive load value for a peak position in the load diagram indicates a positive correlation between the test intensity of that peak position and the component ratio, and vice versa. According to the load diagram, the wavelength range of the characteristic peaks is selected as 400-1800 nm.

[0061] Raman mapping scans were performed on each sample (scanning area 50μm*50μm, scan step size 0.5μm, single scan time 2s) to obtain mapping data, such as... Figure 2 As shown.

[0062] Based on the intensity of characteristic peaks, imaging and mapping of each sample region can yield its proportional distribution image, such as... Figure 3 As shown.

[0063] Without PLS analysis, traditional methods cannot pinpoint the characteristic Raman signal due to the strong internal fluorescence of the sample, thus making it impossible to plot its concentration. The PLS method can not only identify and determine characteristic peaks but also provide the basis for their determination.

[0064] Example 2

[0065] First, six standard samples of ethylene-propylene random copolymers with different proportions (containing ethylene monomers at percentages of 0, 4 wt%, 9 wt%, 11 wt%, 16 wt%, and 18 wt%) and the ethylene-propylene copolymer sample to be characterized were obtained. The Raman spectrometer was turned on and calibrated using silicon standard samples to confirm the accuracy of the peak positions.

[0066] First, a large-spot Raman mode scan was performed on each sample, with a scanning range of 100 μm * 100 μm. The resulting spectra were then preprocessed to obtain Raman spectra of homopolymers with different formulations, such as... Figure 4 As shown.

[0067] The processed spectra are converted into data and analyzed using PLS to obtain the ethylene concentration-Raman spectrum model and loading diagram. The model is as follows: Figure 5 As shown, the load diagram is as follows Figure 6 As shown, the wavelength range of 2800-3000 is selected.

[0068] Raman mapping scans (scanning area 20μm*20μm, scan step size 0.5μm, single scan time 3s) were performed on the ethylene-propylene copolymer sample to be characterized to obtain mapping data, such as... Figure 7 As shown.

[0069] The mapping data consists of 1600 Raman spectra. Substituting the processed Raman data into the above model yields an image of the internal polyethylene distribution and the polyethylene concentration at each point. For example... Figure 8 As shown, a is the actual image under an optical microscope, b is the mapping image, and c is the overlapping effect of a and b.

[0070] Modeling Raman spectra and concentrations using the PLS method not only allows for the mapping of internal ethylene aggregation distribution but also enables the utilization of... Figure 8 The -b scale allows for a direct reading of ethylene distribution data at microscale points.

[0071] Comparative Example 1

[0072] Analysis of PGA-PBAT blends using conventional Raman mapping:

[0073] By consulting relevant materials, the standard Raman spectra of PBAT and PGA can be obtained, such as... Figure 9 and 10As shown. The measured Raman spectroscopy of the PGA-PBAT blend sample was also obtained, as shown below. Figure 11 As shown.

[0074] Depend on Figure 9 , Figure 10 and Figure 11 The comparison revealed that due to fluorescence interference in the actual sample and the overlapping and masking of characteristic peaks in the two substances, the characteristic Raman peaks in the standard spectrum were not visible in the measured Raman spectrum, making it impossible to generate a mapping image using traditional methods. Therefore, it is necessary to use multivariate data analysis methods (such as PLS) to investigate the effect of increasing the proportion of a certain component on the spectrum, thereby finding the peak position that can be used to standardize the component and generate a Raman image using that peak position.

[0075] Comparative Example 2

[0076] like Figure 12 The image shown is an optical image of a polyethylene-polypropylene block copolymer. To investigate the Raman mapping image of a polyethylene-polypropylene block copolymer, the traditional method is to compare it with standard spectra, such as... Figure 13 As shown, the characteristic peak positions of ethylene and propylene are located and imaging is performed based on these peaks.

[0077] The comparison shows that 1150cm -1 The peak position is a characteristic peak of propylene, at 830 cm⁻¹. -1 The peak position is a characteristic peak of polyethylene. Direct Raman mapping scanning was performed on it, with a scanning area of ​​10μm*10μm, a scanning step size of 0.5μm, and a single scan time of 2s, yielding the following results: Figure 14 The image shows the mapping Raman spectrum of the polyethylene-polypropylene block copolymer.

[0078] With 830cm -1 By plotting the intensity of the characteristic peaks, a mapping image can be obtained, such as... Figure 15 As shown in the diagram, the relative concentration changes in each region can be seen from the mapping graph, but since no model has been established, the specific concentration values ​​cannot be determined.

[0079] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the implementation of the present invention. Those skilled in the art can make other variations or modifications based on the above description. It is impossible to exhaustively list all embodiments here. All obvious variations or modifications derived from the technical solutions of the present invention are within the spirit and scope of the present invention.

Claims

1. A Raman imaging method based on PLS, characterized in that, The Raman imaging method includes the following steps: (1) Turn on the Raman test instrument and calibrate it using a standard sample to ensure accurate display of the test peak position; (2) Perform large-spot Raman testing on samples with a single state change; (3) After preprocessing the Raman spectral data obtained in step (2) using Raman spectroscopy analysis technology, PLS is used to analyze it, and a state-spectral model is established accordingly. (4) Raman mapping scan test is performed using the wavelength range of the characteristic peak; wherein, the wavelength range of the characteristic peak is obtained by the load diagram, and the load diagram is derived from the model based on PLS analysis in step (3); (5) The mapping results obtained in step (4) are preprocessed using Raman spectroscopy to make the mapping scanning data compatible with the model obtained in step (3); (6) Plot the mapping scan data obtained in step (5) based on the peak intensity of the characteristic peak or the model to obtain the distribution map.

2. The Raman imaging method according to claim 1, characterized in that, In step (2), the sample is a mixture of two or more components with Raman differences or an apparent homogeneous system composed of the same substance with Raman differences in different states.

3. The Raman imaging method according to claim 1, characterized in that, In step (2), the state changes include changes in component ratio, temperature, pressure, atmosphere, crystallinity, and stress.

4. The Raman imaging method according to claim 2, characterized in that, In step (2), the state changes include changes in component ratio, temperature, pressure, atmosphere, crystallinity, and stress.

5. The Raman imaging method according to any one of claims 1-4, characterized in that, In steps (3) and (5), the preprocessing method is divided into three steps, in the following order: first, "smoothing and denoising" is completed by polynomial moving average filtering method; then, "baseline correction" is completed by polynomial fitting method; and finally, "normalization" is completed by selecting the spectral peak intensity as a reference value.

6. The Raman imaging method according to any one of claims 1-4, characterized in that, In step (2), the sample is a polymer blend with different component ratios; in step (4), the sample in step (2) is subjected to Raman mapping scan test using the wavelength range of the characteristic peaks.

7. The Raman imaging method according to claim 5, characterized in that, In step (2), the sample is a polymer blend with different component ratios; in step (4), the sample in step (2) is subjected to Raman mapping scan test using the wavelength range of the characteristic peaks; the sample is a blend of PGA and PBAT.

8. The Raman imaging method according to any one of claims 1-4, characterized in that, In step (2), the sample is a standard sample of ethylene-propylene random copolymer with different monomer ratios; in step (4), the ethylene-propylene copolymer sample to be tested is subjected to Raman mapping scan using the wavelength range of the characteristic peak.

9. The Raman imaging method according to claim 5, characterized in that, In step (2), the sample is a standard sample of ethylene-propylene random copolymer with different monomer ratios; in step (4), the ethylene-propylene copolymer sample to be tested is subjected to Raman mapping scan using the wavelength range of the characteristic peak.

10. The Raman imaging method according to any one of claims 1-4, 7 and 9, characterized in that, In step (2), the large spot Raman test range is set to between 1μm*1μm and 100μm*100μm; in step (4), the scanning area size is set to between 1μm*1μm and 100μm*100μm.

11. The Raman imaging method according to any one of claims 1-4, 7 and 9, characterized in that, In step (4), the single scan time is 0.1s-50s, the scan step size is 0.1μm to 10μm, and a single point is scanned once or multiple times.

12. The Raman imaging method according to claim 10, characterized in that, In step (4), the single scan time is 0.1s-50s, the scan step size is 0.1μm to 10μm, and a single point is scanned once or multiple times.

13. A quantitative analysis method for polymer materials, characterized in that, The Raman imaging method according to any one of claims 1-12 obtains a distribution map, and the quantitative composition distribution of the local polymer material is obtained based on the distribution map.

Images