Wafer defect detection method

Abstract

The present application relates to a kind of wafer defect detection methods, comprising the following steps: with first detection resolution scanning wafer to carry out first time defect detection to wafer, obtain the scanning image of whole wafer, and generate the first detection report including the size information and position information of defect;First detection report is filtered to obtain the size and position information of first type defect, define the area in the first distance range of first type defect as associated defect area;The edge area of wafer is set as special attention area, and associated defect area and special attention area are set as secondary scanning area;With second detection resolution, secondary scanning area is scanned, and the position information and size information of second type defect are obtained;Secondary scanning image and the position information and size information of second type defect are superimposed with the scanning image of whole wafer and the size information and position information of defect in first detection report, and the final scanning image is obtained.

Description

Technical Field

[0001] This invention relates to the field of defect detection technology, and in particular to a method for detecting wafer defects. Background Technology

[0002] Ultrasonic defect inspection equipment for semiconductor bonded wafers requires placing an ultrasonic probe above the wafer to be inspected. Since ultrasonic waves cannot be transmitted in air, pure water is used as a coupling medium and is introduced between the ultrasonic probe and the wafer. The ultrasonic probe emits ultrasonic signals, which are driven by a drive shaft to reciprocate across the wafer surface. It collects the feedback signals from the ultrasonic waves at the inspection interface, and identifies and analyzes abnormal signals as detected defects.

[0003] Currently, ultrasonic defect inspection equipment for semiconductor bonding wafers can be divided into three types: manual, semi-automatic, and fully automatic. The core component of the inspection equipment is the ultrasonic inspection unit, which consists of a motion mechanism, an ultrasonic probe, pure water, a wafer fixing device, and a pure water storage device, with pure water serving as the coupling medium.

[0004] During testing, the detection resolution can be set. The detection resolution is determined by the stepping distance of the motion mechanism. When the ultrasonic probe detects at a certain point, it emits an ultrasonic signal and receives ultrasonic signals reflected back from different interfaces of the wafer. The reflected ultrasonic signals are then transmitted to the signal processing system of the ultrasonic testing unit for processing, thereby generating an image of that point. Subsequently, according to the set resolution, it steps to the next point, generating an image for that next point as well. As the ultrasonic probe traverses the designated scanning area in a serpentine reciprocating motion with the motion mechanism, a complete image is formed.

[0005] Once the image is formed, the ultrasonic testing equipment uses a specific algorithm to identify defects in the image and outputs a defect analysis report that includes information such as the location, size, and type of the defect.

[0006] The resolution of the complete image is determined by the set detection resolution. A higher detection resolution, larger step distance, and faster motion speed of the moving mechanism result in less detected data and faster data processing. However, the resulting image resolution is lower, and the resolution of the identified defects will also be lower, leading to coarser information about the location, size, and contour of defects during defect identification. Conversely, a lower detection resolution results in a higher-resolution image, with correspondingly higher resolution of the identified defects. This allows for more detailed defect images and more accurate information about the location and size of defects. However, this process takes tens of times longer.

[0007] Defects in bonded wafers or SOI wafers vary in size, ranging from tens to hundreds of micrometers to tens of micrometers. The size distribution of defects is also highly related to the bonding or SOI process.

[0008] The detection resolution of ultrasonic testing equipment is highly dependent on the size of the defect to be detected. For large defects, a high detection resolution is sufficient, resulting in fast detection speed. Conversely, for small defects, a low detection resolution is required to reduce the probability of missed detections, but this slows down the detection process.

[0009] When wafer defects have a large size range, such as from tens of micrometers to hundreds of micrometers, using a small detection resolution can improve the detection rate of small defects, but the detection speed is slow. Using only a large detection resolution can improve the detection speed, but the resulting image is not clear and the defect information is not rich, which may lead to the missed detection of many small defects. Summary of the Invention

[0010] To address at least some of the problems mentioned above in the prior art, the present invention provides a wafer defect detection method, comprising the following steps:

[0011] The wafer is scanned at a first detection resolution to perform the first defect detection on the wafer, resulting in a scanned image of the entire wafer, and a first detection report containing the size and location information of the defects is generated.

[0012] The size and location information of the first type of defect are obtained by filtering the first inspection report. The area within the first distance range of the first type of defect is defined as the associated defect area, which contains the second type of defect. The size of the second type of defect is smaller than that of the first type of defect.

[0013] The edge region of the wafer is set as the special interest region, and the associated defect region and the special interest region are set as the secondary scan region;

[0014] The secondary scanning area is directionally scanned at a second detection resolution to obtain the secondary scan image and the location and size information of the second type of defect, wherein the second detection resolution is smaller than the first detection resolution.

[0015] The final scan image is obtained by overlaying the secondary scan image, the location and size information of the second type of defect, the scan image of the entire wafer, and the size and location information of the defects in the first inspection report; and

[0016] Further analysis of the final scanned image and the location and size information of all defects yields the defect outline, size, and location, and generates a defect report containing all defects.

[0017] Furthermore, it also includes: after generating the first test report, reading the first test report into the test program.

[0018] Furthermore, it also includes: analyzing the second scan area to generate the optimal scan route.

[0019] Furthermore, based on the optimal scanning route, the secondary scanning area is directionally scanned at a second detection resolution.

[0020] Furthermore, the size of the second type of defect is 10 micrometers to 100 micrometers, while the size of the first type of defect is greater than 100 micrometers.

[0021] Furthermore, the first detection resolution is 50 micrometers to 200 micrometers; and / or

[0022] The first distance is 1 micrometer to 10,000 micrometers; and / or

[0023] The second detection resolution is 5 micrometers to 40 micrometers.

[0024] Furthermore, based on the defect definition identification rules obtained from the first defect detection, the size and location information of the first type of defect are obtained by filtering the size information of the defects in the first detection report according to the identification rules.

[0025] Furthermore, the identification rule is to identify defects with a size greater than 100 micrometers.

[0026] Furthermore, the optimal scan route is generated using Dijkstra's algorithm, A* algorithm, or Bellman-Ford algorithm.

[0027] The present invention has at least the following beneficial effects:

[0028] The wafer defect detection method of the present invention first uses a first detection resolution (high detection resolution) to scan the entire wafer for a fast first defect detection. Then, a slower second scan is performed on the associated defect region and the region of special interest around the first type of defect (large defect) to obtain the size and location information of the second type of defect (small defect). The overall time consumed by the two scans is small, which can speed up the detection speed while maximizing the detection rate of small defects. Attached Figure Description

[0029] To further illustrate the above and other advantages and features of the various embodiments of the present invention, a more specific description of the embodiments of the invention will be presented with reference to the accompanying drawings. It is to be understood that these drawings depict only typical embodiments of the invention and are therefore not intended to limit its scope. In the drawings, identical or corresponding parts will be indicated by identical or similar reference numerals for clarity.

[0030] Figure 1The flowchart of a wafer defect detection method according to an embodiment of the present invention is shown. Detailed Implementation

[0031] It should be noted that the components in the accompanying drawings may be shown exaggerated for illustrative purposes and may not be to scale.

[0032] In this invention, the various embodiments are merely intended to illustrate the solutions of the invention and should not be construed as limiting.

[0033] In this invention, unless otherwise specified, the quantifiers “a” and “one” do not exclude scenarios involving multiple elements.

[0034] It should also be noted that, in the embodiments of the present invention, only a portion of the parts or components may be shown for clarity and simplicity. However, those skilled in the art will understand that, under the teachings of the present invention, the required parts or components can be added as needed for specific scenarios.

[0035] It should also be noted that within the scope of this invention, the terms "same", "equal", and "equal to" do not mean that the two values ​​are absolutely equal, but allow for a certain reasonable error. In other words, the terms also cover "substantially the same", "substantially equal", and "substantially equal to".

[0036] It should also be noted that in the description of this invention, the terms "center," "longitudinal," "lateral," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not explicitly or implicitly suggest that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0037] Furthermore, the embodiments of the present invention describe the process steps in a specific order; however, this is only for the convenience of distinguishing each step, and is not intended to limit the order of the steps. In different embodiments of the present invention, the order of each step can be adjusted according to the process.

[0038] When the ultrasonic probe traverses the designated scanning area in a serpentine motion according to the set detection resolution, a complete image is formed. After the image is formed, the ultrasonic testing equipment uses a specific algorithm to identify defects in the image. A typical algorithm uses different gray values ​​in the image to calculate and filter out areas with gray values ​​higher than a threshold. After applying other specific rules, the selected areas are identified as defects. The pixel areas of the identified defects are then colored according to the set rules, and a defect analysis report including information such as defect location, size, and category is output.

[0039] The resolution of the complete image is determined by the set detection resolution. A higher detection resolution, larger step distance, and faster motion mechanism result in less detected data and faster data processing. Therefore, a higher detection resolution leads to faster detection speed. For example, with a 100µm resolution setting, using four probes to simultaneously inspect a 12-inch wafer takes only a few minutes. However, the resulting image resolution is low, and the resolution of identified defects will also be correspondingly low, resulting in coarse information about the location, size, and contour of defects. Conversely, if the detection resolution is set to 10µm, using the same four probes to inspect a 12-inch wafer will yield a higher resolution image, and the resolution of identified defects will also increase accordingly. During defect identification, a more detailed defect image can be obtained, and the location and size information of defects will be more accurate. However, the time required is tens of times longer than with the higher detection resolution.

[0040] For wafer defects with a large size range, such as from tens to hundreds of micrometers, using a low detection resolution can improve the detection rate of small defects, but the detection speed is slow. Using a high detection resolution can improve the detection speed, but the resulting images are unclear, the defect information is not rich, and the detection of small defects is more likely to be missed.

[0041] To detect small defects, it is necessary to maximize the detection rate of small defects. This requires using a low detection resolution, which results in a slow detection speed. However, if a high detection resolution is used, the resulting image will be unclear, the defect information will be insufficient, and small defects may be missed.

[0042] To address the aforementioned technical problems, this invention provides a wafer defect detection method. First, a high detection resolution (first detection resolution) is used to scan the entire wafer. Then, a secondary scan is performed on the associated defect region and the region of special interest surrounding the large defect. The overall time consumption is relatively low, and the detection speed can be accelerated while maximizing the detection rate of small defects.

[0043] like Figure 1 As shown, a wafer defect detection method includes the following steps:

[0044] Step 1: Scan the wafer at a first detection resolution to perform the first defect detection, obtaining a scanned image of the entire wafer, and generating a first inspection report containing the size and location information of the defects. Store the first inspection report in a storage medium, such as the hard drive, cache, CPU, or GPU cache of an industrial control computer.

[0045] In one embodiment, the first detection resolution can be 50 micrometers to 200 micrometers. No further analysis of the defects is required after this scan.

[0046] Step 2: Read the first test report into the test program.

[0047] Step 3: Filter the first inspection report to obtain the size and location information of the first type of defect. Define the area within the first distance range of the first type of defect as the associated defect area, where the associated defect area contains the second type of defect, and the size of the second type of defect is smaller than the size of the first type of defect.

[0048] Based on the defect identification rules obtained in step 1, the size and location information of the first type of defects are identified. The identification rule is to identify defects with a size greater than 100 micrometers.

[0049] In some embodiments, the size of the second type of defect is 10 micrometers to 100 micrometers, and the size of the first type of defect is greater than 100 micrometers. The first distance can be 1 micrometer to 10,000 micrometers. The first distance can be set according to the actual situation.

[0050] The extent of the associated defect area is determined by the manufacturing process, and it is generally within a few centimeters.

[0051] Step 4: Set the edge region of the wafer as the special interest region, and set the associated defect region and the special interest region as the secondary scan region.

[0052] The distribution of Type II defects caused by wafer bonding or SOI processes often follows a pattern, such as in the edge region of the wafer. Therefore, the edge region of the wafer needs to be designated as a region of special concern.

[0053] Step 5: Analyze the second scan area to generate the optimal scan route.

[0054] In some embodiments, algorithms such as Dijkstra's algorithm, A* algorithm, and Bellman-Ford algorithm can be used to generate the optimal scan route.

[0055] Step 6: Based on the optimal scanning route, perform a directional scan of the secondary scanning area at a second detection resolution to obtain the secondary scan image and the location and size information of the second type of defect. Areas outside the secondary scanning area will not be scanned.

[0056] The second detection resolution can be 5 micrometers to 40 micrometers.

[0057] If a type II defect is present in the secondary scanning area, it will be detected.

[0058] Step 7: Overlay the secondary scan image and the location and size information of the second type of defect with the scan image of the whole wafer and the size and location information of the defect in the first inspection report to obtain the final scan image.

[0059] Compared to the scanned image of the entire wafer obtained from the first scan, the image of the second scanned area in the final scanned image after superposition is clearer.

[0060] Step 8: Further analyze the final scanned image and the location and size information of all defects to obtain the defect outline, size and location, and generate a defect report containing all defects.

[0061] Further analysis yields more refined defect profiles and more accurate measurements and locations. The final defect report includes the profiles, dimensions, and locations of all defects.

[0062] By overlaying the results of two scans, an image can be obtained that simultaneously includes both Type I defects (large defects) and Type II defects (small defects) without affecting the integrity of the overall image. The image will also include richer defect information, such as finer defect contours and grayscale variations. Analyzing the overlaid image can generate a defect color map with even finer contour lines. Furthermore, a complete defect report containing more accurate dimensions and locations of all defects can be obtained simultaneously.

[0063] The time required to inspect a wafer using the method of this invention is the sum of the first scan time and the second scan time. Compared to scanning the entire wafer using a small detection resolution, the scanning time of the above method is greatly reduced, and the detection rate of small defects is high.

[0064] While some embodiments of the present invention have been described in this application, those skilled in the art will understand that these embodiments are merely illustrative. Numerous variations, alternatives, and improvements will arise in those skilled in the art under the teachings of this invention without departing from its scope. The appended claims are intended to define the scope of the invention and thereby cover methods and structures within the scope of the claims themselves and their equivalents.

Claims

1. A method for detecting wafer defects, characterized in that, Includes the following steps: The wafer is scanned at a first detection resolution to perform the first defect detection on the wafer, resulting in a scanned image of the entire wafer, and a first detection report containing the size and location information of the defects is generated. The size and location information of the first type of defect are obtained by filtering the first inspection report. The area within a first distance range of the first type of defect is defined as the associated defect area, which contains the second type of defect. The size of the second type of defect is smaller than that of the first type of defect. Identification rules are defined for the defects obtained from the first defect inspection. Based on the identification rules, the size information of the defects in the first inspection report is filtered to obtain the size and location information of the first type of defect. The first distance is 1 micrometer to 10,000 micrometers. The edge region of the wafer is set as the special interest region, and the associated defect region and the special interest region are set as the secondary scan region; The secondary scanning area is scanned at a second detection resolution to obtain the secondary scanning image and the location and size information of the second type of defect, wherein the second detection resolution is smaller than the first detection resolution; The location and size information of the second-type defect are superimposed on the scan image of the whole wafer and the size and location information of the defect in the first inspection report to obtain the final scan image; as well as Further analysis of the final scanned image and the location and size information of all defects yields the defect outline, size, and location, and generates a defect report containing all defects.

2. The wafer defect detection method according to claim 1, characterized in that, Also includes: After generating the first test report, the first test report is read into the test program.

3. The wafer defect detection method according to claim 1, characterized in that, Also includes: Analyze the second scan area to generate the optimal scan route.

4. The wafer defect detection method according to claim 3, characterized in that, Based on the optimal scanning route, the secondary scanning area is directionally scanned at a second detection resolution.

5. The wafer defect detection method according to claim 1, characterized in that, The size of the second type of defect is 10 micrometers to 100 micrometers, while the size of the first type of defect is greater than 100 micrometers.

6. The wafer defect detection method according to claim 1, characterized in that, The first detection resolution is 50 micrometers to 200 micrometers; and / or The second detection resolution is 5 micrometers to 40 micrometers.

7. The wafer defect detection method according to claim 1, characterized in that, The identification rule is to identify defects with a size greater than 100 micrometers.

8. The wafer defect detection method according to claim 4, characterized in that, Using Dijkstra's algorithm, A The algorithm or Bellman-Ford algorithm generates the optimal scan route.

Images