A method for evaluating performance of a diamond coating interface based on cross-section microscopic Raman

Abstract

The present application belongs to the technical field of diamond coating tool performance detection, and particularly relates to a diamond coating interface performance evaluation method based on cross-section microscopic Raman. The method comprises: obtaining a sample containing a flat cross-section of diamond coating; performing Raman area scanning on spatial measurement points containing film-base interfaces on the flat cross-section of the sample, extracting characteristic peak parameters and calculating diamond content parameters and interface residual stress; performing indentation test in the region containing the film-base interface to obtain measured indentation hardness and / or measured indentation modulus; taking the Raman characteristic parameters and the characteristic peak parameters as inputs and the indentation mechanical properties as outputs to construct a quantitative correlation prediction model; and substituting the parameters of the sample to be tested to obtain the mechanical property prediction value. The present application realizes accurate spatial registration and quantitative prediction of the same micro-area phase structure, stress and mechanical property, solves the problems of difficult interface detection of special-shaped tools and lack of quantitative correlation model of microstructure and mechanical property.

Description

Technical Field

[0001] This invention relates to the field of diamond-coated tool performance testing technology, specifically to a method for evaluating the interface performance of diamond coatings based on cross-sectional micro Raman spectroscopy. Background Technology

[0002] Diamond-coated tools, with their ultra-high hardness, excellent wear resistance, and low coefficient of friction, have become ideal tools for machining difficult-to-machine materials such as graphite, ceramics, and carbon fiber composites. However, coating peeling is the main form of failure for diamond-coated tools, the root cause of which lies in insufficient interfacial bonding strength between the coating and the substrate. Therefore, accurately evaluating the interfacial bonding performance is of great significance for coating process optimization and quality control.

[0003] Currently, indentation and scratch testing are the most widely used qualitative and semi-quantitative methods for testing coating adhesion. Indentation involves loading a Rockwell hardness tester and observing the cracks and spalling around the indentation, evaluating the adhesion strength according to HF1-HF6 grades. Scratch testing characterizes adhesion by detecting the critical load corresponding to the point of abrupt change in the friction coefficient. However, both methods have significant limitations: firstly, real cutting tools are often curved or irregularly shaped, making traditional testing methods difficult to implement effectively; secondly, the test results are greatly influenced by the operator's subjective judgment, and the testing process is only performed on the coating surface, failing to obtain microstructural information at the coating-substrate interface. Research shows that the graphite content and microstructure at the interface have a decisive influence on adhesion performance, but existing testing methods often ignore this crucial factor during the evaluation process.

[0004] Raman spectroscopy provides an effective way to overcome the above-mentioned shortcomings. This technique can characterize the content ratio of diamond phase and graphite phase in the coating, and based on the characteristic peak of diamond (approximately...), it can also... The wavenumber shift of Raman spectroscopy can be used to quantitatively assess the residual stress inside the coating. However, existing studies mostly use Raman spectroscopy for independent qualitative analysis of coating quality, and a quantitative evaluation system that correlates microstructural parameters such as residual stress at the interface and the degree of graphitization with the interfacial bonding performance of the film and substrate has not yet been established.

[0005] Metallographic mounting is a standard sample preparation method for analyzing cemented carbide coatings. The coating is cut using a tool and then embedded in resin. After grinding and polishing, a smooth cross-section is obtained. The coating thickness, interface morphology, and elemental distribution can then be observed using techniques such as scanning electron microscopy (SEM), energy-dispersive spectrometry (EDS), and electron backscatter diffraction (EBSD). However, in current techniques, the mounted cross-section is primarily used for morphology observation and compositional analysis. There are no reported technical solutions combining this with micro-Raman surface scanning to obtain microstructural information of the coating at the film-substrate interface and further establish a correlation model with indentation mechanical properties.

[0006] In view of this, the present invention provides a method for evaluating the interfacial performance of diamond coatings based on cross-sectional micro Raman spectroscopy. This method accurately obtains the microstructural parameters of the coating at the film-substrate interface from the cross-sectional direction and establishes a quantitative correlation model with the indentation mechanical properties, aiming to achieve accurate evaluation of the interfacial bonding performance of diamond coatings. Summary of the Invention

[0007] The purpose of this application is to provide a method for evaluating the interface performance of diamond coatings based on cross-sectional micro Raman spectroscopy, so as to solve the above-mentioned technical problems.

[0008] This application proposes a method for evaluating the interface performance of diamond coatings based on cross-sectional micro Raman spectroscopy, the method comprising: S1. Obtain a sample with a flat cross-section containing a diamond coating; S2. Using a micro Raman spectrometer, Raman surface scanning is performed on the spatial measurement points containing the film-substrate interface on the flat cross-section of the sample to obtain Raman spectral data for each spatial measurement point; peak separation, deconvolution, and fitting are performed on the Raman spectral data to extract the characteristic peak parameters of the corresponding spatial measurement points; based on the characteristic peak parameters, the Raman characteristic parameters of the corresponding spatial measurement points are calculated, including diamond content parameters and interface residual stress. S3. Perform indentation testing on the region of the sample containing the membrane-substrate interface to obtain the measured indentation hardness and / or measured indentation modulus in the region containing the membrane-substrate interface. S4. Using Raman characteristic parameters and characteristic peak parameters as input variables, and measured indentation hardness and / or measured indentation modulus as output variables, construct a quantitative correlation prediction model; S5. Substitute the Raman characteristic parameters and characteristic peak parameters of the sample to be tested into the quantitative correlation prediction model to obtain the predicted values ​​of the mechanical properties of the sample to be tested.

[0009] In the above technical solution, by introducing "cross-sectional micro Raman spectroscopy deconvolution" and "in-situ micro-area mechanical testing", a micro-area quantitative mapping between phase structure / micro-stress physical parameters and macroscopic film-substrate bonding mechanical properties is cleverly established, realizing high-precision quantitative prediction of the interface bonding performance of diamond coating.

[0010] Furthermore, step S1 includes: Diamond-coated tools were cut using wire electrical discharge machining to obtain cross-sectional samples containing the interface between the coating and the substrate. Specifically, the cross-sectional sample of the drill bit type special-shaped tool was obtained by cutting along the axial direction to obtain a longitudinal section containing the coating on the cutting edge; the cross-sectional sample of the ball end mill was obtained by cutting along the axial direction through the apex of the ball end mill to obtain a longitudinal section containing the coating on the top of the ball end mill. Place the cross-sectional sample in the mounting mold with the cross-section to be tested facing upwards, and cold mount it by curing epoxy resin at room temperature for 12-24 hours, or hot mount it by using bakelite powder at a temperature of 150-180℃, a pressure of 20-30MPa and a holding time of 5-10 minutes, to prepare a mounting block with a diameter of 30-40mm. The sample block is ground and polished until the roughness of the polished cross-section meets the requirements. Samples with a smooth cross-section containing a diamond coating were prepared.

[0011] In the above technical solution, a sample preparation process combining directional cutting and cold / hot inlay was proposed to address the geometric problem of the non-developable cutting edge area of ​​irregular and complex cutting tools (drills, ball end mills). Under the premise of ensuring the integrity and non-peeling of the interface between the coating and the substrate, an optically flat cross-section with extremely low roughness (eliminating height difference interference) was obtained, which laid a solid sample foundation for high signal-to-noise ratio Raman spectroscopy acquisition and high-precision indentation testing.

[0012] Furthermore, the Raman surface scanning in step S2 includes: Place the sample on the sample stage of the micro Raman spectrometer, and adjust the sample position so that the scanning area completely covers the entire area from the near-film-substrate interface to the outer edge of the coating. The scanning area is centered on the film-substrate interface and extends a predetermined distance towards both the coating and substrate sides. The scanning area is set along a direction parallel to the film-substrate interface. One scan line, and The scan lines are arranged at a set step size in a direction perpendicular to the membrane-substrate interface, and the data is collected along a direction parallel to the membrane-substrate interface on each scan line. A total of [number] spatial measurement points were collected, and a total of [number] data were collected. Raman spectral data of spatial measurement points.

[0013] In the above technical solution, by setting up a cross-interface gridded scanning array with the film-substrate interface as the central reference line and simultaneously covering the coating side and the substrate side, continuous, blind-zone-free acquisition of the micro-region phase structure and stress state gradient from the substrate to the coating surface is achieved, effectively overcoming the measurement deviation caused by the non-uniformity of a single measurement point.

[0014] Furthermore, the extraction process of characteristic peak parameters in step S2 includes: Polynomial fitting or adaptive baseline subtraction methods were used to remove background interference from Raman spectral data. The Lorentzian-Gaussian mixture function was used to perform peak deconvolution fitting on the Raman spectral data. The characteristic peaks obtained by deconvolution included diamond peaks and amorphous peaks. The diamond peak, graphite D peak, and graphite G peak were analyzed to extract characteristic peak parameters, including the diamond characteristic peak position, diamond characteristic half-peak width, diamond peak area, graphite D peak area, graphite G peak area, and amorphous peak area. Peak area.

[0015] In the above technical solution, by limiting the fluorescence background subtraction and the Lorentzian-Gaussian mixed function deconvolution fitting process, the interference of fluorescence background noise on the micro-area spectrum is effectively eliminated, and the high-precision decomposition of overlapping composite Raman peaks is realized. Thus, the characteristic peak parameters reflecting the integrity of the diamond lattice, the degree of graphitization and the composition of the amorphous phase are accurately extracted, providing a high signal-to-noise ratio and high reliability of spectral characteristic data for subsequent calculation of interface physical parameters and correlation of mechanical properties.

[0016] Furthermore, the calculation of Raman characteristic parameters in step S2 includes: The formula for calculating the diamond content parameter is:

[0017] In the formula, This refers to the diamond content parameter; The area of ​​the diamond peak; This is the Raman scattering cross section correction factor; The area of ​​the D peak of graphite; amorphous Peak area; The area of ​​the graphite G peak; The formula for calculating the residual stress at the interface is:

[0018] In the formula, This refers to residual stress at the interface; Stress coefficient; These are characteristic peaks of diamond.

[0019] In the above technical solution, amorphous materials are excluded by defining specific calculation formulas for diamond content parameters and interfacial residual stress. By eliminating the fuzzy interference of hybridization relative to crystal purity and combining the scattering cross section correction factor and stress offset coefficient, the degree of catalytic graphitization and residual stress state at the interface were accurately quantified, providing a reliable microscopic physical structure characterization index for the degradation of interfacial mechanical properties.

[0020] Furthermore, the indentation test in step S3 includes: Indentation tests were performed on the region containing the film-substrate interface using a nanoindenter; the indentation test used a Berkovich pyramidal diamond indenter; the loading mode was set to constant loading rate mode or constant strain rate mode, and the maximum load was set so that the indentation depth did not exceed 5-10% of the diamond coating thickness; Based on the Oliver-Pharr method, the measured indentation hardness and measured indentation modulus at each spatial measurement point are calculated from the load-displacement curves recorded in the indentation test. The formula for calculating the hardness of the measured indentation is: , In the formula, For actual measurement of indentation hardness; Maximum load; The contact projection area; The measured indentation modulus was calculated from the initial slope of the unloading curve.

[0021] In the above technical solution, the indentation depth is strictly limited to not exceeding the coating thickness, which effectively eliminates the substrate softening effect from interfering with the measurement of indentation hardness and modulus. Combined with Oliver-Pharr theory and constant loading rate mode, hardness and modulus parameters that truly reflect the intrinsic mechanical properties of the coating in the near-interface micro-regions are obtained.

[0022] Furthermore, the construction of the quantitative correlation prediction model in step S4 includes: Positioning markers are introduced in the region of the membrane-substrate interface for spatial coordinate registration of Raman surface scanning and indentation testing within the same region; Based on the positioning marks, the measured indentation hardness and / or measured indentation modulus of multiple spatial measurement points in the same area are arithmetically averaged to obtain the mean measured indentation hardness and / or mean measured indentation modulus of the same area; the characteristic peak parameters and Raman characteristic parameters of the corresponding spatial measurement points in the Raman surface scanning area are arithmetically averaged to obtain the mean of each characteristic peak parameter and the mean of each Raman characteristic parameter. Using the mean values ​​of each characteristic peak parameter and the mean values ​​of each Raman characteristic parameter as input variables, and the mean values ​​of the measured indentation hardness and / or the measured indentation modulus in the same region as output variables, a quantitative correlation prediction model is constructed using multiple linear regression, partial least squares regression, or machine learning algorithms.

[0023] In the above technical solution, the microscopic physical spectral parameters (phase composition, stress) and physical and mechanical responses (hardness, modulus) of the same micro-region are precisely aligned in the spatial dimension by using spatial coordinate registration technology, which eliminates data distortion caused by test position deviation and ensures the physical consistency of the input-output regression training set.

[0024] Furthermore, the expression for the quantitative association prediction model is:

[0025] In the formula, As a comprehensive performance evaluation index for diamond coatings; It is a quantitative mapping function; These are characteristic peaks of diamond; The half-width at half-maximum characteristic of diamond; This refers to the diamond content parameter; The ratio of the intensity or area of ​​the graphite D peak to the graphite G peak; This refers to the residual stress at the interface.

[0026] In the above technical solution, the microscopic factors affecting the interface bonding performance (peak position, half-peak width, diamond purity, graphitization degree, residual stress) are sublimated into a unified function mapping system, and a comprehensive evaluation formula for the synergistic effect of multi-physics field parameters is constructed, which greatly broadens the theoretical application scope of interface performance evaluation.

[0027] Furthermore, the quantitative correlation prediction model is a multiple linear regression model:

[0028] or

[0029] in, This is the predicted value for indentation hardness; This is the predicted value of the indentation modulus; The characteristic half-width of diamond , Diamond content parameter , The intensity ratio or area ratio of the graphite D peak to the graphite G peak. , For interfacial residual stress ; , , , , as well as , , , , is the regression coefficient.

[0030] In the above technical solution, by limiting the quantitative correlation prediction model to a multiple linear regression model, an explicit linear analytical expression is provided, which effectively simplifies the calculation process and reduces the complexity of engineering calculations.

[0031] Furthermore, the quantitative association prediction model also includes nonlinear transformation terms applied to the input variables: The nonlinear transformation term is one or more of the following: a quadratic term, a power function transformation term, an exponential transformation term, or a logarithmic transformation term of the input variable.

[0032] In the above technical solution, by introducing nonlinear transformation terms such as quadratic terms, power functions, exponential or logarithmic terms into the model, the nonlinear response trend of microstructure and stress parameters under non-uniform / highly concentrated states is effectively captured, and the synergistic coupling effect between different microphysical parameters is quantified, which significantly improves the quantitative prediction accuracy and robustness of the quantitative correlation prediction model under wide process span and extreme interface states.

[0033] Furthermore, positioning markers are introduced in the region near the membrane-substrate interface for spatial coordinate registration of Raman surface scanning and indentation testing within the same area; Based on the positioning marks, the measured indentation hardness and / or measured indentation modulus of multiple spatial measurement points in the same area are arithmetically averaged to obtain the mean measured indentation hardness and / or mean measured indentation modulus of the same area. At the same time, the characteristic peak parameters and Raman characteristic parameters of the corresponding spatial measurement points in the Raman surface scanning area are arithmetically averaged to obtain the mean of each characteristic peak parameter and the mean of each Raman characteristic parameter in the Raman surface scanning area.

[0034] In the above technical solution, by introducing physical / optical positioning markers near the interface, high-precision spatial overlap of Raman spectral lattice and nanoindentation lattice at the micrometer scale is achieved; the arithmetic mean algorithm is used to smooth out random errors caused by microscopic defects inside the material, ensuring high reliability of cross-instrument data fusion and prediction verification.

[0035] Compared with the prior art, the beneficial results of the present invention are as follows: (1) In this application, on the same flat cross-section sample, micro-Raman surface scanning deconvolution is used to extract micro-physical parameters such as the full width at half maximum (FWHM) of diamond, diamond content, degree of graphitization and residual stress at the interface. Then, in-situ nanoindentation test is combined to obtain the intrinsic hardness and modulus of the coating near the interface. The positioning mark is used to achieve accurate registration of cross-instrument spatial coordinates, and a highly aligned quantitative correlation prediction model is constructed. This realizes high-precision quantitative prediction of the mechanical properties of the coating interface based on the microstructure parameters of the interface.

[0036] (2) This application uses directional electrical discharge wire cutting to cut a longitudinal section containing the main cutting edge or the top of the ball head. Combined with cold / hot inlay and step-by-step polishing technology, the complex irregular tool is transformed into a regular cross-section sample with high flatness and low roughness. This not only eliminates the interference of surface curvature and height difference on Raman optical path and indentation dotting measurement, but also completely preserves the physical bonding state of the film-substrate interface, providing a standardized solution for the detection of coating interfaces of irregular tools.

[0037] (3) This application establishes a two-dimensional lattice scanning grid in the cross-interface region with the film-substrate interface as the central reference line, enabling continuous acquisition of phase composition and stress evolution data from the cemented carbide substrate side to the outer side of the diamond coating. By subtracting the background and deconvolving using the Lorentzian-Gaussian mixed function, the amorphous phase is accurately stripped away. The interference of hybrid phases quantified the degree of catalytic graphitization, graphitized layer thickness, crystal quality variation trend and stress mutation amplitude at the interface. From the perspective of microphysics and materials science, it revealed the microscopic response mechanism of stress concentration and adhesion degradation at the coating interface, providing precise data support for the directional control and optimization of diamond coating preparation process. Attached Figure Description

[0038] The accompanying drawings are included to provide a further understanding of the embodiments and are incorporated in and constitute a part of this specification. The drawings illustrate embodiments and, together with the description, serve to explain the principles of the invention. Many anticipated advantages of the embodiments and other embodiments of the invention will be readily recognized as they become better understood through reference to the following detailed description. Elements in the drawings are not necessarily to scale. The same reference numerals refer to corresponding similar parts.

[0039] Figure 1 This is a flowchart of the method for evaluating the interface performance of diamond coatings based on cross-sectional micro Raman spectroscopy according to an embodiment of this application. Figure 2 This is a technical flowchart of the diamond coating interface performance evaluation method based on cross-sectional micro Raman spectroscopy according to an embodiment of this application; Figure 3This describes the preparation process and cross-sectional morphology of a cemented carbide diamond-coated tool sample according to an embodiment of this application. Figure 4 This is a schematic diagram of a Raman surface scanning test position and measurement point array including a membrane-based interface according to an embodiment of this application; Figure 5 This is a schematic diagram of the load-displacement curve and test position of a nanoindentation test according to an embodiment of this application; Figure 6 This is a typical Raman spectrum peak deconvolution fitting diagram of a diamond coating according to an embodiment of this application; Figure 7 This is a schematic diagram of a quantitative correlation prediction model architecture according to an embodiment of this application. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0041] refer to Figure 1 , Figure 1 The figure shows a flowchart of a method for evaluating the interface performance of diamond coatings based on cross-sectional micro Raman spectroscopy, according to an embodiment of this application. As shown, the method includes the following steps: S1. Obtain a sample with a smooth cross-section containing a diamond coating.

[0042] In some specific embodiments, step S1 includes: S11, a diamond-coated tool is cut using wire electrical discharge machining to obtain a cross-sectional sample containing the complete coating-substrate interface (i.e., film-substrate interface). The sample size is adapted to the mounting mold. To avoid the diamond coating becoming insulated and preventing cutting, an opening is pre-ground at the cutting location using a grinding wheel to expose the cemented carbide substrate. For drill-type irregularly shaped tools, an axial cut is made to obtain a longitudinal section containing the coating on the cutting edge; for ball end mills, an axial cut is made through the apex of the ball to obtain a longitudinal section containing the coating on the top of the ball.

[0043] S12, place the cut cross-section sample in the mounting mold with the test section facing upwards, and cold mount it by curing epoxy resin at room temperature for 12~24h, or hot mount it by using bakelite powder at a temperature of 150~180℃, a pressure of 20~30MPa and a holding temperature and pressure of 5~10min, to prepare a standard-sized mounting block with a diameter of 30~40mm.

[0044] S13, the sample block was ground sequentially with 200-mesh, 600-mesh, 1200-mesh and 2000-mesh sandpaper, and then finely polished with diamond polishing slurry with particle sizes of 3μm, 1μm and 0.5μm respectively, until there were no obvious scratches on the polished cross-section surface and the film-substrate interface was clearly distinguishable, so that the surface roughness of the polished cross-section met Ra≤0.02μm, and a sample with a smooth cross-section containing diamond coating was obtained.

[0045] S2. Using a micro Raman spectrometer, Raman surface scanning is performed on the spatial measurement points containing the film-substrate interface on the flat cross-section of the sample to obtain Raman spectral data for each spatial measurement point; peak separation, deconvolution, and fitting are performed on the Raman spectral data to extract the characteristic peak parameters of the corresponding spatial measurement points; based on the characteristic peak parameters, the Raman characteristic parameters of the corresponding spatial measurement points are calculated, including diamond content parameters and interface residual stress.

[0046] In some specific embodiments, a sample containing a flat cross-section of a diamond coating is placed on the sample stage of a confocal micro Raman spectrometer. The sample position is adjusted so that the diamond coating is centered in the field of view, and positioning marks are introduced in the area near the film-substrate interface for spatial coordinate registration of Raman surface scanning and nanoindentation testing within the same area. This ensures that the scanning area completely covers the entire area from the coating near the film-substrate interface to the outer edge of the coating. The scanning area extends a predetermined distance from the film-substrate interface towards both the coating side and the cemented carbide substrate side. The scanning area is set in predetermined steps along a direction perpendicular to the film-substrate interface. Each scan line acquires data at equal intervals along a direction parallel to the membrane-substrate interface. The system measures multiple spatial measurement points to obtain Raman spectral data containing spatial coordinates and Raman spectral intensity. Specifically, the test parameters are set as follows: excitation wavelength 532 nm, laser power 10~50 mW, grating 1800 gr / mm, and scanning range 1000~1800 mm. -1 The scanning step size is 0.5~2μm, and the integration time is 1~10s / point.

[0047] In some specific embodiments, the step of extracting characteristic peak parameters includes: Step 1: Use polynomial fitting or adaptive baseline subtraction method to remove background interference from the Raman spectra of each spatial measurement point.

[0048] Step two: The Lorentzian-Gaussian mixing function is used to perform peak deconvolution fitting on the Raman spectrum. The characteristic peaks obtained by deconvolution include diamond peaks and amorphous peaks. The diamond peak, graphite D peak, and graphite G peak were analyzed, and characteristic peak parameters were extracted for each spatial measurement point. These characteristic peak parameters include the diamond characteristic peak position, diamond characteristic half-peak width, diamond peak area, graphite D peak area, graphite G peak area, and amorphous peak area. Peak area.

[0049] Specifically, the central wavenumber of the diamond peak is approximately ,correspond Bond structure and extract characteristic peak positions of diamond. Peak height Diamond characteristic half-peak width and peak area Characteristic peak parameters; the central wavenumber of the graphite D peak is approximately ,correspond Defects in the bond structure were identified, and peak positions were extracted. Peak height Half-peak width and peak area Characteristic peak parameters; the central wavenumber of the graphite G peak is approximately And extract peak position Peak height Half-peak width and peak area Characteristic peak parameters; amorphous The center wavenumber of the peak is approximately And extract peak position Peak height Half-peak width and peak area Characteristic peak parameters.

[0050] Furthermore, based on the characteristic peak parameters obtained from the above deconvolution, the Raman characteristic parameters of the diamond content and the interface residual stress at the corresponding spatial measurement points are calculated.

[0051] Specifically, diamond content parameters Based on the diamond peak area at the corresponding spatial measurement point Total carbon peak area at corresponding spatial measurement points The ratio is calculated using the following formula: ,in This is the Raman scattering cross-section correction factor at an excitation wavelength of 532 nm, used to correct... The sensitivity difference between carbon and diamond; a higher ratio indicates a lower degree of graphitization in the diamond coating. Preferably, The value is 0.02; The area of ​​the graphite D peak at the corresponding spatial measurement point; For the corresponding spatial measurement point, the amorphous Peak area; This represents the area of ​​the graphite G peak at the corresponding spatial measurement point.

[0052] Specifically, interfacial residual stress Based on the characteristic peak position of diamond relative to the peak position of standard stress-free diamond The displacement is calculated using the following formula: The unit is GPa, where negative values ​​represent residual compressive stress and positive values ​​represent residual tensile stress. The stress coefficient; preferably, the pressure coefficient determined based on diamond high-pressure Raman spectroscopy. Take the stress coefficient The theoretical value is -0.42. (That is, compressive stress causes the peak position to shift to higher wavenumbers, and the residual stress is negative when the peak position displacement is positive). This represents the characteristic peak position of diamond at the corresponding spatial measurement point. The correction factor is... With stress coefficient These are well-known calibration constants in the field for diamond / graphite Raman spectroscopy analysis under 532nm laser excitation.

[0053] In some specific embodiments, the Raman characteristic parameters of each spatial measurement point within the scanning area are arithmetically averaged to calculate the average diamond content parameter and the average interface residual stress within the scanning area, which are then used for registration analysis with the average nanoindentation mechanical properties within the same area.

[0054] S3. Perform indentation testing on the region of the sample containing the film-substrate interface to obtain the measured indentation hardness and / or measured indentation modulus in the region containing the film-substrate interface.

[0055] In some specific embodiments, spatial coordinate registration of nanoindentation testing and Raman surface scanning is achieved within the same region based on positioning markers introduced in the area near the film-substrate interface. Mechanical properties of the diamond coating cross-section are then tested using a nanoindenter on the same region of the same sample after Raman surface scanning.

[0056] Specifically, the indentation test used a Berkovich pyramidal diamond indenter. The loading mode was set to either constant loading rate (0.5~5 mN / s) or constant strain rate (0.05~0.5 s). - ¹). The maximum load is set based on the thickness of the diamond coating, ensuring the indentation depth does not exceed 5-10% of the coating thickness to eliminate matrix effects. The holding time is set to 5-10 seconds, and the unloading rate is consistent with the loading rate. After the test is completed, mechanical property parameters are calculated and extracted based on the recorded load-displacement curves using the Oliver-Pharr method: using the formula The measured indentation hardness was calculated. ,in For maximum load, The contact projection area is used; the measured indentation modulus is calculated using the initial slope of the unloading curve. The measured indentation hardness and measured indentation modulus data mentioned above can be directly obtained from the testing equipment software.

[0057] Preferably, the arithmetic mean of the measured indentation hardness and measured indentation modulus of multiple spatial measurement points in the same area is calculated to obtain the mean value of the measured indentation hardness and the mean value of the measured indentation modulus in the scanning area, which is then used for registration and regression analysis with the mean value of Raman characteristic parameters in the same area.

[0058] S4. Using Raman characteristic parameters and characteristic peak parameters as input variables, and measured indentation hardness and / or measured indentation modulus as output variables, construct a quantitative correlation prediction model.

[0059] In some specific embodiments, the establishment of the quantitative correlation prediction model includes: S41 involves spatial coordinate registration and statistical analysis of the characteristic peak parameters and Raman characteristic parameters obtained through micro Raman spectroscopy, along with the corresponding nanoindentation mechanical properties at spatial measurement points. The characteristic peak parameters include the diamond characteristic peak position, diamond characteristic half-peak width, and the intensity ratio of the graphite D peak to the graphite G peak; the Raman characteristic parameters are the diamond content parameter and the interfacial residual stress; and the nanoindentation mechanical properties include the measured indentation hardness and the measured indentation modulus.

[0060] Specifically, due to the localized non-uniformity of the diamond coating cross-section in terms of structure and defects at the microscale, and the micro-scale of Raman spectroscopy and nanoindentation testing, as well as the micrometer-level errors in mechanical positioning between the two devices across the testing platform, pursuing absolute physical coordinate correspondence point-to-point easily introduces localized fluctuation interference. Therefore, regional registration of spatial coordinates is achieved by introducing positioning markers in the region near the film-substrate interface. The arithmetic mean of the measured indentation hardness and measured indentation modulus at multiple spatial measurement points within the same region is calculated to obtain the mean measured indentation hardness and mean measured indentation modulus within the scanning region. Simultaneously, the arithmetic mean of the Raman characteristic parameters at each spatial measurement point within the scanning region is calculated to obtain the characteristic peak parameters and the mean of each Raman characteristic parameter within the scanning region. These are then used for subsequent registration and regression analysis with the mean of the nanoindentation mechanical properties within the same region.

[0061] S42. Collect no fewer than 30 sets of sample data under different process conditions. Using characteristic peak parameters and Raman characteristic parameters as input variables, and the measured indentation hardness and measured indentation modulus of the corresponding region as output variables, establish a quantitative correlation prediction model using multiple linear regression, partial least squares regression, or machine learning algorithms. The model expression formula is:

[0062] In the formula, The comprehensive performance evaluation index for diamond coatings includes predicted indentation hardness and predicted indentation modulus. It is a quantitative mapping function constructed based on multiple linear regression, partial least squares regression, or machine learning algorithms; These are characteristic peaks of diamond; The half-width at half-maximum characteristic of diamond; This refers to the diamond content parameter; The ratio of the intensity or area of ​​the D peak to the G peak in graphite; The calculated interfacial residual stress is shown. This model enables quantitative prediction and evaluation of the interfacial mechanical properties of diamond coatings based on characteristic peak parameters and Raman characteristic parameters.

[0063] Specifically, the quantitative correlation prediction model is a multiple linear regression model:

[0064] or

[0065] in, To predict the indentation hardness value; To predict the indentation modulus; The characteristic half-width of diamond , Diamond content parameter , The intensity ratio or area ratio of the graphite D peak to the graphite G peak. , For interfacial residual stress ; , , , , as well as , , , , The regression coefficients are determined by fitting the characteristic peak parameters, Raman characteristic parameters, and nanoindentation test results of multiple sets of calibrated samples.

[0066] Furthermore, when regression analysis shows a non-linear response trend between an input variable and an output variable, a power function transformation is performed on the input variable. Exponential transformation or logarithmic transformation Then, multiple linear regression fitting is performed to achieve high-precision quantitative prediction of the interfacial bonding performance of diamond coating film.

[0067] Specifically, the quantitative correlation prediction model can also be constructed using conventional machine learning algorithms in the field. These machine learning models include, but are not limited to, partial least squares regression, support vector regression, or random forest. It should be understood that the aforementioned machine learning algorithms are mature data modeling and regression analysis tools known and familiar to those skilled in the art. The core innovation of this application lies in extracting specific micro-region physical parameters (including diamond characteristic full width at half maximum, diamond content, degree of graphitization, and interfacial residual stress) through cross-sectional micro Raman spectroscopy deconvolution, and establishing a spatial registration and quantitative mapping relationship between these parameters and the mechanical properties of the film-substrate interface. Regarding the specific algorithm selection, those skilled in the art, based on the characteristic parameter inputs and test data disclosed in this application, can accurately predict the interfacial properties of diamond-coated films using any of the aforementioned well-known machine learning models. Therefore, the specific algorithm itself does not constitute a limitation on the technical solution of this application.

[0068] S5. Substitute the Raman characteristic parameters and characteristic peak parameters of the sample to be tested into the quantitative correlation prediction model to obtain the predicted values ​​of the mechanical properties of the sample to be tested, so as to evaluate the interfacial bonding performance of the diamond coating film of the sample to be tested.

[0069] In some specific embodiments, the process of predicting the interface performance of the sample under test includes: For the unknown diamond-coated tool sample to be tested, repeat the processing steps S1-S3. Perform micro Raman spectroscopy acquisition and peak deconvolution processing on the cross-sectional interface region of the sample to be tested, and extract the corresponding characteristic peak parameters and Raman characteristic parameters.

[0070] By substituting the extracted characteristic peak parameters and Raman characteristic parameters into the established quantitative correlation prediction model, the predicted values ​​of indentation hardness and indentation modulus of the test tool sample at the film-substrate interface can be quantitatively obtained.

[0071] The output mechanical property prediction values ​​enable a rapid quantitative assessment and prediction of the interfacial bonding performance of the diamond coating film on the test tool sample.

[0072] In some specific embodiments, the predicted mechanical properties are the predicted indentation hardness and / or predicted indentation modulus. The essential reason for diamond graphitization is that cobalt in the cemented carbide matrix diffuses to the film-substrate interface during high-temperature deposition, catalyzing the graphitization of diamond. The graphite phase, as a weak structure bound only by van der Waals forces between layers, has an intrinsic hardness and elastic modulus far lower than that of the diamond coating. This is equivalent to introducing a weak interface layer between the diamond coating and the substrate. External loads preferentially initiate cracks at this point, which then propagate along the interlayer structure of the graphite crystals, resulting in a decrease in film-substrate adhesion. The higher the degree of graphitization, the thicker the weak interface layer, and the lower the film-substrate adhesion. Therefore, by quantitatively characterizing the intensity ratio of the graphite D-peak to the graphite G-peak at the film-substrate interface (i.e., the degree of graphitization) and the residual stress at the interface using Raman surface scanning, the degree of mechanical degradation of the weak interface layer can be indirectly reflected. Furthermore, based on the calculated predicted indentation hardness and predicted indentation modulus values, a quantitative prediction and evaluation of the film-substrate interface adhesion performance of the diamond coating can be achieved.

[0073] Example 1 refer to Figures 2-6 , Figure 2 A technical flowchart of a diamond coating interface performance evaluation method based on cross-sectional micro Raman spectroscopy according to an embodiment of this application is shown. As shown in the figure, the technical process includes: During the sample preparation stage, GZ15 series cemented carbide substrate diamond-coated cutting tools were selected, with a coating thickness of approximately 10 mm. A cross-sectional sample of the drill bit head, containing the complete membrane-substrate interface, was obtained by wire electrical discharge machining (EDM) along a direction perpendicular to the drill bit axis. The sample size was approximately 8 mm. 8mm 12mm, such as Figure 3 As shown in the upper left corner; the sample was placed in a 30mm diameter mounting mold with the cross-section to be tested facing upwards, and heat-mounted with bakelite powder at 170℃ and 25MPa for 10 minutes to obtain the mounting block. Subsequently, the cross-section of the mounting block was successively ground with 200-mesh, 600-mesh, 1200-mesh, and 2000-mesh sandpaper, and a grit size of 3... The diamond polishing fluid is used for progressive polishing until the surface roughness of the cross-section is achieved. After mounting and polishing, the sample structure is as follows: Figure 3 As shown.

[0074] The data acquisition phase includes micro Raman surface scanning and nanoindentation testing. Specifically: Micro-Raman surface scanning was performed using a confocal micro-Raman spectrometer with an excitation wavelength set to 532 nm, a laser power set to 20 mW, and a grating set to 1800 gr / mm. Raman surface scanning was conducted on the cross-section of the sample mount. (Reference) Figure 4 , Figure 4A schematic diagram of the Raman surface scanning test location and measurement point array including the membrane-substrate interface according to an embodiment of this application is shown. As shown in the figure, the Raman surface scanning area is defined by a tilted white rectangular frame. The rectangular frame is spatially rotated and registered with the membrane-substrate interface as the central reference line, covering an area of ​​20 mm upwards. The diamond-coated side, covering 5 On the hard alloy substrate side; the 330 white markers arranged in an array inside the rectangular frame are the spatial measurement points, consisting of 10 equally spaced measurement points parallel to the membrane-substrate interface direction (i.e., the long side of the rectangle) and 2 points perpendicular to the membrane-substrate interface direction (i.e., the short side of the rectangle). The system consists of 33 intersecting scan lines with a step size distribution, collecting a total of 330 test points. The integration time for a single point is 5 seconds, enabling continuous lattice spectral acquisition of the micro-region of the membrane-substrate interface.

[0075] Nanoindentation testing was performed using a nanoindenter equipped with a Berkovich diamond indenter. The maximum load was set to 10 mN, and the loading rate was set to 1 mN / s. The test was conducted within the region including the film-substrate interface (specifically, the region at a distance from the film-substrate interface). On the smooth cross-section of the surface (at the location), 33 equally spaced indentation test points were set along a direction parallel to the interface. The load-displacement curves of each test point were recorded, such as... Figure 5 As shown on the left, multiple load-displacement curves exhibit high overlap, indicating good reproducibility of the mechanical response in this test area. The recorded load-displacement curves were calculated using the Oliver-Pharr method (e.g., Figure 5 As shown), the arithmetic mean of the measured indentation hardness is obtained. Average measured indentation modulus .

[0076] The model building and application phase includes Raman spectral peak fitting and feature parameter extraction, establishing a quantitative correlation model, and outputting coating performance predictions. Specifically: Raman spectral peak fitting and feature parameter extraction were performed using a Lorentzian-Gaussian mixture function to perform peak deconvolution fitting on the Raman spectral data for each spectral point. (Reference) Figure 6 , Figure 6 The figure shows a typical Raman spectrum peak deconvolution fitting diagram of a diamond coating according to an embodiment of this application. As shown in the figure, a typical spectrum within the diamond coating is used as an example (its Raman peak fitting spectrum is as follows). Figure 6 As shown), the diamond characteristic peak (1334.0) was obtained through fitting and extraction. ), Graphite D peak (1374.0) ), Graphite G peak (1556.6) ) and amorphous Peak (1487.0) Characteristic peak parameters, such as the ratio of the D peak to the G peak of the graphite in the sample, were calculated. Diamond content parameters Interfacial residual stress (Compressive stress). In the specific calculation logic, amorphous... Peak (approximately 1300.0 cm) - ¹) Used only for the complete reconstruction of the full-spectrum envelope and baseline correction during spectral deconvolution fitting, and not included in the diamond content parameter. The quantitative calculation formula is as follows. Its physical mechanism lies in: amorphous... Carbon, in terms of its structural classification, exists in a transitional state between disordered amorphous networks and ordered crystals at the microscopic level—although its bonding form is... Hybridized, but with a disordered amorphous microstructure, it differs from the complete three-dimensional lattice structure of crystalline diamond, and also from... Graphitized or amorphous network structures of hybrid carbon. If amorphous... Peaks are forcibly classified into the crystalline diamond phase or Phase calculations can lead to blurred phase classification boundaries and distorted light scattering cross-section fitting, thereby reducing the diamond content parameter. The quantitative accuracy and effectiveness of using diamond as an indicator of the purity of crystalline phase.

[0077] A quantitative correlation model was established, and diamond-coated tools with cemented carbide substrates under 12 different process conditions were selected. Cross-sectional inlays were prepared according to the aforementioned method, and micro Raman surface scanning and nanoindentation tests were performed in sequence to obtain data pairs of characteristic peak parameters, Raman characteristic parameters and nanoindentation mechanical properties.

[0078] refer to Figure 7 , Figure 7 A schematic diagram of the quantitative correlation prediction model architecture according to an embodiment of this application is shown. As shown in the figure, the model adopts a three-layer architecture including a feature parameter input layer, a quantitative correlation model core layer, and a mechanical performance output layer. First, the correlation between each input parameter and the output variable is analyzed using univariate linear regression. Experimental results show that the half-peak width of diamond is... Compared with the measured indentation hardness Negative correlation (coefficient of determination) ), interfacial residual compressive stress Compared with the measured indentation hardness Negative correlation (coefficient of determination) ).

[0079] Based on this, the Raman characteristic parameters and characteristic peak parameters of the input layer are used as input variables, and the measured nanoscale indentation hardness of the output layer is used as the input variable. For the output variables, a regression algorithm incorporating nonlinear terms is used in the core layer to construct a quantitative association prediction model. The quantitative association prediction model, determined by fitting data from 12 sets of calibrated samples, is expressed as follows:

[0080]

[0081] In the formula, This is the predicted value for indentation hardness. These are predicted values ​​for indentation modulus, all in units of... ; The characteristic half-width of diamond The unit is ; Diamond content parameter ; The intensity ratio or area ratio of the graphite D peak to the graphite G peak. ; For interfacial residual stress The unit is The regression coefficients of the above model were determined by fitting the characteristic peak parameters, Raman characteristic parameters and nanoindentation test results of multiple sets of calibrated samples.

[0082] Example 2 To further verify the model under different coating conditions To assess the applicability of the model, five samples with different combinations of Raman characteristic parameters were selected for prediction. The input parameters and prediction results are shown in Table 1. Samples numbered 1-4 are conventional samples under different processing conditions, while sample number 5 is an extreme comparison sample with a high degree of graphitization. The Raman characteristic parameters of the five sample groups cover a range from FWHM of 15 cm⁻¹. - ¹ to 26cm - ¹, Diamond content ranges from 0.82 to 0.94, I D / I G A wide range of parameters from 0.060 to 0.750, and residual stress from -1.20 GPa to -0.30 GPa.

[0083] Table 1: Comparison of characteristic parameters and predicted and measured values ​​of nanoindentation hardness of tool samples with different coating processes

[0084] The characteristic parameters of the above samples were substituted into the quantitative correlation prediction model established in Example 1 to calculate the predicted indentation hardness value. Subsequently, actual nanoindentation tests were conducted for verification. Experimental results show that the relative errors between the predicted and measured values ​​are both controlled within a certain range. Within (maximum relative error is only) The results confirm that the quantitative correlation prediction model has good prediction accuracy and robustness for the interfacial bonding performance of diamond coating film over a wide parameter range, and can be used as a reliable method for evaluating the quality of diamond coated tools.

[0085] Example 3 To further verify the model under different coating conditions To assess applicability, five samples with different combinations of Raman characteristic parameters from Example 2 were selected for model prediction. The input parameters and prediction results are shown in Table 2. Samples numbered 1-4 are conventional samples under different process conditions, while sample number 5 is an extreme comparison sample with a high degree of graphitization. The Raman characteristic parameters of the five sample groups cover a range from FWHM of 15 cm⁻¹. - ¹ to 26cm - ¹, Diamond content ranges from 0.82 to 0.94, I D / I G A wide range of parameters from 0.060 to 0.750, and residual stress from -1.20 GPa to -0.30 GPa.

[0086] Table 2: Comparison of characteristic parameters and predicted and measured values ​​of nanoindentation modulus of tool samples with different coating processes

[0087] The indentation modulus prediction value was calculated by substituting the characteristic parameters of each sample into the quantitative correlation prediction model established in Example 1. Subsequently, actual nanoindentation modulus verification was performed. Experimental results show that the relative errors between the predicted and measured values ​​are both controlled within a certain range. Within (maximum relative error is only) This study confirms that the quantitative correlation prediction model has good predictive accuracy and robustness for the interfacial bonding performance of diamond coating films over a wide parameter range, and can be used as an effective means of evaluating the quality of diamond coated tools.

[0088] Example 4 A 6mm diameter irregularly shaped carbide-based diamond-coated drill bit was selected, and a longitudinal section sample (approximately 6mm in size) including the film-substrate interface of the main cutting edge was cut along the axial direction using wire electrical discharge machining. 6mm (10mm). Given the small sample size, epoxy resin was cured at room temperature for 24 hours for cold mounting to obtain a 30mm diameter mounting block. The flat cross-section was ground and polished, and micro Raman surface scanning and nanoindentation tests were performed sequentially. The specific test procedures and data processing methods were the same as in Example 1.

[0089] By performing peak deconvolution fitting on the Raman spectral data from various spatial measurement points, the characteristic peak parameters and Raman characteristic parameter distributions of the film-substrate interface region of the main cutting edge of the drill bit were successfully obtained. Substituting the extracted characteristic peak parameters and Raman characteristic parameters into the established quantitative correlation prediction model, the predicted values ​​of indentation hardness and indentation modulus at the film-substrate interface of the main cutting edge of the drill bit were calculated. Verification through actual nanoindentation tests showed that the relative errors of the predicted indentation hardness and indentation modulus were both controlled within a certain range. This further confirms the applicability and accuracy of the method in this application for predicting the interfacial bonding performance of diamond coating films on different irregularly shaped tool components.

[0090] Although the principles of the present invention have been described in detail above with reference to preferred embodiments, those skilled in the art should understand that the above embodiments are merely illustrative explanations of the implementation of the present invention and are not intended to limit the scope of the present invention. The details in the embodiments do not constitute a limitation on the scope of the present invention. Any obvious changes, such as equivalent transformations or simple substitutions, based on the technical solutions of the present invention without departing from the spirit and scope of the present invention fall within the protection scope of the present invention.

Claims

1. A method for evaluating the interfacial performance of diamond coatings based on cross-sectional micro Raman spectroscopy, characterized in that, The method includes: S1. Obtain a sample with a flat cross-section containing a diamond coating; S2. Using a micro Raman spectrometer, Raman surface scanning is performed on the spatial measurement points containing the film-substrate interface on the flat cross-section of the sample to obtain Raman spectral data for each spatial measurement point; the Raman spectral data is subjected to peak separation, deconvolution, and fitting to extract the characteristic peak parameters of the corresponding spatial measurement points; based on the characteristic peak parameters, the Raman characteristic parameters of the corresponding spatial measurement points are calculated, including diamond content parameters and interface residual stress. S3. Perform an indentation test on the region of the sample containing the film-substrate interface to obtain the measured indentation hardness and / or measured indentation modulus in the region containing the film-substrate interface. S4. Using the Raman characteristic parameters and the characteristic peak parameters as input variables, and the measured indentation hardness and / or the measured indentation modulus as output variables, construct a quantitative correlation prediction model; S5. Substitute the Raman characteristic parameters and characteristic peak parameters of the sample to be tested into the quantitative correlation prediction model to obtain the predicted mechanical properties of the sample to be tested.

2. The method for evaluating the interface performance of diamond coatings based on cross-sectional micro Raman spectroscopy according to claim 1, characterized in that, Step S1 includes: Diamond-coated cutting tools are cut using wire electrical discharge machining to obtain cross-sectional samples containing the interface between the coating and the substrate. Specifically, for drill-type irregular cutting tools, the cross-section to be tested is a longitudinal section containing the cutting edge coating obtained by axial cutting; for ball end mills, the cross-section to be tested is a longitudinal section containing the top coating of the ball end mill obtained by axial cutting through the apex of the ball end mill. The cross-sectional sample is placed in the mounting mold with the cross-section to be tested facing upwards. It is then cold-mounted by curing epoxy resin at room temperature for 12-24 hours, or hot-mounted by using bakelite powder at a temperature of 150-180℃, a pressure of 20-30MPa and a holding time of 5-10 minutes, to prepare a mounting block with a diameter of 30-40mm. The insert is ground and polished until the roughness of the polished cross-section meets the requirements. The sample containing the diamond coating with a smooth cross-section was obtained.

3. The method for evaluating the interface performance of diamond coatings based on cross-sectional micro Raman spectroscopy according to claim 1, characterized in that, The Raman plane scanning in step S2 includes: The sample is placed on the sample stage of the micro Raman spectrometer, and the sample position is adjusted so that the scanning area completely covers the entire area from the near-film-substrate interface region to the outer side of the coating; wherein, the scanning area is centered on the film-substrate interface and extends a predetermined distance towards both the coating side and the substrate side; and is set along a direction parallel to the film-substrate interface. A scan line, and the The scan lines are arranged at a set step size in a direction perpendicular to the membrane-substrate interface, and the data is acquired along a direction parallel to the membrane-substrate interface on each scan line. A total of [number] spatial measurement points were collected, and a total of [number] data were collected. Raman spectral data of spatial measurement points.

4. The method for evaluating the interface performance of diamond coatings based on cross-sectional micro Raman spectroscopy according to claim 1, characterized in that, The extraction process of the characteristic peak parameters in step S2 includes: The Raman spectral data are processed to remove spectral background interference using polynomial fitting or adaptive baseline subtraction methods. The Raman spectral data were subjected to peak deconvolution fitting using a Lorentzian-Gaussian mixture function. The characteristic peaks obtained from the deconvolution included diamond peaks and amorphous peaks. The characteristic peak parameters are obtained by extracting the diamond characteristic peak position, diamond characteristic half-maximum width, diamond peak area, graphite D peak area, graphite G peak area, and amorphous peak area. Peak area.

5. The method for evaluating the interface performance of diamond coatings based on cross-sectional micro Raman spectroscopy according to claim 1, characterized in that, The calculation of the Raman feature parameters in step S2 includes: The formula for calculating the diamond content parameter is as follows: In the formula, This refers to the diamond content parameter; The area of ​​the diamond peak; This is the Raman scattering cross section correction factor; The area of ​​the D peak of graphite; amorphous Peak area; The area of ​​the graphite G peak; The formula for calculating the residual stress at the interface is: In the formula, This refers to residual stress at the interface; Stress coefficient; These are characteristic peaks of diamond.

6. The method for evaluating the interface performance of diamond coatings based on cross-sectional micro Raman spectroscopy according to claim 1, characterized in that, The indentation test in step S3 includes: Indentation tests were performed on the region containing the film-substrate interface using a nanoindenter; wherein the indentation test used a Berkovich pyramidal diamond indenter; the loading mode was set to constant loading rate mode or constant strain rate mode, and the maximum load was set so that the indentation depth did not exceed 5-10% of the diamond coating thickness; Based on the Oliver-Pharr method, the measured indentation hardness and measured indentation modulus at each spatial measurement point are calculated from the load-displacement curves recorded in the indentation test.

7. The method for evaluating the interface performance of diamond coatings based on cross-sectional micro Raman spectroscopy according to claim 1, characterized in that, The construction of the quantitative correlation prediction model in step S4 includes: A positioning mark is introduced in the region of the membrane-substrate interface for spatial coordinate registration of the Raman surface scan and the indentation test in the same region; Based on the positioning marks, the measured indentation hardness and / or measured indentation modulus of multiple spatial measurement points within the same area are arithmetically averaged to obtain the mean measured indentation hardness and / or mean measured indentation modulus within the same area; the characteristic peak parameters and Raman characteristic parameters of the corresponding spatial measurement points within the Raman surface scanning area are arithmetically averaged to obtain the mean of each characteristic peak parameter and the mean of each Raman characteristic parameter. The quantitative correlation prediction model is constructed using the mean values ​​of the characteristic peak parameters and the mean values ​​of the Raman characteristic parameters as input variables, and the mean values ​​of the measured indentation hardness and / or the measured indentation modulus in the same region as output variables, employing multiple linear regression, partial least squares regression, or machine learning algorithms.

8. The method for evaluating the interface performance of diamond coatings based on cross-sectional micro Raman spectroscopy according to claim 1, characterized in that, The expression for the quantitative correlation prediction model is: In the formula, As a comprehensive performance evaluation index for diamond coatings; It is a quantitative mapping function; These are characteristic peaks of diamond; The half-width at half-maximum characteristic of diamond; This refers to the diamond content parameter; The ratio of the intensity or area of ​​the graphite D peak to the graphite G peak; This refers to the residual stress at the interface.

9. The method for evaluating the interface performance of diamond coatings based on cross-sectional micro Raman spectroscopy according to claim 1, characterized in that, The quantitative correlation prediction model is a multiple linear regression model: or in, This is the predicted value for indentation hardness; This is the predicted value of the indentation modulus; The characteristic half-width of diamond , Diamond content parameter , The intensity ratio or area ratio of the graphite D peak to the graphite G peak. , For interfacial residual stress ; , , , , as well as , , , , is the regression coefficient.

10. The method for evaluating the interface performance of diamond coatings based on cross-sectional micro Raman spectroscopy according to claim 9, characterized in that, The quantitative correlation prediction model also includes a nonlinear transformation term for the input variables: The nonlinear transformation term is one or more of the quadratic term, power function transformation term, exponential transformation term, or logarithmic transformation term of the input variable.

Images