JP4885590B2 - Semiconductor wafer inspection method and apparatus - Google Patents

Description

  The present invention detects defects (scratches, foreign matter, pattern collapse, etc.) from image data obtained by imaging the pattern forming surface of a semiconductor wafer, and the detected defects are fatal (normal operation of the semiconductor chip). The present invention relates to a semiconductor wafer inspection method and an apparatus for determining whether or not to inhibit.

2. Description of the Related Art Conventionally, an apparatus for obtaining image data by imaging a pattern forming surface of a semiconductor wafer and detecting defects from the obtained image data has been proposed and put into practical use.
In addition, for example, it has been proposed to detect defects from image data and automatically classify the defects based on information such as the size, shape, and color of the defect, which has been put to practical use.

  However, since these devices cannot determine whether or not the defect is fatal, it is necessary for the operator to observe the defect one by one to determine whether or not it is fatal. The burden on the operator cannot be reduced too much.

In order to eliminate such inconvenience, a semiconductor wafer defect inspection apparatus capable of determining whether or not a defect is fatal has been proposed (see Patent Document 1).
JP 2004-1117016 A

  The semiconductor wafer defect inspection apparatus described in Patent Document 1 measures the distance between a defect and an adjacent wiring on the assumption that the reflected light intensity of the wiring on the semiconductor wafer is high and the reflected light intensity of the defect is low. The measured distance is compared with a set value to determine whether or not the defect is serious.

  Therefore, when the reflected light intensity of the wiring and the reflected light intensity of the defect are substantially equal, the defect and the wiring cannot be distinguished, and as a result, it can be determined whether or not the defect is fatal. It becomes impossible.

  Also, in order to measure the distance, the reflected light intensity in the direction orthogonal to the wiring must be measured, so depending on the orientation of the wiring, it may be necessary to perform rotation processing on the image data, and the entire process Will become complicated.

  The present invention has been made in view of the above problems, and whether or not a defect is fatal without being affected by the brightness of the defect and without performing rotation processing regardless of the direction of the wiring. An object of the present invention is to provide a semiconductor wafer inspection method and apparatus capable of determining the above.

  The semiconductor wafer inspection method according to claim 1 is a semiconductor wafer inspection method in which a pattern forming surface of a semiconductor wafer is imaged by an imaging device, and a defect is detected from image data output from the imaging device, which is output from the imaging device. A step of extracting defects from image data to be obtained, a step of obtaining a contour of a pattern composed of straight lines from image data output from the imaging device, and a contour of a pattern composed of defects and straight lines are superimposed. Scanning the contour of the pattern composed of straight lines, determining whether a defect exists on the contour of the pattern composed of straight lines, and in response to the presence of the defect on the contour, Determining that it is fatal.

  The method for inspecting a semiconductor wafer according to claim 2 further includes a step of expanding the extracted defect and newly adopting the expanded defect as a defect.

  A semiconductor wafer inspection apparatus according to a third aspect of the present invention is a semiconductor wafer inspection apparatus that images a pattern formation surface of a semiconductor wafer with an imaging device and detects defects from image data output from the imaging device, and is output from the imaging device. A defect extracting means for extracting defects from image data to be obtained, an edge acquiring means for obtaining a contour of a pattern composed of straight lines from the image data output from the imaging device, and a contour of the pattern composed of defects and the straight lines. In addition, the contour scanning means for scanning the contour of the pattern composed of straight lines, and whether or not a defect exists on the contour of the pattern composed of straight lines, and the defect exists on the contour And a defect determination means for determining that the defect is fatal in response to.

  The semiconductor wafer inspection apparatus according to claim 4 further includes defect expansion means for expanding the extracted defect and newly adopting the expanded defect as a defect.

  The invention of claim 1 scans the contour of a pattern composed of straight lines to determine whether or not a defect exists on the contour of the pattern composed of straight lines. As a result, the defect is fatal. Since it is determined whether or not there is, it is possible to determine whether or not the defect is fatal without being affected by the brightness of the defect and without performing rotation processing regardless of the direction of the wiring. Has a unique effect.

  The invention of claim 2 can determine that the defect is close to the pattern, and can achieve the same effect as that of claim 1.

  The invention of claim 3 scans the contour of the pattern composed of straight lines to determine whether or not there is a defect on the contour of the pattern composed of straight lines. As a result, the defect is fatal. Since it is determined whether or not there is, it is possible to determine whether or not the defect is fatal without being affected by the brightness of the defect and without performing rotation processing regardless of the direction of the wiring. Has a unique effect.

  The invention of claim 4 can determine that the defect is close to the pattern, and can achieve the same effect as that of claim 3.

  Embodiments of a semiconductor wafer inspection method and apparatus according to the present invention will be described below in detail with reference to the accompanying drawings.

FIG. 1 is a schematic view showing an embodiment of a semiconductor wafer inspection apparatus of the present invention.
This semiconductor wafer inspection apparatus 1 supports an inspection target semiconductor wafer, operates an inspection stage 2 that operates in directions orthogonal to each other and in a predetermined rotation direction, an imaging optical unit 5 that images at least a part of the semiconductor wafer, and focusing. An imaging optical unit driving unit 6 for operating the imaging optical unit 5 for the purpose, a light source 7 for illuminating the semiconductor wafer, an inspection stage 2, a control computer 8 for controlling the imaging optical unit driving unit 6, and the like, and An image processing computer 9 for processing an image obtained by the imaging optical unit 5 is included. Note that the imaging optical unit 5 has a rotatable revolver, an optical system, and an inspection camera on which a plurality of objective lenses are mounted and one distance measuring sensor is mounted.

Since the configuration and operation of the inspection stage 2, the imaging optical unit 5, the imaging optical unit driving unit 6, the illumination light source 7, and the control computer 8 are conventionally known, detailed description thereof will be omitted.
The image processing computer 9 has, for example, a first function for determining the presence or absence of a defect by comparing an image obtained by the imaging optical unit 5 with a reference image held in advance. The image processing computer 9 also has a second function for determining whether or not the defect is fatal.

  Since the configuration and operation of the first function are conventionally known, detailed description is omitted.

  The operation of the second function is, for example, as follows.

  When an image determined to have a defect is given as shown in FIG. 2, an edge (contour line) consisting of only a linear component is obtained and temporarily held in a memory (see FIG. 3), and only the defect is Is extracted and temporarily stored in the memory (see FIG. 4).

  Next, the obtained edge and the extracted defect are overlapped, and the edge is scanned in this state to determine whether or not a defect exists on the edge (see FIG. 5).

  Then, when it is determined that a defect exists on the edge, it is determined that the defect is fatal. Otherwise, it is determined that the defect is not fatal.

  In the above case, instead of obtaining an edge (contour line) consisting only of a linear component and temporarily storing it in the memory (see FIG. 3), an edge (contour line) consisting only of the linear component included in the image of FIG. ) Can be extracted and temporarily stored in the memory. In this case, the end point of the extracted edge is detected, and the defect is detected by a predetermined distance (for example, several pixels) beyond the end point of the edge. The presence or absence may be determined.

  FIG. 6 is a block diagram showing the main part of the image processing computer 9.

  The image processing computer 9 includes a first determination unit 91 that determines whether an image has a defect, a first holding unit 92 that holds an image having a defect, and a defect. An edge acquisition unit 93 that obtains an edge (contour line) consisting only of a straight line component from an existing image and temporarily stores it in a memory; and extracts only a defect from an image having a defect and temporarily stores it in the memory A defect extraction unit 94, and a second determination unit 95 that determines whether or not a defect exists on the edge by superimposing the obtained edge and the extracted defect and scanning the edge in this state; Contains.

  Therefore, when an image determined to have a defect by the first determination unit 91 is held in the first holding unit 92 as shown in FIG. 2, an edge (contour line) consisting of only a linear component is acquired by the edge acquisition unit 93. ) And temporarily stored in the memory (see FIG. 3), and the defect extraction unit 94 extracts only the defect and temporarily stores it in the memory (see FIG. 4).

  Next, the second determination unit 95 superimposes the obtained edge and the extracted defect, and scans the edge in this state, thereby determining whether or not a defect exists on the edge (see FIG. 5). ).

  Then, when it is determined that a defect exists on the edge, it is determined that the defect is fatal. Otherwise, it is determined that the defect is not fatal.

  In the above case, instead of obtaining an edge (contour line) consisting only of a linear component and temporarily storing it in the memory (see FIG. 3), an edge (contour line) consisting only of the linear component included in the image of FIG. ) Can be extracted and temporarily stored in the memory. In this case, the end point of the extracted edge is detected, and the defect is detected by a predetermined distance (for example, several pixels) beyond the end point of the edge. The presence or absence may be determined.

  The embodiment has been described above in which the extracted defect is adopted as it is and whether or not the defect is fatal is determined.

However, instead of adopting the extracted defect as it is, it is possible to determine whether the defect is fatal based on the expanded defect by performing an expansion process.
Specifically, when an image determined to have a defect is given as shown in FIG. 7, an edge (contour line) is extracted (see FIG. 8), and a Hough transform process or the like is performed to obtain an edge (Contour line) Detect only the straight line component from the information (search the straight line in the image and detect each line as one straight line) (see FIG. 9), and compare the detected straight line with the extracted edge Thus, an edge (contour line) consisting only of a straight line component is obtained and temporarily stored in the memory (see FIG. 10). Further, only the defect is extracted from the image shown in FIG. 7 (see FIG. 11), and the extracted defect is subjected to expansion processing to obtain an enlarged defect, which is temporarily held in the memory (FIG. 11). 12).

  Next, the obtained edge composed of only the linear component and the enlarged defect are overlapped, and the edge is scanned in this state to determine whether or not there is a defect on the edge (see FIG. 13).

  Then, when it is determined that there is a magnified defect on the edge, it is determined that the defect is highly fatal. Otherwise, the defect may be fatal. Judge that it is not expensive.

FIG. 14 is a block diagram showing the main part of the image processing computer 9.
The image processing computer 9 includes a first determination unit 91 that determines whether an image has a defect, a first holding unit 92 that holds an image having a defect, and a defect. An edge acquisition unit 93 that obtains an edge (contour line) consisting only of a straight line component from an existing image and temporarily stores it in a memory; and extracts only a defect from an image having a defect and temporarily stores it in the memory The defect extraction unit 94, the expansion processing unit 96 that performs expansion processing on the extracted defect and temporarily holds it in the memory, and the obtained edge and the enlarged defect are overlapped, and the edge is moved in this state. And a second determination unit 95 that determines whether or not a defect exists on the edge by scanning.

  Therefore, when an image determined to have a defect by the first determination unit 91 is held in the first holding unit 92 as shown in FIG. 7, an edge (contour line) consisting only of a linear component is acquired by the edge acquisition unit 93. ) And temporarily stored in the memory (see FIG. 10), the defect extracting unit 94 extracts only the defect and temporarily stores it in the memory (see FIG. 11), and the expansion processing unit 96 extracts the defect. The expanded defect is subjected to expansion processing and temporarily held in the memory (see FIG. 12).

  Next, the second determination unit 95 superimposes the obtained edge and the extracted defect, and scans the edge in this state, thereby determining whether or not a defect exists on the edge (see FIG. 13). ).

Then, when it is determined that a defect exists on the edge, it is determined that the defect is fatal. Otherwise, it is determined that the defect is not fatal.
As can be seen from the above description, according to the embodiment of the present invention, whether or not the defect is fatal without being affected by the brightness of the defect and without performing the rotation process regardless of the direction of the wiring. Can be determined.

  When it is determined that the defect is fatal, it is unnecessary to observe the defect through the microscope by the operator.

It is the schematic which shows one Embodiment of the semiconductor wafer test | inspection apparatus of this invention. It is a figure which shows an example of the image determined to have a defect. It is a figure which shows only the edge which consists of only a linear component obtained from the image of FIG. It is a figure which shows only the defect obtained from the image of FIG. It is a figure which shows the process which determines whether a defect exists on an edge by superimposing the obtained edge and the extracted defect, and scanning an edge in this state. It is a block diagram which shows the principal part of the computer for image processing. It is a figure which shows an example of the image determined to have a defect. It is a figure which shows the extraction result of the edge obtained from the image of FIG. It is a figure which shows the straight line detected from the extraction result of edge. It is a figure which shows only the edge which consists only of a linear component. It is a figure which shows only the defect obtained from the image of FIG. It is a figure which shows the defect in which the expansion process was performed. It is a figure which shows the process which determines whether a defect exists on an edge by superimposing the extracted edge and the enlarged defect, and scanning an edge in this state. It is a block diagram which shows the principal part of the computer for image processing.

Explanation of symbols

93 Edge acquisition unit 94 Defect extraction unit 95 Second determination unit 96 Expansion processing unit

Claims (4)

A semiconductor wafer inspection method for imaging a pattern formation surface of a semiconductor wafer with an imaging device and detecting defects from image data output from the imaging device,
Extracting defects from image data output from the imaging device;
Obtaining a contour of a pattern composed of straight lines from image data output from the imaging device;
Superimposing the defect and the contour of the pattern composed of straight lines, and scanning the contour of the pattern composed of straight lines;
Determining whether a defect exists on the contour of the pattern composed of straight lines, and determining that the defect is fatal in response to the presence of the defect on the contour;
A method for inspecting a semiconductor wafer, comprising: The semiconductor wafer inspection method according to claim 1, further comprising a step of expanding the extracted defect and newly adopting the expanded defect as a defect. A semiconductor wafer inspection device that images a pattern forming surface of a semiconductor wafer with an imaging device and detects defects from image data output from the imaging device,
Defect extraction means (94) for extracting defects from image data output from the imaging device;
Edge acquisition means (93) for obtaining a contour of a pattern composed of straight lines from image data output from the imaging device;
Contour scanning means (95) for superimposing the defect and the contour of the pattern composed of straight lines, and scanning the contour of the pattern composed of straight lines;
Defect determination means (95) for determining whether or not a defect exists on the contour of a pattern constituted by straight lines and determining that the defect is fatal in response to the presence of the defect on the contour When,
A semiconductor wafer inspection apparatus comprising: 4. The semiconductor wafer inspection apparatus according to claim 3, further comprising defect expansion means (96) for expanding the extracted defect and newly adopting the expanded defect as a defect.

Images