Method, device, equipment and medium for identifying line defects of wafers

Abstract

The present disclosure provides a method, device, equipment and medium for identifying line defects of a wafer; the method comprises: collecting stress fluctuation data of each sampling point on a pre-processed wafer to be tested; selecting candidate sampling points from all sampling points according to the stress fluctuation data of each sampling point; constructing a line-shaped connected region based on the candidate sampling points; and determining whether the line-shaped connected region is a line defect according to the stress fluctuation data of all candidate sampling points in the line-shaped connected region.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of semiconductor manufacturing, and particularly relates to a method, device, equipment and medium for identifying line defects of a wafer. BACKGROUND

[0002] The manufacturing process of a silicon wafer includes processes such as crystal pulling, cutting, grinding, etching, polishing, etc. In the process of pulling a single crystal silicon rod by the Czochralski (Cz) method, the arrangement of silicon atoms will appear dislocations due to slippage during the arrangement, and then form line defects (also known as primary line defects). If a wafer with line defects is used to manufacture a semiconductor device, it will cause an increase in the leakage current of the semiconductor device, which will have a very great impact on the quality of the semiconductor device, and even cause the semiconductor device to fail.

[0003] Therefore, it is very important to detect line defects of a crystal rod. SUMMARY

[0004] The present disclosure provides a method, device, equipment and medium for identifying line defects of a wafer, which can accurately identify primary line defects formed in the process of pulling a crystal.

[0005] The technical solution of the present disclosure is implemented as follows:

[0006] In a first aspect, the present disclosure provides a method for identifying line defects of a wafer, the method comprising:

[0007] collecting stress fluctuation data of each sampling point on a pre-processed wafer to be tested;

[0008] selecting candidate sampling points from all sampling points according to the stress fluctuation data of each sampling point;

[0009] constructing a line-shaped connected region based on the candidate sampling points;

[0010] determining whether the line-shaped connected region is a line defect according to the stress fluctuation data of all candidate sampling points in the line-shaped connected region.

[0011] In a second aspect, the present disclosure provides a device for identifying line defects of a wafer, the device comprising a collecting unit, a selecting unit, a constructing unit and a determining unit, wherein:

[0012] The collecting unit is configured to collect stress fluctuation data of each sampling point on a pre-processed wafer to be tested;

[0013] The selecting unit is configured to select candidate sampling points from all sampling points according to the stress fluctuation data of each sampling point;

[0014] The constructing unit is configured to construct a linear connected region based on the candidate sampling points.

[0015] The determining unit is configured to determine whether the linear connected region is a line defect according to the stress fluctuation data of all candidate sampling points in the linear connected region.

[0016] In a third aspect, the present disclosure provides a computing device, comprising a processor and a memory; the processor is configured to execute instructions stored in the memory to implement the method for identifying line defects of a wafer according to the first aspect.

[0017] In a fourth aspect, the present disclosure provides a computer-readable storage medium, which stores at least one instruction for being executed by a processor to implement the method for identifying line defects of a wafer according to the first aspect.

[0018] The present disclosure provides a method, device, equipment and medium for identifying line defects of a wafer; candidate sampling points are selected according to stress fluctuation data of sampling points, and after linear connected regions are constructed from the candidate sampling points, it is determined whether it is a line defect according to average stress fluctuation data of the linear connected regions, which can accurately identify original line defects formed in the wafer pulling process. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 A flowchart of a method for identifying line defects of a wafer is provided in the present disclosure.

[0020] Figure 2 (A) is a line defect direction schematic diagram of a notch with a crystal direction of

[110] provided in the present disclosure.

[0021] Figure 2 (B) is a line defect direction schematic diagram of a notch with a crystal direction of

[100] provided in the present disclosure.

[0022] Figure 3 A stress fluctuation curve schematic diagram is provided in the present disclosure.

[0023] Figure 4 (A) is a SIRD detection data schematic diagram of data collection points on a wafer provided in the present disclosure.

[0024] Figure 4 (B) is a SIRD data schematic diagram of sampling points on a wafer provided in the present disclosure.

[0025] Figure 5 A preprocessing flowchart is provided in the present disclosure.

[0026] Figure 6A schematic diagram of fluctuation curves of multiple sampling points provided by the present disclosure.

[0027] Figure 7 A schematic diagram of the average fluctuation curve provided by the present disclosure.

[0028] Figure 8 (A) Schematic diagram of the peak range of the highest peak provided by the present disclosure.

[0029] Figure 8 (B) Schematic diagram of the FWHM range provided for this disclosure.

[0030] Figure 9 Schematic diagram of the identified line defects provided by the present disclosure.

[0031] Figure 10 A schematic diagram of the composition of a device for identifying linear defects in a wafer provided by the present disclosure.

[0032] Figure 11 A schematic diagram of the structure of a computing device provided by the present disclosure. DETAILED DESCRIPTION

[0033] The technical solutions in the present disclosure will be described clearly and completely below with reference to the accompanying drawings in the present disclosure.

[0034] In this disclosure, linear defects caused by dislocations can cause local stress field fluctuations in the silicon wafer. The amplitude of these fluctuations is related to the defect size, and the observable size of the defect is determined by the thickness and depth of the stacking fault atomic plane. Generally speaking, the small size of linear defects results in weaker stress field fluctuations, which are easily masked by other noise on the wafer surface. For this reason, linear defects are typically detected after the polishing process to reduce missed detections and improve detection accuracy.

[0035] In order to accurately detect line defects as early as possible, for example, accurately detect line defects on wafers after dicing but before grinding, the present disclosure provides a solution for identifying line defects on wafers. Figure 1 , which shows a method flow for identifying linear defects of a wafer provided by the present disclosure, the method comprising steps S101 to S104.

[0036] In step S101 , stress fluctuation data of each sampling point on the pre-processed wafer to be tested is collected.

[0037] In this disclosure, the wafer to be tested can be obtained by wire-slicing a pulled single-crystal silicon crystal, or by grinding or polishing, which is not described in detail in this disclosure. These wafers obtained by wire-slicing or grinding or polishing require pre-processing before implementing the technical solution of this disclosure.

[0038] In some examples, the pre-processing process can include: providing a wafer; rotating the wafer according to a notch direction of the wafer so that a crystal orientation

[110] of the wafer is horizontal or vertical; and performing edge exclusion (EE) processing on the rotated wafer according to a set EE value to obtain a wafer to be tested.

[0039] In the above examples, specifically, the wafer can be obtained by slicing a single crystal silicon wafer obtained by pulling, or can be obtained after a grinding or polishing process. The crystal orientation of the linear defect direction formed based on dislocations is {110}, and the crystal orientation of the notch of the wafer is usually

[110] or

[100] . As shown in FIGS. 2(A) and 2(B), respectively, on the wafer surface, Figure 2 (A) and Figure 2 (B), the possible direction of the linear defect is shown by the dashed line in the figure, which is a horizontal or vertical direction. In Figure 2 (A), the crystal orientation of the notch is

[110] , and on the wafer surface shown in Figure 2 (A), the possible direction of the linear defect is shown by the dashed line in the figure, which is a horizontal or vertical direction. In Figure 2 (B), the crystal orientation of the notch is

[100] , and on the wafer surface shown in Figure 2 (B), the possible direction of the linear defect is shown by the dashed line in the figure, which is a direction at 45 degrees to the horizontal or vertical direction. In the linear defect detection process, the horizontal or vertical direction is easier to detect than the oblique direction, and therefore, the wafer to be tested needs to keep

[110] as the horizontal or vertical direction when performing stress fluctuation data detection. In the specific implementation process, when the crystal orientation of the notch is

[110] , the wafer is not rotated, and when the crystal orientation of the notch is

[100] , the wafer is rotated by 45 degrees, so as to ensure that

[110] is horizontal or vertical. Since there are usually defects on the edge of the wafer that cause abnormal stress fluctuation data detection, and usually the wafer will be subjected to edge exclusion EE before detection. In some examples, the EE value is usually 3 mm or 5 mm. After the above pre-processing, the wafer to be tested that can implement the technical solutions of the present disclosure is obtained.

[0040] In the present disclosure, the stress fluctuation data of the wafer surface can represent the fluctuation degree of the local stress field of the wafer, and can be detected by the scanning infrared depolarization (SIRD) technology. The detection process of the SIRD technology is not described herein. As shown in FIG. 3, the stress fluctuation data of the wafer surface can be obtained by the SIRD technology. Figure 3As shown, in the present disclosure, the stress fluctuation data can be shown as a stress fluctuation curve, the abscissa of which represents the angle of a circle, i.e. from 0-360 degrees, and the ordinate represents the signal intensity, in units of Du. The fluctuation curve has multiple peaks, and the peak value of each peak and the half peak width (FWHM, Full Width at Half Maximum) of the highest peak can represent the stress fluctuation intensity at the point.

[0041] In step S102, candidate sampling points are selected from all sampling points according to the stress fluctuation data of each sampling point.

[0042] In the present disclosure, since the line defect can cause the fluctuation of the local stress field of the wafer, the fluctuation can be determined by the stress fluctuation data, i.e. the stress fluctuation curve, and when the stress fluctuation data at a sampling point can determine that the stress field fluctuation occurs at the sampling point, it indicates that the sampling point can be a sampling point in the line defect. In the present disclosure, these sampling points are used as candidate sampling points for subsequent detection and identification of line defects.

[0043] Specifically, when the stress fluctuation data is a stress fluctuation curve, the fluctuation degree of the stress field can be represented by the peak value of the highest peak and the FWHM of the stress fluctuation curve.

[0044] In step S103, a line-shaped connected region is constructed based on the candidate sampling points.

[0045] In the present disclosure, the line defect is usually composed of multiple points, and in combination with the foregoing preprocessing process, the line defect will appear as a line in the horizontal direction or the vertical direction. Based on this, the coordinates of the candidate sampling points can be used to construct a line-shaped connected region, that is, the candidate sampling points in the connected region can be considered as candidate sampling points connected in a straight line.

[0046] In some examples, the constructing of the line-shaped connected region based on the candidate sampling points comprises:

[0047] searching for candidate sampling point groups capable of forming a straight line in the horizontal or vertical direction from all candidate sampling points by using a depth-first search algorithm;

[0048] forming corresponding line-shaped connected regions for each candidate sampling point group.

[0049] Specifically, first, multiple candidate sampling points capable of forming a horizontal straight line in the horizontal direction are searched for in the candidate sampling points by using a depth-first search (DFS) algorithm, and each horizontal straight line corresponds to a group of candidate sampling points. Then, each group of candidate sampling points forms a horizontal line-shaped connected region.

[0050] Similar to the above, a depth-first search algorithm can also be used to search for multiple candidate sampling points that can form a vertical line in the vertical direction among the candidate sampling points. Each vertical line corresponds to a group of candidate sampling points. Then, each group of candidate sampling points is formed into a vertical line-shaped connected area.

[0051] In the specific implementation of the DFS algorithm, the interval threshold is set to 3, meaning that the maximum length of a continuous line interruption due to noise is 3. The linearity threshold is set to 2, meaning that the threshold for deviation from linearity caused by noise is 2. The straight line length threshold is set to 5, meaning that at least five consecutive candidate sampling points in the horizontal or vertical direction must form a straight line.

[0052] In step S104 , it is determined whether the linear connected region is a line defect based on the stress fluctuation data of all candidate sampling points in the linear connected region.

[0053] In this disclosure, it is still necessary to accurately determine whether linear connected regions are line defects. To this end, for each linear connected region, average stress fluctuation data is used to characterize the overall stress field fluctuations of that connected region. Once the overall stress field fluctuations for each linear connected region are obtained, it is possible to further determine whether each connected region is a line defect.

[0054] In some examples, determining whether the linear connected region is a line defect based on stress fluctuation data of all candidate sampling points in the linear connected region includes:

[0055] Obtaining average stress fluctuation data of the linear connected region based on the stress fluctuation data of all candidate sampling points in the linear connected region;

[0056] Whether the linear connected region is a line defect is determined based on the average stress fluctuation data of the linear connected region.

[0057] Specifically, the average stress fluctuation data can characterize the stress field fluctuation of the entire connected region. In some examples, when the stress fluctuation data is as follows: Figure 3 In the case of the stress fluctuation curve shown, for each linear connected area, the stress fluctuation curves of all candidate sampling points in the connected area can be aligned, and then the signal values ​​corresponding to all the aligned stress fluctuation curves for each horizontal coordinate value are averaged to obtain the average stress fluctuation curve of the linear connected area, that is, the average stress fluctuation data of the linear connected area.

[0058] It should be noted that, for each connected region, in addition to the aforementioned exemplary average stress fluctuation curve obtained by averaging the signal values, the average stress fluctuation curve can also be obtained by other means based on the stress fluctuation curves of all candidate sampling points in the connected region. For example, the stress fluctuation curves of all candidate sampling points can be fitted by data fitting to obtain a stress field fluctuation curve that can represent the overall stress field fluctuation of the connected region. The specific data fitting method can include but is not limited to polynomial fitting, linear fitting, nonlinear fitting, spline function fitting, etc., and the present disclosure does not repeat it. In addition, all stress fluctuation curves of candidate sampling points in each connected region can also be input into a trained machine learning model, and the machine learning model can output an average stress fluctuation curve that can represent the overall stress field fluctuation of the connected region. For example, a random forest regression model, a long short-term memory (LSTM) model, etc.

[0059] In the present disclosure, after obtaining the average stress fluctuation data corresponding to each linear connected region, the stress field fluctuation of the overall connected region can be determined in sequence, and whether the connected region is a line defect can be determined according to the stress field fluctuation.

[0060] Specifically, the peak value of the highest peak of the average stress fluctuation data of the linear connected region, i.e. the average waveform curve, and the FWHM can represent the fluctuation degree of the stress field of the overall connected region.

[0061] For the above technical solution, the candidate sampling points are selected according to the sampling point stress fluctuation data, and after the linear connected region is constructed from the candidate sampling points, whether it is a line defect is determined according to the average stress fluctuation data of the linear connected region, which can accurately identify the original line defects formed in the crystal pulling process.

[0062] For the above technical solution, the candidate sampling points are selected according to the sampling point stress fluctuation data, and after the linear connected region is constructed from the candidate sampling points, whether it is a line defect is determined according to the average stress fluctuation data of the linear connected region, which can accurately identify the original line defects formed in the crystal pulling process. Figure 1 As shown in the technical solution, in some possible implementation manners, the stress fluctuation data of each sampling point on the pre-processed wafer under test is collected, including:

[0063] Selecting data collection points on the surface of the wafer under test in polar coordinates;

[0064] Measuring the stress fluctuation data of each data collection point by SIRD;

[0065] Converting each data collection point into a rectangular coordinate system;

[0066] Interpolating the surface of the wafer under test according to the stress fluctuation data of the data collection points in the rectangular coordinate system and the set rectangular size to obtain interpolation data points and stress fluctuation data of each interpolation data point; wherein the outer contour of the wafer under test is the inscribed circle of the rectangle;

[0067] The data collection points and the interpolation data points constitute the sampling points, and stress fluctuation data of each sampling point is obtained.

[0068] For the above implementation, specifically, the present disclosure adopts SIRD to detect stress fluctuation data, and the detected data is generally presented in the form of polar coordinates, as shown in Figure 4 (A). Data collection points are selected on the surface of the wafer to be measured, and SIRD is performed on each data collection point to obtain stress fluctuation data at each data collection point. In order to facilitate subsequent data processing, the present disclosure converts polar coordinates to rectangular coordinates, and performs linear interpolation on the data collection points on which stress fluctuation data has been detected in the rectangular coordinate system, thereby obtaining interpolation points in the surface of the wafer to be measured and the stress fluctuation data values corresponding to the interpolation points. The data collection points and interpolation points on the surface to be measured are arranged in a circular manner.

[0069] In addition, based on this circular arrangement, the present disclosure extends it to a rectangular arrangement, and the inscribed circle of the rectangle is the outer contour of the wafer to be measured. In this rectangle, the part outside the wafer to be measured is interpolated, and each interpolation point is assigned a value of 0. Through the above-mentioned rectangular extension and interpolation, the above-mentioned data collection points and interpolation points can be integrated to form a collection of sampling points, as shown in Figure 4 (B). This collection can present related data in the form of a matrix, which is conducive to subsequent data processing.

[0070] For the technical solution shown in Figure 1 , the peak value of the highest peak of the stress fluctuation data, i.e. the waveform curve, and the FWHM can represent the fluctuation degree of the stress field, and can be used as an evaluation index to determine whether it is a line defect. Based on this, the present disclosure can obtain the evaluation index range corresponding to the line defect in advance. Specifically, as shown in Figure 5 , it can include steps S51 to S56.

[0071] S51: Prepare multiple wafers with line defects.

[0072] For each wafer, the following steps S52 to S55 are performed:

[0073] S52: After etching the wafer, observe the range where the line defect exists through a microscope.

[0074] Specifically, the wafer surface can be etched with an HF solution to more clearly expose the line defect, facilitating visual observation through a microscope.

[0075] S53: Detect the stress fluctuation curve of each sampling point in the range through SIRD.

[0076] Specifically, the stress fluctuation data obtained by the SIRD method is usually presented in the form of polar coordinates. Similar to the foregoing embodiments, the polar coordinates can be converted to rectangular coordinates and extended to a rectangle, and linear interpolation can be performed within the extended rectangular range. For sampling points within the range of the line defect, the line defect can be rotated to present as a horizontal or vertical direction.

[0077] S54: Obtain the average fluctuation curve of the stress fluctuation curves of all sampling points.

[0078] Specifically, as shown in Figure 6 , it is assumed that there are 15 sampling points within the range of the line defect, and the stress fluctuation curves corresponding to the sampling points are respectively identified as Waveform N in Figure 6 , where N represents the serial number of the sampling point.

[0079] After obtaining the stress fluctuation curves of these sampling points, the stress fluctuation curves are aligned along the horizontal axis coordinates, and the average of the signal values corresponding to each horizontal coordinate value is obtained, thereby obtaining the average waveform curve, as shown in Figure 7 .

[0080] S55: Obtain the peak value of the highest peak of the average fluctuation curve and the FWHM.

[0081] For the average waveform curve shown in Figure 7 , the peak value of the highest peak and the FWHM are obtained.

[0082] S56: Summarize the peak value of the highest peak and the FWHM of the average fluctuation curve within the range of the line defect of all wafers to obtain the peak value range and the FWHM range of the highest peak for evaluating the line defect.

[0083] Specifically, the peak value of the highest peak and the FWHM of the average fluctuation curve within the range of the line defect of all wafers are summarized respectively, and the statistical results shown in Figure 8 (A) and Figure 8 (B) are obtained, which are respectively identified as C2 and C3. From Figure 8 (A), it can be seen that the peak value range of the highest peak for evaluating the line defect in the present disclosure is [20, 40], which can also be referred to as the first determination interval in the present disclosure. From Figure 8 (B), it can be seen that the FWHM range for evaluating the line defect in the present disclosure is [2.2, 2.7], which can also be referred to as the second determination interval in the present disclosure.

[0084] Based on the first determination interval and the second determination interval, in some examples, the candidate sampling points are selected from all sampling points according to the stress fluctuation data of each sampling point, including:

[0085] acquire the peak absolute value and the FWHM of the highest peak corresponding to each sampling point according to the stress fluctuation data of each sampling point;

[0086] compare the peak absolute value and the FWHM of the highest peak corresponding to each sampling point with the first determination interval and the second determination interval respectively;

[0087] determine the sampling point as a candidate sampling point when the peak absolute value and the FWHM of the highest peak corresponding to each sampling point are respectively within the first determination interval and the second determination interval.

[0088] For the above example, it should be noted that the stress fluctuation data of each sampling point is presented as a stress fluctuation curve, and the peak absolute value and the FWHM of the highest peak in the stress fluctuation curve can represent the stress fluctuation. Based on the first determination interval and the second determination interval obtained in the foregoing implementation manner, it is determined whether the peak absolute value and the FWHM of the highest peak corresponding to each sampling point are respectively within the first determination interval and the second determination interval. If yes, the sampling point is a candidate sampling point. Otherwise, the sampling point is not determined as a candidate sampling point.

[0089] Based on the first determination interval and the second determination interval, in some examples, the determination of whether the linear connected region is a line defect according to the average stress fluctuation data of the linear connected region includes:

[0090] acquire the peak absolute value and the FWHM of the highest peak corresponding to each linear connected region according to the average stress fluctuation data of each linear connected region;

[0091] compare the peak absolute value and the FWHM of the highest peak corresponding to each linear connected region with the first determination interval and the second determination interval respectively;

[0092] determine the linear connected region as a line defect when the peak absolute value and the FWHM of the highest peak are respectively within the first determination interval and the second determination interval.

[0093] For the above example, it should be noted that the average stress fluctuation data of each linear connected region is presented as an average waveform curve, and the peak absolute value and FWHM of the highest peak in the average stress fluctuation curve can represent the stress fluctuation of each linear connected region. Based on the first and second determination intervals obtained in the foregoing implementation manner, it is determined whether the peak absolute value and FWHM of the highest peak corresponding to each linear connected region are both within the first and second determination intervals, respectively. If both are within, the linear connected region is a line defect, and the length of the linear connected region is the line defect length. Otherwise, the linear connected region is not determined as a line defect.

[0094] By means of the technical solution, taking the wafer obtained after wire cutting as an example, the example of the line defect identified is as shown in FIG. 1. Figure 9 As can be seen from FIG. 1, the technical solution of the present disclosure can accurately identify the line defect. Figure 9

[0095] Based on the same concept of the foregoing technical solution, referring to FIG. 10, which shows a device 100 for identifying a line defect of a wafer provided by the present disclosure, comprising: an acquisition part 1001, a selection part 1002, a construction part 1003 and a determination part 1004; wherein, Figure 10

[0096] The acquisition part 1001 is configured to acquire stress fluctuation data of each sampling point on the preprocessed wafer to be tested;

[0097] The selection part 1002 is configured to select a candidate sampling point from all sampling points according to the stress fluctuation data of each sampling point;

[0098] The construction part 1003 is configured to construct a linear connected region based on the candidate sampling point;

[0099] The determination part 1004 is configured to determine whether the linear connected region is a line defect according to the stress fluctuation data of all candidate sampling points in the linear connected region.

[0100] In some examples, the selection part 1002 is configured to:

[0101] According to the stress fluctuation data of each sampling point, the peak absolute value and the half-peak width FWHM of the highest peak corresponding to each sampling point are obtained;

[0102] The peak absolute value and the half-peak width FWHM of the highest peak corresponding to each sampling point are compared with the first and second determination intervals, respectively;

[0103] ​​When the peak absolute value and the FWHM of the highest peak corresponding to each sampling point are in the first determination interval and the second determination interval respectively, the sampling point is determined as a candidate sampling point.

[0104] In some examples, the determination unit 1004 is configured to:

[0105] According to the stress fluctuation data of all candidate sampling points in the linear connected region, average stress fluctuation data of the linear connected region is obtained;

[0106] According to the average stress fluctuation data of the linear connected region, it is determined whether the linear connected region is a line defect.

[0107] In some examples, the determination unit 1004 is configured to:

[0108] According to the average stress fluctuation data of each linear connected region, the peak absolute value and the FWHM of the highest peak corresponding to each linear connected region are obtained;

[0109] The peak absolute value and the FWHM of the highest peak corresponding to each linear connected region are compared with the first determination interval and the second determination interval respectively;

[0110] The linear connected region in which the peak absolute value and the FWHM of the highest peak are in the first determination interval and the second determination interval respectively is determined as a line defect.

[0111] In some examples, the first determination interval is [20-40] and the second determination interval is [2.2-2.7].

[0112] In some examples, the determination unit 1004 is configured to:

[0113] When the stress fluctuation data is a stress fluctuation curve, the stress fluctuation curves of all candidate sampling points in each linear connected region are aligned according to the abscissa;

[0114] The average stress fluctuation curve of each linear connected region is obtained by averaging the signal values corresponding to each abscissa value in the aligned stress fluctuation curves.

[0115] In some examples, the construction unit 1003 is configured to:

[0116] A group of candidate sampling points capable of forming a straight line in the horizontal or vertical direction is searched from all candidate sampling points by using a depth-first search algorithm;

[0117] Each group of candidate sampling points forms a corresponding linear connected region.

[0118] In some examples, the determining unit 1004 is configured to:

[0119] The stress fluctuation data of all candidate sampling points in each linear connected region is averaged to obtain average stress fluctuation data of each linear connected region.

[0120] In some examples, the collecting unit 1001 is configured to:

[0121] Select data collection points on the surface of the wafer to be measured in polar coordinates;

[0122] Measure the stress fluctuation data of each data collection point by using SIRD;

[0123] Convert each data collection point into a rectangular coordinate system;

[0124] Interpolate the surface of the wafer to be measured according to the set rectangular size and the stress fluctuation data of the data collection points in the rectangular coordinate system to obtain interpolation data points and stress fluctuation data of each interpolation data point; wherein the outer contour of the wafer to be measured is the inscribed circle of the rectangle;

[0125] The data collection points and the interpolation data points form the sampling points, and the stress fluctuation data of each sampling point is obtained.

[0126] In some examples, the collecting unit 1001 is further configured to:

[0127] Provide a wafer;

[0128] Rotate the wafer according to the notch direction of the wafer so that the crystal direction

[110] of the wafer is horizontal or vertical;

[0129] Edge removal processing is performed on the rotated wafer according to the set EE value to obtain the wafer to be measured.

[0130] Please refer to Figure 11 which shows the structural block diagram of a computing device provided by one exemplary embodiment of the present disclosure. In some examples, the computing device 110 can be at least one of a smart phone, a smart watch, a desktop computer, a laptop computer, a virtual reality terminal, an augmented reality terminal, a wireless terminal, and a laptop computer. The computing device 110 has a communication function and can access a wired network or a wireless network. The computing device 110 can generally refer to one of a plurality of terminals, and those skilled in the art can know that the number of terminals can be more or less. In some examples, the computing device 110 can receive data based on the accessed wired network or wireless network. It can be understood that the computing device 110 undertakes the calculation and processing work of the technical solutions of the present disclosure, which are not limited by the present disclosure.

[0131] like Figure 11 As shown, the computing device in the present disclosure may include one or more of the following components: a processor 1110 and a memory 1120 .

[0132] Optionally, processor 1110 utilizes various interfaces and circuits to connect various components within the computing device. It executes instructions, programs, code sets, or instruction sets stored in memory 1120, as well as accesses data stored in memory 1120, to perform various functions of the computing device and process data. Optionally, processor 1110 can be implemented in at least one hardware form: a digital signal processing (DSP), a field-programmable gate array (FPGA), or a programmable logic array (PLA). Processor 1110 can integrate one or a combination of a central processing unit (CPU), a graphics processing unit (GPU), a neural network processing unit (NPU), and a baseband chip. The CPU primarily processes the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing content displayed on the touchscreen display; the NPU is used to implement artificial intelligence (AI) functions; and the baseband chip handles wireless communications. It is understandable that the above-mentioned baseband chip may not be integrated into the processor 1110, but may be implemented by a separate chip.

[0133] Memory 1120 may include random access memory (RAM) or read-only memory (ROM). Optionally, memory 1120 includes non-transitory computer-readable storage medium. Memory 1120 may be used to store instructions, programs, code, code sets, or instruction sets. Memory 1120 may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), and instructions for implementing each of the above method embodiments. The data storage area may store data created based on the use of the computing device.

[0134] In addition, those skilled in the art can understand that the structure of the computing device shown in the above-described drawings does not constitute a limitation on the computing device, and the computing device can include more or fewer components than those shown in the drawings, or combine certain components, or different component arrangements. For example, the computing device also includes a display screen, a camera assembly, a microphone, a speaker, radio frequency circuitry, an input unit, a sensor (such as an acceleration sensor, an angular velocity sensor, a light sensor, etc.), an audio circuit, a WiFi module, a power supply, a Bluetooth module, and the like, which will not be described here.

[0135] The present disclosure also provides a computer-readable storage medium storing at least one instruction for being executed by a processor to implement the method of identifying line defects of a wafer as described in various embodiments above.

[0136] The present disclosure also provides a computer program product including computer instructions stored in a computer-readable storage medium; a processor of a computing device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions to cause the computing device to perform the method of identifying line defects of a wafer as described in various embodiments above.

[0137] Those skilled in the art should be aware that in one or more examples described above, the functions described in the present disclosure can be implemented in hardware, software, firmware, or any combination thereof. When implemented in software, these functions can be stored in a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium. The computer-readable medium includes computer storage medium and communication medium, wherein the communication medium includes any medium that facilitates the transfer of a computer program from one place to another. The storage medium can be any available medium that can be accessed by a general or special purpose computer.

[0138] It should be noted that the technical solutions described in the present disclosure can be combined arbitrarily without conflict.

[0139] The above description is merely a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered within 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 identifying line defects in a wafer, characterized in that: The method comprises: Collect stress fluctuation data of each sampling point on the pre-processed wafer to be tested; Selecting candidate sampling points from all sampling points according to the stress fluctuation data of each sampling point; Constructing a linear connected area based on the candidate sampling points; Determining whether the linear connected region is a line defect based on stress fluctuation data of all candidate sampling points in the linear connected region; Wherein, the preprocessing includes: providing a wafer; Rotating the wafer according to the notch direction of the wafer so that the crystal direction [110] of the wafer is in a horizontal or vertical direction of the wafer surface; Performing edge removal processing on the rotated wafer according to a set edge removal value to obtain the wafer to be tested; The step of selecting a candidate sampling point from all sampling points based on the stress fluctuation data of each sampling point includes: Obtaining the peak absolute value and half-maximum width (FWHM) of the highest peak corresponding to each sampling point according to the stress fluctuation data of each sampling point; Comparing the peak absolute value and the half-maximum width (FWHM) of the highest peak corresponding to each sampling point with the first determination interval and the second determination interval respectively; When the peak absolute value and the half-maximum width (FWHM) of the highest peak corresponding to each sampling point are respectively within the first determination interval and the second determination interval, the sampling point is determined to be a candidate sampling point.

2. The method according to claim 1, characterized in that The determining whether the linear connected region is a line defect based on the stress fluctuation data of all candidate sampling points in the linear connected region includes: Obtaining average stress fluctuation data of the linear connected region based on the stress fluctuation data of all candidate sampling points in the linear connected region; Whether the linear connected region is a line defect is determined based on the average stress fluctuation data of the linear connected region.

3. The method according to claim 2, characterized in that The determining whether the linear connected region is a line defect based on the average stress fluctuation data of the linear connected region includes: Obtain the peak absolute value and half-maximum width (FWHM) of the highest peak corresponding to each linear connected area according to the average stress fluctuation data of each linear connected area; Comparing the peak absolute value and half-maximum width (FWHM) of the highest peak corresponding to each linear connected area with the first determination interval and the second determination interval respectively; A linear connected region in which the absolute peak value and the half-maximum width (FWHM) of the highest peak are both within the first determination interval and the second determination interval is determined as a line defect.

4. The method according to any one of claims 1 to 3, characterized in that The first determination interval is [20-40], and the second determination interval is [2.2-2.7].

5. The method according to claim 2, characterized in that The step of obtaining average stress fluctuation data of the linear connected region based on the stress fluctuation data of all candidate sampling points in the linear connected region comprises: When the stress fluctuation data is a stress fluctuation curve, aligning the stress fluctuation curves of all candidate sampling points in each linear connected area according to the horizontal axis; The signal values ​​corresponding to all aligned stress fluctuation curves at each abscissa value are averaged to obtain the average stress fluctuation curve of each linear connected area.

6. The method according to claim 1, characterized in that The step of constructing a linear connected area based on the candidate sampling points includes: Using a depth-first search algorithm, a candidate sampling point group that can form a straight line in the horizontal or vertical direction is searched from all candidate sampling points; Each candidate sampling point group is formed into a corresponding linear connected area.

7. The method according to claim 1, characterized in that The collecting of the pre-processed stress fluctuation data of each sampling point on the wafer to be tested includes: Selecting data collection points on the surface of the wafer to be measured in polar coordinates; The stress fluctuation data of each data collection point is measured using the scanning infrared depolarization SIRD method; Convert each data collection point into a rectangular coordinate system; Interpolating the surface of the wafer to be measured according to the set rectangle size and the stress fluctuation data of the data collection points in the rectangular coordinate system to obtain interpolated data points and stress fluctuation data of each interpolated data point; wherein the outer contour of the wafer to be measured is the inscribed circle of the rectangle; The data acquisition points and the interpolation data points constitute the sampling points, and the stress fluctuation data of each sampling point is obtained.

8. A device for identifying line defects in a wafer, characterized in that: The device includes: a collection unit, a selection unit, a construction unit and a determination unit; wherein, The collecting unit is configured to collect stress fluctuation data of each sampling point on the pre-processed wafer to be tested; The selection unit is configured to select a candidate sampling point from all sampling points according to the stress fluctuation data of each sampling point; The constructing unit is configured to construct a linear connected area based on the candidate sampling points; The determination unit is configured to determine whether the linear connected region is a line defect based on stress fluctuation data of all candidate sampling points in the linear connected region; Wherein, the preprocessing includes: providing a wafer; Rotating the wafer according to the notch direction of the wafer so that the crystal direction [110] of the wafer is in a horizontal or vertical direction of the wafer surface; Performing edge removal processing on the rotated wafer according to a set edge removal value to obtain the wafer to be tested; The selection unit is configured to Obtaining the peak absolute value and half-maximum width (FWHM) of the highest peak corresponding to each sampling point according to the stress fluctuation data of each sampling point; Comparing the peak absolute value and the half-maximum width (FWHM) of the highest peak corresponding to each sampling point with the first determination interval and the second determination interval respectively; When the peak absolute value and the half-maximum width (FWHM) of the highest peak corresponding to each sampling point are respectively within the first determination interval and the second determination interval, the sampling point is determined to be a candidate sampling point.

9. A computing device, characterized in that The computing device includes a processor and a memory; the processor is configured to execute instructions stored in the memory to implement the method for identifying wafer line defects according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores at least one instruction, and the at least one instruction is configured to be executed by a processor to implement the method for identifying wafer line defects according to any one of claims 1 to 7.

Images