JP2013200190A - Defect inspection device and defect inspection method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a defect inspection device which detects a defect on a wafer surface without receiving influence of periodic surface unevenness (surface morphology).SOLUTION: The defect inspection device includes: a raw image acquisition part 2 for acquiring a raw image of a wafer surface; an image cutout part 3 for sequentially cutting out image data of one area out of image data of a plurality of divided areas obtained by dividing the raw image of a wafer surface; a Fourier transformation part 4 for performing Fourier transformation processing to the cut out image data of one area; a periodicity discrimination part 5 for discriminating presence/absence of periodicity of image as to image data of one area to which the Fourier transformation processing is applied, and determining that it is an "area without defect" when it is determined that the periodicity is present, and determining that it is an "area with defect" when it is determined that the periodicity is absent; and a coordinate specification part 6 for specifying an XY coordinate position in an entire wafer area where the "area with defect" is located.

Description

  The present invention relates to a defect inspection apparatus for inspecting the presence or absence of defects such as an epitaxial layer on a wafer surface without being affected by periodic surface irregularities (surface morphology), and a defect inspection method using the defect inspection apparatus.

  This type of conventional defect inspection apparatus inspects a defect portion generated in, for example, an epitaxial layer on the wafer surface.

  FIG. 8 is a main part configuration diagram showing an example of a conventional defect inspection apparatus disclosed in Patent Document 1. In FIG.

  As shown in FIG. 8, in a conventional defect inspection apparatus 100, a silicon carbide substrate (SiC substrate) on which an epitaxial layer is formed is used as a single crystal substrate to be inspected. As a single crystal substrate, not only a SiC substrate but also various single crystal substrates on which an epitaxial layer is formed by step flow growth can be inspected. As this conventional defect inspection apparatus 100, a confocal scanning apparatus having a differential interference optical system is used, and the entire surface of the silicon carbide substrate is scanned by the confocal scanning apparatus to capture a confocal differential interference image on the entire surface of the substrate. The obtained confocal differential interference image is subjected to various image processes to detect defects and their coordinates (addresses) and also measure the distribution density of step bunching. Since the differential interference optical system detects a change in the height of the sample surface as a change in the phase difference of the scanning light, it is possible to detect a concave and convex defect of several nm to several tens of nm as a luminance change. .

  Here, a mercury lamp is used as the illumination light source 101, but various illumination light sources other than a mercury lamp such as a xenon lamp can also be used. The illumination beam emitted from the illumination light source 101 is incident on an optical fiber bundle 102 in which a plurality of optical fibers are stacked in a circle, propagates through the optical fiber, and is emitted as a divergent beam having a substantially circular cross section. Incident. The filter 103 emits green wavelength light (e-line: wavelength 546 nm) from the incident light beam. The light beam emitted from the filter 103 is converted into a parallel beam by the converging lens 104 and enters the slit 105. The slit 105 is disposed at the pupil position of the converging lens 104, and has an elongated opening extending in a first direction (a direction orthogonal to the paper surface). Here, the first direction is referred to as the X direction. The width of the opening of the slit 105 is set to 10 to 20 μm, for example. Therefore, an elongated line-shaped light beam extending from the slit 105 in the first direction is emitted. The linear light beam emitted from the slit 105 enters the polarizer 106 and is converted into polarized light having a single vibration surface. This line-shaped polarized beam is reflected by the half mirror 107 that functions as a beam splitter, and enters a vibrating mirror 109 that functions as a scanning device via a relay lens 108.

  A driving circuit 110 is connected to the vibrating mirror 109, and the driving circuit 110 drives the vibrating mirror 109 based on a control signal supplied from the signal processing device 111. The oscillating mirror 109 deflects the incident line-shaped light beam in a second direction (Y direction) orthogonal to the first direction. The signal processing device 111 has position information of the light beam in the Y direction based on the angle information of the vibrating mirror 109. The vibrating mirror 109 is used when performing defect detection by step-and-repeat, or is used for reviewing defects. The line-shaped light beam emitted from the vibration mirror 109 enters the differential interference optical system 114 via the relay lenses 112 and 113. Here, a Nomarski prism is used as the differential interference optical system. The linear polarized beam incident on the Nomarski prism 114 is converted into two sub beams whose vibration planes are orthogonal to each other. A phase difference of (2m + 1) π / 2 is given between these two sub-beams, where m is a natural number. Therefore, it is possible to detect a defect having a height change of several nm formed on the surface of the epitaxial layer of the SiC substrate as a luminance image. A motor 115 is connected to the Nomarski prism 114, and the shearing direction of the Nomarski prism 114 can be freely set, and the shearing direction is set according to the extending direction of the step bunching existing in the epitaxial layer formed on the SiC substrate. Can be set.

  The two sub beams emitted from the Nomarski prism 114 are incident on the objective lens 116. The objective lens 116 focuses the incident two line-shaped sub-beams and projects the sub-beams toward the surface of the epitaxial layer of the silicon carbide substrate 118 (SiC substrate) to be observed arranged on the stage 117. Therefore, the surface of SiC substrate 118 is scanned in a second direction (Y direction) perpendicular to each other by two line-shaped sub-beams extending in the first direction (X direction).

  The stage 117 is configured by an XY stage that can move in the X direction and the Y direction orthogonal to the X direction, and is a stage that can rotate in a plane orthogonal to the optical axis of the objective lens 116. Position information of the stage 117 in the X direction and Y direction is detected by a position sensor (not shown), and the position information of the stage is supplied to the signal processing device 111. The rotation angle of the stage 117 is also detected by a sensor (not shown) and supplied to the signal processing device 111. When a defect is detected by scanning the entire surface of the SiC substrate 118, the entire surface of the SiC substrate can be scanned by moving the stage 117 in a zigzag manner in the Y direction and the X direction while maintaining the oscillating mirror 109 in a stationary state. . Alternatively, the stage 117 can be continuously moved in the first direction, and combined with scanning in the second direction by the vibrating mirror 109, defect inspection can be performed on the entire surface of the SiC substrate. Further, the entire surface of the substrate can be scanned by step-and-repeat combining scanning by the vibrating mirror 109 and stage movement.

  A motor 119 and a motor drive circuit 120 are connected to the objective lens 116, and can be moved along the optical axis direction by a drive control signal supplied from the signal processing device 111. The position of the objective lens 116 in the optical axis direction is detected by the position sensor 121 and supplied to the signal processing device 111. Here, the motor 119 is a relative distance in the optical axis direction between the objective lens 116 and the SiC substrate 118 on the stage 117, that is, a relative distance between the focal point of the light beam that scans the substrate surface and the substrate surface. It functions as a means to change. The objective lens 116 can move in the optical axis direction with a resolution of 10 nm.

  The two reflected beams reflected by the surface of the SiC substrate 118 are collected by the objective lens 116 and enter the differential interference optical system (Nomarski prism 114). The two reflected sub-beams are synthesized by the differential interference optical system 114 to form an interference beam including the height change of the SiC substrate surface as phase difference information. For example, when a concave or convex defect of about several nanometers exists on the surface of the SiC substrate 118, one of the two sub beams incident on the SiC substrate surface scans on the defect, and the other sub beam is Since the normal surface portion is scanned, a phase difference corresponding to the height of the defect is introduced between the two sub beams. Further, when step bunching is formed on the surface of the epitaxial layer, step bunching having a height of about 1 nm can be detected as a linear high luminance image or low luminance image. As a result, the interference beam emitted from the differential interference optical system 114 includes, as phase difference information, unevenness changes of about several nanometers appearing on the surface of the SiC substrate due to crystal defects and steps due to step bunching.

  The interference beam emitted from the Nomarski prism 114 propagates in the opposite direction along the original optical path, enters the oscillating mirror 109 via the relay lenses 113 and 112, and is descanned by the oscillating mirror 109. The interference beam emitted from the oscillating mirror 109 passes through the lens 108 acting as an imaging lens, passes through the half mirror 107, and enters the analyzer 122. The analyzer 122 is arranged in a crossed Nicols relationship with respect to the polarizer 106. Accordingly, light other than the polarized light synthesized in the Nomarski prism 114 is blocked, and only the light constituting the differential interference image is transmitted through the analyzer 122.

  The linear interference beam transmitted through the analyzer 122 enters the linear image sensor 124 through the positioner 123. The linear image sensor 124 includes a plurality of light receiving elements arranged in a direction corresponding to the first direction, and receives the incident line-shaped interference beam. Each light receiving element of the linear image sensor 124 converts the phase difference information included in the interference beam into luminance information. Accordingly, irregularities and step bunching of about several nanometers formed on the surface of SiC substrate 118 or the surface of the epitaxial layer are displayed as luminance images. The light receiving element array arranged in a line shape of the linear image sensor 124 has an entrance opening limited by a frame, and is therefore almost equivalent to a structure in which a pinhole is disposed on the front surface of each light receiving element. Therefore, the confocal optical system having the differential interference optical system is configured by receiving the reflected light from the surface of the SiC substrate by the linear image sensor 124.

  The electric charge accumulated in each light receiving element of the linear image sensor 124 is sequentially read by a read control signal supplied from the signal processing device 111, and is output as a one-dimensional image signal on the surface of the SiC substrate. The one-dimensional image signal output from the linear image sensor 124 is amplified by the amplifier 125 and supplied to the signal processing device 111 via the camera link. The signal processing device 111 includes an image processing board, and generates a two-dimensional image of the surface of the SiC substrate using the received one-dimensional image signal, position information of the vibrating mirror 109, position information of the stage 117, and the like. In addition, various image processes including a filtering process, a binarization process, and a threshold comparison process are performed on the generated two-dimensional image to detect a defect and acquire coordinates of the defect position. Step bunching can be detected separately from other defects such as lattice defects.

  SiC substrate 118 is transparent in the wavelength range of visible light. For this reason, when the surface of the SiC substrate 118 is scanned with a light beam, the incident light beam passes through the inside of the SiC substrate 118, and the reflected light reflected by the back surface of the SiC substrate 118 enters the detector, thereby reducing the resolution. There is an inconvenience. For this reason, both the case where the SiC substrate 118 is imaged by the differential interference microscope and the case where the defect is detected by the laser scattering method have a drawback that the resolution is low and the accuracy of the defect detection is lowered. On the other hand, this conventional confocal defect inspection apparatus 100 has a configuration that is almost equivalent to that in which a pinhole is arranged on the front surface of the linear image sensor 124, so that it passes through the SiC substrate 118 and on the back surface. The reflected light deviates from the optical path and does not enter the light receiving element of the linear image sensor 124, and only the reflected light reflected by the surface of the SiC substrate 118 enters the linear image sensor 124. As a result, by using the defect inspection apparatus 100, it is possible to capture a confocal differential interference image with a resolution higher than that of a differential interference image obtained by a general differential interference microscope, and detect defects with higher detection accuracy. It is possible.

  The differential interference optical system detects minute unevenness of about several nanometers formed on the sample surface as a phase difference, so that the epitaxial layer is caused by crystal defects existing on the substrate such as edge dislocation defects and dislocations in the basal plane. When unevenness of about several nm is formed on the surface, these unevenness can be detected as a luminance image. Furthermore, with respect to protrusion defects and concave defects, upward and downward slopes are detected as low-intensity images or high-intensity images on the differential interference image, so the brightness changes in brightness displayed on the captured differential interference image It is possible to determine whether it is a concave defect or a convex defect based on the above. Therefore, it is possible to easily determine whether the defect is a pit defect or a bump defect based on the captured bright and dark differential interference image.

  Furthermore, the step bunching is a linear step defect extending in a direction orthogonal to the orientation flat and having a height of about several nanometers to several tens of nanometers. Therefore, by capturing a confocal differential interference image on the surface of the epitaxial layer, it is detected as a linear luminance image of low luminance (dark) or high luminance (bright). In the case where both lattice defects and step bunching are formed in the epitaxial layer, on the confocal differential interference image, a form in which a point-like or specific luminance image and a linear luminance image are mixed. Detected as

JP 2011-220757 A

  In the conventional defect inspection apparatus 100 disclosed in Patent Document 1, the shearing direction of the differential interference optical system is stepped by adjusting the shearing direction of the differential interference optical system or by adjusting the rotation angle of the stage that supports the substrate. Set parallel to the extending direction of bunching (surface morphology). Thus, it is necessary to set the shearing direction in advance in the extending direction of the step bunching (surface morphology) of the workpiece (SiC substrate 118). Further, when the extending direction of step bunching (surface morphology) in the workpiece (SiC substrate 118) is changed or jagged, there is a problem that it is difficult to cope with defect inspection.

  As shown in the raw image and the defect map in FIG. 9, when the extending direction of step bunching (surface morphology) changes or becomes jagged, the surface morphology F (periodic surface) of the defective portion and the surrounding jagged surface is obtained. It is difficult to obtain true defect information. In short, the above-described conventional defect inspection method that directly generates a defect map from a raw image has a problem that it is difficult to obtain true defect information because it is easily influenced by surface morphology to detect a defect.

  The present invention solves the above-described conventional problems, and uses a defect inspection apparatus capable of detecting defects on the wafer surface without being affected by periodic surface irregularities (surface morphology) and the defect inspection apparatus. An object is to provide a defect inspection method.

  A defect inspection apparatus according to the present invention is a defect inspection apparatus that inspects a defect portion generated on a wafer surface, a raw image acquisition unit that acquires raw image data of the wafer surface, and a plurality of divisions obtained by dividing the raw image of the wafer surface An image cutout unit that sequentially cuts out image data of one area from the image data of the area, a Fourier transform unit that performs a Fourier transform process on the image data of the one area that is sequentially cut out, and one area that has undergone the Fourier transform process It is determined whether or not the image has periodicity with respect to the frequency data, and when it is determined that there is periodicity, it is determined that the area is “defect-free”, or when it is determined that there is no periodicity, “defect-free area” And the periodicity discriminating unit for discriminating that the above-mentioned object is achieved.

  Preferably, the defect inspection apparatus of the present invention further includes a coordinate specifying unit that specifies an XY coordinate position in the entire wafer area where the “defect-present area” determined by the periodicity determination unit is located.

  Still preferably, in a defect inspection apparatus according to the present invention, the raw image acquisition unit obtains raw image data of the wafer surface by scanning the wafer surface with laser light and photographing the reflected light.

  Still preferably, in a defect inspection apparatus according to the present invention, the raw image acquisition unit photographs the wafer surface using an optical microscope to obtain raw image data of the wafer surface.

  Furthermore, it is preferable that the image cutout unit in the defect inspection apparatus of the present invention adjusts the size of the divided area so that at least the defective portion of the defect portion having no periodicity and its peripheral portion is included in the divided area. The ratio of at least the defect portion of the defect area having no periodicity and its peripheral portion to the divided area is set to 10% to 100%.

  Still preferably, in a defect inspection apparatus of the present invention, the wafer surface is an epitaxial layer surface grown on the wafer.

  Further preferably, the periodicity determination unit in the defect inspection apparatus of the present invention is configured such that the frequency component of the frequency data in the frequency data after the Fourier transform processing is at least higher than the threshold with reference to a threshold higher than the noise level. When the signal level is high, it is determined that there is periodicity, and when the signal level of the frequency component is less than the threshold, it is determined that there is no periodicity.

  The defect inspection method of the present invention is a defect inspection method for inspecting a defect portion occurring on a wafer surface, wherein a raw image acquisition step in which a raw image acquisition unit acquires raw image data on the wafer surface, and an image cutout unit includes the wafer An image cutout process for sequentially cutting out image data of one area from image data of a plurality of divided areas obtained by dividing the raw image on the surface, and a Fourier transform unit performs a Fourier transform process on the image data of one area cut out sequentially. When the Fourier transform process and the periodicity discriminating unit determine that the periodicity is present by determining the presence or absence of the periodicity of the image with respect to the frequency data of one area subjected to the Fourier transform process, It has a periodicity determination step for determining that the area is “no area” or determining that the area is “defect area” when it is determined that there is no periodicity. The objects can be achieved.

  Preferably, the coordinate specifying unit in the defect inspection method of the present invention further includes a coordinate specifying step of specifying an XY coordinate position in the entire wafer area where the “defect-present area” is located.

  Further preferably, in the raw image acquisition step in the defect inspection method of the present invention, the raw image acquisition unit scans the wafer surface with a laser beam and images the reflected light to reflect the raw surface of the wafer. Obtain image data.

  Further preferably, in the raw image acquisition step in the defect inspection method of the present invention, the raw image acquisition unit images the wafer surface using an optical microscope to obtain raw image data of the wafer surface.

  Further preferably, in the image cutout step in the defect inspection method of the present invention, the image cutout unit is configured so that at least the defect portion of the defect portion having no periodicity and its peripheral portion is included in the divided area. The size of the divided area is set, and the proportion of at least the defective portion of the defective area having no periodicity and its peripheral portion in the divided area is set to 10% to 100%.

  Further preferably, the wafer surface in the defect inspection method of the present invention is an epitaxial layer surface grown on the wafer.

  Further preferably, in the periodicity determining step in the defect inspection method of the present invention, the periodicity determining unit is configured based on a threshold at which the signal level of the frequency component in the frequency data after the Fourier transform processing is at least higher than the noise level. When the signal level of the frequency component is higher than the threshold, it is determined that there is periodicity, and when the signal level of the frequency component is less than the threshold, it is determined that there is no periodicity.

  With the above configuration, the operation of the present invention will be described below.

  In the present invention, in a defect inspection apparatus for inspecting the presence / absence of a defect occurring on the wafer surface, a raw image acquisition unit for acquiring raw image data on the wafer surface, and images of a plurality of divided areas obtained by dividing the raw image on the wafer surface An image cutout unit that sequentially cuts out image data of one area from data, a Fourier transform unit that performs Fourier transform processing on the image data of one area that is cut out sequentially, and frequency data of one area that has been subjected to Fourier transform processing On the other hand, the presence or absence of periodicity of the image is determined, and when it is determined that there is periodicity, it is determined that the area is “defect free”, or when it is determined that there is no periodicity, it is determined that it is “defect free area”. And a periodicity discriminating unit.

  As described above, when the periodicity determination unit 5 determines the presence or absence of periodicity of the image with respect to the frequency data of one area subjected to the Fourier transform processing, and determines that there is periodicity, the “defect-free area” When it is determined that there is no periodicity, it is easily determined that the area has a defect, so that defects on the wafer surface are detected without being affected by periodic surface irregularities (surface morphology). It becomes possible.

  As described above, according to the present invention, when the periodicity determination unit 5 determines the presence / absence of periodicity by determining the presence / absence of periodicity of the image with respect to the frequency data of one area subjected to the Fourier transform process, Wafer surface without being affected by periodic surface irregularities (surface morphology) because it is easily determined that the area is “defect free area” or is “defect free area” when it is determined that there is no periodicity. Defects can be detected.

It is a block diagram which shows the principal part control structural example of the defect inspection apparatus in Embodiment 1 of this invention. It is a figure for demonstrating operation | movement of the defect inspection apparatus of FIG. (A) is a plan view schematically showing an image of the coordinate position of a defective area in the entire screen of the wafer, and (b) is a diagram after Fourier transform processing of the AA ′ line image portion of the “defect-free area” in FIG. FIG. 8C is a diagram showing a signal level with respect to the frequency, and FIG. 8C is a diagram showing a signal level with respect to the frequency after the Fourier transform processing of the BB ′ line image portion of the “defect area” in FIG. (A) is a plan view of a photograph showing a raw image of the wafer surface in a predetermined divided area obtained by scanning the surface of the wafer with a laser beam and photographing the surface morphology, and (b) is a Fourier transform of the image data of the predetermined divided area in (a). It is a FTT image top view after processing. (A) is a plan view of a photograph showing a raw image of the wafer surface in a predetermined divided area obtained by scanning the wafer surface with a laser beam to photograph epi defects, and (b) is a Fourier transform of the image data in the predetermined divided area of (a). It is a FTT image top view after processing. (A) is a photographic plan view showing a raw image of the wafer surface in a predetermined divided area obtained by photographing the surface morphology of the wafer with an optical microscope, and (b) is after Fourier transform processing of image data of the predetermined divided area in (a). It is a FTT image top view of. (A) is a photographic plan view showing a raw image of the wafer surface in a predetermined divided area obtained by photographing an epi defect on the wafer surface with an optical microscope, and (b) is a Fourier transform process on the image data of the predetermined divided area in (a). It is a FTT image top view of. It is a principal part block diagram which shows an example of the conventional defect inspection apparatus currently disclosed by patent document 1. FIG. It is a top view which shows typically the predetermined area of the raw image in which a defect exists in jagged step bunching (surface morphology), and the defect map of all the images.

  Hereinafter, a first embodiment of the defect inspection apparatus of the present invention will be described in detail with reference to the drawings. In addition, each thickness, length, etc. of the structural member in each figure are not limited to the structure to illustrate from a viewpoint on drawing preparation.

(Embodiment 1)
FIG. 1 is a block diagram illustrating a configuration example of a main part control of the defect inspection apparatus according to the first embodiment of the present invention.

  In FIG. 1, a defect inspection apparatus 1 according to the first embodiment includes a raw image acquisition unit 2 that acquires raw image data on a wafer surface, and image data of one area from image data of a plurality of areas obtained by dividing the raw image of a wafer surface. The image cutout unit 3 that sequentially cuts out, the Fourier transform unit 4 that performs the Fourier transform process on the image data of one area that is sequentially cut out, and the frequency data that has been subjected to the Fourier transform process according to the presence or absence of periodicity A periodicity determining unit 5 for determining whether the area is “defect free area” or “defect present area”, and a coordinate specifying unit for specifying the XY coordinate position in the entire wafer area where the “defect area” is located 6 and inspecting the presence or absence of defects such as epitaxial layers on the wafer surface and the position thereof without being affected by periodic surface irregularities (surface morphology).

  In the defect inspection apparatus 1 according to the first embodiment, for example, a defect portion such as an epitaxial layer such as a concave portion or a convex portion and a peripheral portion thereof are different from other surface morphology portions having periodicity, and have periodic surface morphology. Paying attention to the fact that there is no part, inspect the presence / absence of defects and their position by inspecting the frequency data obtained by performing Fourier transform on the sequentially extracted image data of one area, and checking the presence / absence of periodicity. To do.

  The raw image acquisition unit 2 obtains raw image data on the wafer surface by light scattered light reflected by scanning the laser beam with the laser beam. The raw image acquisition unit 2 may obtain raw image data on the wafer surface by photographing the wafer surface with an optical microscope. In addition to the method for obtaining the raw image data, the conventional defect inspection apparatus 100 shown in FIG. 8 can be used. In short, the raw image data on the surface of the epitaxial layer may be obtained.

  The image cutout unit 3 divides the entire wafer area into a plurality of areas, and the size of the divided areas is set to such a size that the periodicity of the surface morphology can be confirmed. On the other hand, the smaller the divided area, the larger the proportion of the defective portion in the divided area, and the more easily the defective portion is detected. In short, the size of the divided area is good enough that the defective part of the concave part or convex part having no periodicity and its peripheral part can be accommodated in the divided area. The image cutout unit 3 sets the size of the divided area so as to include a defective portion having no periodicity and its peripheral portion, and at least the defective portion of the defective portion having no periodicity and its peripheral portion is a divided area. The ratio to 10% to 100%.

  When Fourier transform processing is performed on the surface morphology of the periodic shape in the image data of the divided area of a predetermined size from the raw image data on the entire surface of the wafer, the Fourier transform unit 4 periodically converts the FFT frequency image data (frequency data) after the Fourier transform processing. A pattern in which dots D are arranged (dot D in FIG. 2) appears. On the other hand, even if the image data of the divided area including the defect portion having no periodicity and the peripheral portion thereof is subjected to the fast Fourier transform process, only the noise N appears as shown in FIG. It does not appear.

  The periodicity discriminating unit 5 discriminates the presence / absence of periodicity in the FFT frequency image data (frequency data) subjected to the Fourier transform process. That is, the image data determined to have periodicity is determined as the “defect-free area”, and the image data determined to have no periodicity is determined as the “defect-free area”. Whether or not there is periodicity is the presence or absence of periodic dots D appearing in the FFT frequency image data (frequency data) after Fourier transform processing, but a threshold is set where the signal level of the FFT frequency component is higher than the noise level When the signal level of the FFT frequency component is higher than the threshold with reference to the threshold, it is determined that there is periodicity, and when the signal level of the FFT frequency component is less than the threshold, it is determined that there is no periodicity.

  The coordinate specifying unit 6 integrates each area after the periodicity determination processing and the determination result, and specifies the XY coordinate information of the “defect area” as the XY coordinates (defect address information) in the entire area of the wafer before the division. After that, it is stored in a memory (not shown).

  The operation of the above configuration will be described in detail below.

  FIG. 2 is a diagram for explaining the operation of the defect inspection apparatus of FIG.

  As shown in FIG. 2, the raw image acquisition unit 2 acquires a raw image of the wafer surface, and the image cutout unit 3 divides the raw image of the wafer surface into image data of a plurality of areas, and the divided image data of one area. Are sequentially cut out. The Fourier transform unit 4 performs Fourier transform processing on the sequentially cut image data, and the periodicity determination unit 5 performs periodicity on the FFT frequency image data (frequency data) subjected to the Fourier transform processing. The presence / absence of the image data is discriminated. The image data for which the periodicity is determined to be “present” is determined as the “defect free area”, and the image data for which the periodicity is determined to be “not present” is determined to be the “defect present area”. Further, after integrating the areas after the periodicity discrimination processing and the discrimination results, the coordinate specifying unit 6 uses the XY coordinate information of the “defect area” as the XY coordinates (defect address information) in the entire wafer area before division. Identify. This defective address information is stored in the memory.

  In this way, a raw image is divided into constituent elements (divided areas), and FFT processing (fast Fourier transform processing) is performed to identify constituent elements (divided areas) with inferior periodicity as defect address information of defective areas. .

  A surface morphology in which surface irregularities appear periodically in an epitaxial layer or the like used for a light emitting layer of a semiconductor light emitting element is generated. In contrast to the surface morphologies with periodic steps, defective parts such as voids themselves are inferior in periodicity although they are holes (concaves) or steps (convex parts). Is inferior in periodicity. When the defective portion having no periodicity and the peripheral portion thereof are Fourier transformed, almost no frequency component appears. On the other hand, when a periodic surface morphology portion is Fourier-transformed, dots D (patterns in which dots D are periodically arranged) appear as frequency components. A threshold (reference value for determining that there is periodicity) for the level of the frequency component corresponding to the dot D is set to determine the presence or absence of periodicity. Thereby, the presence or absence of a defect can be determined. The XY coordinate position of the defect area in which the defect has occurred can be stored in the memory as a defect MAP.

  3A is a plan view schematically showing an image of the XY coordinate position of the defective area in the entire screen of the wafer, and FIG. 3B is an AA ′ line image portion of the “defect-free area” in FIG. FIG. 3C is a diagram showing the signal level with respect to the frequency after the Fourier transform process, and FIG. 3C is a diagram showing the signal level with respect to the frequency after the Fourier transform process of the BB ′ line image portion of the “defect area” in FIG.

  As shown in FIG. 3A, the coordinate information (defect MAP) of the defective area is specified as the XY coordinates in the entire wafer area before division. The XY coordinates may be the center coordinate position of the “defect area”, or may be the upper right coordinate position or the upper left coordinate position of the divided defect area.

   As shown in FIG. 3B, in the frequency data after Fourier transform processing of the AA ′ line portion of the “defect-free area” in FIG. 2, the SN ratio between the noise N and the frequency component S is 1 or more and a predetermined value. If it is equal to or greater than the threshold value, it can be determined as “no defect area” because of periodicity.

  As shown in FIG. 3C, the frequency data after the Fourier transform processing of the BB ′ line portion of the “defect area” in FIG. 2 has only noise N, no frequency component S, and the SN ratio is less than a predetermined threshold value. Therefore, it can be determined that there is no periodicity in the image area as an “defective area”.

  4A is a photographic plan view showing a raw image of the wafer surface in a predetermined divided area obtained by scanning the surface of the wafer with a laser beam and photographing the surface morphology, and FIG. 4B is a predetermined divided area of FIG. 4A. It is a FTT image top view after carrying out the Fourier-transform process of the image data of.

  As shown in FIG. 4A, a raw image of the wafer surface is captured by light scattered light reflected by scanning the entire wafer area with a laser beam, and a predetermined divided area sequentially cut out from the entire wafer area. In the plan view of the photograph, a surface morphology with a step is periodically photographed.

  As shown in FIG. 4B, dots D appear in the FFT image after the Fourier transform processing is performed on the periodic image data of the surface morphology in FIG. The presence or absence of periodicity can be determined by setting a threshold for the signal level of the dot D.

  FIG. 5A is a plan view of a photograph showing a raw image of the wafer surface in a predetermined divided area obtained by scanning the wafer surface with a laser beam to photograph an epi defect, and FIG. 5B is a predetermined divided area in FIG. 5A. It is a FTT image top view after carrying out the Fourier-transform process of the image data of.

  As shown in FIG. 5A, a raw image of the wafer surface is captured by light scattered light reflected by scanning the entire wafer area with a laser beam, and a photograph of a predetermined divided area cut out from the entire wafer area. In the plan view, a concave portion of a defect portion on the surface of the epitaxial layer is photographed at the center portion. The concave portion of the defective portion has almost no periodicity.

  As shown in FIG. 5B, the FFT image after the Fourier transform processing of the image data of the defect portion on the surface of the epitaxial layer in FIG. 5A shows a clear periodicity as shown in FIG. The dot D does not appear, and it is determined that there is no periodicity, and it can also be determined that there is a defect in this predetermined divided area.

  6A is a photograph plan view showing a raw image of the wafer surface in a predetermined divided area obtained by photographing the surface morphology of the wafer with an optical microscope, and FIG. 6B is image data of the predetermined divided area in FIG. 6A. It is a FTT image top view after carrying out a Fourier-transform process.

  As shown in FIG. 6 (a), the surface morphology of the wafer surface is photographed with an optical microscope. In the photograph plan view of the image data of the predetermined divided area sequentially cut out from the image data of the whole wafer region photographed, A surface morphology with periodic steps is captured in the raw image.

  As shown in FIG. 6B, a dot D appears in the FFT image after the Fourier transform processing is performed on the surface morphology image data in FIG. 6A. The presence or absence of periodicity can be determined by setting a threshold for the signal level of the dot D.

  7A is a plan view of a photograph showing a raw image of the wafer surface in a predetermined divided area obtained by photographing an epi defect on the wafer surface with an optical microscope, and FIG. 7B is image data of the predetermined divided area in FIG. 7A. It is a FTT image top view after carrying out a Fourier-transform process.

  As shown in FIG. 7 (a), a raw image of the surface morphology of the entire wafer area is taken with an optical microscope. In the photograph plan view of predetermined divided areas sequentially cut out from the image data of the taken wafer whole area. The concave portion of the defect portion on the surface of the epitaxial layer is photographed in the central portion. There is almost no periodicity in the concave portion of the defective portion and its peripheral portion.

  As shown in FIG. 7B, the FFT image after Fourier transform processing of the image data of the defect portion on the surface of the epitaxial layer in FIG. 7A shows a clear periodicity as shown in FIG. 7A. Dots do not appear, and it is determined that there is no periodicity, so that it can be determined that there is a defect in this predetermined divided area.

  Here, the defect inspection method of the first embodiment will be described.

  The defect inspection method according to the first embodiment is a defect inspection method for inspecting a defect portion generated on the wafer surface. The raw image acquisition step in which the raw image acquisition unit 2 acquires a raw image of the wafer surface; However, the image cut-out step of sequentially cutting out image data of one area from the image data of a plurality of divided areas obtained by dividing the raw image of the wafer surface, and the Fourier transform unit 4 performs Fourier processing on the image data of one area cut out sequentially. The Fourier transform process for performing the transform process and the periodicity determination unit 5 determine that there is periodicity by determining the presence or absence of the periodicity of the image with respect to the frequency image data of one area subjected to the Fourier transform process. A periodicity determining step for determining that the area is “defect free area” or determining that the area is “defect free” when it is determined that there is no periodicity; And a coordinate specifying step of specifying the XY coordinate positions in the whole wafer region position "with defect area".

  In the raw image acquisition process, the raw image acquisition unit 2 can acquire raw image data of the wafer surface by light scattered light reflected by scanning the wafer surface with laser light. Alternatively, in the raw image acquisition step, the raw image acquisition unit 2 may acquire raw image data of the wafer surface by photographing the wafer surface with an optical microscope.

  As described above, according to the first embodiment, the raw image acquisition unit 2 that acquires the raw image data of the wafer surface and the image data of one area are sequentially obtained from the image data of a plurality of divided areas obtained by dividing the raw image of the wafer surface. The image cutout unit 3 to be cut out, the Fourier transform unit 4 that performs Fourier transform processing on the frequency image data (frequency data) of one area that is sequentially cut out, and the image data of one area that has been subjected to the Fourier transform processing The period for determining whether the image has periodicity, determining that the area is periodic with no periodicity, or determining that the area is defective with no periodicity And a coordinate specifying unit 6 for specifying the XY coordinate position in the entire wafer area where the “defect area” is located.

  As described above, when the periodicity determination unit 5 determines the presence or absence of the periodicity of the image with respect to the image data of one area that has been subjected to the Fourier transform process, Detects defects on the wafer surface without being affected by periodic surface irregularities (surface morphology) in order to easily determine that there is a “defect area” when it is determined that there is no periodicity. be able to. In the future, the present invention can also be applied to defects on the surface of fine irregularities for improving the light extraction efficiency.

Although not particularly described in the first embodiment, an image of one area from the raw image acquisition unit 2 that acquires a raw image of the wafer surface and image data of a plurality of divided areas obtained by dividing the raw image of the wafer surface. An image cutout unit 3 that sequentially cuts out data, a Fourier transform unit that performs a Fourier transform process on the cut out image data of one area, and an image periodicity of the image data of one area that has been subjected to the Fourier transform process A periodicity discriminating unit that discriminates the presence or absence of the periodicity and discriminates that the area has no defect, or discriminates the absence of the periodicity and discriminates that the area has a defect. Thus, the object of the present invention for detecting defects on the wafer surface without being affected by periodic surface irregularities (surface morphology) can be achieved.
In the method of the present invention, various SiC and GaN single crystal substrates on which an epitaxial layer is formed by step flow growth can be inspected.

  As mentioned above, although this invention has been illustrated using preferable Embodiment 1 of this invention, this invention should not be limited and limited to this Embodiment 1. It is understood that the scope of the present invention should be construed only by the claims. It is understood that those skilled in the art can implement an equivalent range from the description of the specific preferred embodiment 1 of the present invention based on the description of the present invention and the common general technical knowledge. Patents, patent applications, and documents cited herein should be incorporated by reference in their entirety, as if the contents themselves were specifically described herein. Understood.

  The present invention relates to a defect inspection apparatus for inspecting the presence or absence of defects such as an epitaxial layer on a wafer surface without being affected by periodic surface irregularities (surface morphology) and a defect inspection method using the defect inspection apparatus. When the discriminating unit 5 discriminates the presence or absence of periodicity of the image with respect to the image data of one area subjected to the Fourier transform processing and discriminates that there is a periodicity, When it is determined that there is no periodicity, it is easily determined that the area has a defect, so that a defect on the wafer surface can be detected without being affected by periodic surface irregularities (surface morphology).

DESCRIPTION OF SYMBOLS 1 Defect inspection apparatus 2 Raw image acquisition part 3 Image clipping part 4 Fourier transformation part 5 Periodicity discrimination part 6 Coordinate specification part D Dot F Jagged surface morphology

Claims (14)

In the defect inspection device that inspects the defects that occur on the wafer surface
A raw image acquisition unit that acquires raw image data of the wafer surface, an image cutout unit that sequentially cuts out image data of one area from image data of a plurality of divided areas obtained by dividing the raw image of the wafer surface, and cut out sequentially A Fourier transform unit that performs Fourier transform processing on image data in one area, and the presence / absence of periodicity in the frequency data of one area that has undergone Fourier transform processing is determined. A defect inspection apparatus including a periodicity determining unit that determines that the area is “defect free area” when it is detected, or that it is determined as “area with defect” when it is determined that there is no periodicity.   2. The defect inspection apparatus according to claim 1, further comprising a coordinate specifying unit that specifies an XY coordinate position in the entire wafer region where the “defect-present area” determined by the periodicity determination unit is located.   The defect inspection apparatus according to claim 1, wherein the raw image acquisition unit obtains raw image data of the wafer surface by scanning the wafer surface with laser light and photographing the reflected light.   The defect inspection apparatus according to claim 1, wherein the raw image acquisition unit obtains raw image data of the wafer surface by photographing the wafer surface using an optical microscope.   The image cutout unit sets the size of the division area so that at least the defect portion of the defect portion having no periodicity and its peripheral portion is included in the division area, and the defect area having no periodicity 2. The defect inspection apparatus according to claim 1, wherein a ratio of at least the defective portion of the peripheral portion to the divided area is 10 percent or more and 100 percent or less.   The defect inspection apparatus according to claim 1, wherein the wafer surface is an epitaxial layer surface grown on the wafer.   The periodicity determination unit has periodicity when the signal level of the frequency component in the frequency data after the Fourier transform processing is higher than the threshold with reference to a threshold that is at least higher than the noise level. The defect inspection apparatus according to claim 1, wherein the defect inspection apparatus determines that there is no periodicity when the signal level of the frequency component is less than the threshold value. In a defect inspection method for inspecting a defect portion generated on a wafer surface,
A raw image acquisition step in which a raw image acquisition unit acquires raw image data on the wafer surface, and an image cutout unit sequentially outputs image data of one area from image data of a plurality of divided areas obtained by dividing the raw image on the wafer surface. An image cutout step to be cut out, a Fourier transform step in which a Fourier transform unit performs a Fourier transform process on the image data of one area that has been cut out sequentially, and a periodicity discriminating unit in one area on which the Fourier transform process has been performed. By determining the presence or absence of periodicity of the image with respect to the frequency data, it is determined that the area has no periodicity, and it is determined that the area has no defect, or if the periodicity is determined to have no periodicity A defect inspection method including a periodicity determining step of determining that the area is “area”.   The defect inspection method according to claim 8, further comprising: a coordinate specifying step in which the coordinate specifying unit specifies an XY coordinate position in an entire wafer area where the “defect area” is located.   The defect according to claim 8, wherein in the raw image acquisition step, the raw image acquisition unit obtains raw image data of the wafer surface by scanning the wafer surface with laser light and photographing the reflected light. Inspection method.   The defect inspection method according to claim 8, wherein in the raw image acquisition step, the raw image acquisition unit captures the wafer surface using an optical microscope to obtain raw image data of the wafer surface.   In the image cutout step, the image cutout unit sets the size of the divided area so that at least the defective portion of the defect portion having no periodicity and the peripheral portion thereof is included in the divided area; and The defect inspection method according to claim 8, wherein a ratio of at least the defect portion of the defect area having no periodicity and its peripheral portion to the divided area is 10% to 100%.   The defect inspection method according to claim 8, wherein the wafer surface is an epitaxial layer surface grown on the wafer.   In the periodicity determining step, the periodicity determining unit determines that the signal level of the frequency component is higher than the threshold with reference to a threshold at which the signal level of the frequency component in the frequency data after Fourier transform processing is at least higher than the noise level. The defect inspection method according to claim 8, wherein when it is high, it is determined that there is periodicity, and when the signal level of the frequency component is less than the threshold value, it is determined that there is no periodicity.