JP2014062837A - Defect inspection device and defect reviewing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To solve the problem in which a conventional device has a limit as to an array on a wafer manageable by defect inspection or inspection sensitivity.SOLUTION: As an inspection device, the inspection device includes: a color image imaging device that images a color image of an inspected object; and a calculation part that selects at least two color component images of a plurality of color component images generated from one color image, plots a group of a color component having the same coordinate value on color space with a concentration difference gradation value between the two color component images as each axis, separates each of plotted points into a normal area and a defect area on the basis of a distribution of the plotted points and determines a pixel having a group of the same concentration difference gradation value as a point classified into the defect area as a defect.

Description

  The present invention relates to an optical defect inspection apparatus and defect review apparatus.

  In the wafer process, a large number of semiconductor elements (or chips) are formed in a wafer surface composed of, for example, a silicon single crystal. This process includes a process of transferring a pattern to the wafer surface by lithography and various processing processes. By executing these plural steps, a multilayer structure in which a predetermined number of patterns are overlapped is formed.

  In order to improve the yield of semiconductor elements and to achieve predetermined characteristics, it is extremely important that individual patterns are formed in a predetermined shape. Therefore, a nondestructive inspection is usually performed every time an individual pattern is formed.

  In non-destructive inspection, two grayscale image data are acquired from an inspection chip and a comparison chip that are separated by an integral multiple of the pitch interval from among chips regularly arranged in a lattice pattern on the wafer surface. If the density difference is larger than a preset defect determination threshold, the location is extracted as a defect. Patent Document 1 discloses a method for comparing two adjacent chips.

  FIG. 1 shows an outline of the conventional method. FIG. 1 shows a case where one chip is inspected for the sake of simplicity. Here, the chip B102 located in the middle of the three adjacent chips A, B, and C is the inspection target. At this time, the inspection apparatus scans the chip A101, the chip B102, and the chip C103 in order to acquire each image, and calculates a difference image between two adjacent images. That is, the difference image 105 between the chip A101 and the chip B102 and the difference image 106 between the chip B102 and the chip C103 are calculated. When the defect 104 is continuously detected in the two difference images 105 and 106, the inspection apparatus determines that the chip B102 has a defect.

  FIG. 2 shows a conventional example of an optical wafer pattern appearance inspection apparatus (hereinafter referred to as “appearance inspection apparatus”) for inspecting defects by such a procedure.

  Prior to the inspection, an inspection wafer 204 is placed on the surface of the XY stage 201. After the start of the inspection, the appearance inspection apparatus irradiates the inspection wafer 204 with inspection light from the light source 206. For the light source 206, for example, a xenon lamp is used. The inspection light is reflected by the inspection wafer 204, passes through the objective lens 205 positioned above the inspection wafer 204, and reaches the imaging surface of the one-dimensional black and white CCD camera (imaging device) 202. In this state, the XY stage 201 is scanned in the X direction, for example. Thereby, a two-dimensional gray image corresponding to the inspection target area is acquired.

  A two-dimensional grayscale image (that is, a monochrome image) is output from the one-dimensional monochrome CCD camera 202 to the image comparison unit 203. The image comparison unit 203 is realized through an image processing function of a personal computer, for example.

  The image comparison unit 203 includes an image memory unit 209 and a subtracter 210. The image memory unit 209 delays the two-dimensional inspection image (current image) from the one-dimensional monochrome CCD camera 202 by the time required for the XY stage 201 to scan one chip. The image memory unit 209 is configured by a shift register circuit, for example. As a result, the two-dimensional inspection image (delayed image) one chip before is input to the subtracter 210.

  The subtractor 210 calculates the difference between the same points of the delayed image and the current image in synchronization with the scanning of the XY stage 201 in the X direction. Thereby, a difference image between the inspection chip 207 and the comparison chip 208 is calculated.

  Patent Document 2 discloses a configuration in which a two-dimensional monochrome CCD camera is used to acquire an inspection image in a semiconductor wafer inspection apparatus.

JP 2006-133025 A JP 2001-068517 A

  However, the conventional appearance inspection apparatus has restrictions on the arrangement of inspection chips on the wafer and limits on inspection sensitivity.

  First, restrictions on arrangement will be described using a wafer map shown in FIG. As described above, for the inspection of one chip, it is necessary that a total of three chips including two chips on the left and right are present on the same line. That is, as shown in the uppermost row of the wafer map shown in FIG. 3, it is necessary to have three or more chips in one row. When there are only one or two chips in one row as in the lowermost row, the inspection is performed. I can't do it.

  Next, the limit relating to the inspection sensitivity will be described with reference to FIGS. The limit shown in FIG. 4 is a limit caused by using a difference image for defect inspection. Even when the accuracy of the XY stage 201 is high and the amount of shift between the images of the chip A 101 and the chip B 102 is less than one pixel, generally a sampling error remains between the delayed image and the current image. When this sampling error generates noise larger than that of the defect 104, it becomes impossible to detect only a true defect. Next, the second limit will be explained. FIG. 5 shows a chip image including a metal wiring pattern. The example of FIG. 5 is an example in which grains exist in the wiring of the chip A and the chip B. It is assumed that there is no grain in the wiring of chip C. In the case of the example in FIG. 5, a grayscale difference larger than that of the defect 104 appears in the difference image 105 in the grain portion. As a result, it becomes impossible to detect only true defects 104.

  In order to solve the above problems, a defect inspection apparatus and a defect review apparatus according to the present invention include a color image imaging apparatus that captures a color image of a region to be inspected, and a plurality of color component images generated from one color image. At least two color component images are selected, and a set of color components having the same coordinate value is plotted on the color space with the grayscale values as axes, and the plotted points are distributed in their distribution. And a calculation unit that divides the pixel into a normal region and a defective region and determines a pixel having the same grayscale value set as the point classified as the defective region as a defect.

  According to the present invention, chips in arbitrary rows and positions on a wafer can be inspected or reviewed. Further, according to the present invention, since a sampling error does not occur, only a true defect can be detected. In addition, according to the present invention, only true defects can be detected even when grains are included in the inspection region. Problems, configurations, and effects other than those described above will become apparent from the following description of embodiments.

The figure explaining the conventional test | inspection method. The block diagram explaining the conventional external appearance inspection apparatus. The figure which shows the restriction | limiting regarding the chip arrangement | sequence in the conventional external appearance inspection apparatus. The figure explaining the limit regarding the inspection sensitivity in the conventional appearance inspection apparatus (the 1). The figure explaining the limit regarding the inspection sensitivity in the conventional appearance inspection apparatus (the 2). 1 is a diagram illustrating a configuration of an appearance inspection apparatus according to Embodiment 1. FIG. 5 is a flowchart for explaining a defect inspection operation according to the first embodiment. The figure which shows the color image of the chip | tip B. The figure which shows R component image, G component image, and B component image. The figure explaining the principle of the defect detection based on the pixel distribution on a RG component space figure. The figure explaining an example of the pre-processing performed when detecting a defect area | region. The figure which shows the color image of the chip | tip B containing a grain. The figure which shows R component image, G component image, and B component image containing a grain. The figure explaining the principle of the defect detection based on the pixel distribution on a RG component space figure. FIG. 6 is a diagram illustrating a configuration of an appearance inspection apparatus with a review function according to a second embodiment. 9 is a flowchart for explaining defect inspection and review operations according to the second embodiment.

  Embodiments of the present invention will be described below with reference to the accompanying drawings. In addition, this invention is not limited to the Example mentioned later, A various deformation | transformation is possible in the range of the technical thought.

[Example 1]
(Device configuration)
FIG. 6 shows a configuration example of the optical wafer pattern appearance inspection apparatus according to the embodiment. FIG. 6 shows parts corresponding to those in FIG. The difference between the embodiment apparatus and the conventional apparatus (FIG. 2) is that a two-dimensional color CCD camera 601 is installed instead of the one-dimensional monochrome CCD camera (imaging apparatus) 202, and image processing is performed instead of the image comparison unit 203. The point is that the personal computer 602 is connected. If color imaging is possible, it is not necessary to be a two-dimensional color CCD camera. For example, a one-dimensional color CCD camera, a two-dimensional color CMOS camera, a one-dimensional color CMOS camera, or the like may be used.

  In this embodiment, the defect detection process is provided through image processing by a program executed on the image processing personal computer 602. The image processing personal computer 602 according to the present embodiment separates the R component image, the G component image, and the B component image from one color image acquired from the same imaging region, and among them, the colors defined for the R component image and the G component image The presence or absence of a defect is determined based on the component distribution in space. Note that all or part of the processing content provided through the program may be realized by hardware through a dedicated processing board, ASIC, or other semiconductor integrated circuit.

(Processing operation)
FIG. 7 shows an outline of the inspection operation executed in the appearance inspection apparatus according to the present embodiment. In this embodiment, the inspection object is a wafer. First, the inspection wafer 204 is placed on the XY stage 201. On the inspection wafer 204, chips are regularly arranged in a grid pattern.

  Thereafter, the appearance inspection apparatus moves the chip to be inspected (inspection chip 207) to the inspection area through control of the XY stage 201 by a drive control unit (not shown) (step 701). The inspection area is an area irradiated with light source light. After starting the movement, the image processing personal computer 602 determines whether or not the positioning (stage movement) of the inspection chip 207 has been completed (step 702).

  When the stage movement is complete, the inspection wafer 204 is irradiated with inspection light from a light source 206 (for example, a xenon lamp). The inspection light is reflected by the inspection chip 207. This reflected light enters the imaging area of the two-dimensional color CCD camera 601 through the objective lens 205 disposed above the inspection wafer 204. The two-dimensional color CCD camera 601 outputs the reflected light image to the image processing personal computer 602 as a two-dimensional color image.

  The image processing personal computer 602 captures the two-dimensional color image input from the two-dimensional color CCD camera 601 on a capture board or the like (step 703). An example of a captured color image is shown in FIG. FIG. 8A shows a case where the inspection chip 207 is the chip B. The image of the inspection chip 207 illustrated in FIG. 8A is apparently the same as the image of the chip B illustrated in FIG. 4, but is a color image instead of a monochrome image. The field of view of the captured image is, for example, 100 × 100 μm when the pixel size is 1 μm.

  Next, the image processing personal computer 602 separates the captured color image into RGB components, and generates three color component images (R component image, G component image, and B component image) (step 704). FIG. 8-2 illustrates an example of the R component image, the G component image, and the B component image separated from the color image. The color image captured on the capture board is often in a 24-bit bitmap format, but it is easy to convert this to an 8-bit 256 grayscale RGB image. Of course, the color image data format and the image conversion method are not limited to this, and any known method can be applied.

  Thereafter, the image processing personal computer 602 selects two color component images from the generated three color component images (step 705). Which color component image is selected is determined in advance based on past experience before inspection. In the case of the present embodiment, two of an R component image and a G component image are selected.

  Next, the image processing personal computer 602 plots the R component grayscale value of the pixel having the same coordinate value as the vertical axis component and the G component grayscale value as the horizontal axis component on the RG component space diagram. (Step 706). Points corresponding to all the pixels constituting the inspection chip 207 are plotted on the RG component space diagram. In the present embodiment, since one side of the inspection chip 207 is 100 pixels, 10000 points are plotted. Fig. 8-3 shows a plot example. In FIG. 8C, the horizontal axis represents the G component grayscale value, and the vertical axis represents the R component grayscale value.

  Thereafter, the image processing personal computer 602 extracts a plot point region appearing at a certain distance from the lump region including the center of gravity of all plot points on the RG component space diagram as a defect region (step 707). The defect is generally a portion where the material and the level difference are different from the surroundings. These differences are detected as differences in chromatic aberration with respect to the surrounding area. The inventors pay attention to this characteristic. Note that the value of the constant distance used as the threshold value between the normal area and the defective area is determined in advance before inspection based on past experience. Note that the certain distance varies depending on the material of the inspection region and the illumination light.

  FIG. 9 shows a concept of a process for obtaining a lump area including the center of gravity of all the pixels plotted in the RG component space diagram and a process for extracting a pixel separated from the area by a certain distance or more. In the case of the present embodiment, the image processing personal computer 602 executes a process of filling a certain distance from the center of each plot point. For example, the range of the radius 8 level from the center of the plot point is filled. Next, the image processing personal computer 602 is an area formed by connecting a plurality of plot points after the filling process directly or indirectly, and includes an area including the centroids of all plot points. Defined as a "region".

  The radius 8 level is an example. For example, the optimum filling range may be set for each process by comparing the automatic determination result using the set filling range with the operator's review result for the same region.

  In the example of FIG. 9, a black area is a “lumped area including the center of gravity”. In the example of FIG. 9, there are three regions separated from the “lumped region including the center of gravity” by a certain distance or more. These areas are determined as “defect areas”. In FIG. 9, these defect areas are shown in gray.

  In the image processing personal computer 602, a defect in which a combination of a gray level value of a G component and a gray level value of an R component is extracted first among 10000 pixels (= 100 pixels × 100 pixels) constituting the inspection chip 207. If the area matches, the coordinates of the pixel (coordinates within 100 × 100 pixels) are output as defect coordinates. The inventor has confirmed that the output defect coordinates are actually defects. This means that the color of the defective part is different from the normal pattern.

  Note that the image processing personal computer 602 repeatedly executes the processing from step 701 to step 707 until the inspection of all the inspection areas is completed (step 708). For example, when inspecting the entire inspection chip 207 given a size of 1000 μm × 1000 μm, the image processing personal computer 602 repeats the processing from step 701 to step 707 100 times, and then ends the inspection.

  The inventor confirmed that the detection method of this example is also effective when the inspection chip 207 includes grains. That is, the inventor confirmed that grains were not detected and only true defects could be detected.

  For reference, FIG. 10A shows an example in which grains are included in a captured color image. The image illustrated in FIG. 10A is the image of the chip B as in FIG. 5, but is not a monochrome image but a color image. FIG. 10-2 illustrates an example of an R component image, a G component image, and a B component image separated from a color image. As indicated by the circles in the figure, the R component image and the G component image contain grains. FIG. 10-3 is a diagram plotted on the RG component space diagram with the grayscale value of the R component of the pixels having the same coordinate value as the vertical axis component and the grayscale value of the G component as the horizontal axis component. is there. As described above, the inventor has confirmed that the area indicated by a circle in the figure is only a true defect.

(Summary)
As described above, in the appearance inspection apparatus according to the present embodiment, a defect can be determined only with a color image captured from one inspection chip. For this reason, a defect can be inspected even in a column in which no contrast pixel exists on the left and right, for example, a column having only one chip. In addition, even in the case of a column including three or more chips in one column, it is possible to inspect defects in chips at both end positions that do not have contrast pixels on the left and right.

  In addition, in the appearance inspection apparatus according to the present embodiment, no sampling error is generated in principle. For this reason, defects are not buried in noise, and only true defects can be detected. In addition, the visual inspection apparatus according to the present embodiment can detect only true defects even when grains are included in the imaging region.

[Example 2]
(Device configuration)
Here, a case where the above-described appearance inspection apparatus is used as a defect review apparatus will be described. For inspection of defects to be reviewed, an optical wafer pattern appearance inspection apparatus having the configuration shown in FIG. 2 described as a conventional apparatus is used.

  FIG. 11 shows a configuration example of an appearance inspection apparatus with a review function according to the second embodiment. In FIG. 11, parts corresponding to those in FIGS. 2 and 6 are denoted by the same reference numerals. In this embodiment, the imaging field of view of the one-dimensional monochrome CCD camera 202 is wider than that of the two-dimensional color CCD camera 601. Therefore, the time required for visual inspection of the entire inspection wafer 204 is shorter when the one-dimensional monochrome CCD camera 202 is used than when the two-dimensional color CCD camera 601 is used.

  Therefore, in this embodiment, a one-dimensional monochrome CCD camera 202 is used for appearance inspection, and a two-dimensional color CCD camera 601 is used for review of detected defects. That is, the present embodiment is a hybrid system of the conventional apparatus and the first embodiment, and the apparatus configuration used in the first embodiment is used as a review apparatus.

  In FIG. 11, a defect detection unit 1101 that detects the defect 104 based on a comparison between two consecutive difference images and outputs the detected defect coordinates to the image processing personal computer 602 for review is added. Of course, the defect detection processing by the defect detection unit 1101 may also be realized as a function of the image processing personal computer 602.

(Processing operation)
FIG. 12 shows an outline of the inspection operation executed in the appearance inspection apparatus with a review function according to the present embodiment. Note that, in FIG. 12, the same reference numerals are given to portions corresponding to those in FIG.

  In the appearance inspection apparatus with a review function according to the present embodiment, first, the XY stage 201 on which the inspection wafer 204 is placed is scanned in the X direction, and a scan image of the entire detection area is captured by the one-dimensional monochrome CCD camera 202. (Step 1201).

  At this time, the two-dimensional inspection image (current image) captured by the one-dimensional monochrome CCD camera 202 is output to the image comparison unit 203 and compared with the delayed image by the subtractor 210. These difference images between successive chips are given to the defect detection unit 1101 to determine a defect. Incidentally, the defect detection unit 1101 determines the presence or absence of a defect according to the principle described in FIG. Information on the defect coordinates, which is the determination result, is output from the defect detection unit 1101 to the image processing personal computer 602. The defect coordinates may be output sequentially each time a defect is detected, or may be output collectively after the end of scanning of all inspection areas.

  As described above, there are chips that cannot be inspected by the defect inspection using the one-dimensional monochrome CCD camera 202. Such chips may be collectively inspected at the time of review inspection to be described later.

  Next, the image processing personal computer 602 gives the acquired defect coordinate information to the drive control unit (not shown) as the coordinate position of the inspection chip 207. The drive control unit (not shown) moves the XY stage 201 so that the inspection chip 207 with the defect coordinates is located in the inspection area (step 1202).

  When the stage movement is completed, the image processing personal computer 602 captures the two-dimensional color image input from the two-dimensional color CCD camera 601 (step 703). Thereafter, the same processing as in the first embodiment, that is, the processing from step 704 to step 707 is executed in order. When a defect area is extracted in step 707, it is determined that the defect detected in step 1201 is a true defect. If a defect area is not extracted in step 707, it is determined that the defect detected in step 1201 is false information.

  Thereafter, the image processing personal computer 602 determines whether or not the determination process has been completed for all defect coordinates (step 1203). While the undetermined coordinates remain among the defect coordinates notified from the defect detection unit 1101, the image processing personal computer 602 repeats the processing of Step 1202 and Steps 703 to 707. On the other hand, when the determination is completed for all defect coordinates, the image processing personal computer 602 ends the review inspection.

(Summary)
As described above, the review function (defect review apparatus) according to the present embodiment can obtain the same effects as those of the first embodiment. That is, the defect review apparatus according to the present embodiment reviews whether there is a defect even in a chip in a column where no contrast pixel exists on the left and right, and in a column including three or more chips, and a chip located at both ends that does not have a contrast pixel on the left and right. can do.

  In addition, the defect review apparatus according to the present embodiment does not generate a sampling error in principle. Thus, only true defects can be reviewed. In addition, the defect review apparatus according to the present embodiment can review only true defects even when grains are included in the imaging region.

  In addition, the appearance inspection apparatus with a review function according to the present embodiment can reduce the time required for detecting a defect as compared with the first embodiment by combining the review function described above.

[Other embodiments]
(1) In the above-described embodiments, the wafer pattern appearance inspection apparatus has been described. However, the inspection / review principle described in each embodiment can be applied to defect inspection and review other than for wafers. For example, it can be applied to a bump inspection device and a substrate inspection device.
(2) In the above-described embodiment, the color component of each pixel is mapped on a two-dimensional color component space to obtain a “lumped region including the center of gravity”, and a region separated from the region by a certain distance is defined as a defective region. However, the color component of each pixel may be mapped on a three-dimensional color component space.
(3) In the above-described embodiment, the color image is separated into the R component image, the G component image, and the B component image. However, the color image may be separated into other color component images.
(4) The present invention is not limited to the above-described embodiments, and includes various modifications. For example, the above-described embodiments have been described in detail for easy understanding of the present invention, and are not necessarily limited to the case where all the configurations described are provided.

  In addition, a part of the configuration of a certain embodiment can be replaced with the configuration of another embodiment. It is also possible to add the configuration of another embodiment to the configuration of one embodiment. Further, with respect to a part of the configuration of each embodiment, another configuration can be added, deleted, or replaced.

  Each of the above-described configurations, functions, processing units, processing means, and the like may be realized by hardware by designing a part or all of them with, for example, an integrated circuit. Further, each of the above-described configurations, functions, and the like may be realized by software by the CPU interpreting and executing a program that realizes each function.

  Information such as programs, tables, and files for realizing each function can be stored in a recording device such as a memory, a hard disk, an SSD (Solid State Drive), or a recording medium such as an IC card, an SD card, or a DVD.

  Further, the control lines and information lines indicate what is considered necessary for the explanation, and not all the control lines and information lines on the product are necessarily shown. Actually, it may be considered that almost all the components are connected to each other.

101: Chip A
102: Chip B
103: Chip C
104: Defect 105: Difference image (chip A-chip B)
106: Difference image (chip B-chip C)
201: XY stage 202: one-dimensional monochrome CCD camera 203: image comparison unit 204: inspection wafer 205: objective lens 206: light source 207: inspection chip 208: comparison chip 209: image memory unit 210: subtractor 601: two-dimensional Color CCD camera 602: Image processing personal computer 1101: Defect detection unit

Claims (9)

A light source that generates illumination light;
An illumination optical system for illuminating a desired light beam on the surface of the object to be inspected;
A condensing optical system for condensing the reflected light from the surface of the inspection object;
A color image capturing device that receives reflected light collected by the condensing optical system and captures a color image;
A set of color components having the same coordinate value in a color space having at least two color component images selected from among a plurality of color component images generated from one color image and having the grayscale values as axes. And plotting each of the plotted points into a normal area and a defective area based on their distribution, and determining a pixel having the same grayscale value set as the point classified into the defective area as a defect An inspection apparatus having a calculation unit. The inspection apparatus according to claim 1,
The calculation unit uses a predetermined value for the size of each point on the color space, and is an area formed by connecting the points directly or indirectly, and includes all the plotted points. An inspection apparatus characterized in that an area including a center of gravity is determined as the normal area. The inspection apparatus according to claim 2,
The said calculation part determines the point which left | separated more than predetermined distance from the said normal area | region on the said color space as the said defect area | region. The inspection apparatus characterized by the above-mentioned. The inspection apparatus according to claim 1,
The said calculation part determines the point which left | separated more than predetermined distance from the said normal area | region on the said color space as the said defect area | region. The inspection apparatus characterized by the above-mentioned. A light source that generates illumination light;
An illumination optical system for illuminating a desired light beam on the surface of the object to be inspected;
A condensing optical system for condensing the reflected light from the surface of the inspection object;
A color image imaging device that receives the reflected light collected by the condensing optical system and images a color image in a state where the inspection object is positioned in a region corresponding to a defect coordinate;
A set of color components having the same coordinate value in a color space having at least two color component images selected from among a plurality of color component images generated from one color image and having the grayscale values as axes. And plotting each of the plotted points into a normal area and a defective area based on their distribution, and determining a pixel having the same grayscale value set as the point classified into the defective area as a defect A review device having a calculation unit. The review device according to claim 5,
The calculation unit uses a predetermined value for the size of each point on the color space, and is an area formed by connecting the points directly or indirectly, and includes all the plotted points. The review apparatus characterized by determining the area including the center of gravity as the normal area. The review device according to claim 6,
The review unit is characterized in that a point separated from the normal area by a predetermined distance or more on the color space is determined as the defect area. The review device according to claim 5,
The said calculation part determines the plot point away from the said normal area | region more than the predetermined distance on the said color space as the said defect area. The review apparatus characterized by the above-mentioned. The review device according to claim 5,
A black and white image capturing apparatus that receives reflected light collected by the condensing optical system and captures a black and white image;
A review apparatus comprising: a defect detection unit that detects a defect based on the black-and-white imaging and outputs coordinates of the detected region as the detection coordinate.

Images