TW202348988A - Defect inspection device and defect inspection method - Google Patents

Abstract

Provided is a defect inspection device making it possible, with a simple method, to automatically detect optimum optical conditions for imaging a region that includes a defect. A defect inspection device 1 comprises: an optical microscope 10 provided with an imaging unit 14 for capturing an image of a chip 31; a control unit 20 for controlling optical conditions for capturing the image of the chip; and a storage unit 25 for storing the location of a defect in the chip that has been detected in advance. The imaging unit is provided with a detecting means for: capturing, while changing the optical conditions, a first image of a region that includes the defect stored in the storage unit as well as a second image of a region that does not include the defect in the same region as the region that includes the defect; and detecting the optimum optical conditions for imaging the region that includes the defect by comparing a feature amount of the first image and a feature amount of the second image.

Description

Defect inspection device and defect inspection method

The present invention relates to a defect inspection device and a defect inspection method for inspecting defects in a wafer formed on a wafer.

Semiconductor devices are manufactured by repeatedly forming a plurality of semiconductor device circuits (chips) in a matrix on a semiconductor wafer, then singulating them into individual wafers, and packaging the singulated wafers.

Before singulating each wafer, the appearance patterns of each wafer formed on the wafer are photographed one after another, while the inspection image and the reference image are compared in pixel units, and the wafer is detected based on the difference in the brightness values of the compared pixels. internal defects (for example, Patent Document 1). [Prior technical literature] [Patent Document]

[Patent Document 1] Japanese Patent Application Laid-Open No. 2007-155610

[Problem to be solved by the invention]

The defect inspection in the wafer is carried out while taking the image of the wafer with an optical microscope, but it is necessary to set the optical conditions of the optical microscope such that the defects are clearly visible.

When a person manually sets the optical conditions, even if the location of the defect is known, the optical microscope is moved to the area containing the defect, the optical conditions are changed, and the inspection image is visually confirmed to explore the optical conditions under which the defect is clearly visible.

However, since there are multiple optical conditions, it is difficult for humans to explore all the optical conditions. Furthermore, since humans make visual judgments, there is a risk that the detection of defects may be biased by humans.

The present invention was made in view of this point, and its main purpose is to provide a defect inspection device that uses an optical microscope to inspect defects in a wafer formed on a wafer, and can automatically detect in a simple way the area containing the defect by photographing it. Defect inspection device and defect inspection method under optimal optical conditions. [Technical means to solve problems]

The defect inspection device of the present invention inspects defects in a wafer formed on a wafer, and is provided with: an optical microscope equipped with an imaging unit that takes an image of the wafer; and a control unit that controls the optical microscope to take an image of the wafer. the optical conditions of the image; and a memory unit that memorizes the locations of previously detected defects in the wafer; the imaging unit changes the optical conditions with the control unit while capturing the first image of the area containing the defects memorized in the memory unit, and a second image of an area that does not contain a defect within the same area as the area that contains the defect; and the defect detection device is capable of detecting by comparing the feature quantity of the first image with the feature quantity of the second image. Inspection mechanism for optimal optical conditions for photographing areas containing defects. [Effects of the invention]

According to the present invention, it is possible to provide a defect inspection device that uses an optical microscope to inspect defects in a wafer formed on a wafer, and can automatically detect the optimal optical conditions for photographing an area containing a defect in a simple manner, and Defect inspection methods.

Hereinafter, embodiments of the present invention will be described in detail based on the drawings. In addition, the present invention is not limited to the following embodiments. In addition, appropriate changes can be made without departing from the scope of exerting the effects of the present invention. (First Embodiment) FIG. 1 is a diagram schematically showing the structure of a defect inspection device 1 according to the first embodiment of the present invention.

As shown in FIG. 1 , the defect inspection device 1 of this embodiment is used to inspect defects in a wafer formed on a wafer. It includes an optical microscope 10 and a control unit 20 that controls the optical system for capturing an image of the wafer. Conditions; the memory unit 25, which memorizes the locations of previously detected defects in the wafer; and the detection mechanism 40, which detects the optimal optical conditions for photographing the area containing the defects with the optical microscope 10.

The optical microscope 10 includes a stage 11 that mounts the wafer 30 , a microscope body 12 that holds the stage 11 , an illumination unit 13 that irradiates the wafer 30 with illumination light L, and an imaging unit 14 that photographs the wafer. image.

The illumination light L irradiated from the illumination unit 13 is condensed by the condenser lens 15 , is deflected by the half-reflecting mirror 16 along the optical axis of the objective lens 17 , and is irradiated to the surface of the wafer 30 . The reflected light from the wafer 30 enters the imaging element 19 via the objective lens 17 , the half mirror 16 and the coupling lens 18 . The image captured by the imaging element 19 is subjected to image processing by the image processing unit 35 .

In this embodiment, the detection mechanism 40 detects the optimal optical conditions for photographing a region containing defects based on the image information of the wafer captured by the imaging unit 14 . The control unit 20 sets the optical microscope 10 to the optimal optical condition detected by the detection mechanism 40 , photographs the wafer, and detects defects in the wafer 30 .

In addition, in this embodiment, the optical conditions of the optical microscope 10 controlled by the control unit 20 relate to the magnification of the objective lens 17, the type (coaxial, oblique light, transmission) or brightness of the illumination light L, and the differential interference filter (not shown in the figure). ), the filter (not shown) of the illuminating light L, and many other aspects.

FIG. 2 is a flowchart showing a method of detecting optimal optical conditions for photographing a region containing defects using the defect inspection device 1 of this embodiment.

First, as shown in FIG. 3(A) , the wafer 31 formed on the wafer 30 is photographed by the optical microscope 10 in advance, defects in the wafer 31 are detected, and the position of the detected defect in the wafer 31a is registered in the memory unit 25 ( Step S101).

Next, as shown in FIG. 3(B) , the control unit 20 changes the optical conditions of the optical microscope 10 (step S102), and captures an image (first image) of the area (defect area) A including the defect 50 (step S103). ).

Next, as shown in FIG. 3(B) , under the same optical conditions, in the same area as the area A including the defect 50, an image of the area B (good product area) of the wafer 31b that does not include the defect 50 is captured (No. 2 images) (step S104). Here, the defective area A and the good area B are the same areas within the wafer.

Next, the brightness value (feature amount) of the image of defective area A (first image) is compared with the brightness value (feature amount) of the image of defective area B (second image) (step S105).

Next, when the search for optimal optical conditions has not ended (No in step S106), the optical conditions of the optical microscope 10 are changed (step S102), and steps S103 to S105 are repeated. When the search for optimal optical conditions ends (Yes in step S106 ), the optical microscope 10 is set to the optimal optical conditions to detect defects in the wafer 30 .

Table 1 shows an example of the brightness value (A) of the defective area A and the brightness value (A) of the defective area B ( B) and the table of the difference between the brightness values of the two (B-A). In addition, the numerical value of the brightness value represents an arbitrary unit.

[Table 1] Optical conditions 1 2 3 4 Brightness value of defective area (A) 20 10 25 15 Brightness value of good product area (B) 95 100 85 90 Difference in brightness values (BA) 75 90 60 75

Even if the same areas A and B are photographed, if the optical conditions change, the brightness values of the images captured in areas A and B will also change. The higher the definition of the image, the greater the difference between the two. Therefore, the difference (B-A) between the brightness value (A) of the defective area A and the brightness value (B) of the defective area B becomes an evaluation value for obtaining the optimal optical conditions.

In the example shown in Table 1, among the four optical conditions, the difference (B-A) between the brightness values of optical condition 2 is the largest. Therefore, in this case, optical condition 2 can be said to be the optimal optical condition.

In this way, in this embodiment, detection can be performed by comparing the brightness value of the image of the defective area A (the first image) with the brightness value of the image of the defective area B (the second image). Optimum optical conditions for areas containing defects. Thereby, there is a margin for the threshold when detecting defects, so the detection accuracy when detecting defects in the wafer can be improved.

In addition, in Table 1, the optimal optical conditions are obtained from the difference (B-A) between the brightness value (A) of the defective area A and the brightness value (B) of the good product area B. However, the brightness value (A) of the defective area A is different from the brightness value (B) of the good product. The comparison of the brightness value (B) of area B can also be performed as the ratio of the two (B/A) as shown in Table 2.

In the example shown in Table 2, among the four optical conditions, the ratio of the brightness values (B/A) between the two in optical condition 2 is the largest. Therefore, in this case, optical condition 2 can be said to be the optimal optical condition.

[Table 2] Optical conditions 1 2 3 4 Brightness value of defective area (A) 20 10 25 15 Brightness value of good product area (B) 95 100 85 90 Brightness value ratio (B/A) 4.75 10.0 3.4 6.0

(Modification example of the first embodiment) In the above embodiment, the brightness value of the image of the defective area A (the first image) is compared with the brightness value of the image of the defective area B (the second image), and the maximum value for photographing the area containing the defect is detected. Optimal optical conditions can be determined, but the optimal optical conditions can also be detected by adding the difference in brightness values between images of different good product areas to the evaluation value.

As shown in FIG. 4 , in the same area as the area A including the defect 50 shown in FIG. 3(B) , the area of the wafer 31c that is different from the wafer 31b shown in FIG. 3(C) and does not include the defect 50 (good product) is photographed. Image of area) B' (second image). Here, the defective area A and the good area B' are the same area in the wafer.

Next, the brightness value (B1) of the image of the good product area (the first good product area) B shown in Figure 3(C) and the brightness of the image of the good product area (the second good product area) B' shown in Figure 4 value (B2) for comparison.

Table 3 is the following table. Like Table 1, while changing the optical conditions, it illustrates the brightness value of the defective area A (A), the brightness value of the first good product area B (B1), and the second good product under each optical condition. The brightness value (B2), the difference between the brightness value (A) and the brightness value (B1) of the area B' (the difference between the first brightness value) (D1=B1-A), the brightness value (B1) and the brightness value (B2) (Difference in the second brightness value) (D2=B1-B2), and the difference between the difference in the first brightness value (D1) and the difference in the second brightness value (D2) (D1-D2). In addition, the numerical value of the brightness value represents an arbitrary unit.

Here, the difference in the first brightness value (D1=B1-A) is the same as the difference (B-A) between the brightness value (A) of the defective area A and the brightness value (B) of the defective area B shown in Table 1.

[table 3] Optical conditions 1 2 3 4 Brightness value of defective area (A) 20 10 25 15 Brightness value of the 1st good product area (B1) 95 100 85 90 Brightness value of the second good product area (B2) 85 90 80 85 The difference between the first brightness value (D1=B1-A) 75 90 60 75 The difference between the 2nd brightness value (D2=B1-B2) 10 10 5 5 The difference between the 1st and 2nd brightness values (D1-D2) 65 80 55 70

Even if the good product areas B and B' are photographed under the same optical conditions, due to the deviation between chips, the brightness values B1 and B2 of the images taken in the good product areas B and B' will also change. However, the higher the definition of the image, the higher the resolution of the two images. The smaller the difference. Therefore, the difference (B1-B2) between the brightness values of the good product areas also becomes an evaluation value for obtaining the optimal optical conditions. Therefore, the difference between the brightness value (A) of the defective area A and the brightness value (B1) of the defective area B (the first difference in brightness value: D1), and the difference between the brightness values of the defective area B and B' (the second difference in brightness value) The difference in brightness value: D2) (D1-D2) becomes the evaluation value for obtaining the best optical conditions.

In the example shown in Table 3, the difference in brightness value (D1-D2) of optical condition 2 among the four optical conditions is the largest. Therefore, in this case, optical condition 2 can be said to be the optimal optical condition. In addition, by reducing the difference in brightness values between good product areas, detection of suspected defects that treat good products as defects can also be suppressed.

(Second Embodiment) In the first embodiment, the brightness value of the image captured in the defective area and the defective area is used as the characteristic quantity for obtaining the evaluation value of the optimal optical condition. However, the "sensitivity value" explained below may also be used as the characteristic quantity.

As a defect inspection method, a good product learning method called DSI (Die-to-Statistical Image) comparison method is known.

FIG. 5 is a diagram illustrating a method of producing learning materials for a plurality of good quality wafers.

First, as shown in FIG. 5(A) , a plurality of defective wafers are photographed to obtain images of the same area as the defective area B shown in FIG. 3(C) . Here, the number of pixels P in area B is set to 12 (3×4).

Since images of the same area are acquired, the distribution of brightness values of each pixel forms a regular distribution that reflects the deviation of good products. Therefore, by performing statistical processing for each pixel, as shown in Figure 5(B) and (C), the average value Gi (i=1~12) and the standard deviation σi (i=1~12) of each pixel can be obtained 12).

Using the learning data obtained in this way, for the wafer to be inspected, the sensitivity value Ai of each pixel in the defective area A and the good area B shown in Figures 3(B) and (C) is calculated based on the following formula (1) . Here, gi represents the brightness value of each pixel.

[Number 1]

Figure 6(A) shows the brightness value gi of each pixel in the defective area A, and Figure 6(B) shows the sensitivity value Ai of each pixel calculated using equation (1).

Similarly, Figure 7(A) shows the brightness value g'i of each pixel in the defective area B, and Figure 7(B) shows the sensitivity value A'i of each pixel calculated using equation (1).

In this way, while changing the optical conditions of the optical microscope, the sensitivity value Ai of each pixel in the defective area A and the sensitivity value A'i of each pixel in the defective area B are obtained.

Then, for each optical condition, the maximum value Amax of the sensitivity value Ai of each pixel in the defective area A is compared with the maximum value A'max of the sensitivity value Ai of each pixel in the defective area B.

Table 4 shows an example of changing the optical conditions 4 times. Under each optical condition, the sensitivity value of defective area A (Amax), the sensitivity value of defective area B (A'max) and the difference between the two sensitivity values (A'maxB -Amax) table. In addition, the numerical value of the sensitivity value represents an arbitrary unit.

[Table 4] Optical conditions 1 2 3 4 Sensitivity value of defective area (Amax) 200 500 250 200 Sensitivity value of good product area (A'max) 80 50 180 210 Difference in sensitivity values (Amax'-Amax) 120 450 70 -10

Even if the same areas A and B are photographed, if the optical conditions change, the sensitivity values of the images captured in areas A and B will also change. The higher the definition of the defective part of area A, the greater the difference with the learning data, and the greater the difference with the area. The greater the difference between B. Therefore, the difference (Amax-A'max) between the sensitivity value (Amax) of the defective area A and the sensitivity value (A'max) of the defective area B becomes the evaluation value for obtaining the optimal optical conditions. The sensitivity value becomes the threshold used to detect defects in the object during inspection. Therefore, the evaluation value is the same as the actual inspection, so a more accurate evaluation can be performed than using the "brightness value".

In the example shown in Table 4, the difference in brightness value (A'max-Amax) between the two under optical condition 2 among the four optical conditions is the largest. Therefore, in this case, optical condition 2 can be said to be the optimal optical condition.

In addition, in Table 4, the optimal optical conditions are obtained from the difference (A'max-Amax) between the sensitivity value (Amax) of the defective area A and the sensitivity value (A'max) of the good area B. However, the sensitivity of the defective area A The comparison between the value (Amax) and the sensitivity value (A'max) of the good product area B can also be done by the ratio of the two (A'max/Amax).

As mentioned above, the present invention has been described based on the preferred embodiments. However, this description is not a limitation, and various changes can of course be made. For example, in the above-mentioned embodiment, the brightness value or sensitivity value of the image taken in the defective area and the defective area is used as the feature quantity to obtain the evaluation value of the optimal optical condition of the optical microscope as an example, but it is not limited to this. Here, as long as the feature quantity changes according to the sharpness of the image, other feature quantities such as a dispersion value or a brightness value after edge enhancement filter processing may also be used.

1: Defect inspection device 10: Optical microscope 11:Carrying stage 12:Microscope body 13:Lighting Department 14:Camera Department 15: condenser lens 16: Half mirror 17:Objective lens 18:Coupling lens 19:Camera components 20:Control Department 25:Memory Department 30:wafer 31:wafer 31a:wafer 31b:wafer 31c:wafer 35:Image processing department 40:Testing agency 50: Defects A: Defect area A1～A12: sensitivity value A'1～A'12: sensitivity value B: Good product area B': Good product area G1～G12: average value g1～g12: brightness value g'1～g'12: brightness value L: illumination light P: pixel S101～S107: steps σ1～σ12: standard deviation

FIG. 1 is a diagram schematically showing the structure of a defect inspection device 1 according to the first embodiment of the present invention. FIG. 2 is a flowchart showing the use of the defect inspection device of the first embodiment to detect optimal optical conditions for photographing a region containing defects. 3(A) to 3(C) are diagrams illustrating a method of testing optimal optical conditions for photographing a region containing defects according to the flowchart shown in FIG. 2 . FIG. 4 is a diagram illustrating a method of detecting optimal optical conditions in a variation of the first embodiment. 5(A) to 5(C) are diagrams illustrating a method of creating learning materials for a plurality of good-quality wafers in the second embodiment of the present invention. 6(A) and (B) are diagrams showing the brightness value gi and sensitivity value Ai of each pixel in the defective area A. 7(A) and (B) are diagrams showing the brightness value g'i and sensitivity value A'i of each pixel in the good product area B.

1: Defect inspection device

10: Optical microscope

11:Carrying stage

12:Microscope body

13:Lighting Department

14:Camera Department

15: condenser lens

16: Half mirror

17:Objective lens

18:Coupling lens

19:Camera components

20:Control Department

25:Memory department

30:wafer

35:Image processing department

40:Testing agency

L: illumination light

Claims (10)

A defect inspection device that inspects defects formed in wafers and has: An optical microscope equipped with a camera unit for capturing images of the above-mentioned wafer; A control unit that controls the optical conditions used by the optical microscope to capture images of the wafer; and A memory unit that memorizes the locations of previously detected defects in the wafer; and While changing the optical conditions by the control unit, the imaging unit captures the first image of the area including the defect stored in the memory unit, and the same area as the area including the defect, but does not include the area of the defect 2nd image; and the defect inspection device has: A detection mechanism detects the optimal optical conditions for photographing the area containing the defect by comparing the feature quantity of the first image with the feature quantity of the second image. The defect inspection device of claim 1, wherein the imaging unit includes capturing a plurality of the second images, The above-mentioned detection mechanism not only compares the feature quantities of the above-mentioned first image and the above-mentioned second image, but also compares the feature quantities of the plurality of second images with each other, thereby detecting the area used to photograph the area containing the above-mentioned defect. Optimum optical conditions. The defect inspection device of claim 1, wherein the optical condition with the largest difference between the feature quantity of the first image and the feature quantity of the second image is detected as the best optical condition for photographing the area containing the defect. The defect inspection device of claim 2, wherein the optical condition with the largest difference between the feature quantity of the first image and the feature quantity of the second image is detected as the best optical condition for photographing the area containing the defect. The defect inspection device of claim 1, wherein the difference between the feature quantities of the first image and the feature quantities of the second image is the first difference, and the difference between the feature quantities of the plurality of second images is As the second difference, the optical condition in which the difference or ratio between the above-mentioned first difference and the above-mentioned second difference is the largest is detected as the best optical condition for photographing the area containing the above-mentioned defect. The defect inspection device of Claim 2, wherein the difference between the feature quantities of the above-mentioned first image and the feature quantities of the above-mentioned second image is set as the first difference, and the difference between the feature quantities of the plurality of second images is As the second difference, the optical condition in which the difference or ratio between the first difference and the second difference is the largest is detected as the best optical condition for photographing the area containing the defect. The defect inspection device of any one of claims 1 to 6, wherein the characteristic quantity of the first image and the characteristic quantity of the second image are brightness values or sensitivity values. The defect inspection device as claimed in any one of items 1 to 6, wherein the above-mentioned optical conditions include the magnification of the objective lens for capturing the image of the above-mentioned wafer, the type of the illumination light irradiated to the above-mentioned wafer, the brightness of the above-mentioned illumination light and the above-mentioned illumination light At least one of the optical filters. The defect inspection device of claim 7, wherein the above-mentioned optical conditions include at least the magnification of the objective lens for capturing the image of the above-mentioned wafer, the type of the illumination light irradiated to the above-mentioned wafer, the brightness of the above-mentioned illumination light and the filter of the above-mentioned illumination light. Either. A defect inspection method, which uses an optical microscope to take images of wafers formed on wafers and inspect defects in the wafers, and includes the following steps: Pre-register the locations of defects on the above-mentioned wafers; While changing the optical conditions of the above-mentioned optical microscope, take a first image of the area containing the above-mentioned defects, and a second image of the area not containing the above-mentioned defects within the same area as the area containing the above-mentioned defects; and By comparing the feature quantities of the first image and the feature quantities of the second image, the optimal optical conditions for photographing the area including the defect are detected.