Detection method and apparatus for edge defect of silicon wafer, processing method, and product

Abstract

The present application relates to the technical field of semiconductor manufacturing, and provides a detection method and apparatus for an edge defect of a silicon wafer, a processing method, and a product. The detection method for an edge defect of a silicon wafer comprises: after a polishing process, using laser emitted by an optical detection device to detect the edge of a silicon wafer, and determining at least one target region, the target region being identified as an abnormal defect; determining at least one delocalization value of the target region on the basis of the feedback signal intensity of the laser in the target region, wherein each delocalization value is the difference value between the feedback signal intensity of a position point and the feedback signal intensity of an adjacent position point, and the silicon wafer circumferential angle corresponding to the position point differs from the silicon wafer circumferential angle corresponding to the adjacent position point by a preset angle; and when the delocalization value of the target region satisfies a preset condition, determining that the target region is an edge haze defect.

Description

Method and device for detecting edge defects of silicon wafer, processing method and product

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to Chinese Patent Application No. 202410862272.6, filed on June 28, 2024, the contents of which are incorporated herein by reference in its entirety. TECHNICAL FIELD

[0003] The present application relates to the technical field of semiconductor manufacturing, in particular to a method and device for detecting edge defects of a silicon wafer, a processing method and a product. BACKGROUND

[0004] In the field of semiconductor silicon wafer manufacturing, defects in the edge region of a silicon wafer are identified by an automatic optical inspection device. As the precision of integrated circuits improves, the requirements for silicon substrates also increase, and particular attention is paid to the edge region of the silicon wafer, which requires more precise detection and classification of the edge region of the silicon wafer.

[0005] Defects in the edge region of a silicon wafer are generally classified into three categories: chips, cracks, and edge scratches. Through defect detection and classification, the risk of damage to the silicon wafer during subsequent processes can be avoided, and silicon wafers that do not pose a risk of damage can be prevented from being scrapped, thereby reducing costs and losses. SUMMARY

[0006] To solve the above technical problems, the present application provides a method and device for detecting edge defects of a silicon wafer, a processing method and a product, which can avoid product waste and loss.

[0007] To achieve the above purpose, the technical solution adopted by the embodiments of the present application is as follows:

[0008] A method for detecting edge defects of a silicon wafer, comprising:

[0009] After a polishing process, a laser emitted by an optical inspection device is used to detect the edge of the silicon wafer to determine at least one target region, which is identified as an abnormal defect;

[0010] At least one delocalization value of the target region is determined according to the feedback signal intensity of the laser in the target region, each delocalization value being the difference in feedback signal intensity between a position point and an adjacent position point, the position point corresponding to a silicon wafer circumference angle that differs from the silicon wafer circumference angle corresponding to the adjacent position point by a predetermined angle;

[0011] When the delocalization values of the target region satisfy a predetermined condition, the target region is determined to be an edge mist defect.

[0012] In some embodiments, the preset condition comprises:

[0013] The target region comprises N delocalization values, and M of the N delocalization values exceed a set threshold, and M / N is greater than a preset ratio, wherein M is an integer greater than or equal to 0, and N is an integer greater than 0.

[0014] In some embodiments, the set threshold is 800-1500 mW, and the preset ratio is 2 / 7-2 / 5.

[0015] In some embodiments, the determining the at least one delocalization value of the target region according to the feedback signal intensity of the laser on the target region comprises:

[0016] The target region is divided into a plurality of position points according to the preset angle, and the circumferential angles of the silicon wafer corresponding to adjacent position points differ by the preset angle;

[0017] In a clockwise direction or a counterclockwise direction, for each position point, a difference between the feedback signal intensity of the position point and the feedback signal intensity of the next adjacent position point is calculated, and the difference is taken as the delocalization value.

[0018] In some embodiments, the preset angle is 0.1°-0.3°.

[0019] Embodiments of the present application also provide a device for detecting edge defects of a silicon wafer, for implementing the method described above, comprising:

[0020] The detection module is configured to detect the edge of the silicon wafer by using the laser emitted by the optical detection device after the polishing process, and determine at least one target region, wherein the target region is identified as an abnormal defect.

[0021] The calculation module is configured to determine at least one delocalization value of the target region according to the feedback signal intensity of the laser on the target region, wherein each delocalization value is a difference between the feedback signal intensity of a position point and the feedback signal intensity of an adjacent position point, and the circumferential angle of the silicon wafer corresponding to the position point differs by a preset angle from the circumferential angle of the silicon wafer corresponding to the adjacent position point.

[0022] The judgment module is configured to judge that the target region is an edge mist defect when the delocalization value of the target region meets a preset condition.

[0023] Embodiments of the present application also provide a processing method of a silicon wafer, wherein the silicon wafer has an edge mist defect detected by the method described above, and the silicon wafer with the edge mist defect is subjected to edge polishing,

[0024] The edge polishing parameters are: polishing pressure 20-80 N, high-speed polishing time 30-80 s, high-speed polishing speed 300-700 rpm, low-speed polishing time 10-30 s, and low-speed polishing speed 100-300 rpm.

[0025] In some embodiments, the silicon wafer is edge polished so that the delocalization value of the edge region of the polished silicon wafer product is less than a set threshold value, the delocalization value being the difference between the feedback signal intensity of each position point and the adjacent position point when the edge region of the silicon wafer is scanned by laser, the silicon wafer circumferential angle corresponding to the position point and the silicon wafer circumferential angle corresponding to the adjacent position point differ by a preset angle, and the set threshold value is 800-1500 mW, and the preset angle is 0.1°-0.3°.

[0026] The application also provides a silicon wafer product, which is obtained by the processing method described above, and the silicon wafer does not include the edge mist defect detected by the detection method.

[0027] The application has the following beneficial effects:

[0028] The laser emitted by the optical detection equipment is used to identify the edge defects of the silicon wafer, at least one target region is determined, the delocalization value of the target region is determined according to the feedback signal intensity of the laser in the target region, and the target region is determined as an edge mist defect according to the multiple delocalization values of the target region. The edge mist defect can be identified according to the delocalization value of the target region in this embodiment, and after it is determined that the silicon wafer has the edge mist defect, the defect can be removed by edge polishing of the silicon wafer, thereby improving the product quality and yield. BRIEF DESCRIPTION OF DRAWINGS

[0029] FIG. 1 shows a schematic diagram of the edge defects of a silicon wafer;

[0030] FIGS. 2a-2d show schematic diagrams of various defects in the edge region of a silicon wafer;

[0031] FIG. 3 shows a schematic diagram of the parameters of a defect;

[0032] FIG. 4 shows a schematic diagram of a Chip defect in an actual production process;

[0033] FIG. 5 shows a schematic diagram of an edge mist defect of a silicon wafer;

[0034] FIG. 6 shows a flowchart of the detection method of the edge defects of a silicon wafer according to an embodiment of the application;

[0035] FIG. 7 shows a schematic diagram of an edge mist defect;

[0036] FIG. 8 shows an AC signal value diagram of the edge mist defect shown in FIG. 7;

[0037] FIG. 9 shows a schematic diagram of a notch defect;

[0038] FIG. 10 shows a schematic diagram of an AC signal value of the notch defect shown in FIG. 9;

[0039] FIG. 11 shows a schematic diagram of a structure of an edge defect detection device of a silicon wafer according to an embodiment of the present application;

[0040] FIGS. 12 and 13 show schematic diagrams of a silicon wafer with edge haze defects. DETAILED DESCRIPTION

[0041] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions of the embodiments of the present application will be described below in connection with the drawings of the embodiments of the present application. Obviously, the described embodiments are part of, rather than all of, the embodiments of the present application. Based on the described embodiments of the present application, all other embodiments obtained by a person of ordinary skill in the art belong to the scope of protection of the present application.

[0042] In the field of semiconductor silicon wafer manufacturing, an automatic optical inspection device (EBFIS) is used to identify edge defects of a silicon wafer, as shown in FIG. 1. The edge defects of a silicon wafer are generally classified into three categories: Chip, as shown in FIG. 2a; Crack, as shown in FIG. 2b; and Edge-Scratch, as shown in FIG. 2c. Because these defects have a risk of breakage in the later process of a wafer, such as Chip defects that are prone to cause notch or crack of a back-end fragment, wafers with these three types of defects are all discarded in a wafer factory. However, there is a type of edge haze defect, as shown in FIG. 2d, in the production process. This type of defect does not cause a risk of breakage, but cannot be detected and determined using the detection means in the related art, so it can only be determined according to the determination level parameters similar to Chip and other defects, resulting in product waste and loss.

[0043] Specifically, the automatic optical inspection in the related art identifies and classifies defects through defect parameters, as shown in FIG. 3 and Table 1.

[0044] Therefore, the defect parameter classification in the related art cannot distinguish Chip and edge mist defects. FIG. 4 is a schematic diagram of Chip defects in an actual production process, and FIG. 5 is a schematic diagram of wafer edge mist defects. However, it can be seen from the comparison between FIG. 4 and FIG. 5 that Chip defects are mostly a whole piece, while edge mist defects are in a dispersed state, that is, it can be extracted from the image that Chip defects and edge mist defects have a certain difference in dispersion, for example, the transverse delocalization value of Chip defects in FIG. 4 is relatively small, and the transverse delocalization value of edge mist defects in FIG. 5 is relatively large. Therefore, the delocalization value can be added to the detected defect parameters, so as to distinguish Chip defects and edge mist defects according to the transverse delocalization value. The delocalization value is the difference in transverse AC (laser signal intensity) fluctuation of the defects in the image.

[0045] Table 1

[0046] The embodiments of the present application provide a method and device for detecting edge defects of a silicon wafer, a processing method and a product, which can accurately detect edge mist defects, thereby avoiding product waste and loss.

[0047] The embodiments of the present application provide a method for detecting a silicon wafer, as shown in FIG. 6, which comprises the following steps.

[0048] Step 101: After a polishing process, a laser emitted by an optical detection device is used to detect the edge of the silicon wafer to determine at least one target area, which is identified as an abnormal defect.

[0049] In this embodiment, after the silicon wafer is polished and cleaned, the laser emitted by the optical detection device is used to identify defects of the edge of the silicon wafer. After the optical detection device detects the defects, the parameters such as the frame width, length, frame area, defect area and the ratio of the frame area to the defect area (i.e., the filling ratio), and the lengths of the major axis and the minor axis (used to determine the aspect ratio) are used to determine the type of the defects, so that Chip, Crack and Edge-Scratch defects can be identified. In this embodiment, the area identified as a notch defect by the optical detection device is defined as the target area, and the target area needs to be further determined to determine whether it is an edge mist defect or a notch defect.

[0050] Step 102: At least one delocalization value of the target area is determined according to the feedback signal intensity of the laser in the target area, and each delocalization value is the difference in feedback signal intensity between a position point and an adjacent position point, the circumferential angle of the silicon wafer corresponding to the position point and the circumferential angle of the silicon wafer corresponding to the adjacent position point differ by a preset angle.

[0051] After scanning the edge of the silicon wafer by the laser, a feedback signal (AC) of the laser at each position point can be obtained, and a delocalization value of the target region can be determined according to the feedback signal strength of each position point, wherein the feedback signal strength is the reflected laser signal energy strength, and the unit is milliwatt. Specifically, the target region can be divided into a plurality of position points according to the preset angle, and the circumferential angles of the silicon wafer corresponding to adjacent position points differ by the preset angle; in a clockwise direction or a counterclockwise direction, for each position point, the difference between the feedback signal strength of the position point and the feedback signal strength of the next adjacent position point is calculated, and the difference is taken as the delocalization value.

[0052] In some embodiments, the preset angle can be 0.1°-0.3°, such as 0.1°, 0.2° or 0.3°.

[0053] FIG. 8 is an AC signal value diagram of the edge mist defect shown in FIG. 7, wherein the abscissa represents the circumferential angle of the silicon wafer, and the ordinate is the AC signal value; and FIG. 10 is an AC signal value diagram of the notch defect shown in FIG. 9, wherein the abscissa represents the circumferential angle of the silicon wafer (unit: 0.1°), and the ordinate is the AC signal value. The delocalization value is defined as DE, and is calculated every 0.1° (i.e. the preset angle is 0.1°), as shown in FIG. 10, DE1=AC2-AC1, DE2=AC3-AC2.

[0054] As can be seen in FIG. 8, when the preset angle is 0.1°, the AC signal value difference of adjacent two position points can reach 3900 mW (milliwatt), i.e. the delocalization value can reach 3900 mW; in FIG. 10, the AC signal value difference of adjacent two position points does not exceed 500 mW, i.e. the delocalization value does not exceed 500 mW, so the delocalization value can be used to distinguish the notch defect and the edge mist defect.

[0055] In a specific example, the edge region of the silicon wafer is divided into 3600 position points according to the circumferential angle of the silicon wafer, taking 0.1° as the preset angle. The 3600 position points are sequentially ordered as the 1st position point, the 2nd position point, the 3rd position point, …, and the 3600th position point in the clockwise direction. Assuming that the target region includes K position points, K is an integer greater than 1. For the kth position point in the K position points, the AC signal value of the kth position point is subtracted from the AC signal value of the k+1th position point, and the difference is taken as a delocalization value of the target region. In this way, all delocalization values included in the target region are obtained. Alternatively, the 3600 position points are sequentially ordered as the 1st position point, the 2nd position point, the 3rd position point, …, and the 3600th position point in the counterclockwise direction. Assuming that the target region includes K position points, K is an integer greater than 1. For the kth position point in the K position points, the AC signal value of the kth position point is subtracted from the AC signal value of the k+1th position point, and the difference is taken as a delocalization value of the target region. In this way, all delocalization values included in the target region are obtained.

[0056] Step 103: When the delocalization value of the target region satisfies a preset condition, the target region is determined as an edge mist defect.

[0057] In this embodiment, a set threshold value can be set in advance. When one or more delocalization values of the target region are greater than the set threshold value, the target region is determined as an edge mist defect, otherwise the target region is determined as a notch defect. The set threshold value can be set according to the actual production situation.

[0058] When the silicon wafer edge is scanned by the laser, the greater the AC signal value, the greater the possibility of defect, and the longer the AC peak duration, the greater the defect. Generally, when the AC signal value is greater than 1500 mW, it can be considered that there is an edge mist defect. FIG. 8 is an AC signal value schematic diagram of the edge mist defect shown in FIG. 7, wherein the abscissa is the circumferential angle of the silicon wafer, and the ordinate is the AC signal value. FIG. 10 is an AC signal value schematic diagram of the notch defect shown in FIG. 9, wherein the abscissa is the circumferential angle of the silicon wafer, and the ordinate is the AC signal value. As can be seen, when the preset angle is 0.1°, the delocalization value of the edge mist defect can reach more than 3900 mW; and the delocalization value of the notch defect is less than 500 mW, so the set threshold value can be set to 800-1500 mW, such as 800 mW, 900 mW, 1000 mW, 1100 mW, 1200 mW, 1300 mW, 1400 mW, or 1500 mW.

[0059] In a specific example, when the preset angle is 0.1°, the threshold value can be 1000 mW, and when one or more of the delocalization values of the target region is greater than 1000 mW, it can be considered that the target region is an edge mist defect or discontinuous contamination; when all the delocalization values of the target region are not greater than 1000 mW, it can be considered that the target region is an uninterrupted defect, i.e., a Chip.

[0060] Further, in order to improve the judgment accuracy, when all the delocalization values of the target region exceed the set threshold value, it can be considered that the target region is an edge mist defect or discontinuous contamination, otherwise it can be considered that the target region is an uninterrupted defect, i.e., a Chip.

[0061] In some embodiments, if the target region includes N delocalization values, M delocalization values exceed the set threshold value, and M / N is greater than a preset ratio, it is determined that the target region is an edge mist defect, otherwise it is determined that the target region is a notch defect, wherein M is an integer greater than or equal to 0, and N is an integer greater than 0. Specifically, the preset ratio can be 2 / 7-2 / 5. In a specific example, the range of the target region is 1°, and the preset angle is 0.1°. The target region includes 10 position points, corresponding to 10 AC signal values, and 9 delocalization values can be obtained. If more than 1 / 3 of the delocalization values are greater than the set threshold value, it is determined that the target region is an edge mist defect, otherwise it is determined that the target region is a notch defect. Because if only one delocalization value of the target region exceeds the set threshold value, there is a risk of missing detection because the delocalization value is at the edge of the defect, and if more than 1 / 3 of the delocalization values are greater than the set threshold value, it indicates that the defect is not coherent, which is an edge mist defect.

[0062] In the embodiment, the laser emitted by the optical detection device is used to identify the edge of the silicon wafer to determine at least one target region, the feedback signal intensity of the laser in the target region is used to determine the delocalization value of the target region, and the target region is determined to be an edge mist defect according to the multiple delocalization values of the target region. The embodiment can identify the edge mist defect according to the delocalization value of the target region, and after determining that the silicon wafer is an edge mist defect, the edge of the silicon wafer can be polished to improve product quality and yield.

[0063] The embodiment of the present application also provides a silicon wafer processing device, as shown in FIG. 11, which comprises:

[0064] The detection module 21 is configured to use the laser emitted by the optical detection device to detect the edge of the silicon wafer after the polishing process to determine at least one target region, and the target region is identified as an abnormal defect.

[0065] In this embodiment, after the silicon wafer is polished and cleaned, the edge of the silicon wafer is identified by a laser emitted by an optical detection device. After the optical detection device detects a defect, the type of the defect is determined according to parameters such as the frame width, length, frame area, defect area and the ratio of the frame area to the defect area (i.e. the filling ratio), and the length of the major axis and the minor axis (for determining the aspect ratio). Chip, Crack and Edge-Scratch defects can be identified. However, these parameters cannot distinguish Chip defects from edge mist defects, and the edge mist defects will also be determined as Chip defects. In this embodiment, the area identified as a notch defect by the optical detection device is set as a target area, and the target area needs to be further determined to determine whether it is an edge mist defect or a notch defect.

[0066] The computing module 22 is configured to determine at least one delocalization value of the target area according to the feedback signal intensity of the laser on the target area. Each delocalization value is the difference between the feedback signal intensity of a position point and that of an adjacent position point, and the circumferential angle of the silicon wafer corresponding to the position point differs from that of the adjacent position point by a preset angle.

[0067] After the edge of the silicon wafer is scanned by the laser, the feedback signal (AC) of each position point to the laser can be obtained, and the delocalization value of the target area can be determined according to the feedback signal intensity of each position point. In some embodiments, the computing module 22 is specifically configured to divide the target area into a plurality of position points according to the preset angle, and the circumferential angle of the silicon wafer corresponding to adjacent position points differs by the preset angle. In a clockwise direction or a counterclockwise direction, for each position point, the difference between the feedback signal intensity of the position point and that of the next adjacent position point is calculated, and the difference is taken as the delocalization value.

[0068] In some embodiments, the preset angle can be 0.1°-0.3°, such as 0.1°, 0.2° or 0.3°.

[0069] FIG. 8 is an AC signal value diagram of the edge mist defect shown in FIG. 7, in which the abscissa represents the circumferential angle of the silicon wafer, and the ordinate is the AC signal value. FIG. 10 is an AC signal value diagram of the notch defect shown in FIG. 9, in which the abscissa represents the circumferential angle of the silicon wafer (unit: 0.1°), and the ordinate is the AC signal value. The delocalization value is defined as DE, and is calculated every 0.1° (i.e. the preset angle is 0.1°). As shown in FIG. 10, DE1 = AC2-AC1, DE2 = AC3-AC2.

[0070] It can be seen that in FIG. 8, when the preset angle is 0.1°, the AC signal value difference between two adjacent position points can reach 3900 mW, i.e., the delocalization value can reach 3900 mW; in FIG. 10, the AC signal value difference between two adjacent position points is not more than 500 mW, i.e., the delocalization value is not more than 500 mW, and thus the delocalization value can be used to distinguish the gap defect and the edge mist defect.

[0071] The processing module 23 is configured to determine that the target region is the edge mist defect when the delocalization value of the target region satisfies a preset condition.

[0072] In this embodiment, a set threshold value can be preset, and when one or more delocalization values of the target region are greater than the set threshold value, it can be determined that the target region is the edge mist defect, otherwise, it can be determined that the target region is the gap defect. The set threshold value can be set according to actual production conditions.

[0073] When the edge of the silicon wafer is scanned by the laser, the greater the AC signal value, the greater the possibility of defect, and the longer the AC peak duration, the greater the defect. Generally, when the AC signal value is greater than 1500 mW, it can be considered that there is an edge mist defect. FIG. 8 is an AC signal value diagram of the edge mist defect shown in FIG. 7, wherein the abscissa is the circumferential angle of the silicon wafer, and the ordinate is the AC signal value; FIG. 10 is an AC signal value diagram of the gap defect shown in FIG. 9, wherein the abscissa is the circumferential angle of the silicon wafer, and the ordinate is the AC signal value. It can be seen that when the preset angle is 0.1°, the delocalization value of the edge mist defect can reach more than 3900 mW; and the delocalization value of the gap defect is less than 500 mW, and thus the set threshold value can be set to 800-1500 mW, such as 800 mW, 900 mW, 1000 mW, 1100 mW, 1200 mW, 1300 mW, 1400 mW or 1500 mW.

[0074] Further, in order to improve the judgment accuracy, when the delocalization values of the target region all exceed the set threshold value, it can be considered that the target region is the edge mist defect or the discontinuous dirt, otherwise, it can be considered that the target region is the uninterrupted defect, i.e., Chip.

[0075] In some embodiments, the processing module 23 is specifically configured to determine that the target region is an edge mist defect if the target region includes N out-of-range values, M out-of-range values exceed the set threshold, and M / N is greater than a preset ratio, and determine that the target region is a notch defect otherwise, where M is an integer greater than or equal to 0, and N is an integer greater than 0. Specifically, the preset ratio can be 2 / 7-2 / 5. In a specific example, the range of the target region is 1°, the preset angle is 0.1°, the target region includes 10 position points, and 10 AC signal values are obtained. If more than 1 / 3 of the out-of-range values are greater than the set threshold, it is determined that the target region is an edge mist defect, and otherwise, it is determined that the target region is a notch defect. Because if only one out-of-range value of the target region exceeds the set threshold, there is a risk of missing detection because the out-of-range value is at the edge of the defect. If more than 1 / 3 of the out-of-range values are greater than the set threshold, it indicates that the defect is not coherent, and it is an edge mist defect.

[0076] In this embodiment, the laser emitted by the optical detection device is used to identify the edge of the silicon wafer for defect recognition, at least one target region is determined, the out-of-range values of the target region are determined according to the feedback signal intensity of the laser in the target region, and the target region is determined to be an edge mist defect according to the multiple out-of-range values of the target region. This embodiment can identify edge mist defects according to the out-of-range values of the target region. After determining that the silicon wafer is an edge mist defect, the silicon wafer can be edge polished, thereby improving product quality and yield.

[0077] The embodiments of the present application also provide a processing method of a silicon wafer having an edge mist defect detected by the method described above, and the method comprises:

[0078] polishing the edge of the silicon wafer having the edge mist defect,

[0079] The parameters of the edge polishing are as follows: the polishing pressure is 20-80N, the high-speed polishing time is 30-80s, the high-speed polishing speed is 300-700rpm, the low-speed polishing time is 10-30s, and the low-speed polishing speed is 100-300rpm.

[0080] When the silicon wafer is edge polished by using the above parameters, the edge mist defect of the silicon wafer can be effectively eliminated. Specifically, the polishing pressure can be 20N, 30N, 40N, 50N, 60N, 70N or 80N; the high-speed polishing time can be 30s, 40s, 50s, 60s, 70s or 80s; the high-speed polishing speed can be 300rpm, 400rpm, 500rpm, 600rpm or 700rpm; the low-speed polishing time can be 10s, 20s or 30s, and the low-speed polishing speed can be 100rpm, 200rpm or 300rpm.

[0081] In this embodiment, the parameters of edge polishing can be adjusted according to the hardness of the silicon wafer. The greater the hardness of the silicon wafer, the greater the polishing pressure and the longer the polishing time. In addition, the pressure for edge polishing of the silicon wafer should be less than the pressure for polishing of the silicon wafer in the polishing process, so as to avoid the formation of new edge haze defects.

[0082] In this embodiment, the edge of the silicon wafer is polished, so that the delocalization value of the edge region of the polished silicon wafer product is less than a set threshold value. The delocalization value is the difference between the feedback signal intensity of each position point and the adjacent position point when the edge region of the silicon wafer is scanned by laser. The circumferential angle of the silicon wafer corresponding to the position point and the circumferential angle of the silicon wafer corresponding to the adjacent position point differ by a preset angle. The set threshold value is 800-1500 mW, and the preset angle is 0.1°-0.3°. In this way, the silicon wafer can be recovered by edge polishing, avoiding product waste and loss.

[0083] The application also provides a silicon wafer product processed by the above processing method. The silicon wafer does not include edge haze defects detected by the above detection method.

[0084] The silicon wafer does not include edge haze defects detected by the above detection method, i.e. after the silicon wafer with edge haze defects is processed by the above processing method, the silicon wafer no longer has edge haze defects.

[0085] The silicon wafer does not include edge haze defects detected by the above detection method, i.e. the edge region of the silicon wafer product satisfies the following condition: the delocalization value of the edge region of the silicon wafer product is less than a set threshold value. The delocalization value is the difference between the feedback signal intensity of each position point and the adjacent position point when the edge region of the silicon wafer product is scanned by laser. The circumferential angle of the silicon wafer corresponding to the position point and the circumferential angle of the silicon wafer corresponding to the adjacent position point differ by a preset angle.

[0086] In this embodiment, after the gap defect is identified by the optical detection device, the region identified as the gap defect by the optical detection device is set as a target region, and the target region is further judged to determine whether the target region is an edge haze defect or a gap defect. After the silicon wafer is determined to be an edge haze defect, the silicon wafer can be edge polished so that the delocalization value of the edge region of the silicon wafer is less than a set threshold value, and the silicon wafer no longer has edge haze defects, which can meet the normal product delivery standard and meet the product needs.

[0087] In some embodiments, the set threshold value can be 800-1500 mW. When the delocalization value of the edge region of the silicon wafer is less than the set threshold value, the edge haze defect of the silicon wafer can be avoided.

[0088] In some embodiments, the preset angle can be 0.1°-0.3°, such as 0.1°, 0.2° or 0.3°. In a specific example, taking the preset angle of 0.1° as an example, according to the circumferential angle of the silicon wafer, the edge region of the silicon wafer can be divided into 3600 position points. In a clockwise direction, the 3600 position points are sequentially sorted as the first position point, the second position point, the third position point, …, and the 3600th position point. Assuming that the target region includes K position points, K is an integer greater than 1. For the kth position point in the K position points, the AC signal value of the kth position point and the AC signal value of the k+1th position point are subtracted, and the difference value is taken as a delocalization value of the target region. In this way, all delocalization values included in the target region are obtained. Alternatively, in a counterclockwise direction, the 3600 position points are sequentially sorted as the first position point, the second position point, the third position point, …, and the 3600th position point. Assuming that the target region includes K position points, K is an integer greater than 1. For the kth position point in the K position points, the AC signal value of the kth position point and the AC signal value of the k+1th position point are subtracted, and the difference value is taken as a delocalization value of the target region. In this way, all delocalization values included in the target region are obtained.

[0089] In a specific example, before the silicon wafer is judged by using the delocalization value, the silicon wafer shown in FIGS. 12 and 13 is judged as a Chip defect, and will be directly disposed of. After the silicon wafer is judged by using the delocalization value in the technical solution of the present embodiment, the target region of the silicon wafer shown in FIG. 12 has a total of 8 delocalization values, and 6 delocalization values exceed 1000 mW. Therefore, it is judged that the silicon wafer is an edge fog defect. After the edge polishing of the silicon wafer, it does not meet the Chip judgment requirement, and is judged as OK by the equipment. In this way, the yield of the silicon wafer is improved, and there is no risk of fragments in the product backend.

[0090] It should be noted that each of the embodiments in the present specification is described in a progressive manner, and the same and similar parts between each embodiment can be referred to each other. Each embodiment focuses on the difference from other embodiments. In particular, for the embodiment, since it is basically similar to the product embodiment, it is described more simply, and the relevant part can be referred to the part of the product embodiment.

[0091] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The materials, methods, and examples provided herein are illustrative only and are not intended to be limiting. Unless otherwise defined, technical terms and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. Unless specifically set forth herein, the terms "a," "an" and "the" are not limited and do not warrant that a single reference will comprise more than one item. Singular terms do not exclude the plural. "Consisting essentially of can be construed as open terms, meaning including additional steps, ingredients, components or elements unless otherwise defined herein. "Connected" or "coupled" or terms similar thereto are not limited to physical or mechanical connections or links, but also include electrical connections or links, whether direct or indirect. "Over," "under," "between," and like terms are used only to indicate relative relationships between components, and such relationships can change when the absolute positions of the components change.

[0092] It is understood that when an element such as a layer, film, region, or substrate is referred to as being "on" or "under" another element, it can be directly on or under the other element, or intervening elements can also be present.

[0093] In the description of the above embodiments, the specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0094] The above description is only specific embodiments of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for detecting edge defects of a silicon wafer, comprising: detecting an edge of the silicon wafer by using a laser emitted by an optical detection device after a polishing process to determine at least one target region, the target region being identified as an abnormal defect; determining at least one delocalization value of the target region according to feedback signal intensity of the laser at the target region, each of the delocalization values being a difference between feedback signal intensity of a position point and that of an adjacent position point, the position point corresponding to a silicon wafer circumference angle that differs from a silicon wafer circumference angle corresponding to the adjacent position point by a preset angle; when the delocalization values of the target region satisfy a preset condition, judging that the target region is an edge mist defect.

2. The method of claim 1, wherein, The preset condition comprises: of N delocalization values included in the target region, M delocalization values exceed a set threshold value, and M / N is greater than a preset ratio, wherein M is an integer greater than or equal to 0, and N is an integer greater than 0.

3. The method of claim 2, wherein, The set threshold value is 800-1500 mW, and the preset ratio is 2 / 7-2 / 5.

4. The method of claim 1, wherein, The determining at least one delocalization value of the target region according to feedback signal intensity of the laser at the target region comprises: dividing the target region into a plurality of position points according to the preset angle, adjacent position points corresponding to silicon wafer circumference angles that differ by the preset angle; in a clockwise direction or a counterclockwise direction, for each position point, calculating a difference between the feedback signal intensity of the position point and that of the next adjacent position point, and taking the difference as the delocalization value.

5. The method of claim 4, wherein, The preset angle is 0.1°-0.3°. 6.A device for detecting edge defects of a silicon wafer, configured to implement the method of any one of claims 1-5, the device for detecting edge defects of a silicon wafer comprising: a detection module configured to detect an edge of the silicon wafer by using a laser emitted by an optical detection device after a polishing process to determine at least one target region, the target region being identified as an abnormal defect; a calculation module configured to determine at least one delocalization value of the target region according to feedback signal intensity of the laser at the target region, each of the delocalization values being a difference between feedback signal intensity of a position point and that of an adjacent position point, the position point corresponding to a silicon wafer circumference angle that differs from a silicon wafer circumference angle corresponding to the adjacent position point by a preset angle; a judgment module configured to, when the delocalization values of the target region satisfy a preset condition, judge that the target region is an edge mist defect. 7.A method for processing a silicon wafer having an edge mist defect detected by the method of any one of claims 1-5, the method comprising edge polishing the silicon wafer having the edge mist defect detected, parameters of the edge polishing comprising: a polishing pressure of 20-80 N, a high-speed polishing time of 30-80 s, a high-speed polishing rotation speed of 300-700 rpm, a low-speed polishing time of 10-30 s, and a low-speed polishing rotation speed of 100-300 rpm.

8. The method of claim 7, wherein, The silicon wafer is edge polished so that the delocalization value of the edge region of the polished silicon wafer product is less than a set threshold value, the delocalization value being the difference between the feedback signal intensity of each position point and the adjacent position point when the edge region of the silicon wafer is scanned by laser, the silicon wafer circumferential angle corresponding to the position point and the silicon wafer circumferential angle corresponding to the adjacent position point differ by a preset angle, the set threshold value being 800-1500 mW, and the preset angle being 0.1°-0.3°. 9.A silicon wafer product treated by the treatment method of claim 7 or 8, wherein the silicon wafer does not include the edge haze defects detected by the method of any one of claims 1-5.

Images