Silicon carbide lapped wafer and method for non-destructively detecting lapping damage layer of lapped wafer

Abstract

Disclosed are a silicon carbide lapped wafer and a method for non-destructively detecting a lapping damage layer of the lapped wafer, belonging to the technical field of silicon carbide production and processing. The detection method comprises the following steps: (1) extending inward 20-30 μm from a surface of the lapped wafer, so as to perform Raman testing on a surface layer of the lapped wafer, to obtain a stress distribution diagram of the surface layer of the lapped wafer; (2) dividing the stress distribution diagram into first square areas and second square areas having a side length of 100 microns, the first square areas being areas having stress absolute values greater than 35 MPa, and the second square areas being areas having stress absolute values less than or equal to 30 MPa, and determining whether the lapping damage layer of the lapped wafer is completely removed according to a proportion of first square areas. The present application can directly perform non-destructive testing of a lapped wafer by means of Raman testing, thereby reflecting the size of a lapping damage layer of the lapped wafer, and can also be directly used to guide improvement of the lapping process and polishing process based on test results.

Description

Silicon carbide grinding sheet and method for grinding damaged layer of nondestructive testing grinding sheet Technical Field

[0001] The present application relates to a silicon carbide grinding sheet and a method for grinding a damaged layer of the grinding sheet for non-destructive testing, belonging to the technical field of silicon carbide production and processing. Background Art

[0002] Silicon carbide crystals are cut and separated using multi-wire sawing to obtain silicon carbide wafer substrates. However, due to the presence of numerous wire marks and a damaged layer on the surface after sawing, grinding is required to remove this damaged layer and produce abrasive sheets. During the grinding process, the abrasive and the silicon carbide substrate are subjected to pressure and hard contact friction via the grinding disc to achieve the purpose of thickening and removing the damaged layer. However, this process inevitably reintroduces the damaged layer, so the abrasive sheets generally require a further polishing process to further remove the damaged layer.

[0003] The introduction of a grinding damage layer will affect the surface roughness and other physical parameters of the silicon carbide substrate. Currently, the KOH etching method is often used to verify the presence of a grinding damage layer on the wafer. By etching the wafer with a hot KOH solution, the damage to the polished surface after polishing can be confirmed. In addition to normal dislocation corrosion pits, if there are a large number of dense teardrop-shaped corrosion pits, it means that the grinding damage layer still remains after polishing. If only dislocation corrosion pits are visible under an optical microscope after corrosion, and other areas are flat and smooth, it means that the grinding damage layer has been basically removed. Since the surface of the grinding wafer is in a non-mirror state, no defects can be seen by corrosion, so this method is only performed on polished samples. Moreover, this method is a destructive method. If the damage layer is not completely removed, it needs to be reworked to the grinding stage for secondary removal, which further consumes manpower and material resources. Summary of the Invention

[0004] In order to solve the above problems, a method for non-destructive detection of the grinding damage layer of the grinding sheet is provided. The grinding damage layer of the grinding sheet can be directly tested by Raman, and the stress of the surface layer of the grinding sheet can be characterized, thereby reflecting the size of the grinding damage layer of the grinding sheet. This detection method is a non-destructive test, and the test results can, on the one hand, guide the improvement of the grinding process to reduce the introduction of grinding damage stress, and on the other hand, guide the subsequent polishing process to further remove the grinding damage layer of the grinding sheet.

[0005] According to one aspect of the present application, a method for non-destructive detection of a damaged layer of a polishing sheet is provided, comprising the following steps:

[0006] (1) Extending 20-30 μm inward from the surface of the polishing sheet to perform Raman testing on the surface of the polishing sheet to obtain a stress distribution map of the surface of the polishing sheet;

[0007] (2) The stress distribution diagram is divided into a first square area and a second square area, both of which have a side length of 100 μm. The first square area is an area where the absolute value of the stress value is greater than 35 MPa, and the second square area is an area where the absolute value of the stress value is less than or equal to 30 MPa. If the overall proportion of the stress distribution diagram in the first square area is greater than 3.5% and is irregularly scattered, it indicates that the grinding damage layer has not been completely removed. If the overall proportion of the stress distribution diagram in the first square area is less than or equal to 3.5% and there is no obvious abnormal scattered distribution, it indicates that the grinding damage layer has been completely removed.

[0008] The Raman detection method can be used to perform non-destructive testing on the grinding sheet, without having to wait until the grinding sheet is polished before testing, and can more intuitively reflect the quality of the grinding process; if the grinding damage layer is not completely removed in the test results, it can be determined whether to rework to the grinding stage or directly proceed to the next polishing process based on the state of the grinding damage layer of the grinding sheet. Since the polishing removal rate is slower than grinding, when the detection method detects that there is less residual grinding damage layer on the grinding sheet, the polishing process can be directly carried out to remove the excess grinding damage layer. When the grinding damage layer remains seriously, it cannot be completely removed by polishing alone, and the grinding process needs to be optimized to perform secondary grinding to reduce the grinding damage layer before polishing to remove it. Therefore, this detection method can save the processing steps of the wafer and avoid problems such as thinning or uneven wafer thickness caused by secondary grinding, polishing and other processes.

[0009] Optionally, the stress distribution diagram is calculated as follows:

[0010] Coefficient determination: Before the Raman test, calculate the peak position shift and stress conversion coefficient of the polished sheet;

[0011] Test calculation: The peak value and peak offset of the wafer are obtained according to the Raman test, and the stress of the surface layer of the polishing sheet is calculated by the peak offset and the peak offset in step S1 and the stress conversion coefficient, so as to obtain the stress distribution diagram.

[0012] Optionally, the diameter of the grinding sheet is greater than 150 mm, preferably greater than 200 mm.

[0013] Optionally, the steps for calculating the peak position offset and the stress conversion coefficient are:

[0014] S1: Use a tester to test and calculate the interplanar spacing dx and interplanar spacing difference △d at different positions of the polishing sheet, and use Hooke's law or stress-strain formula to calculate the stress value at different positions;

[0015] S2: Use Raman to test the peak position value of the polishing sheet and calculate the peak position offset at different positions. The ratio of the peak position offset to the stress value is the peak position offset to stress conversion coefficient.

[0016] Optionally, the tester in S1 is selected from any one or more of an XRD diffractometer, a synchrotron radiation tester, a white light tester, a neutron diffractometer, a scanning electron microscope, and a transmission electron microscope.

[0017] When the substrate is tested using an XRD diffractometer, the interplanar spacing dx at different positions on the substrate is calculated using the Bragg diffraction formula.

[0018] In the above substrate surface stress test and substrate internal stress test, the peak position value and peak position offset in step S2 are obtained by fitting, and the fitting uses functions including Gaussian, Lorentz, GaussLor, AGauss, Aloren and AgaussLor.

[0019] Optionally, an equidistant focusing mode is used in the Raman test of step (1), wherein the equidistant focusing mode controls the distance between the Raman laser spot at each test point and the surface of the polishing sheet to be equal, and the peak position value and peak position offset are obtained after fitting.

[0020] The Raman tester's automatic focusing function ensures that the laser spot hits the surface of the abrasive sheet, so that the test is not affected by the surface undulations of the abrasive sheet, thereby accurately testing the stress condition of the surface of the abrasive sheet.

[0021] Optionally, the grinding sheet is a silicon carbide grinding sheet;

[0022] Preferably, the crystal form of the silicon carbide grinding sheet is one of 2H-SiC, 4H-SiC, 6H-SiC, 3C-SiC, and 15R-SiC.

[0023] According to another aspect of the present application, a silicon carbide grinding sheet is provided, wherein the silicon carbide grinding sheet is obtained by cutting and grinding a silicon carbide crystal, and the silicon carbide grinding sheet comprises a first main surface and a second main surface opposite to the first main surface;

[0024] The silicon carbide polishing sheet is divided into a first square region having a surface stress absolute value greater than 35 MPa and a length of 100 μm, and a second square region having a surface stress absolute value less than or equal to 30 MPa and a length of 100 μm, wherein the ratio of the number of the first square regions to the number of the second square regions is 1:(30-100), preferably 1:(31-100), and more preferably 1:(67-100);

[0025] The absolute value of the surface stress is the absolute value of the stress detected in a region extending 20-30 μm inward from the first main surface or the second main surface.

[0026] The smaller the ratio of the number of the first square regions to the number of the second square regions, the larger the proportion of the second square regions, the smaller the proportion of the unremoved grinding damage layer contained in the silicon carbide abrasive sheet, and the better the quality of the silicon carbide abrasive sheet, which is more conducive to subsequent polishing and processing. Preferably, the ratio of the number of the first square regions to the number of the second square regions is 1:100.

[0027] Due to the shape limitation of the silicon carbide grinding wheel, for the first square area and the second square area at the edge of the silicon carbide grinding wheel, there may be a situation where the area of ​​one first square area or the second square area is not satisfied. Therefore, the number of first square areas and second square areas may not be an integer. For example, if the edge of the silicon carbide grinding wheel can only be divided into 50μm areas, but the absolute value of the surface stress in this area is less than or equal to 20MPa, it is calculated as 0.25 second square areas according to the area ratio.

[0028] Optionally, at any plane parallel to the first main surface and / or the second main surface, S1 represents the average value of the absolute value of the surface stress in all the first square regions, S2 represents the average value of the absolute value of the surface stress in any of the first square regions, and 0.79≤S1 / S2≤1.35.

[0029] Optionally, the diameter of the silicon carbide grinding disc is greater than 150 mm, preferably greater than 200 mm.

[0030] Optionally, at any plane parallel to the first main surface and / or the second main surface, S3 represents the average value of the absolute values ​​of the surface stress in all the second square regions, S4 represents the average value of the absolute values ​​of the surface stress in any second square region, and 0.21≤S3 / S4≤1.50;

[0031] Preferably, 1.5≤S1 / S3≤10.

[0032] More preferably, 1.5≤S1 / S3≤9.9.

[0033] Multi-point stress detection is performed on the silicon carbide polishing sheet. The average value of the absolute value of the surface stress in the first square area or the second square area refers to the value obtained by dividing the sum of the absolute values ​​of the surface stress at all detection points in the area by the number of detection points. The multi-point stress detection can be performed by selecting multiple points at different positions in the area for detection, or by performing a mapping test on the substrate.

[0034] The above-mentioned ratio of S1 / S2 represents the difference between any first square area and the average value of all first square areas in the radial direction. The closer the ratio is to 1, the smaller the difference between the surface stresses of all first square areas in the radial direction, that is, the smaller the regional difference of the grinding damage layer, which means that the distribution of the grinding damage layer on the silicon carbide grinding sheet is more uniform.

[0035] Similarly, the above-mentioned S3 / S4 ratio represents the difference between any second square area and the average value of all second square areas in the radial direction. The closer the ratio is to 1, the smaller the difference between the surface stresses of all second square areas in the radial direction, and the more uniform the stress distribution of the silicon carbide grinding sheet is, that is, the better the quality of the silicon carbide grinding sheet; the S1 / S3 ratio represents the average radial internal stress ratio of the area with a grinding damage layer and the area without a grinding damage layer of the silicon carbide substrate in the radial direction, representing the difference between all the first square areas and the second square areas. The more grinding damage layers remain in the first square area, the greater the stress value of the first square area, and the greater the S1 / S3 ratio, which means that the stress difference of the silicon carbide grinding sheet is greater, that is, the worse the quality is, and the less conducive it is to subsequent polishing and production processing.

[0036] Optionally, at any plane parallel to the first main surface and / or the second main surface, S max1 S is the maximum value of the surface stress in any of the first square regions. min1 Denote the minimum value of the surface stress in any of the first square regions, △S1= S max1 -S min1 , 10MPa≤△S1≤20MPa.

[0037] Preferably, 10.9 MPa≤ΔS1≤20 MPa.

[0038] Optionally, at any plane parallel to the first main surface and / or the second main surface, S max2 S is the maximum internal stress in any second square region. min2 It is denoted as the minimum surface stress in the second square region, △S2= S max2 -S min2 , 10MPa≤△S2≤30MPa.

[0039] Preferably, 10.2MPa≤ΔS2≤30MPa

[0040] Since the silicon carbide grinding sheet is subjected to tensile and compressive stress, the interplanar spacing d undergoes corresponding expansion and contraction changes. Therefore, when using Raman to test stress, the Raman peak intensity will appear to move toward low frequency or high frequency. When the silicon carbide grinding sheet is subjected to tensile stress, the Raman peak shifts toward low frequency, and the obtained stress value is positive. When the silicon carbide grinding sheet is subjected to compressive stress, the Raman peak shifts toward high frequency, and the obtained stress value is negative. Therefore, the positive and negative before the stress value in this application represent the force direction of the silicon carbide grinding sheet, and the absolute value of the value represents the magnitude of the stress. For example, when the surface stress in the first square area is -20MPa and the surface stress in the second square area is -10MPa, it means that the surface stresses in the first square area and the second square area are both compressive stresses, and the surface stress in the first square area is greater than the surface stress in the second square area.

[0041] The above S max1 、S min1 、S max2 and S min2 The meaning is the true value of the radial surface stress, which distinguishes compressive stress and tensile stress in the radial direction. For example, at any plane parallel to the first main surface and / or the second main surface, the radial surface stress of the first square area is -10~5MPa, then the maximum value of the radial surface stress is 5MPa, and the minimum value of the radial surface stress is -10MPa. △S1 is 15MPa. The △S1 and △S2 respectively represent the degree of change in the interplanar spacing of the silicon carbide grinding sheet in the radial direction in the first square area and the second square area. The smaller the above △S1 and △S2 are, the smaller the change in the interplanar spacing of the silicon carbide grinding sheet in the radial direction is, which means that the stress distribution of the silicon carbide grinding sheet is more uniform and the quality of the silicon carbide grinding sheet is better.

[0042] The above stress includes absolute stress and relative stress. Absolute stress can reflect the difference between the wafer and a perfect SiC crystal without defects and is used to determine the stress level of the SiC wafer. However, since perfect defect-free wafers are difficult to achieve with current technology, relative stress can be used to determine the relative stress distribution within the wafer surface.

[0043] For example, when the absolute stress of the chip is large and the relative stress is small, it means that the in-plane lattice distortion causes large stress, but the stress distribution is uniform, the crystal quality is poor, and the stress quality in each area of ​​the chip is relatively uniform; if the absolute stress is small and the relative stress is large, the overall chip quality is good, but there are abnormal stress quality deviations in some areas.

[0044] In this application, the reference value of absolute stress is the reference stress calculated based on the standard Raman peak position obtained based on the perfect lattice parameters of SiC, which is marked as 0; the relative stress is to take a reference value on the entire test plane, and the value of each test point is correlated with the reference value to obtain the relative stress value of the test point. The reference value includes but is not limited to any one of the median, mean, mode or other statistical function calculation results of all stress values ​​on the entire test plane.

[0045] The beneficial effects of this application include but are not limited to:

[0046] 1. The present invention provides a method for detecting abrasive damage layers on abrasive sheets, allowing for direct testing of the sheet. This method not only intuitively characterizes the extent to which the grinding process has removed the damaged layer, but also enables non-destructive testing. This allows for selection of appropriate post-processing procedures based on the residual damaged layer, thus saving testing time and avoiding waste of the sheet.

[0047] 2. The results obtained from the method for detecting the grinding damage layer of an abrasive sheet according to the present application can, on the one hand, guide improvements to the grinding process to reduce the introduction of grinding damage stress, and on the other hand, guide subsequent polishing processes to further remove the grinding damage layer of the abrasive sheet.

[0048] 3. The silicon carbide abrasive sheet of the present application has a low residual surface damage layer, which facilitates subsequent polishing and processing. Its use in the preparation of silicon carbide devices can improve the performance and yield of silicon carbide devices, facilitate mass production and processing, and reduce the processing costs of silicon carbide devices.

[0049] 4. According to the silicon carbide abrasive sheet of the present application, the radial difference between the first square region and the second square region is small. Furthermore, the ratio of the number and surface stress of all first square regions to all second square regions indicates that the first square regions account for a smaller proportion, thus demonstrating the good quality of the silicon carbide abrasive sheet. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0051] FIG1 is a flow chart of calculation of peak position offset and stress conversion coefficient involved in Example 1 of the present application.

[0052] FIG2 is a flow chart of the Raman stress test involved in Example 1 of the present application.

[0053] FIG3 is a schematic diagram of the test of surface stress of the polishing sheet in the Raman equidistant focusing mode.

[0054] Figure 4 is a Raman test image of the polished sheet.

[0055] FIG5 is a schematic diagram of the polishing wafer after being etched with KOH solution. Implementation Method

[0056] The present application is described in detail below with reference to embodiments, but the present application is not limited to these embodiments.

[0057] Unless otherwise specified, the raw materials in the examples of this application were purchased through commercial channels. Example

[0058] This embodiment relates to a method for detecting a grinding damage layer of a grinding sheet, comprising the following steps:

[0059] Referring to Figure 1, we first use a single crystal XRD diffractometer to test the silicon carbide wafer and analyze the substrate based on the reciprocal lattice space, calculate the parameters of the standard silicon carbide wafer and compare them with the standard parameters. Then, according to the Bragg diffraction equation: 2dsinθ=nλ, where d is the interplanar spacing, θ is the diffraction half-angle, n is the diffraction order, and λ is the wavelength, we test the θ and d of different silicon carbide wafers. It can be seen that the tensile and compressive stress inside the wafer causes the interplanar spacing d to undergo corresponding expansion and contraction changes. According to the different interplanar spacings d0 and d x , d x It is divided into d1 and d2, where d0 is the theoretical interplanar spacing, d1 is the interplanar spacing after tensile stress, and d1 is greater than d0; d2 is the interplanar spacing after compressive stress, and d2 is less than d0. The interplanar spacing difference △d at different positions is calculated, △d = d1- d0 or △d = d0- d2.

[0060] The actual peak position of the silicon carbide wafer is obtained according to the Raman test, and the peak position offset △v at different test points is calculated compared with the standard peak position; the stress value σ of the silicon carbide wafer at different test points is calculated in combination with Hooke's law or stress-strain formula. The ratio of the peak position offset to the stress value at different test points is the peak position offset to stress conversion coefficient μ, which ranges from -125 to -2500.

[0061] The calculation formula of Hooke's law is σ / S=E×(△d / d0), where S is the area where σ acts, which is a fixed value, E is the Young's modulus of the substrate, which is a fixed parameter of the substrate, △d is the difference in crystal plane spacing at different positions, d0 is the theoretical crystal plane spacing, and σ is the stress value that needs to be calculated.

[0062] The calculation formula for stress and strain is σ = E × △d, where E is the Young's modulus of the substrate, which is a fixed parameter of the substrate, △d is the difference in crystal plane spacing at different positions, and σ is the stress value to be calculated.

[0063] Refer to Figure 2. First, set the Raman test points. You can perform a single-point test or a mapping scan of the entire polishing wafer. Single-point testing involves setting up discrete coordinate points, while mapping scans can range from 10x10 to 50x50 or more points. You can choose the appropriate test method based on the properties of the substrate to determine the Raman test coordinate diagram for the polishing wafer.

[0064] The test results are then output, including but not limited to editable peak positions, peak intensity data, and mapping.

[0065] The output data is then subjected to peak position fitting, with fitting functions including but not limited to Gaussian, Lorentz, GaussLor, AGauss, Aloren, and AGaussLor, to obtain precise peak position values ​​and peak position offsets, with the peak position accurate to at least one decimal place.

[0066] Finally, according to the peak shift and stress conversion coefficient μ calculated in (1) and the peak shift △v obtained in the Raman test, the stress value of silicon carbide is calculated as follows: σ (MPa) = μ × △v (cm -1 ).

[0067] The Raman test adopts an equidistant focusing test method. The test principle is shown in Figure 3. Figure 3 (b) is a schematic diagram of the focus of the Raman laser on the damaged layer on the substrate surface. Figure 3 (a) is a diagram of the test principle. Under this test method, the focal depth (d) of the Raman laser focused on the damaged layer on the surface of the polishing wafer at different test points remains unchanged.

[0068] By debugging the parameters of the Raman tester and using the automatic focusing function to focus the laser on the surface of the grinding sheet, the surface damage layer of the grinding sheet falls within the range of the laser spot, thereby achieving equidistant focusing tests at different test points of the grinding sheet, that is, the distance between the focusing plane and the substrate surface is always equal. This method is not affected by the bending and warping of the grinding sheet surface, so that the obtained Raman peak intensity is always maintained at the axial maximum value at the test point. According to the above-mentioned stress quantitative test method, the stress test image of the grinding sheet is obtained as shown in Figure 4. Figure 4 (a) is a schematic diagram of the grinding damage layer of the grinding sheet that is not completely removed, and Figure 4 (b) is a schematic diagram of the grinding damage layer of the grinding sheet that is completely removed.

[0069] The polished wafer obtained after the same grinding process was etched with hot KOH solution and polished is shown in Figure 5. Under an optical microscope, Figure 5 (a) shows a large number of dense teardrop-shaped corrosion pits, which is a schematic diagram of the incomplete removal of the grinding damage layer of the grinding wafer. Figure 5 (b) shows only dislocation corrosion pits, and the other areas are flat and smooth, which is a schematic diagram of the complete removal of the grinding damage layer of the grinding wafer. Example

[0070] This embodiment relates to a grinding process for silicon carbide wafers. The cut silicon carbide wafers are placed in a grinder and ground using a mixture of nanofluid, abrasive, and slurry. The grinding process uses a grinding pad inlaid with diamond particles of W5. The specific grinding process is as follows:

[0071] 1. First, prepare the nanofluid, select dispersant-modified SiO2, Si, and SiC as nanopowders, the concentration of the nanopowder is 1.5%-3.5%, and the particle size of the nanopowder is 10-1000 nm; use heat-conducting oil or a mixed solution of ethylene glycol and water as the nanofluid carrier liquid, and add 10%-11% of the dispersant and 3%-5% of the modifier. Use an ultrasonic stirrer to ultrasonically stir the above-mentioned liquid until the nanopowder is evenly dispersed in the carrier liquid. Finally, use diluted HCl and NaOH as pH regulators to adjust the pH to 7.5-9 to obtain the nanofluid.

[0072] In the preparation of nanofluids, the above “%” refers to the weight percentage of the added substance in the nanofluid carrier liquid.

[0073] The abrasive material is then prepared, using a nanofluid as the abrasive carrier. A 20%-30% abrasive of boron carbide and diamond powder is added to the nanofluid, where the weight ratio of boron carbide to diamond powder is 1.5-4:1, the boron carbide particle size is 5-6 μm, and the diamond particle size is 6-8 μm. A 1%-3% suspending agent is then added, followed by a dispersant, so that the volume ratio of suspending agent to dispersant in the system is 1:20-25, thus obtaining the abrasive material.

[0074] In the preparation of the abrasive material, the above “%” refers to the weight percentage of the added substance in the nanofluid.

[0075] The above-mentioned dispersants include but are not limited to glycerol, polyethylene glycol, sodium hexametaphosphate, sodium pyrophosphate, and sodium orthophosphate; the modifiers include but are not limited to soft sodium alkylbenzene sulfonate and taurate; and the suspending agents include but are not limited to polyvinyl alcohol and BYK.

[0076] 2. Then, the cut SiC wafer is roughly ground to remove the surface marks of the cut piece. The diamond powder particle size is W7 and the pressure is 30g / cm 2 The speed ratio of the inner and outer ring gears is 1:2 for rough grinding, and the rough grinding time is 1 hour.

[0077] 3. After the rough grinding, the middle grinding is carried out to remove the micro cracks under the cutting line and thin the cutting damage layer. The process used is to select diamond powder as W5 and adjust the pressure to 70g / cm 2 The speed ratio of the inner and outer gear rings is 1:2 for mid-section grinding, and the grinding time is 1h.

[0078] 4. Finally, the sample is finely ground to remove the damage layer introduced by the previous grinding as much as possible, so that the subsequent polishing can remove the grinding damage layer as much as possible. The diamond powder particle size is W3.5 and the pressure is 50g / cm 2 The speed ratio of the inner and outer ring gears is 1:4 for fine grinding, and the fine grinding time is 3 hours to obtain silicon carbide grinding discs.

[0079] Silicon carbide polishing sheets 1#-8# were prepared using the above-mentioned polishing process, wherein steps 2-4 were made using the above-mentioned process, and the differences in step 1 were as follows:

[0080] First, a nanofluid was prepared, and SiC surface-modified with sodium hexametaphosphate was selected as a nanopowder with a concentration of 3.5% and a particle size of 300 nm. A mixed solution of ethylene glycol and water was used as a carrier liquid for the nanofluid grinding liquid with a ratio of ethylene glycol to water of 3:5. In addition, 10% glycerol and 0.1% sodium hexametaphosphate were added as dispersants, and 5% soft sodium alkylbenzene sulfonate was added as a modifier. The above-mentioned liquid was ultrasonically stirred using an ultrasonic stirrer until the nanopowder was evenly dispersed in the carrier liquid. Finally, diluted HCl and NaOH were used as pH regulators to adjust the pH to 8 to obtain a nanofluid.

[0081] An abrasive is then prepared, using a nanofluid as an abrasive carrier. A 25% abrasive of boron carbide and diamond powder is added to the nanofluid, where the weight ratio of boron carbide to diamond powder is 3:1, the boron carbide particle size is 6 μm, and the diamond particle size is 7 μm. 1.4% BYK is then added as a suspending agent, and glycerin is further added as a dispersant, resulting in a suspending agent to dispersant volume ratio of 1:25 in the entire system. This yields abrasive material 1#, which is then used to grind the cut wafer in steps 2-4 to obtain silicon carbide abrasive sheet 1#.

[0082] First, a nanofluid was prepared, and SiO2 surface-modified with sodium pyrophosphate was selected as the nanopowder. The concentration of the nanopowder was 3.5%, and the particle size of the nanopowder was 10 nm. A mixed solution of ethylene glycol and water was used as the nanofluid grinding liquid carrier, and the ratio of ethylene glycol to water was 3:5. In addition, 10% glycerol and 0.1% sodium pyrophosphate were added as dispersants, and 3% soft alkylbenzene sulfonate was added as a modifier. The above-mentioned liquid was ultrasonically stirred using an ultrasonic stirrer until the nanopowder was evenly dispersed in the carrier liquid. Finally, diluted HCl and NaOH were used as pH regulators to adjust the pH to 7.5 to obtain the nanofluid.

[0083] An abrasive is then prepared, using a nanofluid as an abrasive carrier. A 20% abrasive of boron carbide and diamond powder is added to the nanofluid, where the weight ratio of boron carbide to diamond powder is 4:1, the boron carbide particle size is 5 μm, and the diamond particle size is 8 μm. 3% BYK is then added as a suspending agent, and glycerin is further added as a dispersant, resulting in a suspending agent to dispersant volume ratio of 1:20. This yields abrasive material 2#, which is then used to grind the cut wafer in steps 2-4 to obtain silicon carbide abrasive sheet 2#.

[0084] First, a nanofluid was prepared, and Si surface-modified with sodium orthophosphate was selected as the nanopowder. The concentration of the nanopowder was 1.5%, and the particle size of the nanopowder was 1000 nm. A mixed solution of ethylene glycol and water was used as the nanofluid grinding liquid carrier, and the ratio of ethylene glycol to water was 3:5. In addition, 10% glycerol and 1% sodium orthophosphate were added as dispersants, and 5% taurate was added as a modifier. The above-mentioned liquid was ultrasonically stirred using an ultrasonic stirrer until the nanopowder was evenly dispersed in the carrier liquid. Finally, diluted HCl and NaOH were used as pH regulators to adjust the pH to 9 to obtain the nanofluid.

[0085] Next, an abrasive is prepared, using a nanofluid as an abrasive carrier. A 30% abrasive of boron carbide and diamond powder is added to the nanofluid, where the weight ratio of boron carbide to diamond powder is 2:1, and the boron carbide particle size is 6 μm, while the diamond particle size is 6 μm. 1% BYK is then added as a suspending agent, and glycerin is further added as a dispersant, resulting in a suspending agent to dispersant volume ratio of 1:25. This yields abrasive material 3#, which is then used to grind the cut wafer in steps 2-4 to obtain silicon carbide abrasive sheet 3#.

[0086] The difference between this silicon carbide grinding sheet 4# and silicon carbide grinding sheet 1# is that SiO2 surface-modified with sodium pyrophosphate and Si surface-modified with sodium orthophosphate are used as nanopowders, with concentrations of 2% and 1% respectively. The remaining substances and grinding steps are the same as those of silicon carbide grinding sheet 1#, thus obtaining silicon carbide grinding sheet 4#.

[0087] The difference between this silicon carbide grinding sheet 5# and silicon carbide grinding sheet 1# is that heat transfer oil is used as the nanofluid carrier liquid, and the other materials and grinding steps are the same as those of silicon carbide grinding sheet 1#, so as to obtain silicon carbide grinding sheet 5#.

[0088] The difference between this silicon carbide grinding sheet 6# and silicon carbide grinding sheet 1# is that the particle size of the nano powder is 800nm, and the remaining materials and grinding steps are the same as those of silicon carbide grinding sheet 1#, thus obtaining silicon carbide grinding sheet 6#.

[0089] The difference between the comparative silicon carbide grinding sheet D1# and the silicon carbide grinding sheet 1# is that the concentration of the nanopowder used is 5%, and the remaining materials and grinding steps are the same as those of the silicon carbide grinding sheet 1#, thus obtaining the comparative silicon carbide grinding sheet D1#.

[0090] The difference between the comparative silicon carbide grinding sheet D2# and the silicon carbide grinding sheet 1# is that the particle size of boron carbide is 10μm, the particle size of diamond is 3μm, and the remaining materials and grinding steps are the same as those of the silicon carbide grinding sheet 1#, thus obtaining the comparative silicon carbide grinding sheet D2#.

[0091] The above silicon carbide polishing sheet was characterized using the detection method of Example 1, and the data is shown in Table 1 below.

[0092] Table 1

[0093] The ratio of the number of the first square area to the number of the second square area of ​​the sample △S1△S2S1 / S2S3 / S4S1 / S3Silicon carbide polishing sheet 1#1:9514.315.70.91-1.230.75-1.323.4Silicon carbide polishing sheet 2#1:6716.519.90.84-1.260.68-1.376.3Silicon carbide polishing sheet 3#1:3120.024.30.79-1.350.21-1.509.9Silicon carbide polishing sheet 4#1:10010.910.20.9 3-1.200.78-1.291.5Silicon carbide grinding sheet 5#1:8816.430.00.84-1.240.62-1.375.2Silicon carbide grinding sheet 6#1:8115.317.40.81-1.290.60-1.394.9Comparison Silicon carbide grinding sheet D1#1:2724.928.70.73-1.420.44-1.5810.2Comparison Silicon carbide grinding sheet D2#1:2428.432.40.68-1.610.42-1.7912.1

[0094] The performance parameters of the grinding sheet in Table 1 are all calculated through relative stress values, and the data in Table 1 are only the grinding sheets prepared by combining some of the processing parameters in Example 2. When the grinding sheet is prepared using other processing parameters in the grinding process of Example 2, and the performance parameters of the grinding sheet in Table 1 are characterized by absolute stress, the grinding sheet can still meet the requirements that the ratio of the number of the first square area to the number of the second square area is 1: (30-100); 0.79≤S1 / S2≤1.35; 0.21≤S3 / S4≤1.50; 1.5≤S1 / S3≤10; 10MPa≤△S1≤20MPa; 10MPa≤△S2≤30MPa.

[0095] The foregoing is merely an embodiment of the present application, and the scope of protection of the present application is not limited by these specific embodiments, but is determined by the claims of the present application. For those skilled in the art, the present application may have various modifications and variations. Any modifications, equivalent substitutions, improvements, etc. made within the technical ideas and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A method for nondestructively testing a grinding damage layer of a grinding sheet, characterized in that: The steps include: (1) Extending 20-30 μm inward from the surface of the abrasive sheet to perform Raman testing on the surface of the abrasive sheet to obtain a stress distribution map of the surface of the abrasive sheet; (2) The stress distribution diagram is divided into a first square area and a second square area, both of which have a side length of 100 μm. The first square area is an area where the absolute value of the stress value is greater than 35 MPa, and the second square area is an area where the stress value is less than or equal to 30 MPa. If the overall proportion of the stress distribution diagram in the first square area is greater than 3.5% and is irregularly scattered, it indicates that the grinding damage layer has not been completely removed. If the overall proportion of the stress distribution diagram in the first square area is less than or equal to 3.5% and there is no obvious abnormal scattered distribution, it indicates that the grinding damage layer has been completely removed.

2. The method according to claim 1, characterized in that The calculation method of the stress distribution diagram is: Coefficient determination: Before the Raman test, calculate the peak position shift and stress conversion coefficient of the polished sheet; Test calculation: obtaining the peak value and peak offset of the wafer according to the Raman test, calculating the stress of the surface layer of the polishing sheet by the peak offset and the peak offset and stress conversion coefficient in step S1, that is, obtaining the stress distribution diagram; and / or The diameter of the grinding sheet is greater than 150 mm.

3. The method according to claim 2, characterized in that The calculation steps of the peak position offset and the stress conversion coefficient are as follows: S1: Use a tester to test and calculate the interplanar spacing dx and interplanar spacing difference △d at different positions of the grinding sheet, and use Hooke's law or stress-strain formula to calculate the stress value at different positions; S2: The peak position value of the polishing sheet is tested by Raman, and the peak position offset at different positions is calculated. The ratio of the peak position offset to the stress value is the peak position offset and stress conversion coefficient.

4. The method according to claim 3, characterized in that The tester in S1 is selected from any one or more of an XRD diffractometer, a synchrotron radiation tester, a white light tester, a neutron diffractometer, a scanning electron microscope, and a transmission electron microscope.

5. The method according to claim 1, characterized in that In the Raman test of step (1), an equidistant focusing mode is used, wherein the equidistant focusing mode controls the distance between the Raman laser spot at each test point and the surface of the polishing sheet to be equal, and the peak position value and the peak position offset are obtained after fitting.

6. A silicon carbide grinding sheet, characterized in that: The silicon carbide grinding sheet is obtained by cutting and grinding silicon carbide crystals, and the silicon carbide grinding sheet includes a first main surface and a second main surface opposite to the first main surface; The silicon carbide polishing sheet is divided into a first square region with a surface stress absolute value greater than 35 MPa and a length of 100 μm, and a second square region with a surface stress absolute value less than or equal to 30 MPa and a length of 100 μm, wherein the ratio of the number of the first square regions to the number of the second square regions is 1:(30-100); The absolute value of the surface stress is the absolute value of the stress value detected in a region extending 20-30 μm inward from the first main surface or the second main surface.

7. The silicon carbide polishing sheet according to claim 6, characterized in that: At any plane parallel to the first main surface and / or the second main surface, S1 represents the average value of the absolute values ​​of the surface stress in all the first square regions, S2 represents the average value of the absolute values ​​of the surface stress in any of the first square regions, and 0.79≤S1 / S2≤1.35; The diameter of the silicon carbide grinding sheet is greater than 150 mm.

8. The silicon carbide polishing sheet according to claim 7, characterized in that: At any plane parallel to the first main surface and / or the second main surface, S3 represents the average value of the absolute values ​​of the surface stress in all the second square regions, S4 represents the average value of the absolute values ​​of the surface stress in any second square region, and 0.21≤S3 / S4≤1.50; Preferably, 1.5≤S1 / S3≤10.

9. The silicon carbide polishing sheet according to claim 6, characterized in that: At any plane parallel to the first main surface and / or the second main surface, S max1 S is the maximum value of the surface stress in any of the first square regions. min1 Denote as the minimum value of the surface stress in any of the first square regions, △S1= S max1 -S min1 , 10MPa≤△S1≤20MPa.

10. The silicon carbide polishing sheet according to claim 9, characterized in that: At any plane parallel to the first main surface and / or the second main surface, S max2 S is the maximum surface stress in any second square region. min2 Denote as the minimum surface stress in any second square region, △S2= S max2 -S min2 , 10MPa≤△S2≤30MPa.

Images