A method for cutting semiconductor wafer

Abstract

The present invention relates to the technical field of semiconductor wafer cutting, and discloses a semiconductor wafer cutting method. The method helps to judge the surface flatness and defects of a wafer to be tested by setting a wafer that meets the standards as a reference. After the laser is collimated, a parallel light beam is formed, which effectively reduces the scattering and divergence of light during the propagation process, and ensures the accuracy of interference fringes generated. The parallel light beam is divided into two beams by a beam splitter, which are respectively irradiated on the surface of a standard wafer and a wafer to be tested, and interference fringes are formed by reflection. The angle adjustment of the reflector ensures that the light beam is accurately irradiated and returns to the beam splitter, generating a clear interference fringe image. Subsequently, by analyzing the interference fringes, the microscopic differences on the wafer surface can be intuitively presented, the uneven areas can be identified and preliminary adjustments can be made. This process provides an accurate data basis for the further processing of subsequent wafers, ensuring that the surface flatness of the wafer meets the subsequent process requirements.

Description

Technical Field

[0001] The invention relates to the technical field of semiconductor wafer cutting, and in particular to a semiconductor wafer cutting method. Background Art

[0002] In the semiconductor manufacturing process, wafer cutting is one of the key links, which directly affects the final quality of the chip. Wafers are prepared through a series of complex processes, including epitaxial growth, photolithography, etching, ion implantation, etc., and finally the complete wafer needs to be cut into independent chips. In this process, how to ensure the cutting accuracy and the surface quality of the wafer after cutting is an important technical challenge.

[0003] Wafer cutting usually uses methods such as laser cutting, mechanical cutting or plasma cutting. Different cutting methods will affect the quality of the wafer edge and may cause problems such as edge collapse and cracks. Especially in scenarios with high precision requirements, the surface quality of the wafer is crucial to the positioning of the cutting line. If there are slight unevenness on the surface of the wafer, it will affect the subsequent scribing accuracy and cause cutting deviation. Therefore, how to accurately detect and adjust the wafer surface before cutting becomes a crucial step. To address this problem, this paper proposes a semiconductor wafer cutting method, which includes wafer surface unevenness detection and wafer division surface adjustment strategy. Summary of the invention

[0004] The present invention provides a semiconductor wafer cutting method, which helps solve the problems mentioned in the above background technology.

[0005] The present invention provides the following technical solution: a method for cutting a semiconductor wafer, comprising:

[0006] Obtain a wafer whose surface meets the standard and record it as a standard wafer;

[0007] Obtain a wafer whose surface is to be inspected, recorded as the wafer to be inspected;

[0008] Instrument setup and adjustment:

[0009] The laser is placed on the optical axis of the interferometer, and the emitted light beam is collimated to form a parallel light beam;

[0010] Direct the parallel beam of light vertically onto the beam splitter;

[0011] The beam splitter is used to split the light beam into two beams, one beam of light is vertically irradiated to the surface of the standard wafer, and the other beam of light is vertically irradiated to the surface of the wafer to be tested;

[0012] Place the standard wafer and the wafer to be tested in parallel on the testing table of the interferometer;

[0013] The standard wafer is placed on one of the reflection paths of the beam and serves as a reflection plane for the reference beam;

[0014] The wafer to be inspected is placed on another reflection path, serving as a reflection plane for the measurement beam;

[0015] Adjust the angle of the reflector so that the light beam separated from the beam splitter can be irradiated onto the standard wafer and the wafer to be inspected respectively, and reflected back to the beam splitter;

[0016] Wafer surface measurement:

[0017] Start the laser light source, and after the light beam is collimated, it is split into two beams by a beam splitter, which are irradiated onto the surface of the standard wafer and the wafer to be tested respectively;

[0018] The two beams of light are reflected on the surfaces of the standard wafer and the wafer to be tested respectively and then merged into interference fringes;

[0019] Adjust the angle of the reflector so that the two beams overlap at the beam splitter;

[0020] Acquire interference fringe images formed by reflected light;

[0021] The interference fringe image is analyzed, the concave and convex positions on the surface of the wafer to be inspected are marked, and preliminary adjustments are made to the wafer to be inspected.

[0022] Optionally, the preliminary adjustment of the wafer to be inspected specifically includes:

[0023] Set the highest convexity threshold;

[0024] Acquire a raised area on a wafer to be inspected;

[0025] Get the highest point of the raised area;

[0026] Obtain the distance from the highest point to the wafer surface, which is recorded as the highest value of the protrusion;

[0027] If the highest bulge value is less than or equal to the highest bulge value threshold, no adjustment is made;

[0028] If the highest value of the bulge is greater than the highest value threshold of the bulge, the bulge area is polished until the highest value of the bulge area is less than or equal to the highest value threshold of the bulge;

[0029] Perform initial division on the wafers to be inspected after preliminary adjustment.

[0030] Optionally, the initial division of the wafers to be inspected after the preliminary adjustment specifically includes:

[0031] Get a sample chip;

[0032] The circular surface formed by the wafer to be inspected is recorded as the divided circular surface;

[0033] Obtain any diameter of the divided circular surface and record it as the first divided diameter;

[0034] Obtain a diameter perpendicular to the first dividing diameter and record it as the second dividing diameter;

[0035] Obtain the lengths of two intersecting sides of the rectangle where the sample chip is located and record them as the first side length and the second side length respectively;

[0036] Divide the first dividing diameter from the center of the dividing circle into N parts on both sides of the center according to the first side length, and mark the dividing points;

[0037] The second dividing diameter starts from the center of the dividing circle, and is divided into M parts according to the second side length on both sides of the center, and the dividing points are marked;

[0038] Line segments parallel to the second dividing diameter are drawn sequentially through the dividing points on the first dividing diameter, and the endpoints of the line segments intersect on the circumference of the dividing circular surface;

[0039] Line segments parallel to the first dividing diameter are drawn successively through the dividing points on the second dividing diameter, and the endpoints of the line segments intersect on the circumference of the dividing circular surface;

[0040] The partitioning circle is divided into a number of partitioning surfaces by interlaced line segments;

[0041] Obtaining the areas of all segmentation surfaces within the segmentation circle, and calculating the ratio of the area of ​​each segmentation surface to the area of ​​the sample chip, which is recorded as the segmentation ratio;

[0042] The segmentation faces with segmentation ratio less than 1 are recorded as corner faces;

[0043] The split surface with a split ratio equal to 1 is recorded as the standard surface;

[0044] The divided circular surface is recorded as the initial divided circular surface.

[0045] Optionally, the initial division of the wafers to be inspected after the preliminary adjustment specifically includes:

[0046] Obtain two endpoints of the first divided diameter, which are respectively recorded as the first endpoint and the second endpoint;

[0047] The line segments passing through the first dividing diameter are recorded in order from the nearest to the first endpoint as the first horizontal dividing line, the second horizontal dividing line, ... the Nth horizontal dividing line;

[0048] Obtain two endpoints of the second divided diameter, which are respectively recorded as the third endpoint and the fourth endpoint;

[0049] The line segments passing through the first dividing diameter are recorded in order from the nearest to the third endpoint as the first vertical dividing line, the second vertical dividing line, ... the Mth vertical dividing line;

[0050] The position of the concave portion of the wafer surface to be inspected on the initial dividing circular surface is obtained, and the dividing line path is adjusted to form a secondary division.

[0051] Optionally, the step of obtaining the position of the concave portion of the surface of the wafer to be inspected on the initial dividing circular surface and adjusting the dividing line path to form a secondary division specifically includes:

[0052] If the concave part is within the standard surface, it is recorded as the concave standard surface;

[0053] Get the midpoints of two adjacent sides of the concave standard surface;

[0054] Draw perpendicular lines from the two midpoints to the opposite sides, and the intersection of the two perpendicular lines divides the concave standard surface into four concave detection areas, which are respectively recorded as the first concave detection area, the second concave detection area, the third concave detection area, and the fourth concave detection area;

[0055] Obtaining the area of ​​the concave portion and the areas of four concave detection regions;

[0056] Obtaining the area occupied by the concave portion in the four concave detection regions respectively, and calculating the ratio of the area to the area of ​​the concave detection region, and comparing them;

[0057] Selecting the largest area ratio among the four area ratios, and obtaining the concave detection area corresponding to the area ratio, which is recorded as the concave adjustment area;

[0058] Obtain a horizontal dividing line and a vertical dividing line where the concave adjustment area is located;

[0059] Use the concave adjustment area vertex strategy to obtain the row area and vertical area where the concave area is located;

[0060] Get the lengths of the two horizontal sides of the row area;

[0061] Get the number of all standard faces in the row area;

[0062] Calculate the shortest horizontal side length minus the horizontal side length of the concave adjustment area, and record it as the pre-adjusted horizontal side length;

[0063] Calculate the required horizontal length of all standard surfaces in the row area, and record it as the total required horizontal length;

[0064] If the pre-adjusted horizontal side length is greater than or equal to the total required horizontal side length, the vertical dividing line of the standard surface in the row area is moved away from the recessed adjustment area by a distance equal to the horizontal side length of the recessed adjustment area.

[0065] Optionally, the step of obtaining the position of the concave portion of the surface of the wafer to be inspected on the initial dividing circular surface and adjusting the dividing line path to form a secondary division specifically includes:

[0066] If the pre-adjusted horizontal side length is less than the total required horizontal side length, obtain the two vertical side lengths of the vertical area;

[0067] Get the number of all standard faces in the vertical area;

[0068] The length of the shortest vertical side minus the length of the vertical side of the concave adjustment area is recorded as the pre-adjusted vertical side length;

[0069] Calculate the required vertical side lengths of all standard surfaces in the vertical area, and record it as the total required vertical side length;

[0070] If the pre-adjusted vertical side length is greater than or equal to the total required vertical side length, the horizontal dividing line of the standard surface in the vertical area is moved away from the concave adjustment area by the length of the vertical side of the concave adjustment area;

[0071] If the pre-adjusted vertical side length is less than the total required vertical side length, no adjustment will be made.

[0072] Optionally, the step of using the concave adjustment area vertex strategy to obtain the row area and column area where the concave area is located specifically includes:

[0073] Obtain a horizontal side length and a vertical side length of the concave adjustment area corresponding to the concave standard surface;

[0074] Get the two vertices of the horizontal length, which are marked as point A and point B respectively;

[0075] Take point A as the vertex, and draw a line segment along the vertical dividing line where the concave standard surface is located, and the intersection with the circle is recorded as point C;

[0076] Take point B as the vertex, and draw a line segment along the direction of the vertical dividing line where the concave standard surface is located, and the intersection with the circle is recorded as point D;

[0077] The area enclosed by ABCD is recorded as the vertical area;

[0078] Get the two vertices of the vertical side length, which are marked as point E and point F respectively;

[0079] Take point E as the vertex, and draw a line segment along the direction of the horizontal dividing line where the concave standard surface is located, and the intersection with the circle is recorded as point G;

[0080] Take point F as the vertex, and draw a line segment along the direction of the horizontal dividing line where the concave standard surface is located, and the intersection with the circle is recorded as point H;

[0081] The area enclosed by EFGH is recorded as the horizontal area.

[0082] Optionally, the step of obtaining the position of the concave portion of the surface of the wafer to be inspected on the initial dividing circular surface and adjusting the dividing line path to form a secondary division specifically includes:

[0083] If the concave part is between the two standard surfaces and located on the vertical dividing line;

[0084] Calculate the proportion of the concave part in the two standard surfaces respectively and compare them, and obtain the standard surface with the smaller proportion, which is recorded as the concave standard surface;

[0085] Use the concave adjustment area vertex strategy to obtain the row area where the concave area is located;

[0086] Get the lengths of the two horizontal sides of the row area;

[0087] Get the number of all standard surfaces in the horizontal area;

[0088] The shortest horizontal side length - the horizontal side length of the concave adjustment area is recorded as the pre-adjusted horizontal side length;

[0089] Calculate the required horizontal length of all standard surfaces in the row area, and record it as the total required horizontal length;

[0090] If the pre-adjusted horizontal side length is greater than or equal to the total required horizontal side length, the vertical dividing line of the standard surface in the horizontal area is moved away from the concave adjustment area by the length of the horizontal side of the concave adjustment area;

[0091] If the pre-adjusted horizontal length is less than the total required horizontal length, no adjustment will be made.

[0092] Optionally, the step of obtaining the position of the concave portion of the surface of the wafer to be inspected on the initial dividing circular surface and adjusting the dividing line path to form a secondary division specifically includes:

[0093] If the concave part is between the two standard surfaces and located on the horizontal dividing line;

[0094] Calculate the proportion of the concave part in the two standard surfaces respectively and compare them, and obtain the standard surface with the smaller proportion, which is recorded as the concave standard surface;

[0095] Use the concave adjustment area vertex strategy to obtain the vertical area where the concave area is located;

[0096] Get the lengths of the two vertical sides of the vertical area;

[0097] Get the number of all standard faces in the vertical area;

[0098] The length of the shortest vertical side minus the vertical side length of the concave adjustment area is recorded as the pre-adjusted vertical side length;

[0099] Calculate the required vertical side lengths of all standard surfaces in the vertical area, and record it as the total required vertical side length;

[0100] If the pre-adjusted vertical side length is greater than or equal to the total required vertical side length, the horizontal dividing line of the standard surface in the vertical area is moved away from the concave adjustment area by the length of the vertical side of the concave adjustment area;

[0101] If the pre-adjusted vertical side length is less than the total required vertical side length, no adjustment will be made.

[0102] The present invention has the following beneficial effects:

[0103] 1. By setting a wafer that meets the quality standards as a reference, it can help determine the surface flatness and defects of the wafer to be tested. By setting the laser on the optical axis of the interferometer and collimating the light beam to form a parallel beam, the scattering and divergence of the light during propagation can be reduced, and the consistency of the light path can be improved. The parallelism of the light beam is crucial for the generation of interference fringes, which helps to ensure that the subsequent reflected light can accurately overlap, thereby improving the accuracy of the measurement and analysis results. By irradiating the parallel light beam vertically onto the beam splitter, the light beam can be effectively separated to ensure that the paths of the two separated light beams are parallel and the energy is evenly distributed. The light beam is divided into two beams through the beam splitter, one beam irradiates the standard wafer and the other beam irradiates the standard wafer. A beam of light is irradiated onto the wafer to be inspected, and a contrast can be established between the two. The beneficial effect of this process is that it can simultaneously obtain the reflected light of the two wafers, and use the interference phenomenon for comparison to generate an interference fringe image. The standard wafer and the wafer to be inspected are placed in parallel to ensure that the optical paths of the two are consistent, avoiding inaccurate measurement results due to angle differences. By adjusting the angle of the reflector, the light beam is accurately irradiated onto the wafer surface and reflected back to the spectrometer. The beneficial effect of this operation is that it ensures the accuracy of the light beam reflection and the symmetry of the optical path, reduces the error on the optical path, so that the generated interference fringes can clearly reflect the differences on the wafer surface, and ensure the stability and reliability of the measurement.

[0104] 2. After starting the light source, ensure that the light beam reaches the beam splitter accurately and is divided into two beams of light through collimation. This step ensures that the energy and direction of the light beam are stable. The two beams of light are reflected on different wafer surfaces and merged into interference fringes at the beam splitter. The beneficial effect of this step is that the interference fringe image can intuitively present the microscopic differences on the wafer surface, so that the surface defects of the wafer to be tested can be visualized, providing a basis for subsequent adjustments. Obtaining the interference fringe image provides a data basis for subsequent analysis. Its beneficial effect is that by observing the changes in the interference fringes, the unevenness of the surface of the wafer to be tested can be accurately judged. The changes in the fringes directly reflect the height difference of the wafer surface, providing a scientific basis for the preliminary adjustment of the wafer. The final analysis step is to interpret the interference fringes, mark the irregular surface area of ​​the wafer to be tested, and make preliminary adjustments. The beneficial effect of this step is that by accurately analyzing the interference fringes, the defective area on the wafer surface can be identified, helping to formulate an adjustment strategy, thereby improving the flatness of the wafer surface and ensuring the smooth progress of subsequent cutting work.

[0105] 3. Setting the maximum protrusion threshold is the key to determining whether the uneven surface of the wafer needs to be adjusted. By setting a threshold, it can be clear which protrusions on the wafer surface need to be polished, thereby improving the adjustment efficiency and accuracy. If the detected maximum protrusion is lower than or equal to the set threshold, it means that the surface flatness of the wafer meets the standard and no adjustment is required. The beneficial effect of this process is to avoid unnecessary operations. When the detected protrusion exceeds the set threshold, the area needs to be polished until it meets the standard. The beneficial effect of this process is that it can accurately eliminate surface protrusions and ensure that the wafer meets the expected flatness standard.

[0106] 4. Obtain any diameter as the first dividing diameter to provide a reference for subsequent dividing operations. Obtain a diameter perpendicular to the first dividing diameter as the second dividing diameter to help further construct a dividing framework. This perpendicular diameter provides a second important reference point for wafer segmentation to ensure the uniformity of the divided area in two directions. The side length of the rectangle where the sample chip is located is obtained to divide the dividing circular surface in proportion to the size of the sample chip. Dividing the first dividing diameter according to the first side length of the sample chip can ensure that the area on the dividing circular surface matches the size of the chip. By evenly dividing the diameter into N parts, the segmentation can be made more precise and regular, which is helpful for subsequent cutting and processing. Dividing the second dividing diameter according to the second side length is to form a segmentation point that matches the size of the sample chip in the other direction. Drawing a line segment parallel to the second dividing diameter through the segmentation point of the first dividing diameter can construct a preliminary segmentation framework. The beneficial effect of this step is to ensure that each line segment The parallelism in the same direction makes the divided area more regular. The line segments parallel to the first dividing diameter are drawn through the dividing points of the second dividing diameter to form the final grid structure. The beneficial effect of this operation is that a uniform grid is formed by interlacing parallel line segments, and the divided circular surface is divided into multiple independent areas. The circular surface is divided into multiple divided surfaces by interlaced line segments, which provides a basis for subsequent segmentation ratio calculation and processing. The beneficial effect of this step is that through uniform line segmentation, the area of ​​each divided surface can be ensured to be consistent or close, which is convenient for subsequent processing. By calculating the ratio of the dividing surface area to the sample chip area, it can be determined whether the size of each divided surface meets the standard. The dividing surface with a dividing ratio less than 1 is marked as a corner surface to distinguish those edge areas that are not suitable for chips. The dividing surface with a dividing ratio equal to 1 is marked as a standard surface, which means that these areas can be directly used for chip cutting. The divided circular surface is defined as the initial divided circular surface, marking the completion of the entire preliminary division.

[0107] 5. Obtain the two endpoints of the first dividing diameter as reference points to facilitate the sorting and marking of subsequent dividing lines. By clarifying the positions of the endpoints, the division operation can be made more orderly. Marking the horizontal dividing lines in order of distance from the first endpoint is helpful for systematic management of each line segment of the divided circular surface. The beneficial effect of this step is that each horizontal dividing line has a clear positioning through standardized naming and sorting. Obtain the two endpoints of the second dividing diameter as reference points to provide corresponding starting points for the vertical dividing line. The beneficial effect of this step is that by clarifying the endpoints of the second dividing diameter, the symmetry and regularity of the dividing lines can be maintained in different directions. According to the first Marking the vertical dividing lines in order of the distance of the three endpoints helps to systematically manage each vertical line segment. The beneficial effect of this step is that through unified naming and sorting, it ensures that the vertical dividing lines and the horizontal dividing lines form a clear staggered grid. The position of the recessed part of the surface of the wafer to be inspected is obtained in order to make adjustments based on the initial division to avoid the recessed part affecting the cutting quality. By adjusting the dividing line path, the division can be made more adapted to the actual surface conditions of the wafer. The beneficial effect of this process is that by identifying and adjusting the recessed area, it can ensure the accuracy of the division results and reduce the cutting error caused by the uneven surface, thereby improving the final processing quality of the wafer.

[0108] 6. Identifying whether the depressed part is located within the standard surface is to distinguish the impact range of the depression. Recording it as the "depression standard surface" can clearly identify the problem area, which is convenient for the precise positioning of subsequent adjustment operations. The depression standard surface is divided into four detection areas by drawing a vertical line through two midpoints, aiming to decompose the depression area more finely. The beneficial effect of this step is that after decomposing a larger area into multiple small areas, the impact of the depression on each area can be analyzed more accurately, making subsequent adjustments more targeted, reducing unnecessary errors, and improving the operation accuracy when repairing depressions. The area data of the depressed part and the detection area are obtained, laying the foundation for subsequent calculations. The beneficial effect of this step is that by quantitatively grasping the size and distribution of the depression, the impact of the depression on each detection area can be better evaluated, which is convenient for subsequent effective adjustments and corrections to the most serious depression area, thereby improving the flatness of the wafer surface, and obtaining the dividing line where the depression adjustment area is located, which is convenient for the precise positioning of the subsequent adjustment line. This step The beneficial effect is that by clarifying the position of the dividing line, a more accurate division basis can be provided for subsequent operations, ensuring that the adjustment path matches the actual concave position, and using the vertex strategy of the concave adjustment area to locate the row area and vertical area of ​​the concave, providing a coordinate basis for subsequent adjustments. The beneficial effect of this step is that through the application of the vertex strategy, the position of the concave in the two-dimensional plane can be more accurately determined, ensuring that the adjustment operation can be concentrated in specific row and column areas, improving the accuracy and effectiveness of the repair process, and by calculating the pre-adjusted horizontal side length, the specific side length adjustment amount can be determined, and the total horizontal side length of the standard surface in the row area is calculated to ensure that the adjustment does not affect the overall division structure. The beneficial effect of this step is that by calculating the required total horizontal side length, the problem of uneven overall division caused by adjusting the concave area can be avoided, thereby maintaining the geometric integrity and rationality of the divided surface while repairing the concave. This step adjusts the division structure by moving the vertical dividing line to ensure that the concave repair does not affect the consistency of the standard surface.

[0109] 7. The beneficial effect of this sentence is that by judging whether the pre-adjusted horizontal side length is less than the total required horizontal side length, it can be decided whether to continue the adjustment operation of the vertical area to avoid unnecessary complex calculations. If the horizontal side length is insufficient, it is necessary to transfer to the adjustment of the vertical area. The role of this step is to accurately evaluate the feasibility of horizontal adjustment. The pre-adjusted vertical side length is calculated to judge the feasibility and necessity of vertical adjustment. By subtracting the vertical side length of the concave adjustment area, the pre-adjusted vertical side length data can be obtained to provide a basis for the subsequent judgment. By calculating the total vertical side length of the standard surface, it can be ensured that the segmentation structure of the standard surface will not be negatively affected when the vertical adjustment is performed, thereby maintaining the integrity and consistency of the standard surface within the divided circular surface. If the pre-adjusted vertical side length is greater than or equal to the total required vertical side length, it means that the space in the vertical area is sufficient and will not affect the overall standard surface division. This step can effectively solve the concave problem by moving the horizontal dividing line, while maintaining the integrity of the vertical area and ensuring that the adjustment operation will not affect the division results of other parts. When the pre-adjusted vertical side length is less than the total required vertical side length, no adjustment is performed, which can avoid the problem of uneven division caused by insufficient vertical adjustment.

[0110] 8. The horizontal and vertical lengths of the concave adjustment area are obtained to define the boundary of the area and determine the scope of the adjustment operation. The beneficial effect of this step is that by clarifying the horizontal and vertical boundaries, a clear geometric basis can be provided for the subsequent segmentation line adjustment; the boundary of the vertical area is clarified by extending the line segment from point A to the circumference, and the boundary of the vertical area is further determined by extending from point B to the circumference. The beneficial effect of this step is that it cooperates with the previous point A-point C operation to complete the overall definition of the vertical area. By connecting points A, B, C, and D, the boundary range of the vertical area is determined. The beneficial effect of this step is that by constructing a complete vertical area, it can ensure that the concave adjustment operation is performed within a specific range; obtaining the two vertices of the vertical side length, points E and F, helps to clarify the two key points of the horizontal boundary, extending from point E as the vertex to the circumference to further determine the boundary of the horizontal area, and extending from point F to the circumference to determine the boundary of the horizontal area to form a complete horizontal area. The beneficial effect of this step is that it cooperates with the aforementioned point E-point G operation to completely define the boundary of the horizontal area, providing a clear operating range for the concave adjustment.

[0111] 9. By obtaining the position of the concave part, the setting of this condition helps to determine the specific position of the concave part, especially between the standard surfaces and on the vertical dividing line. By clarifying this condition, the special position of the concave area between the two standard surfaces can be accurately identified, providing a basis for subsequent segmentation adjustments. By calculating the proportion of the concave part on the two standard surfaces and selecting the standard surface with a smaller proportion as the concave standard surface, it can be effectively determined which area needs to be adjusted; use the vertex strategy to obtain the row area of ​​the concave area to ensure that the concave area can be accurately located in the row and column division of the wafer. The beneficial effect of this step is that the vertex strategy can be used to more accurately locate the row area where the concave is located, thereby ensuring that the subsequent adjustment operation can be performed in the correct position; the horizontal length of the row area is obtained to clarify the boundary of the adjustment range. The beneficial effect of this step is that by mastering the horizontal length of the row area, a data basis can be provided for the subsequent horizontal dividing line adjustment, so as to better optimize the adjustment path; by obtaining the number of all standard surfaces in the horizontal area, a benchmark can be provided for the adjustment operation. The beneficial effect of this step is that by calculating the number of standard surfaces, the total number of standard surfaces that need to be adjusted during the repair process can be accurately evaluated, so as to ensure the rationality and accuracy of the repair range; by subtracting the horizontal length of the concave adjustment area, the pre-adjusted horizontal length can be calculated to evaluate whether the dividing line needs to be adjusted; calculating the required horizontal lengths for all standard surfaces can help evaluate the space requirements in the row area.

[0112] 10. By obtaining the position of the sunken part, the setting of this condition clarifies the specific position of the sunken part, especially in the case of the horizontal dividing line, it helps to determine the position of the sunken area between the two standard surfaces, providing a positioning basis for the subsequent segmentation path adjustment. The beneficial effect of this step is to make the segmentation adjustment more targeted; by calculating the proportion of the sunken part in the two standard surfaces, the area with less sunken impact can be determined as the key area for adjustment; the vertex strategy can be used to accurately locate the specific position of the sunken area in the vertical area. The beneficial effect of this step is that through the vertex strategy, the sunken area can be efficiently locked to ensure that the subsequent adjustment can be carried out on the correct vertical area; obtaining the lengths of the two vertical sides of the vertical area helps to determine the range and spatial size of the vertical area. The beneficial effect of this step is By accurately grasping the side length of the vertical area, it is possible to provide necessary geometric parameters for subsequent vertical adjustments to ensure a reasonable adjustment range; obtaining the number of standard surfaces in the vertical area can help evaluate the number of standard surfaces that need to be adjusted and provide a reference for the subsequent repair process. The beneficial effect of this step is that it can accurately calculate the repair range of the vertical area and ensure reasonable allocation during the adjustment process; calculating the pre-adjusted vertical side length is to evaluate whether there is enough space for adjustment. The beneficial effect of this step is that by calculating the pre-adjusted vertical side length, it is possible to predict whether the adjustment can be completed in the existing space and prevent repair errors caused by insufficient space; by calculating the vertical side lengths required for all standard surfaces in the vertical area, the space requirements for the overall repair can be evaluated. The beneficial effect of this step is that by calculating the total required vertical side lengths, the space in the vertical area can be reasonably allocated. BRIEF DESCRIPTION OF THE DRAWINGS

[0113] Figure 1 This is a schematic diagram of the initial division of the circular surface of the present invention.

[0114] Figure 2 It is a schematic diagram of the concave portion of the present invention within the standard plane. DETAILED DESCRIPTION

[0115] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0116] Embodiment, a method for cutting a semiconductor wafer, comprising:

[0117] Obtain a wafer whose surface meets the standard and record it as a standard wafer;

[0118] Obtain a wafer whose surface is to be inspected, recorded as the wafer to be inspected;

[0119] Instrument setup and adjustment:

[0120] The laser is placed on the optical axis of the interferometer, and the emitted light beam is collimated to form a parallel light beam. The collimation refers to the process of converting the light beam into a parallel light beam, so that the light remains parallel during propagation, reducing scattering and divergence. Collimation is to maintain the consistency of the optical path, improve measurement accuracy and reduce beam diffraction;

[0121] Direct the parallel beam of light vertically onto the beam splitter;

[0122] The beam splitter is used to split the light beam into two beams, one beam of light is vertically irradiated to the surface of the standard wafer, and the other beam of light is vertically irradiated to the surface of the wafer to be tested;

[0123] Place the standard wafer and the wafer to be tested in parallel on the testing table of the interferometer;

[0124] The standard wafer is placed on one of the reflection paths of the beam and serves as a reflection plane for the reference beam;

[0125] The wafer to be inspected is placed on another reflection path, serving as a reflection plane for the measurement beam;

[0126] Adjust the angle of the reflector so that the light beam separated from the beam splitter can be irradiated onto the standard wafer and the wafer to be inspected respectively, and reflected back to the beam splitter;

[0127] Wafer surface measurement:

[0128] Start the laser light source, and after the light beam is collimated, it is split into two beams by a beam splitter, which are irradiated onto the surface of the standard wafer and the wafer to be tested respectively;

[0129] The two beams of light are reflected on the surfaces of the standard wafer and the wafer to be tested respectively and then merged into interference fringes;

[0130] Adjust the angle of the reflector so that the two beams overlap at the beam splitter;

[0131] Acquire interference fringe images formed by reflected light;

[0132] The interference fringe image is analyzed, the concave and convex positions on the surface of the wafer to be inspected are marked, and preliminary adjustments are made to the wafer to be inspected.

[0133] By setting a wafer that meets the quality standards as a reference, it can help determine the surface flatness and defects of the wafer to be tested. By setting the laser on the optical axis of the interferometer and collimating the light beam to form a parallel beam, the scattering and divergence of the light during propagation can be reduced, and the consistency of the light path can be improved. The parallelism of the light beam is crucial to the generation of interference fringes, which helps to ensure that the subsequent reflected light can accurately overlap, thereby improving the accuracy of the measurement and analysis results. By irradiating the parallel light beam vertically onto the beam splitter, the light beam can be effectively separated to ensure that the paths of the two separated light beams are parallel and the energy is evenly distributed. The light beam is divided into two beams through the beam splitter, one beam irradiates the standard wafer and the other beam irradiates the standard wafer. The beam is irradiated onto the wafer to be inspected, and a contrast can be established between the two. The beneficial effect of this process is that it can simultaneously obtain the reflected light of the two wafers, and use the interference phenomenon for comparison, thereby generating an interference fringe image. The standard wafer and the wafer to be inspected are placed in parallel to ensure that the optical paths of the two are consistent, avoiding inaccurate measurement results due to angle differences. By adjusting the angle of the reflector, the beam is ensured to be accurately irradiated onto the wafer surface and reflected back to the spectrometer. The beneficial effect of this operation is that it ensures the accuracy of the beam reflection and the symmetry of the optical path, reduces the error on the optical path, so that the generated interference fringes can clearly reflect the differences on the wafer surface, and ensure the stability and reliability of the measurement.

[0134] After starting the light source, collimation is used to ensure that the light beam reaches the beam splitter accurately and is divided into two beams of light. This step ensures that the energy and direction of the light beam are stable. The two beams of light are reflected on different wafer surfaces and merged into interference fringes at the beam splitter. The beneficial effect of this step is that the interference fringe image can intuitively present the microscopic differences on the wafer surface, so that the surface defects of the wafer to be tested can be visualized, providing a basis for subsequent adjustments. Obtaining the interference fringe image provides a data basis for subsequent analysis. Its beneficial effect is that by observing the changes in the interference fringes, the unevenness of the surface of the wafer to be tested can be accurately judged. The changes in the fringes directly reflect the height difference of the wafer surface, providing a scientific basis for the preliminary adjustment of the wafer. The final analysis step is to interpret the interference fringes, mark the irregular surface area of ​​the wafer to be tested, and make preliminary adjustments. The beneficial effect of this step is that by accurately analyzing the interference fringes, the defective area on the wafer surface can be identified, helping to formulate an adjustment strategy, thereby improving the flatness of the wafer surface and ensuring the smooth progress of subsequent cutting work.

[0135] The preliminary adjustment of the wafer to be inspected specifically includes:

[0136] Set the highest convexity threshold. Setting the highest convexity threshold is used to determine whether the convexity on the crystal surface needs to be adjusted. The smaller the highest convexity threshold, the lower the convexity on the crystal surface, and vice versa.

[0137] Acquire a raised area on a wafer to be inspected;

[0138] Get the highest point of the raised area;

[0139] Obtain the distance from the highest point to the wafer surface, which is recorded as the highest value of the protrusion;

[0140] If the highest bulge value is less than or equal to the highest bulge value threshold, no adjustment is made;

[0141] If the highest value of the bulge is greater than the highest value threshold of the bulge, the bulge area is polished until the highest value of the bulge area is less than or equal to the highest value threshold of the bulge;

[0142] Perform initial division on the wafers to be inspected after preliminary adjustment.

[0143] Setting the threshold for the maximum value of the protrusions is the key to determining whether the unevenness of the wafer surface needs to be adjusted. By setting a threshold, it can be clear which protrusions on the wafer surface need to be polished, thereby improving the adjustment efficiency and accuracy. If the detected maximum value of the protrusion is lower than or equal to the set threshold, it means that the surface flatness of the wafer meets the standard and no adjustment is required. The beneficial effect of this process is to avoid unnecessary operations. When the detected protrusion exceeds the set threshold, the area needs to be polished until it meets the standard. The beneficial effect of this process is that it can accurately eliminate surface protrusions and ensure that the wafer meets the expected flatness standard.

[0144] Reference Figure 1 The initial division of the wafers to be inspected after the preliminary adjustment specifically includes:

[0145] Get a sample chip;

[0146] The circular surface formed by the wafer to be inspected is recorded as the divided circular surface;

[0147] Obtain any diameter of the divided circular surface and record it as the first divided diameter;

[0148] Obtain a diameter perpendicular to the first dividing diameter and record it as the second dividing diameter;

[0149] Obtain the lengths of two intersecting sides of the rectangle where the sample chip is located and record them as the first side length and the second side length respectively;

[0150] Divide the first dividing diameter from the center of the dividing circle into N parts on both sides of the center according to the first side length, and mark the dividing points;

[0151] The second dividing diameter starts from the center of the dividing circle, and is divided into M parts according to the second side length on both sides of the center, and the dividing points are marked;

[0152] Line segments parallel to the second dividing diameter are drawn sequentially through the dividing points on the first dividing diameter, and the endpoints of the line segments intersect on the circumference of the dividing circular surface;

[0153] Line segments parallel to the first dividing diameter are drawn successively through the dividing points on the second dividing diameter, and the endpoints of the line segments intersect on the circumference of the dividing circular surface;

[0154] The partitioning circle is divided into a number of partitioning surfaces by interlaced line segments;

[0155] Obtaining the areas of all segmentation surfaces within the segmentation circle, and calculating the ratio of the area of ​​each segmentation surface to the area of ​​the sample chip, which is recorded as the segmentation ratio;

[0156] The segmentation faces with segmentation ratio less than 1 are recorded as corner faces;

[0157] The split surface with a split ratio equal to 1 is recorded as the standard surface;

[0158] The divided circular surface is recorded as the initial divided circular surface.

[0159] Acquiring any diameter as the first dividing diameter is convenient for providing a reference for subsequent dividing operations. Acquiring a diameter perpendicular to the first dividing diameter as the second dividing diameter is helpful for further constructing a dividing framework. This perpendicular diameter provides a second important reference point for wafer segmentation, ensuring the uniformity of the divided area in two directions. Acquiring the side length of the rectangle where the sample chip is located is to divide the dividing circular surface in equal proportion according to the size of the sample chip. Dividing the first dividing diameter according to the first side length of the sample chip can ensure that the area on the dividing circular surface matches the size of the chip. By evenly dividing the diameter into N parts, the segmentation can be made more precise and regular, which is helpful for subsequent cutting and processing. Dividing the second dividing diameter according to the second side length is to form a segmentation point that matches the size of the sample chip in the other direction. Drawing a line segment parallel to the second dividing diameter through the segmentation point of the first dividing diameter can construct a preliminary segmentation framework. The beneficial effect of this step is to ensure that each line segment is The parallelism in the same direction makes the divided area more regular. The line segments parallel to the first divided diameter are drawn through the division points of the second divided diameter to form the final grid structure. The beneficial effect of this operation is that a uniform grid is formed by interlacing parallel line segments, and the divided circular surface is divided into multiple independent areas. The circular surface is divided into multiple divided surfaces by interlaced line segments, which provides a basis for subsequent segmentation ratio calculation and processing. The beneficial effect of this step is that through uniform line segmentation, the area of ​​each divided surface can be ensured to be consistent or close, which is convenient for subsequent processing. By calculating the ratio of the divided surface area to the sample chip area, it can be determined whether the size of each divided surface meets the standard. The dividing surface with a dividing ratio less than 1 is marked as a corner surface to distinguish those edge areas that are not suitable for chips. The dividing surface with a dividing ratio equal to 1 is marked as a standard surface, which means that these areas can be directly used for chip cutting. The divided circular surface is defined as the initial divided circular surface, marking the completion of the entire preliminary division.

[0160] The initial division of the wafers to be inspected after the preliminary adjustment specifically includes:

[0161] Obtain two endpoints of the first divided diameter, which are respectively recorded as the first endpoint and the second endpoint;

[0162] The line segments passing through the first dividing diameter are recorded in order from the nearest to the first endpoint as the first horizontal dividing line, the second horizontal dividing line, ... the Nth horizontal dividing line;

[0163] Obtain two endpoints of the second divided diameter, which are respectively recorded as the third endpoint and the fourth endpoint;

[0164] The line segments passing through the first dividing diameter are recorded in order from the nearest to the third endpoint as the first vertical dividing line, the second vertical dividing line, ... the Mth vertical dividing line;

[0165] The position of the concave portion of the wafer surface to be inspected on the initial dividing circular surface is obtained, and the dividing line path is adjusted to form a secondary division.

[0166] The two endpoints of the first dividing diameter are obtained as reference points to facilitate the sorting and marking of subsequent dividing lines. By clarifying the positions of the endpoints, the division operation can be made more orderly. The horizontal dividing lines are marked in order of distance from the first endpoint, which is helpful for systematic management of each line segment of the divided circular surface. The beneficial effect of this step is that each horizontal dividing line has a clear positioning through standardized naming and sorting. The two endpoints of the second dividing diameter are obtained as reference points to provide corresponding starting points for the vertical dividing line. The beneficial effect of this step is that by clarifying the endpoints of the second dividing diameter, the symmetry and regularity of the dividing lines can be maintained in different directions. According to the third Marking the vertical dividing lines in order of the distance of the endpoints helps to systematically manage each vertical line segment. The beneficial effect of this step is that through unified naming and sorting, it ensures that the vertical dividing lines and the horizontal dividing lines form a clear staggered grid. The position of the recessed part of the surface of the wafer to be inspected is obtained in order to make adjustments based on the initial division to avoid the recessed part affecting the cutting quality. By adjusting the dividing line path, the division can be made more adapted to the actual surface conditions of the wafer. The beneficial effect of this process is that by identifying and adjusting the recessed area, it can ensure the accuracy of the division results and reduce the cutting error caused by the uneven surface, thereby improving the final processing quality of the wafer.

[0167] Reference Figure 2 The step of obtaining the position of the concave portion of the wafer surface to be detected on the initial dividing circular surface and adjusting the dividing line path to form a secondary division specifically includes:

[0168] If the concave part is within the standard surface, it is recorded as the concave standard surface;

[0169] Get the midpoints of two adjacent sides of the concave standard surface;

[0170] Draw perpendicular lines from the two midpoints to the opposite sides, and the intersection of the two perpendicular lines divides the concave standard surface into four concave detection areas, which are respectively recorded as the first concave detection area, the second concave detection area, the third concave detection area, and the fourth concave detection area;

[0171] Obtaining the area of ​​the concave portion and the areas of four concave detection regions;

[0172] Obtaining the area occupied by the concave portion in the four concave detection regions respectively, and calculating the ratio of the area to the area of ​​the concave detection region, and comparing them;

[0173] Selecting the largest area ratio among the four area ratios, and obtaining the concave detection area corresponding to the area ratio, which is recorded as the concave adjustment area;

[0174] Obtain a horizontal dividing line and a vertical dividing line where the concave adjustment area is located;

[0175] Use the concave adjustment area vertex strategy to obtain the row area and vertical area where the concave area is located;

[0176] Get the lengths of the two horizontal sides of the row area;

[0177] Get the number of all standard faces in the row area;

[0178] Calculate the shortest horizontal side length minus the horizontal side length of the concave adjustment area, and record it as the pre-adjusted horizontal side length;

[0179] Calculate the required transverse length of all standard surfaces in the row area. The required transverse length of all standard surfaces is the number of standard surfaces in the transverse area × the length of one transverse side of the standard surface, which is recorded as the total required transverse length.

[0180] If the pre-adjusted horizontal side length is greater than or equal to the total required horizontal side length, the vertical dividing line of the standard surface in the row area is moved away from the recessed adjustment area by a distance equal to the horizontal side length of the recessed adjustment area.

[0181] Identifying whether the depressed part is located within the standard surface is to distinguish the impact range of the depression. Recording it as the "depressed standard surface" can clearly identify the problem area, which is convenient for the precise positioning of subsequent adjustment operations. The depressed standard surface is divided into four detection areas by drawing a vertical line through two midpoints, aiming to decompose the depressed area more finely. The beneficial effect of this step is that after decomposing a larger area into multiple small areas, the impact of the depression on each area can be analyzed more accurately, making subsequent adjustments more targeted, reducing unnecessary errors, and improving the operational accuracy when repairing depressions. The area data of the depressed part and the detection area are obtained, laying the foundation for subsequent calculations. The beneficial effect of this step is that by quantitatively grasping the size and distribution of the depression, the impact of the depression on each detection area can be better evaluated, which is convenient for subsequent effective adjustments and corrections to the most serious depression areas, thereby improving the flatness of the wafer surface, and obtaining the dividing line where the depression adjustment area is located, which is convenient for the precise positioning of the subsequent adjustment line. This step The beneficial effect is that by clarifying the position of the dividing line, a more accurate division basis can be provided for subsequent operations, ensuring that the adjustment path matches the actual concave position, and using the vertex strategy of the concave adjustment area to locate the row area and vertical area of ​​the concave, providing a coordinate basis for subsequent adjustments. The beneficial effect of this step is that through the application of the vertex strategy, the position of the concave in the two-dimensional plane can be more accurately determined, ensuring that the adjustment operation can be concentrated in specific row and column areas, improving the accuracy and effectiveness of the repair process, and by calculating the pre-adjusted horizontal side length, the specific side length adjustment amount can be determined, and the total horizontal side length of the standard surface in the row area is calculated to ensure that the adjustment does not affect the overall division structure. The beneficial effect of this step is that by calculating the required total horizontal side length, the problem of uneven overall division caused by adjusting the concave area can be avoided, thereby maintaining the geometric integrity and rationality of the divided surface while repairing the concave. This step adjusts the division structure by moving the vertical dividing line to ensure that the concave repair does not affect the consistency of the standard surface.

[0182] The step of obtaining the position of the concave portion of the surface of the wafer to be inspected on the initial dividing circular surface and adjusting the dividing line path to form a secondary division specifically includes:

[0183] If the pre-adjusted horizontal side length is less than the total required horizontal side length, obtain the two vertical side lengths of the vertical area;

[0184] Get the number of all standard faces in the vertical area;

[0185] The length of the shortest vertical side minus the length of the vertical side of the concave adjustment area is recorded as the pre-adjusted vertical side length;

[0186] Calculate the required vertical side lengths of all standard surfaces in the vertical area. The required vertical side lengths of all standard surfaces are the number of standard surfaces in the vertical area × the length of one vertical side of the standard surface, which is recorded as the total required vertical side length.

[0187] If the pre-adjusted vertical side length is greater than or equal to the total required vertical side length, the horizontal dividing line of the standard surface in the vertical area is moved away from the concave adjustment area by the length of the vertical side of the concave adjustment area;

[0188] If the pre-adjusted vertical side length is less than the total required vertical side length, no adjustment will be made.

[0189] The beneficial effect of this sentence is that by judging whether the pre-adjusted horizontal side length is less than the total required horizontal side length, it can be decided whether to continue the adjustment operation of the vertical area to avoid unnecessary complex calculations. If the horizontal side length is insufficient, it is necessary to transfer to the adjustment of the vertical area. The role of this step is to accurately evaluate the feasibility of horizontal adjustment. The pre-adjusted vertical side length is calculated to judge the feasibility and necessity of vertical adjustment. By subtracting the vertical side length of the concave adjustment area, the pre-adjusted vertical side length data can be obtained to provide a basis for the subsequent judgment. By calculating the total vertical side length of the standard surface, it can be ensured that the segmentation structure of the standard surface will not be negatively affected when the vertical adjustment is performed, thereby maintaining the integrity and consistency of the standard surface within the divided circular surface. If the pre-adjusted vertical side length is greater than or equal to the total required vertical side length, it means that the space in the vertical area is sufficient and will not affect the overall standard surface division. This step can effectively solve the concave problem by moving the horizontal dividing line, while maintaining the integrity of the vertical area, ensuring that the adjustment operation will not affect the division results of other parts. When the pre-adjusted vertical side length is less than the total required vertical side length, no adjustment is performed, which can avoid the problem of uneven division caused by insufficient vertical adjustment.

[0190] The method of using the concave adjustment area vertex strategy to obtain the row area and column area where the concave area is located specifically includes:

[0191] Obtain a horizontal side length and a vertical side length of the concave adjustment area corresponding to the concave standard surface;

[0192] Get the two vertices of the horizontal length, which are marked as point A and point B respectively;

[0193] Take point A as the vertex, and draw a line segment along the vertical dividing line where the concave standard surface is located, and the intersection with the circle is recorded as point C;

[0194] Take point B as the vertex, and draw a line segment along the direction of the vertical dividing line where the concave standard surface is located, and the intersection with the circle is recorded as point D;

[0195] The area enclosed by ABCD is recorded as the vertical area;

[0196] Get the two vertices of the vertical side length, which are marked as point E and point F respectively;

[0197] Take point E as the vertex, and draw a line segment along the direction of the horizontal dividing line where the concave standard surface is located, and the intersection with the circle is recorded as point G;

[0198] Take point F as the vertex, and draw a line segment along the direction of the horizontal dividing line where the concave standard surface is located, and the intersection with the circle is recorded as point H;

[0199] The area enclosed by EFGH is recorded as the horizontal area.

[0200] The purpose of obtaining the horizontal and vertical lengths of the concave adjustment area is to define the boundaries of the area and determine the scope of the adjustment operation. The beneficial effect of this step is that by clarifying the horizontal and vertical boundaries, a clear geometric basis can be provided for the subsequent segmentation line adjustment; the boundary of the vertical area is clarified by extending the line segment from point A to the circumference, and the boundary of the vertical area is further determined by extending from point B to the circumference. The beneficial effect of this step is that it cooperates with the previous point A-point C operation to complete the overall definition of the vertical area. By connecting points A, B, C, and D, the boundary range of the vertical area is determined. The beneficial effect of this step is that by constructing a complete vertical area, it can ensure that the concave adjustment operation is performed within a specific range; obtaining the two vertices of the vertical side length, points E and F, helps to clarify the two key points of the horizontal boundary, extending from point E as the vertex to the circumference to further determine the boundary of the horizontal area, and extending from point F to the circumference to determine the boundary of the horizontal area to form a complete horizontal area. The beneficial effect of this step is that it cooperates with the aforementioned point E-point G operation to completely define the boundary of the horizontal area, providing a clear operating range for the concave adjustment.

[0201] The step of obtaining the position of the concave portion of the surface of the wafer to be inspected on the initial dividing circular surface and adjusting the dividing line path to form a secondary division specifically includes:

[0202] If the concave part is between the two standard surfaces and located on the vertical dividing line;

[0203] Calculate the proportion of the concave part in the two standard surfaces respectively and compare them, and obtain the standard surface with the smaller proportion, which is recorded as the concave standard surface;

[0204] Use the concave adjustment area vertex strategy to obtain the row area where the concave area is located;

[0205] Get the lengths of the two horizontal sides of the row area;

[0206] Get the number of all standard surfaces in the horizontal area;

[0207] The shortest horizontal side length - the horizontal side length of the concave adjustment area is recorded as the pre-adjusted horizontal side length;

[0208] Calculate the required transverse length of all standard surfaces in the row area. The required transverse length of all standard surfaces is the number of standard surfaces in the transverse area × the length of one transverse side of the standard surface, which is recorded as the total required transverse length.

[0209] If the pre-adjusted horizontal side length is greater than or equal to the total required horizontal side length, the vertical dividing line of the standard surface in the horizontal area is moved away from the concave adjustment area by the length of the horizontal side of the concave adjustment area;

[0210] If the pre-adjusted horizontal length is less than the total required horizontal length, no adjustment will be made.

[0211] By obtaining the position of the concave part, the setting of this condition helps to determine the specific position of the concave part, especially between the standard surfaces and on the vertical dividing line. By clarifying this condition, the special position of the concave area between the two standard surfaces can be accurately identified, providing a basis for subsequent segmentation adjustments. By calculating the proportion of the concave part on the two standard surfaces and selecting the standard surface with a smaller proportion as the concave standard surface, it is possible to effectively determine which area needs to be adjusted; the vertex strategy is used to obtain the row area of ​​the concave area to ensure that the concave area can be accurately located in the row and column division of the wafer. The beneficial effect of this step is that the vertex strategy can be used to more accurately locate the row area where the concave is located, thereby ensuring that the subsequent adjustment operation can be performed at the correct position. The purpose of obtaining the horizontal length of the row area is to clarify the boundary of the adjustment range. The beneficial effect of this step is that by mastering the horizontal length of the row area, a data basis can be provided for the subsequent adjustment of the horizontal dividing line, so as to better optimize the adjustment path; by obtaining the number of all standard surfaces in the horizontal area, a benchmark can be provided for the adjustment operation. The beneficial effect of this step is that by calculating the number of standard surfaces, the total number of standard surfaces that need to be adjusted during the repair process can be accurately evaluated, so as to ensure the rationality and accuracy of the repair range; by subtracting the horizontal length of the concave adjustment area, the pre-adjusted horizontal length can be calculated to evaluate whether the dividing line needs to be adjusted; calculating the required horizontal length of all standard surfaces can help evaluate the space requirements in the row area.

[0212] The step of obtaining the position of the concave portion of the surface of the wafer to be inspected on the initial dividing circular surface and adjusting the dividing line path to form a secondary division specifically includes:

[0213] If the concave part is between the two standard surfaces and located on the horizontal dividing line;

[0214] Calculate the proportion of the concave part in the two standard surfaces respectively and compare them, and obtain the standard surface with the smaller proportion, which is recorded as the concave standard surface;

[0215] Use the concave adjustment area vertex strategy to obtain the vertical area where the concave area is located;

[0216] Get the lengths of the two vertical sides of the vertical area;

[0217] Get the number of all standard faces in the vertical area;

[0218] The length of the shortest vertical side minus the vertical side length of the concave adjustment area is recorded as the pre-adjusted vertical side length;

[0219] Calculate the required vertical side lengths of all standard surfaces in the vertical area. The required vertical side lengths of all standard surfaces are the number of standard surfaces in the vertical area × the length of one vertical side of the standard surface, which is recorded as the total required vertical side length.

[0220] If the pre-adjusted vertical side length is greater than or equal to the total required vertical side length, the horizontal dividing line of the standard surface in the vertical area is moved away from the concave adjustment area by the length of the vertical side of the concave adjustment area;

[0221] If the pre-adjusted vertical side length is less than the total required vertical side length, no adjustment will be made.

[0222] By obtaining the position of the sunken part, the setting of this condition clarifies the specific position of the sunken part, especially in the case of the horizontal dividing line, it helps to determine the position of the sunken area between the two standard surfaces, and provides a positioning basis for the subsequent segmentation path adjustment. The beneficial effect of this step is to make the segmentation adjustment more targeted; by calculating the proportion of the sunken part in the two standard surfaces, the area with less impact from the sunken part can be determined as the key area for adjustment; the vertex strategy can be used to accurately locate the specific position of the sunken area in the vertical area. The beneficial effect of this step is that through the vertex strategy, the sunken area can be efficiently locked to ensure that the subsequent adjustment can be carried out on the correct vertical area; obtaining the lengths of the two vertical sides of the vertical area helps to determine the range and spatial size of the vertical area. The beneficial effect of this step is that through Accurately grasping the side length of the vertical area can provide the necessary geometric parameters for subsequent vertical adjustments to ensure a reasonable adjustment range; obtaining the number of standard surfaces in the vertical area can help evaluate the number of standard surfaces that need to be adjusted and provide a reference for the subsequent repair process. The beneficial effect of this step is that it can accurately calculate the repair range of the vertical area and ensure reasonable allocation during the adjustment process; calculating the pre-adjusted vertical side length is to evaluate whether there is enough space for adjustment. The beneficial effect of this step is that by calculating the pre-adjusted vertical side length, it is possible to predict whether the adjustment can be completed in the existing space and prevent repair errors caused by insufficient space; by calculating the required vertical side lengths of all standard surfaces in the vertical area, the space requirements for the overall repair can be evaluated. The beneficial effect of this step is that by calculating the total required vertical side lengths, the space in the vertical area can be reasonably allocated.

[0223] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0224] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for cutting a semiconductor wafer, characterized in that: include: Obtain a wafer whose surface meets the standard and record it as a standard wafer; Obtain a wafer whose surface is to be inspected, recorded as the wafer to be inspected; Instrument setup and adjustment: The laser is placed on the optical axis of the interferometer, and the emitted light beam is collimated to form a parallel light beam; Direct the parallel beam of light vertically onto the beam splitter; The beam splitter is used to split the light beam into two beams, one beam of light is vertically irradiated to the surface of the standard wafer, and the other beam of light is vertically irradiated to the surface of the wafer to be tested; Place the standard wafer and the wafer to be tested in parallel on the testing table of the interferometer; The standard wafer is placed on one of the reflection paths of the beam and serves as a reflection plane for the reference beam; The wafer to be inspected is placed on another reflection path, serving as a reflection plane for the measurement beam; Adjust the angle of the reflector so that the light beam separated from the beam splitter can be irradiated onto the standard wafer and the wafer to be inspected respectively, and reflected back to the beam splitter; Wafer surface measurement: Start the laser light source, and after the light beam is collimated, it is split into two beams by a beam splitter, which are irradiated onto the surface of the standard wafer and the wafer to be tested respectively; The two beams of light are reflected on the surfaces of the standard wafer and the wafer to be tested respectively and then merged into interference fringes; Adjust the angle of the reflector so that the two beams overlap at the beam splitter; Acquire interference fringe images formed by reflected light; Analyze the interference fringe image, mark the concave and convex positions on the surface of the wafer to be inspected, and make preliminary adjustments to the wafer to be inspected; The preliminary adjustment of the wafer to be inspected specifically includes: Set the highest convexity threshold; Acquire a raised area on a wafer to be inspected; Get the highest point of the raised area; Obtain the distance from the highest point to the wafer surface, which is recorded as the highest value of the protrusion; If the highest bulge value is less than or equal to the highest bulge value threshold, no adjustment is made; If the highest value of the bulge is greater than the highest value threshold of the bulge, the bulge area is polished until the highest value of the bulge area is less than or equal to the highest value threshold of the bulge; Perform initial division on the wafers to be inspected after preliminary adjustment; The initial division of the wafers to be inspected after the preliminary adjustment specifically includes: Get a sample chip; The circular surface formed by the wafer to be inspected is recorded as the divided circular surface; Obtain any diameter of the divided circular surface and record it as the first divided diameter; Obtain a diameter perpendicular to the first dividing diameter and record it as the second dividing diameter; Obtain the lengths of two intersecting sides of the rectangle where the sample chip is located and record them as the first side length and the second side length respectively; Divide the first dividing diameter from the center of the dividing circle into N parts on both sides of the center according to the first side length, and mark the dividing points; The second dividing diameter starts from the center of the dividing circle, and is divided into M parts according to the second side length on both sides of the center, and the dividing points are marked; Line segments parallel to the second dividing diameter are drawn sequentially through the dividing points on the first dividing diameter, and the endpoints of the line segments intersect on the circumference of the dividing circular surface; Line segments parallel to the first dividing diameter are drawn successively through the dividing points on the second dividing diameter, and the endpoints of the line segments intersect on the circumference of the dividing circular surface; The partitioning circle is divided into a number of partitioning surfaces by interlaced line segments; Obtaining the areas of all segmentation surfaces within the segmentation circle, and calculating the ratio of the area of ​​each segmentation surface to the area of ​​the sample chip, which is recorded as the segmentation ratio; The segmentation faces with segmentation ratio less than 1 are recorded as corner faces; The split surface with a split ratio equal to 1 is recorded as the standard surface; The divided circular surface is recorded as the initial divided circular surface; Obtain two endpoints of the first divided diameter, which are respectively recorded as the first endpoint and the second endpoint; The line segments passing through the first dividing diameter are recorded in order from the nearest to the first endpoint as the first horizontal dividing line, the second horizontal dividing line, ... the Nth horizontal dividing line; Obtain two endpoints of the second divided diameter, which are respectively recorded as the third endpoint and the fourth endpoint; The line segments passing through the second dividing diameter are recorded in order from the nearest to the third endpoint as the first vertical dividing line, the second vertical dividing line, ... the Mth vertical dividing line; Obtaining the position of the concave portion of the surface of the wafer to be inspected on the initial dividing circular surface, and adjusting the dividing line path to form a secondary division; The step of obtaining the position of the concave portion of the surface of the wafer to be inspected on the initial dividing circular surface and adjusting the dividing line path to form a secondary division specifically includes: If the concave part is within the standard surface, it is recorded as the concave standard surface; Get the midpoints of two adjacent sides of the concave standard surface; Draw perpendicular lines from the two midpoints to the opposite sides, and the intersection of the two perpendicular lines divides the concave standard surface into four concave detection areas, which are respectively recorded as the first concave detection area, the second concave detection area, the third concave detection area, and the fourth concave detection area; Obtaining the area of ​​the concave portion and the areas of four concave detection regions; Obtaining the area occupied by the concave portion in the four concave detection regions respectively, and calculating the ratio of the area to the area of ​​the concave detection region, and comparing them; Selecting the largest area ratio among the four area ratios, and obtaining the concave detection area corresponding to the area ratio, which is recorded as the concave adjustment area; Obtain a horizontal dividing line and a vertical dividing line where the concave adjustment area is located; Use the concave adjustment area vertex strategy to obtain the row area and vertical area where the concave area is located; Get the lengths of the two horizontal sides of the row area; Get the number of all standard faces in the row area; Calculate the shortest horizontal side length minus the horizontal side length of the concave adjustment area, and record it as the pre-adjusted horizontal side length; Calculate the required horizontal length of all standard surfaces in the row area, and record it as the total required horizontal length; If the pre-adjusted horizontal side length is greater than or equal to the total required horizontal side length, the vertical dividing line of the standard surface in the row area is moved away from the concave adjustment area by a distance equal to the horizontal side length of the concave adjustment area; If the pre-adjusted horizontal side length is less than the total required horizontal side length, obtain the two vertical side lengths of the vertical area; Get the number of all standard faces in the vertical area; The length of the shortest vertical side minus the length of the vertical side of the concave adjustment area is recorded as the pre-adjusted vertical side length; Calculate the required vertical side lengths of all standard surfaces in the vertical area, and record it as the total required vertical side length; If the pre-adjusted vertical side length is greater than or equal to the total required vertical side length, the horizontal dividing line of the standard surface in the vertical area is moved away from the concave adjustment area by the length of the vertical side of the concave adjustment area; If the pre-adjusted vertical side length is less than the total required vertical side length, no adjustment will be made; If the concave part is between the two standard surfaces and located on the vertical dividing line; Calculate the proportion of the concave part in the two standard surfaces respectively and compare them, and obtain the standard surface with the smaller proportion, which is recorded as the concave standard surface; Use the concave adjustment area vertex strategy to obtain the row area where the concave area is located; Get the lengths of the two horizontal sides of the row area; Get the number of all standard surfaces in the horizontal area; The shortest horizontal side length - the horizontal side length of the concave adjustment area is recorded as the pre-adjusted horizontal side length; Calculate the required horizontal length of all standard surfaces in the row area, and record it as the total required horizontal length; If the pre-adjusted horizontal side length is greater than or equal to the total required horizontal side length, the vertical dividing line of the standard surface in the horizontal area is moved away from the concave adjustment area by the length of the horizontal side of the concave adjustment area; If the pre-adjusted horizontal length is less than the total required horizontal length, no adjustment will be made; If the concave part is between the two standard surfaces and located on the horizontal dividing line; Calculate the proportion of the concave part in the two standard surfaces respectively and compare them, and obtain the standard surface with the smaller proportion, which is recorded as the concave standard surface; Use the concave adjustment area vertex strategy to obtain the vertical area where the concave area is located; Get the lengths of the two vertical sides of the vertical area; Get the number of all standard faces in the vertical area; The length of the shortest vertical side minus the vertical side length of the concave adjustment area is recorded as the pre-adjusted vertical side length; Calculate the required vertical side lengths of all standard surfaces in the vertical area, and record it as the total required vertical side length; If the pre-adjusted vertical side length is greater than or equal to the total required vertical side length, the horizontal dividing line of the standard surface in the vertical area is moved away from the concave adjustment area by the length of the vertical side of the concave adjustment area; If the pre-adjusted vertical side length is less than the total required vertical side length, no adjustment will be made; The method of using the concave adjustment area vertex strategy to obtain the row area and column area where the concave area is located specifically includes: Obtain a horizontal side length and a vertical side length of the concave adjustment area corresponding to the concave standard surface; Get the two vertices of the horizontal length, which are marked as point A and point B respectively; Take point A as the vertex, and draw a line segment along the vertical dividing line where the concave standard surface is located, and the intersection with the circle is recorded as point C; Take point B as the vertex, and draw a line segment along the direction of the vertical dividing line where the concave standard surface is located, and the intersection with the circle is recorded as point D; The area enclosed by ABCD is recorded as the vertical area; Get the two vertices of the vertical side length, which are marked as point E and point F respectively; Take point E as the vertex, and draw a line segment along the direction of the horizontal dividing line where the concave standard surface is located, and the intersection with the circle is recorded as point G; Take point F as the vertex, and draw a line segment along the direction of the horizontal dividing line where the concave standard surface is located, and the intersection with the circle is recorded as point H; The area enclosed by EFGH is recorded as the horizontal area.

Images