Crystal ingot cutting detection method

Abstract

The invention relates to the technical field of semiconductor manufacturing, and provides a crystal ingot cutting detection method which comprises the following steps: dividing the end face of a crystal ingot into a plurality of cells; an included angle is formed between the irradiation direction of the detection light beam and the normal direction of the end face; receiving reflected light generated after the detection light beam irradiates the end face of the crystal ingot, and performing spectral analysis on the reflected light to obtain the cutting thickness of each cell; if the cutting thickness of the cell is within the thickness threshold range, judging that the cell is a valid cell, and if the cutting thickness of the cell is outside the thickness threshold range, judging that the cell is an invalid cell; calculating a cell cutting effective rate, wherein the cell cutting effective rate is a ratio of the number of effective cells to the total number of the cells; if the cell cutting effective rate is higher than an effective rate threshold value, it is judged that cutting is effective. According to the detection method, a standardized stripping judgment standard is provided for crystal ingot cutting, the detection progress can be improved, and the fragment rate can be reduced.

Description

Technical Field

[0001] The present invention relates to the field of semiconductor manufacturing technology, and in particular to a crystal ingot cutting detection method. Background Art

[0002] Wafer preparation is an important step in semiconductor processing, which refers to the process of cutting the prepared ingot into wafers.

[0003] The ingot cutting process has an important impact on the performance of the wafer. Taking silicon carbide as an example, silicon carbide is a commonly used preparation material for semiconductors. Silicon carbide has high-quality properties such as high energy gap, high thermal conductivity, and high electron saturation drift rate. At present, the production cost of silicon carbide devices has remained high, mainly due to two aspects: one is the long production cycle and limited output; the other is the large loss and low yield of the ingot cutting process. At present, the cutting technology of silicon carbide ingots mainly includes traditional cutting processes such as mortar wire cutting and diamond wire cutting, as well as new cutting processes such as laser cutting.

[0004] The above traditional cutting technologies have a high loss rate and long working hours. Taking mortar wire cutting as an example, up to 40% of the silicon carbide ingot is wasted as dust during the cutting process, and the high-speed running of the cutting wire will also cause 20-50μm roughness and surface / subsurface structure damage.

[0005] The principle of laser cutting technology is to use laser to focus below the surface of the wafer, and scan layer by layer at different depths inside the ingot to generate a single or multiple modified layers. Afterwards, under the action of external tension, the cracks in the modified layer extend in a direction perpendicular to the surface of the wafer, splitting the wafer from the inside to the outside, thereby peeling off the wafer from the silicon carbide ingot.

[0006] Currently, there is no suitable method for detecting peeling conditions in laser cutting, and one can only rely on experience to judge, which can easily cause peeling fragments.

[0007] Therefore, a method for detecting ingot cutting is needed to provide a method for detecting peeling conditions after laser cutting. Summary of the invention

[0008] The object of the present invention is to provide a crystal ingot cutting detection method, which divides the end face of the crystal ingot into multiple cells and uses spectrum analysis of optical reflection to measure the thickness of the peeling surface. By judging the peeling thickness of each cell, it is judged whether the peeling of each cell is effective, thereby judging whether the crystal ingot meets the peeling requirements, providing a standardized peeling judgment standard for crystal ingot cutting.

[0009] In order to solve the above technical problems, the present invention provides a method for detecting ingot cutting, comprising the following steps:

[0010] Divide cells: divide the end surface of the ingot into several cells;

[0011] Irradiating the detection beam: irradiating the detection beam onto the end face of the ingot, with the irradiation direction of the detection beam forming an angle with the normal direction of the end face;

[0012] Calculate the cell cutting thickness: receive the reflected light after the detection beam is irradiated on the end face of the ingot, and perform spectrum analysis on the reflected light to obtain the cutting thickness of each cell;

[0013] Determine valid cells: if the cutting thickness of a cell is within the thickness threshold range, the cell is determined to be a valid cell; if the cutting thickness of a cell is outside the thickness threshold range, the cell is determined to be an invalid cell;

[0014] Determine the cutting effect: calculate the cell cutting efficiency, which is the ratio of the number of effective cells to the total number of cells; if the cell cutting efficiency is higher than the efficiency threshold, the cutting is determined to be effective.

[0015] In the above-mentioned ingot cutting detection method, the end face of the ingot is divided into multiple cells, and the thickness of the peeling surface is measured by using the spectrum analysis of optical reflection. The peeling thickness of each cell is judged to determine whether the peeling of each cell is effective, so as to determine whether the ingot meets the peeling requirements, and provide a standardized peeling judgment standard for ingot cutting, which helps to reduce the peeling breakage rate, reduce labor costs, and improve cutting efficiency. Moreover, the detection method can be applied to the cutting detection of ingots of different specifications, for example, it can be applied to the cutting detection of existing ingots with diameters of 6 inches, 8 inches, and 12 inches. The cutting detection method has a wide range of applications.

[0016] Optionally, determining the cutting effect further includes:

[0017] Calculate the proportion of connected domains: Identify the connected domains, which are the areas of connected invalid cells, and calculate the number of invalid cells contained in the largest connected domain. The proportion of connected domains is the ratio of the number of invalid cells contained in the largest connected domain to the total number of cells. If the proportion of connected domains is less than the threshold of the proportion of invalid connected domains, the cutting is determined to be effective.

[0018] Optionally, determining the cutting effect further includes:

[0019] Calculating the proportion of large-surface connected domains, where the proportion of large-surface connected domains is: the ratio of the number of invalid cells contained in the largest connected domain to the total number of cells in the large surface of the end surface of the ingot;

[0020] If the proportion of the large surface connected domain is less than the large surface connected domain proportion threshold, it is determined that the cutting is valid.

[0021] Optionally, determining the cutting effect further includes:

[0022] Calculating the proportion of small face connected domains, where the proportion of small face connected domains is: the ratio of the number of invalid cells contained in the largest connected domain to the total number of cells in the small face of the end face of the ingot;

[0023] If the proportion of the small face connected domain is less than the small face connected domain proportion threshold, it is determined that the cutting is valid.

[0024] Optionally, determining the cutting effect further includes:

[0025] Calculate the connected domain cutting efficiency: identify the connected domain, which is the area of ​​connected invalid cells, and count the total number of invalid cells in all connected domains. The connected domain cutting efficiency is the ratio of the number of cells remaining after deducting the total number of invalid cells in all connected domains from the connected domain cutting efficiency to the total number of cells;

[0026] If the connected domain cutting efficiency is greater than the connected domain cutting efficiency threshold, it is determined that the cutting is effective.

[0027] Optionally, determining the cutting effect further includes:

[0028] Calculate the effective rate of cutting the large surface connected domain; count the total number of invalid cells of all connected domains in the large surface of the end face of the ingot, and the effective rate of cutting the large surface connected domain is the ratio of the total number of cells in the large surface minus the total number of invalid cells in all connected domains in the large surface to the total number of cells in the large surface;

[0029] If the large surface connected domain cutting efficiency is greater than the large surface connected domain cutting efficiency threshold, it is determined that the cutting is effective.

[0030] Optionally, determining the cutting effect further includes:

[0031] Calculate the effective rate of cutting the small face connected domain: count the total number of invalid cells in all connected domains in the small face of the end face of the ingot, and the effective rate of cutting the small face connected domain is the ratio of the total number of cells in the small face minus the total number of invalid cells in all connected domains in the small face to the total number of cells in the small face;

[0032] If the facet connected domain cutting efficiency is greater than the facet connected domain cutting efficiency threshold, it is determined that the cutting is effective.

[0033] Optionally, determining the cutting effect further includes:

[0034] Calculate the total cutting efficiency; the total cutting efficiency is the sum of the cell cutting efficiency weight and the connected domain cutting efficiency weight;

[0035] The cell cutting efficiency weight is the product of the cell cutting efficiency and the cell weight coefficient;

[0036] The connected domain cutting efficiency weight is the product of the connected domain cutting comprehensive efficiency and the connected domain weight coefficient;

[0037] The comprehensive efficiency of the connected domain cutting is the ratio of the difference between the total number of cells and the total number of invalid cells in all connected domains multiplied by the weighting coefficient to the total number of cells;

[0038] If the total cutting effectiveness is greater than the total effectiveness threshold, it is determined that the cutting effect is effective.

[0039] Optionally, determining the cutting effect further includes:

[0040] Calculate the total cutting efficiency of the large surface; the total cutting efficiency of the large surface is the sum of the cutting efficiency weight of the large surface unit cell in the large surface of the ingot end face and the cutting efficiency weight of the large surface connected domain;

[0041] The large surface unit cell cutting efficiency weight is the product of the large surface unit cell cutting efficiency and the large surface unit cell weight coefficient;

[0042] The large surface connected domain cutting efficiency weight is the product of the large surface connected domain cutting comprehensive efficiency and the large surface connected domain weight coefficient;

[0043] The comprehensive efficiency of the large-surface connected domain cutting is the ratio of the total number of cells in the large surface minus the total number of invalid cells in all connected domains in the large surface multiplied by the weighted coefficient of the large surface to the total number of cells in the large surface;

[0044] If the large-surface total cutting efficiency is greater than the large-surface total cutting efficiency threshold, it is determined that the cutting effect is effective.

[0045] Optionally, determining the cutting effect further includes:

[0046] Calculate the total facet cutting efficiency; the total facet cutting efficiency is the sum of the facet unit cell cutting efficiency weight and the facet connected domain cutting efficiency weight in the facet at the end face of the ingot;

[0047] The facet cell cutting efficiency weight is the product of the facet cell cutting efficiency and the facet cell weight coefficient;

[0048] The small face connected domain cutting efficiency weight is the product of the small face connected domain cutting comprehensive efficiency and the small face connected domain weight coefficient;

[0049] The comprehensive efficiency of the small face connected domain cutting is the ratio of the total number of cells in the small face minus the total number of invalid cells in all connected domains in the small face multiplied by the small face weighting coefficient to the total number of cells in the small face;

[0050] If the total facet cutting efficiency is greater than the total facet cutting efficiency threshold, it is determined that the cutting effect is effective.

[0051] In the above-mentioned ingot cutting detection method, the end face of the ingot is divided into multiple cells, and the thickness of the peeling surface is measured by using the spectrum analysis of optical reflection. The peeling thickness of each cell is judged to determine whether the peeling of each cell is effective, so as to determine whether the ingot meets the peeling requirements, and provide a standardized peeling judgment standard for ingot cutting, which helps to reduce the peeling breakage rate, reduce labor costs, and improve cutting efficiency. Moreover, the detection method can be applied to the cutting detection of ingots of different specifications, for example, it can be applied to the cutting detection of existing ingots with diameters of 6 inches, 8 inches, and 12 inches. The cutting detection method has a wide range of applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 A schematic diagram of ingot cutting according to an embodiment of the present invention;

[0053] Figure 2 FIG. 4 is a schematic diagram of the detection beam irradiation principle of an embodiment of the present invention.

[0054] Among them, in the attached drawings:

[0055] 10- ingot;

[0056] 20- wafer;

[0057] d1-end face;

[0058] d2-cutting surface;

[0059] L-detection beam;

[0060] L1-first reflected beam;

[0061] L2 - second reflected beam;

[0062] T-cut thickness. DETAILED DESCRIPTION

[0063] The ingot cutting detection method proposed in the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. According to the following description, the advantages and features of the present invention will become clearer. It should be noted that the accompanying drawings are all in a very simplified form and are not in precise proportions, which are only used to conveniently and clearly assist in explaining the purpose of the embodiments of the present invention.

[0064] As used in the present invention, the singular forms "one", "an" and "the" include plural objects, the term "or" is usually used to include the meaning of "and / or", the term "several" is usually used to include the meaning of "at least one", and the terms "at least two" or "multiple" are usually used to include the meaning of "two or more". In addition, the terms "first", "second" and "third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first", "second" and "third" can explicitly or implicitly include one or at least two of the features. In addition, as used in the present invention, "installed", "connected", "connected", and one element "set" on another element should be understood in a broad sense, usually only indicating that there is a connection, coupling, matching or transmission relationship between the two elements, and the connection, coupling, matching or transmission between the two elements can be direct or indirect through an intermediate element, and cannot be understood as indicating or implying the spatial position relationship between the two elements, that is, one element can be in any orientation such as inside, outside, above, below or on one side of another element, unless the content clearly indicates otherwise. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. In addition, directional terms such as above, below, up, down, upward, downward, left, right, etc. are used with respect to the exemplary embodiments as they are shown in the figures, with the upward or upper direction toward the top of the corresponding figure, and the downward or lower direction toward the bottom of the corresponding figure.

[0065] The present invention provides a crystal ingot cutting detection method, which involves a variety of cutting data. In order to clearly explain the technical solution of the present invention, various cutting data are uniformly summarized here.

[0066] C1-cell cutting efficiency;

[0067] C11- large surface unit cell cutting efficiency;

[0068] C12-facet cell cutting efficiency;

[0069] C1w-cell cutting efficiency weight;

[0070] C11w-large surface unit cell cutting efficiency weight;

[0071] C12w-facet cell cutting efficiency weight;

[0072] γ1-cell weight coefficient;

[0073] γ11-large surface unit cell weight coefficient;

[0074] γ12-facet cell weight coefficient;

[0075] C2-connected domain cutting efficiency;

[0076] C2w-connected domain cutting efficiency weight;

[0077] C21w- large surface connected domain cutting efficiency weight;

[0078] C22w- small face connected domain cutting efficiency weight;

[0079] γ2-connected domain weight coefficient;

[0080] γ21-weight coefficient of large surface connected domain;

[0081] γ22- weight coefficient of small surface connected domain;

[0082] C21- large surface connected domain cutting efficiency;

[0083] C22-Efficiency of cutting small face connected domains;

[0084] C3-total cleavage efficiency;

[0085] C31-large surface total cutting efficiency;

[0086] C32- total cutting efficiency of small face;

[0087] Cz-comprehensive efficiency of connected domain cutting;

[0088] Cz1- comprehensive efficiency of cutting large connected domains;

[0089] Cz2- comprehensive efficiency of cutting small face connected domain;

[0090] α-fixed weight coefficient;

[0091] α1-large surface fixed weight coefficient;

[0092] α2-facet fixed weight coefficient;

[0093] β-dynamic weighting coefficient;

[0094] β1-large surface dynamic weighting coefficient;

[0095] β2-facet dynamic weighting coefficient;

[0096] V-connected domain proportion;

[0097] V1-large surface connected domain;

[0098] V2- small surface connected domain;

[0099] N - total number of cells;

[0100] N1-total number of cells in the large surface;

[0101] N2-total number of cells in the facet;

[0102] Nx - the number of valid cells;

[0103] Nx1 - the number of valid cells in the large face;

[0104] Nx2 - the number of valid cells in the facet;

[0105] Ny - the number of invalid cells contained in the largest connected domain in the end face;

[0106] Ny1-the number of invalid cells contained in the largest connected domain in the large surface;

[0107] Ny2-the number of invalid cells contained in the largest connected domain in the facet;

[0108] Nz - total number of invalid cells in all connected domains;

[0109] Nz1 - the total number of invalid cells in all connected domains in the large surface;

[0110] Nz2 - Total number of invalid cells in all connected domains in the facet.

[0111] The present invention mainly involves four cutting effectiveness judgment data, namely, cell cutting efficiency C1, connected domain cutting efficiency C2, total cutting efficiency C3 and connected domain proportion V.

[0112] The present invention discusses the ingot cutting detection method through the following embodiments 1, 2 and 3.

[0113] Example 1 uses the cell cutting efficiency C1 and the connected domain ratio V as the judgment conditions;

[0114] In the second embodiment, the cell cutting efficiency C1 and the connected domain cutting efficiency C2 are used as the judgment conditions;

[0115] In Example 3, the cell cutting efficiency C1 and the total cutting efficiency C3 are used as the judgment conditions.

[0116] The above four determination data can also be combined in other ways, for example, using the cell cutting efficiency C1, the connected domain cutting efficiency C2 and the total cutting efficiency C3 as determination conditions; or using the cell cutting efficiency C1, the connected domain cutting efficiency C2 and the connected domain proportion V as determination conditions; or using the cell cutting efficiency C1, the total cutting efficiency C3 and the connected domain proportion V as determination conditions; or using the cell cutting efficiency C1, the connected domain cutting efficiency C2, the total cutting efficiency C3 and the connected domain proportion V as determination conditions; or using only the cell cutting efficiency C1 as the determination condition. The above four cutting determination data can be flexibly combined based on actual usage requirements.

[0117] In addition, in the following embodiments, in order to clearly illustrate the detection method, each step in the method is numbered, and the numbering of the step does not limit the execution order of the steps. During the specific execution process, the execution order of the steps can be adjusted based on the actual detection conditions.

[0118] Embodiment 1:

[0119] The ingot cutting detection method comprises the following steps:

[0120] S1: Divide the cells; divide the end face of the ingot into a number of cells; in this embodiment, the cells are preferably square, and the cell pitch is 40um, that is, the size of the cells is 40um×40um, and the total number of cells N can reach tens of millions or hundreds of millions. The size and shape of the cells can be set based on actual usage requirements, for example, the size of the cells can also be set to 20um×20um.

[0121] S2: irradiate a detection beam; the detection beam can be visible light or broadband infrared; irradiate the detection beam L onto the end face d1 of the ingot (the end face here refers to the end face of the cut end of the ingot), and the irradiation direction of the detection beam L forms an angle with the normal direction of the end face d1. The angle is limited to ensure that the detection beam L is irradiated obliquely onto the end face d1, rather than vertically onto the end face d1. The specific size of the angle can be set based on actual needs;

[0122] Combination Figure 1 As shown, the upper end of the ingot 10 is cut, and when the cutting is effective, the wafer 20 can be separated from the ingot.

[0123] During the laser cutting process, the laser is irradiated on the end face d1 of the ingot 10, and the laser is focused through the end face d1 of the ingot 10 to below the end face d1 of the wafer. The laser scans to generate a single or multiple modified layers, and then forms a cutting surface d2. The distance between the end face d1 and the cutting surface d2 is the cutting thickness T.

[0124] S3: Calculate the cell cutting thickness; receive the reflected light after the detection light beam is irradiated on the end face of the ingot, and perform spectrum analysis on the reflected light to obtain the cutting thickness of each cell.

[0125] Please refer to Figure 1 and Figure 2 As shown, when the detection light beam is irradiated on the end surface d1, different reflected lights will be generated depending on whether the cutting is effective.

[0126] Combination Figure 2 As shown, taking one of the cells as an example, when the bottom of the cell is cut by laser to form a cutting surface d2, after the detection beam L is irradiated on the cell, since the detection beam L and the normal direction of the end surface d1 form a certain angle, the detection beam L will irradiate the end surface d1 to form a first reflected beam L1, and the detection beam L will pass through the end surface d1 to irradiate the cutting surface d2 and form a second reflected beam L2. By analyzing the first reflected beam L1 and the second reflected beam L2, the cutting thickness T can be calculated. If the bottom of the cell is not cut by laser, there is no cutting surface d2 under the cell, and the cutting thickness T can also be calculated based on the analysis of the reflected light, and the cutting thickness T = 0.

[0127] By detecting the beam scanning and receiving the reflected light through the optical probe for frequency analysis, the cutting thickness corresponding to each cell can be obtained through spectral analysis.

[0128] S4: Determine a valid cell; if the cutting thickness of the cell is within the thickness threshold range, the cell is determined to be a valid cell; if the cutting thickness of the cell is outside the thickness threshold range, the cell is determined to be an invalid cell;

[0129] The threshold range is artificially set based on the target thickness of the cutting. For example, taking the target thickness of 90um as an example, the thickness threshold range is set to be greater than 90um. When the cell cutting thickness is greater than 90um, it is determined to be a valid cell, and when the cell cutting thickness is less than 90um, it is determined to be an invalid cell. Alternatively, the thickness threshold range can be set between two thickness values. When the cell cutting thickness is between the two thickness values, it is determined to be a valid cell, otherwise it is determined to be an invalid cell.

[0130] S5: Determine the cutting effect;

[0131] Calculate the cell cutting efficiency C1, which is the ratio of the number of effective cells Nx to the total number of cells N; the number of effective cells Nx and the total number of cells N here are counted for the entire end face;

[0132] That is, C1 = Nx / N;

[0133] If the unit cell cutting efficiency C1 is higher than the efficiency threshold, the cutting is determined to be effective. Effective cutting means that the peeling requirements of the wafer are met. At this time, the wafer can be separated from the ingot by a small external force.

[0134] The efficiency threshold is determined based on factors such as the material and size of the ingot. For example, for an ingot with a diameter of 6 inches, the efficiency threshold can be 80%. When the cell cutting efficiency C1 is greater than 80%, it means that 80% of the cells are cut and the entire wafer can be peeled off with a small external force, so the cutting is considered effective. For another example, for an ingot with a diameter of 8 inches, due to the increase in the area of ​​the end face d1, the efficiency threshold can be appropriately increased to 85%.

[0135] If the cell cutting efficiency C1 is not higher than the efficiency threshold, it is determined that the cutting is invalid. At this time, the laser needs to scan and cut again to improve the cell cutting efficiency C1.

[0136] In the above-mentioned ingot cutting detection method, the end face of the ingot is divided into multiple cells, and the thickness of the peeling surface is measured by using the spectrum analysis of optical reflection. The peeling thickness of each cell is judged to determine whether the peeling of each cell is effective, so as to determine whether the ingot meets the peeling requirements, and provide a standardized peeling judgment standard for ingot cutting, which helps to reduce the peeling breakage rate, reduce labor costs, and improve cutting efficiency. Moreover, the detection method can be applied to the cutting detection of ingots of different specifications, for example, it can be applied to the cutting detection of existing ingots with diameters of 6 inches, 8 inches, and 12 inches. The cutting detection method has a wide range of applications.

[0137] Considering that there may be a large local adhesion area during the ingot cutting process, even if the unit cell cutting efficiency C1 is greater than the efficiency threshold, there may be a local adhesion area that causes a large local adhesion stress on the cutting surface. At this time, if the wafer is forcibly peeled off by external force, it is very likely that the wafer will have defects or even break at the adhesion point. Therefore, in this embodiment, the connected domain factor is further introduced to judge whether the cutting is effective.

[0138] Furthermore, the determining of the cutting effect in step S5 further includes:

[0139] Calculate the proportion of connected domains V; identify the connected domain, which is the area of ​​connected invalid cells. When multiple invalid cells are connected, a large local adhesion area will be formed at the cutting surface. Calculate the number of invalid cells Ny contained in the largest connected domain, that is, identify the largest adhesion area. The proportion of connected domains V is the ratio of the number of invalid cells Ny contained in the largest connected domain to the total number of cells N. Here, the number of invalid cells Ny contained in the largest connected domain, the total number of cells N, and the proportion of connected domains V are statistics for the end face;

[0140] That is, V = Ny / N;

[0141] If the connected domain proportion V is less than the invalid connected domain proportion threshold, it is determined that the cutting is valid.

[0142] By calculating the proportion of connected domains, the percentage of the largest adhesion area in the end face to the entire end face cutting area is identified. By limiting the maximum value of this percentage to improve the peeling defects caused by excessively large connected domains, it helps to improve the yield and ensure better cutting effects.

[0143] In addition, the end faces of the ingot can be divided into small faces and large faces. Taking the silicon carbide ingot as an example, the small face of the silicon carbide ingot refers to the specific growth characteristic surface formed during the growth of the silicon carbide crystal, which is commonly known as the "small face" or "growth small face" in the field of crystal growth, and the rest of the area is commonly known as the "large face".

[0144] When silicon carbide crystals are prepared by physical vapor transport (PVT), this growth feature surface appears due to the inherent properties of crystal growth. During the crystal growth process, the growth rates of different crystal faces are different, resulting in some crystal faces growing faster and others growing relatively slowly, thus forming a small face structure. The presence of silicon carbide small faces may affect the quality of the crystal, such as causing uneven stress distribution inside the crystal, which in turn affects the mechanical and electrical properties of the crystal. The defect density at the small face may be different from that in other areas, which will also affect the overall quality of the crystal.

[0145] The color depth of the large surface and the small surface is different. The large surface area and the small surface area in the end surface can be identified through optical illumination. The identification process belongs to the existing technology and will not be repeated here.

[0146] Therefore, the mechanical properties of the small surface area and the large surface area in the end surface of the ingot are different, which leads to the further introduction of the proportion of the large surface connected domain and the proportion of the small surface connected domain in this embodiment.

[0147] Furthermore, the determination of the cutting effect in step S5 also includes the calculation of the proportion V1 of the large surface connected domain;

[0148] The large surface connected domain ratio V1 is: in the large surface of the end surface of the ingot, the ratio of the number of invalid cells Ny1 contained in the largest connected domain to the total number of cells N in the end surface;

[0149] That is, V1 = Ny1 / N;

[0150] If the large surface connected domain proportion V1 is less than the large surface connected domain proportion threshold, it is determined that the cutting is effective.

[0151] In addition, the determination of the cutting effect in step S5 also includes the calculation of the proportion V2 of the small face connected domain;

[0152] The small face connected domain ratio V2 is: in the small face of the end face of the ingot, the ratio of the number of invalid cells Ny2 contained in the largest connected domain to the total number of cells N in the end face;

[0153] That is, V2=Ny2 / N;

[0154] If the ratio of the small face connected domain to V2 is less than the small face connected domain ratio threshold, the cutting is determined to be effective.

[0155] Due to the different material mechanical properties of the large surface area and the small surface area, the critical stress for safe peeling in the large surface area and the small surface area is different. Therefore, the threshold value of the large surface connected domain ratio is usually different from the threshold value of the small surface connected domain ratio. The specific threshold value is based on the adaptability of the mechanical properties of the large surface area and the small surface area.

[0156] By calculating the proportion of large surface connected domains V1 and small surface connected domains V2 as mentioned above to determine whether the cutting standard is met, the interference of the difference in mechanical properties caused by large surface areas and small surface areas on the effective judgment of cutting can be eliminated, thereby further improving the accuracy of cutting detection.

[0157] Embodiment 2

[0158] In the first embodiment, the cell cutting efficiency C1 and the connected domain proportion V (as well as the large surface connected domain proportion V1 and the small surface connected domain proportion V2) are used as judgment conditions, and the cutting is deemed to be effective when both conditions are met.

[0159] In this embodiment, the cell cutting efficiency C1 is retained, and the connected domain cutting efficiency C2 replaces the connected domain proportion V.

[0160] In this embodiment, the determining of the cutting effect in step S5 further includes:

[0161] Calculate the connected domain cutting efficiency C2; identify the connected domain, the connected domain is the area of ​​connected invalid cells, count the total number of invalid cells Nz of all connected domains, the connected domain cutting efficiency C2 is the ratio of the number of remaining cells from the total number of cells N minus the total number of invalid cells Nz of all connected domains to the total number of cells N, and the connected domain cutting efficiency C2 is counted for the end face;

[0162] That is, C2 = (N-Nz) / N;

[0163] If the connected domain cutting efficiency C2 is greater than the connected domain cutting efficiency threshold, it is determined that the cutting is effective.

[0164] The above-mentioned connected domain cutting efficiency C2 takes into account the influence of all connected domains on the entire end face cutting effectiveness.

[0165] Furthermore, in this embodiment, the difference in mechanical properties between the large surface area and the small surface area is taken into consideration, and the large surface connected domain cutting efficiency C21 and the small surface connected domain cutting efficiency C22 are derived.

[0166] The large surface connected domain cutting efficiency C21 is: in the large surface of the end surface of the ingot, the total number of invalid cells Nz1 of all connected domains is counted, and the ratio of the number of remaining cells from the total number of cells N1 in the large surface minus the total number of invalid cells Nz1 of all connected domains in the large surface to the total number of cells N1 in the large surface; the large surface connected domain cutting efficiency C21 is counted for the large surface in the end surface;

[0167] That is, C21 = (N1 - Nz1) / N1;

[0168] If the large surface connected domain cutting efficiency C2 is greater than the large surface connected domain cutting efficiency threshold, it is determined that the cutting is effective.

[0169] The small face connected domain cutting efficiency C22 is: in the small face of the end face of the ingot, the total number of invalid cells Nz2 of all connected domains is counted, and the ratio of the number of cells remaining from the total number of cells in the small face N2 minus the total number of invalid cells Nz2 of all connected domains in the small face to the total number of cells in the small face N2; the small face connected domain cutting efficiency C22 is counted for the small face in the end face;

[0170] That is, C22 = (N2 - Nz2) / N2;

[0171] If the facet connected domain cutting efficiency C22 is greater than the facet connected domain cutting efficiency threshold, it is determined that the cutting is effective.

[0172] Due to the different mechanical properties of the materials in the large and small surface areas, the critical stresses for safe peeling in the large and small surface areas are different. Therefore, the effective threshold for cutting the large connected domain is usually different from the effective threshold for cutting the small connected domain, and the specific threshold is set based on the adaptability of the mechanical properties of the large and small surface areas.

[0173] By calculating the above-mentioned large surface connected domain cutting efficiency C21 and small surface connected domain cutting efficiency C22 to determine whether the cutting standard is met, the interference of the difference in mechanical properties caused by the large surface area and the small surface area on the effective cutting judgment can be eliminated, and the accuracy of cutting detection is further improved.

[0174] Embodiment 3

[0175] In the first embodiment, the cell cutting efficiency C1 and the connected domain proportion V (including the large surface connected domain proportion V1 and the small surface connected domain proportion V2) are used as the judgment conditions, and the cutting is deemed to be effective when both conditions are met.

[0176] In this embodiment, the cell cutting efficiency C1 is retained, and the total cutting efficiency C3 replaces the connected domain proportion V.

[0177] In this embodiment, the determining of the cutting effect in step S5 further includes:

[0178] Calculate the total cutting efficiency C3; the total cutting efficiency C3 is the sum of the cell cutting efficiency weight C1w and the connected domain cutting efficiency weight C2w;

[0179] That is, C3=C1w+C2w;

[0180] The cell cutting efficiency weight C1w is the product of the cell cutting efficiency C1 and the cell weight coefficient γ1;

[0181] That is, C1w=C1·γ1;

[0182] The connected domain cutting efficiency weight C2w is the product of the connected domain cutting comprehensive efficiency Cz and the connected domain weight coefficient γ2;

[0183] That is, C2w = Cz·γ2;

[0184] The comprehensive efficiency Cz of the connected domain cutting is the ratio of the difference between the total number of cells N minus the total number of invalid cells Nz of all connected domains multiplied by the weighting coefficient to the total number of cells N;

[0185] That is, Cz = (N-Nz·α·β) / N;

[0186] The weighting coefficients include a fixed weight coefficient α and a dynamic weight coefficient β;

[0187] The fixed weight coefficient α is a fixed coefficient; the dynamic weight coefficient β is a dynamically adjusted coefficient, which is dynamically adjusted in combination with specific working conditions.

[0188] In summary, C3=C1·γ1+(N-Nz·α·β) / N·γ2;

[0189] The above total cutting efficiency C3 is calculated based on the end face.

[0190] If the total cutting effectiveness C3 is greater than the total effectiveness threshold, it is determined that the cutting effect is effective.

[0191] The above-mentioned total cutting efficiency C3 comprehensively considers the influence of the effective number of cells and the connected domain on the cutting effectiveness, which helps to improve the detection accuracy.

[0192] Furthermore, in this embodiment, the difference in mechanical properties between the large surface area and the small surface area is taken into consideration, and the total cutting efficiency C31 of the large surface and the total cutting efficiency C32 of the small surface are further derived.

[0193] The total cutting efficiency of the large surface C31 is the sum of the cutting efficiency weight C11w of the large surface unit cell in the large surface of the ingot end face and the cutting efficiency weight C21w of the large surface connected domain;

[0194] That is, C31 = C11w + C21w;

[0195] The large surface cell cutting efficiency weight C11w is the product of the large surface cell cutting efficiency C11 and the large surface cell weight coefficient γ11; wherein the large surface cell cutting efficiency C11 is the ratio of the number of effective cells Nx1 in the large surface to the total number of cells N1 in the large surface, C11 = Nx1 / N1;

[0196] That is, C11w=C11·γ11;

[0197] The large surface connected domain cutting efficiency weight C21w is the product of the large surface connected domain cutting comprehensive efficiency Cz1 and the large surface connected domain weight coefficient γ21;

[0198] That is, C21w=Cz1·γ21;

[0199] The comprehensive efficiency Cz1 of the large surface connected domain cutting is the ratio of the total number of cells N1 in the large surface minus the total number of invalid cells Nz1 in all connected domains in the large surface multiplied by the weighted coefficient of the large surface to the total number of cells N1 in the large surface;

[0200] That is, Cz1 ​​= (N1-Nz1·α1·β1) / N1;

[0201] The large surface weighting coefficient includes the large surface fixed weighting coefficient α1 and the large surface dynamic weighting coefficient β1;

[0202] The large surface fixed weight coefficient α1 is a fixed coefficient; the large surface dynamic weight coefficient β1 is a dynamically adjusted coefficient, which is dynamically adjusted in combination with specific working conditions.

[0203] In summary, C31 = C11·γ11 + (N1-Nz1·α1·β1) / N1·γ21;

[0204] The total large surface cutting efficiency C31 is calculated based on the statistics of the large surface of the end face.

[0205] If the large-surface total cutting efficiency C31 is greater than the large-surface total cutting efficiency threshold, it is determined that the cutting effect is effective.

[0206] The total facet cutting efficiency C32 is the sum of the facet unit cell cutting efficiency weight C12w and the facet connected domain cutting efficiency weight C22w in the facet at the end face of the ingot;

[0207] That is, C32 = C12w + C22w;

[0208] The facet cell cutting efficiency weight C12w is the product of the facet cell cutting efficiency C12 and the facet cell weight coefficient γ12; the facet cell cutting efficiency C12 is the ratio of the number of effective cells Nx2 in the facet to the total number of cells N2 in the facet, C12=Nx2 / N2;

[0209] That is, C12w = C12·γ12;

[0210] The small face connected domain cutting efficiency weight C22w is the product of the small face connected domain cutting comprehensive efficiency Cz2 and the small face connected domain weight coefficient γ21;

[0211] That is, weight C22w = Cz2·γ22;

[0212] The comprehensive efficiency Cz2 of the small face connected domain cutting is the ratio of the total number of cells N2 in the small face minus the total number of invalid cells Nz2 in all connected domains in the small face multiplied by the small face weighting coefficient to the total number of cells N2 in the small face;

[0213] That is, Cz2 = (N2-Nz2·α2·β2) / N2;

[0214] The small face weighting coefficient includes the large face fixed weighting coefficient α2 and the small face dynamic weighting coefficient β2;

[0215] The small surface fixed weight coefficient α2 is a fixed coefficient; the small surface dynamic weight coefficient β2 is a dynamically adjusted coefficient, which is dynamically adjusted in combination with specific working conditions.

[0216] In summary, C32 = C12·γ12 + (N2-Nz2·α2·β2) / N2·γ22;

[0217] The above-mentioned total small face cutting efficiency C32 is calculated for small faces.

[0218] If the small face total cutting efficiency C32 is greater than the small face total cutting efficiency threshold, it is determined that the cutting effect is effective.

[0219] The material mechanical properties of the large surface area and the small surface area are different, which leads to different critical stresses for safe peeling in the large surface area and the small surface area. Therefore, the total effective rate threshold of the large surface is usually different from the total effective rate threshold of the small surface, and the specific threshold value is set based on the adaptability of the mechanical properties of the large surface area and the small surface area.

[0220] By calculating the above-mentioned total cutting efficiency C31 of the large surface and the total cutting efficiency C32 of the small surface to determine whether the cutting standard is met, the interference of the difference in mechanical properties caused by the large surface area and the small surface area on the effective cutting judgment can be eliminated, and the accuracy of cutting detection is further improved.

[0221] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0222] The above description is only a description of the preferred embodiments of the present invention, and is not intended to limit the scope of the present invention. Any changes or modifications made by a person skilled in the art in the field of the present invention based on the above disclosure shall fall within the scope of protection of the claims.

Claims

1. A method for detecting ingot cutting, characterized in that: The following steps are involved: Divide cells: divide the end surface of the ingot into several cells; Irradiating the detection beam: irradiating the detection beam onto the end face of the ingot, with the irradiation direction of the detection beam forming an angle with the normal direction of the end face; Calculate the cell cutting thickness: receive the reflected light after the detection beam is irradiated on the end face of the ingot, and perform spectrum analysis on the reflected light to obtain the cutting thickness of each cell; Determine valid cells: if the cutting thickness of a cell is within the thickness threshold range, the cell is determined to be a valid cell; if the cutting thickness of a cell is outside the thickness threshold range, the cell is determined to be an invalid cell; Determine the cutting effect: calculate the cell cutting efficiency, which is the ratio of the number of effective cells to the total number of cells; If the cell cutting efficiency is higher than the efficiency threshold, it is determined that the cutting is effective.

2. The ingot cutting detection method according to claim 1, characterized in that: The determining of the cutting effect further comprises: Calculate the proportion of connected domains: Identify the connected domains, which are the areas of connected invalid cells, and calculate the number of invalid cells contained in the largest connected domain. The proportion of connected domains is the ratio of the number of invalid cells contained in the largest connected domain to the total number of cells. If the proportion of connected domains is less than the threshold of the proportion of invalid connected domains, the cutting is determined to be effective.

3. The ingot cutting detection method according to claim 2, characterized in that: The determining of the cutting effect further comprises: Calculating the proportion of large-surface connected domains, where the proportion of large-surface connected domains is: the ratio of the number of invalid cells contained in the largest connected domain to the total number of cells in the large surface of the end surface of the ingot; If the proportion of the large surface connected domain is less than the large surface connected domain proportion threshold, it is determined that the cutting is valid.

4. The ingot cutting detection method according to claim 2 or 3, characterized in that: The determining of the cutting effect further comprises: Calculating the proportion of small face connected domains, where the proportion of small face connected domains is: the ratio of the number of invalid cells contained in the largest connected domain to the total number of cells in the small face of the end face of the ingot; If the proportion of the small face connected domain is less than the small face connected domain proportion threshold, it is determined that the cutting is valid.

5. The ingot cutting detection method according to claim 1, characterized in that: The determining of the cutting effect further comprises: Calculate the connected domain cutting efficiency: identify the connected domain, which is the area of ​​connected invalid cells, and count the total number of invalid cells in all connected domains. The connected domain cutting efficiency is the ratio of the number of cells remaining after deducting the total number of invalid cells in all connected domains from the connected domain cutting efficiency to the total number of cells; If the connected domain cutting efficiency is greater than the connected domain cutting efficiency threshold, it is determined that the cutting is effective.

6. The ingot cutting detection method according to claim 5, characterized in that: The determining of the cutting effect further comprises: Calculate the effective rate of cutting the large surface connected domain; count the total number of invalid cells of all connected domains in the large surface of the end face of the ingot, and the effective rate of cutting the large surface connected domain is the ratio of the total number of cells in the large surface minus the total number of invalid cells in all connected domains in the large surface to the total number of cells in the large surface; If the large surface connected domain cutting efficiency is greater than the large surface connected domain cutting efficiency threshold, it is determined that the cutting is effective.

7. The ingot cutting detection method according to claim 5 or 6, characterized in that: The determining of the cutting effect further comprises: Calculate the effective rate of cutting the small face connected domain: count the total number of invalid cells in all connected domains in the small face of the end face of the ingot, and the effective rate of cutting the small face connected domain is the ratio of the total number of cells in the small face minus the total number of invalid cells in all connected domains in the small face to the total number of cells in the small face; If the facet connected domain cutting efficiency is greater than the facet connected domain cutting efficiency threshold, it is determined that the cutting is effective.

8. The ingot cutting detection method according to claim 1, characterized in that: The determining of the cutting effect further comprises: Calculate the total cutting efficiency; the total cutting efficiency is the sum of the cell cutting efficiency weight and the connected domain cutting efficiency weight; The cell cutting efficiency weight is the product of the cell cutting efficiency and the cell weight coefficient; The connected domain cutting efficiency weight is the product of the connected domain cutting comprehensive efficiency and the connected domain weight coefficient; The comprehensive efficiency of the connected domain cutting is the ratio of the difference between the total number of cells and the total number of invalid cells in all connected domains multiplied by the weighting coefficient to the total number of cells; If the total cutting effectiveness is greater than the total effectiveness threshold, it is determined that the cutting effect is effective.

9. The ingot cutting detection method according to claim 8, characterized in that: The determining of the cutting effect further comprises: Calculate the total cutting efficiency of the large surface; the total cutting efficiency of the large surface is the sum of the cutting efficiency weight of the large surface unit cell in the large surface of the ingot end face and the cutting efficiency weight of the large surface connected domain; The large surface unit cell cutting efficiency weight is the product of the large surface unit cell cutting efficiency and the large surface unit cell weight coefficient; The large surface connected domain cutting efficiency weight is the product of the large surface connected domain cutting comprehensive efficiency and the large surface connected domain weight coefficient; The comprehensive efficiency of the large-surface connected domain cutting is the ratio of the total number of cells in the large surface minus the total number of invalid cells in all connected domains in the large surface multiplied by the weighted coefficient of the large surface to the total number of cells in the large surface; If the large-surface total cutting efficiency is greater than the large-surface total cutting efficiency threshold, it is determined that the cutting effect is effective.

10. The ingot cutting detection method according to claim 8 or 9, characterized in that: The determining of the cutting effect further comprises: Calculate the total facet cutting efficiency; the total facet cutting efficiency is the sum of the facet unit cell cutting efficiency weight and the facet connected domain cutting efficiency weight in the facet at the end face of the ingot; The facet cell cutting efficiency weight is the product of the facet cell cutting efficiency and the facet cell weight coefficient; The small face connected domain cutting efficiency weight is the product of the small face connected domain cutting comprehensive efficiency and the small face connected domain weight coefficient; The comprehensive efficiency of the small face connected domain cutting is the ratio of the total number of cells in the small face minus the total number of invalid cells in all connected domains in the small face multiplied by the small face weighting coefficient to the total number of cells in the small face; If the total facet cutting efficiency is greater than the total facet cutting efficiency threshold, it is determined that the cutting effect is effective.

Images