Defect classification method, analysis method, and inspection apparatus

Abstract

This invention provides a defect classification method, analysis method, and inspection apparatus that improve the detection accuracy of defects, including stacking faults, in SiC substrates. [Solution] The defect classification method includes the steps of: projecting an illumination beam onto a silicon carbide substrate and scanning the position where the illumination beam illuminates the silicon carbide substrate; detecting reflected light and photoluminescence light including the visible range emitted from the silicon carbide substrate; and classifying defects based on the detection results of the reflected light and the detection results of the photoluminescence light including the visible range. The defect classification step includes classifying the defect as a 3C-SF (Stacking Fault) based on the detection of a first defect image by reflected light, and classifying the defect as an SSF (Shockley-type Stacking Fault) based on the detection of a second defect image by photoluminescence light including the visible range.

Description

Technical Field

[0001] The present disclosure relates to a defect classification method, an analysis method, and an inspection apparatus.

Background Art

[0002] A wafer obtained by homoepitaxially growing SiC on a silicon carbide (hereinafter also referred to as SiC) wafer is used as a power semiconductor. The epitaxial film is used as a drift layer of a device, and crystal defects existing in the epitaxial film layer cause device failures. However, not all crystal defects cause device failures, and it is known that specific types of crystal defects cause device failures.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Inspection using a near-infrared light receiving filter that transmits photoluminescence (hereinafter also referred to as PL) light from a basal plane dislocation (hereinafter also referred to as BPD), which is a killer defect, is known. However, there is a need for accurate detection of stacking faults and classification of stacking faults by type.

[0005] The present disclosure has been made in view of such problems, and provides a defect classification method, an analysis method, and an inspection apparatus that improve the classification accuracy of defects including stacking faults in a SiC substrate.

Means for Solving the Problems

[0006] A defect classification method according to one aspect of this embodiment is: A defect classification method for classifying the types of defects present in a silicon carbide substrate, A step of projecting an illumination beam onto the silicon carbide substrate and scanning the position on the silicon carbide substrate to be illuminated by the illumination beam, A step of detecting reflected light and photoluminescent light including the visible range emitted from the silicon carbide substrate, The process includes classifying the defects based on the detection results of the reflected light and the detection results of the photoluminescence light including the visible range, The step of classifying the defects includes classifying the defects as 3C-SF (Stacking Fault) based on the detection of a first defect image by the reflected light, and classifying the defects as SSF (Shockley-type Stacking Fault) based on the detection of a second defect image by photoluminescence light including the visible range.

[0007] An analysis method according to one aspect of this embodiment is: A method for analyzing defects present in a silicon carbide substrate, A step of projecting an illumination beam toward the silicon carbide substrate and scanning the position on the silicon carbide substrate to be irradiated by the illumination beam, A step of detecting photoluminescent light, including at least the visible and infrared regions, emitted from the silicon carbide substrate, A process to analyze the cause of SSF (Shockley-type Stacking Fault) based on the detection results of photoluminescence light in the visible range and photoluminescence light in the infrared range. Includes.

[0008] An inspection apparatus according to one aspect of this embodiment is: An inspection device for classifying the types of defects present in a silicon carbide substrate, An illumination means for projecting an illumination beam toward the silicon carbide substrate, A scanning means for scanning the position where the silicon carbide substrate is illuminated by the illumination beam, A photodetection means for detecting reflected light and photoluminescent light including the visible range emitted from the silicon carbide substrate, The system includes a classification means for classifying the defects based on the detection results of the reflected light and the detection results of the photoluminescence light including the visible range, The classification means classifies the defect as a 3C-SF (Stacking Fault) based on the detection of a first defect image by the reflected light, and classifies the defect as an SSF (Shockley-type Stacking Fault) based on the detection of a second defect image by photoluminescence light including the visible range.

[0009] An inspection apparatus according to one aspect of this embodiment is: An inspection device for analyzing defects present in a silicon carbide substrate, An illumination means for projecting an illumination beam toward the silicon carbide substrate, A scanning means for scanning the position where the illumination beam irradiates the silicon carbide substrate, A detection means for detecting photoluminescent light, including at least the visible and infrared regions, emitted from the silicon carbide substrate, An analytical means for analyzing the cause of SSF (Shockley-type Stacking Fault) based on the detection results of photoluminescence light in the visible range and photoluminescence light in the infrared range. It is equipped with. [Effects of the Invention]

[0010] According to this disclosure, it is possible to provide a defect classification method, an analysis method, and an inspection apparatus that improve the accuracy of classifying defects, including stacking faults, in SiC substrates. [Brief explanation of the drawing]

[0011] [Figure 1] This is a diagram illustrating the configuration of the light source device according to Embodiment 1. [Figure 2] This is a diagram illustrating the configuration of the signal processing device according to Embodiment 1. [Figure 3] It is a diagram for explaining an example of a method for classifying SSF. [Figure 4] It is a diagram for explaining SSF derived from BPD. [Figure 5] It is a flowchart showing a defect classification method according to Embodiment 1. [Figure 6] It is a diagram showing an example of an inspection apparatus for implementing a defect classification method according to Embodiment 2. [Figure 7] It is a flowchart showing a defect classification method according to Embodiment 2.

Mode for Carrying Out the Invention

[0012] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. The following description shows preferred embodiments of the present disclosure, and the scope of the present disclosure is not limited to the following embodiments. In the following description, those denoted by the same reference numerals indicate substantially the same content.

[0013] Embodiment 1 FIG. 1 is a diagram showing an example of an inspection apparatus for implementing a defect classification method according to Embodiment 1. In this example, a confocal scanning apparatus having a differential interference optical system is used to scan the entire surface of the SiC substrate to be inspected, and the reflected light and PL light emitted from the SiC substrate are detected individually, and the reflected light inspection and the PL light inspection are performed in parallel. A SiC substrate on which an epitaxial layer is formed is used as an inspection object, and defects formed in the epitaxial layer are detected and classified. Note that a SiC substrate on which no epitaxial layer is formed is also used as an inspection object. The SiC substrate may be a 4H-SiC substrate.

[0014] In this example, a first illumination system for forming a reflected image and a second illumination system for forming a PL image are used. In the first illumination system, the substrate surface is scanned using an illumination beam in the visible region, and in the second illumination system, the substrate surface is scanned using an illumination beam in the ultraviolet region. The first and second illumination beams illuminate the same point on the SiC substrate.

[0015] In this example, a mercury xenon lamp is used as the illumination light source 50, and the 546 nm visible light emitted from the mercury xenon lamp is used as the illumination beam of the first illumination system to perform reflected light inspection. In addition, ultraviolet light with a wavelength of 313 nm is used as the illumination beam of the second illumination system to perform PL light inspection. The light beam emitted from the illumination light source 50 propagates through the first optical fiber 51. An optical fiber coupler 52 is coupled to the output end of the first optical fiber. The second and third optical fibers 53 and 54 are coupled to the output side of the optical fiber coupler 52. The light beam is split into two by the optical fiber coupler 52; one beam propagates through the second optical fiber 53 and is used for reflected light inspection, and the other beam propagates through the third optical fiber 54 and is used for PL light inspection. Note that the light source emitting the light used for reflected light inspection and the light source emitting the light used for PL light inspection may be provided separately. In this case, the inspection device does not need to be equipped with an optical fiber coupler 52.

[0016] The light beam emitted from the second optical fiber 53 is incident on a filter 55 that transmits light with a wavelength of 546 nm, and a first illumination beam with a wavelength of 546 nm is emitted. The light beam emitted from the third optical fiber 54 passes through a filter 56 that transmits light with a wavelength of 313 nm and is incident on a polarizer 57. The polarizer 57 emits a light beam that is P-polarized from the incident light beam. Therefore, a second illumination beam with a wavelength of 313 nm and P-polarized is emitted from the polarizer 57. The second illumination beam is projected onto the SiC substrate 16 as a focused illumination beam via a focusing lens 58.

[0017] The second illumination system projects a P-polarized illumination beam toward the SiC substrate 16 at an incident angle equal to the Brewster angle. Therefore, the optical axes of the third optical fiber 54, filter 56, polarizer 57, and lens 58 constituting the second illumination system are arranged at an angle equal to the Brewster angle. When a P-polarized illumination beam is projected at an incident angle equal to the Brewster angle, reflected light is almost zero, and most of the illumination beam incident on the SiC substrate 16 penetrates into the interior of the SiC substrate 16. Therefore, a higher intensity PL light can be generated, and a clear PL image is formed. As a result, the accuracy of defect image detection is improved, and the advantage of highly accurate defect classification is achieved.

[0018] The first illumination beam emitted from the filter 55 is converted into a parallel beam by the focusing lens 4 and incident on the slit 5. The slit 5 is positioned at the pupil of the lens 4 and has an elongated opening extending in a first direction (a direction perpendicular to the plane of the paper). Here, the first direction is referred to as the X direction. The width of the opening of the slit 5 is set to, for example, 10 to 20 μm. Therefore, an elongated linear light beam extending in the first direction is emitted from the slit 5. The linear light beam emitted from the slit 5 is incident on the polarizer 6 and converted into polarized light with a single plane of vibration. This linear polarized beam is reflected by the half mirror 7, which functions as a beam splitter, and incident on the vibrating mirror 9, which functions as a scanning device, via the lens 8.

[0019] The vibrating mirror 9 functions as a beam scanning device that deflects the illumination beam in a second direction (Y direction) perpendicular to the first direction when reviewing the surface image of the SiC substrate 16. During inspection, the entire surface of the SiC substrate 16 is scanned by the two-dimensional movement of the stage 15 supporting the SiC substrate 16, so the vibrating mirror 9 functions as a total internal reflection mirror during inspection. The linear light beam emitted from the vibrating mirror 9 passes through lenses 10 and 11 and is incident on the dichroic mirror 59. The dichroic mirror 59 performs the function of separating the reflected light and PL light emitted from the SiC substrate 16. Therefore, a dichroic mirror that transmits light with a wavelength of 546 nm ± 20 nm and reflects light of other wavelengths is used. The illumination beam that has passed through the dichroic mirror 59 is incident on the Nomarski prism 13. In this example, the Nomarski prism 13 is used as a differential interference optical system. The linear illumination beam incident on the Nomarski prism 13 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 defects with height variations of several nm formed on the surface of the SiC substrate 16 as a brightness image of light and dark. The shearing amount of the Nomarski prism 13 is set to, for example, 2 μm. The Nomarski prism 13 is positioned to be insertable and detachable from the optical path. It is inserted into the optical path when capturing a confocal differential interference image of the SiC substrate 16, and removed from the optical path in other cases, such as when capturing a three-dimensional confocal image of the SiC substrate 16 or when capturing a surface contour image of the SiC substrate 16.

[0020] Two sub-beams emitted from the Nomarski prism 13 are incident on the objective lens 14. The objective lens 14 focuses the two incident line-shaped sub-beams and projects them onto the SiC substrate 16 placed on the stage 15. In this example, a SiC substrate 16 with an epitaxial layer is used as the substrate to be inspected, and defects present in the epitaxial layer are detected. The stage 15 is composed of an XY stage, and its position information is detected by a position sensor 17, and the position information of the stage 15 is supplied to the signal processing device 22. During inspection, the stage 15 moves in a zigzag pattern in the Y and X directions, and the entire surface of the SiC substrate 16 is scanned by the illumination beam. When reviewing a defect using the address of the detected defect, the stage is moved in the X and Y directions based on the coordinate information of the defect to move the defect into the field of view, and a confocal differential interference image of the defect and its vicinity can be acquired by scanning the vibrating mirror 9 in a second direction.

[0021] When a defect appears on the surface of the SiC substrate 16 as irregularities of a few nanometers to several hundred nanometers, a phase difference corresponding to the change in the height of the defect is introduced between the two subbeams reflected from the surface of the SiC substrate 16. This forms two reflected subbeams containing phase difference information corresponding to the height of the defect present on the surface of the SiC substrate 16. These two reflected subbeams are focused by the objective lens 14 and incident on the Nomarski prism 13. The two reflected subbeams are then combined by the Nomarski prism 13 to form an interference beam containing phase difference information indicating the change in the height of the surface of the SiC substrate 16. For example, if there is a concave or convex defect of a few nanometers on the surface of the SiC substrate 16, one of the two subbeams incident on the surface of the SiC substrate 16 scans over the defect, while the other subbeam scans a normal surface area. As a result, a phase difference corresponding to the height of the defect is introduced between the two subbeams. Consequently, the defect appearing on the surface of the SiC substrate 16 is detected as a brightness image.

[0022] If crystal defects exist within the epitaxial layer of the SiC substrate 16, the second illumination beam projected through the lens 58 penetrates the surface of the SiC substrate 16, enters the interior, and is incident on the defects. When ultraviolet light is incident on a defect, PL light of various wavelengths is generated depending on the type of defect. According to the description in IEC 63068-3 ANNEX B, the wavelength of PL light from BPD (Basal Plane Dislocation), one of the defects in the SiC substrate 16, is approximately 700 nm. SSF (Shockley-type Stacking Fault), another type of defect in the SiC substrate 16, is classified into 1SSF to 4SSF. The wavelength of PL light from 1SSF is approximately 420 nm. The wavelength of PL light from 2SSF is approximately 500 nm. The wavelength of PL light from 3SSF is approximately 480 nm. The wavelength of PL light from 4SSF is approximately 455 nm.

[0023] The reflected light and PL light emitted from the SiC substrate 16 are focused by the objective lens 14, passed through the Nomarski prism 13, and incident on the dichroic mirror 59. The dichroic mirror 59 transmits light with a wavelength of 546 nm ± 20 nm and reflects light of other wavelengths. Therefore, the PL light generated by the BPD and SSF is reflected by the dichroic mirror 59. On the other hand, the reflected light reflected from the surface of the SiC substrate 16 is transmitted through the dichroic mirror 59.

[0024] The reflected light that passes through the dichroic mirror 59 passes through lenses 11 and 10, is reflected by the vibrating mirror 9, and enters lens 8. This lens 8 acts as an imaging lens for the reflected light from the SiC substrate 16. The reflected light that has passed through lens 8 passes through the half mirror 7 and enters the first photodetector 20 via the analyzer 18 and positioner 19. In this example, the first photodetector 20 is composed of a line sensor. The image signal output from the line sensor, which is an imaging device, is supplied to the signal processing device 22 via the amplifier 21.

[0025] The PL light reflected by the dichroic mirror 59 enters the filter 24 via the lens 23. The filter 24 transmits light with wavelengths including the visible and infrared regions. The infrared region represents the wavelength range greater than 700 nm.

[0026] Filter 24 transmits light of at least the wavelength of PL light from the SSF and light of the wavelength of PL light from the BPD. The emission intensity of PL light from 1SSF, 2SSF, 3SSF, and 4SSF peaks in the range of 420-500 nm, and the wavelength of PL light from the BPD peaks at approximately 700 nm. Therefore, filter 24 is configured to transmit light of wavelengths from 400-520 nm and light of wavelengths above 680 nm.

[0027] The second light detection means 25 is comprised of a line sensor. The image signal output from the second light detection means 25 is supplied to the signal processing device 22 via the amplifier 28.

[0028] Figure 2 shows an example of a signal processing device 22. In this example, defect detection by reflected light inspection and defect detection by PL light inspection are performed in parallel, and the detected defects are classified based on the results of these inspections. The image signals output from the first and second optical detection means 20 and 25 are supplied to A / D converters 30 and 31, respectively, and converted into digital signals.

[0029] The image signal output from the first light detection means 20 is supplied to the first image forming means 33 to form a two-dimensional image (reflection image) using reflected light. The second image signal output from the second light detection means 25 is supplied to the second image forming means 34 to form a two-dimensional image (PL image) using PL light including the visible range.

[0030] The reflected image signal and the PL image signal are supplied to the defect image detection means 36. The defect image detection means 36 compares the input reflected image signal and PL image signal with a reference brightness value for each pixel and detects pixels that show brightness values ​​outside the range of the reference brightness value to form a defect image. For example, a defect image is formed by mapping pixels that are outside the range of the reference brightness value. A defect image may also be detected by mapping pixels in which an image signal with brightness exceeding the reference value is detected.

[0031] The defect image detection means 36 is also supplied with address information indicating the position of the SiC substrate 16 onto which the illumination beam is incident. This address information may be determined using position information of the stage supporting the SiC substrate 16 and the pixel information of the first and second optical detection means 20 and 25. The detected defect image is stored in the defect data memory 37 along with the address information. After the inspection of the entire surface of the SiC substrate 16 is completed, the operator can access the defect data memory 37 using the defect address to display and observe the desired defect image on the monitor.

[0032] The detected defect image is supplied to the defect classification means 40. If a triangular defect image is detected in the reflected image, the defect classification means 40 classifies the defect as 3C-SF (also called polyinclusion). The defect classification means may also classify the defect as 3C-SF by taking into account the presence of particles at the vertices of the triangle.

[0033] The defect classification means 40 may classify defects that show a triangular defect image in the reflection image and a dark triangular defect image in the PL image as 3C-SF. If the peak wavelength of the PL light from 3C-SF is not included in the transmission wavelength range of the filter 24, the defect image of 3C-SF in the PL image will appear darker against the background.

[0034] The defect classification means 40 classifies a defect as an SSF (Single Segmented Failure) if a bright triangular defect image is detected in the PL (Plant-Light) image. SSFs include 1SSF to 4SSF. Since the filter 24 transmits PL light from the SSFs, the SSF defect images in the PL image appear brighter against the background.

[0035] In related technologies, the filter 24 did not transmit PL light from the SSF, so the SSF was detected as a dark triangular region. However, this resulted in low detection accuracy due to low contrast. Embodiment 1 uses a filter 24 that transmits PL light from the SSF, thus improving the detection accuracy of the SSF.

[0036] Furthermore, if a rectangular (e.g., trapezoidal, rhombus-shaped) defect image is detected in the PL image, the defect classification means 40 may classify that defect as an SSF. It is known that BPDs can grow into SSFs, in which case a rectangular defect image may be detected in the PL image. Because the contrast of the defect image has been improved, Embodiment 1 can detect SSFs originating from BPDs.

[0037] Detecting SSF from reflection images is difficult, but Embodiment 1 can detect SSF using PL images. The defect classification means 40 may also classify defects as SSF, taking into account the detection of defect images representing one or two sides of a triangle in the reflection image.

[0038] The defect classification means 40 may classify SSFs into 1SSF to 4SSF. The defect classification means 40 can classify SSFs into 1SSF to 4SSF based on the detection results for each wavelength of PL light.

[0039] The defect classification means 40 can classify SSFs into 1SSF to 4SSF based on PL images for each wavelength. For example, the second photodetector 25 may be a spectral camera. Alternatively, by placing a filter corresponding to one of the PL wavelengths from 1SSF to 4SSF between the filter 24 and the second photodetector 25 in Figure 1, PL images for each wavelength can be captured. Alternatively, a prism or color filter may be provided with the second photodetector 25 to capture a color image, and the SSFs may be classified into 1SSF to 4SSF based on the color differences of each defect image.

[0040] As shown in Figure 3, by increasing the number of sensors, it is also possible to classify SSFs into 1SSF to 4SSF. The dichroic mirror 60 transmits light with wavelengths greater than A and reflects light with wavelengths less than A. A is set to be greater than the PL wavelength of 3SSF (e.g., 480nm) and less than the PL wavelength of 2SSF (e.g., 500nm). In this case, the SSF detected using the second photodetection means 25 is classified as 2SSF. The dichroic mirror 61 transmits light with wavelengths greater than B and reflects light with wavelengths less than B. B is set to be greater than the PL wavelength of 1SSF (e.g., 420nm) and less than the PL wavelength of 4SSF (455nm). The SSF detected using sensor 25a is classified as 1SSF. A filter that transmits PL light from 1SSF may be placed between sensor 25a and dichroic mirror 61. The dichroic mirror 62 transmits light with wavelengths greater than C and reflects light with wavelengths less than C. C is set to be greater than the PL wavelength of 4SSF (e.g., 455 nm) and less than the PL wavelength of 3SSF (e.g., 480 nm). SSF detected using sensor 25b is classified as 4SSF. A filter that transmits light from 4SSF may be placed between sensor 25b and the dichroic mirror 62. SSF detected using sensor 25c is classified as 3SSF. A filter that transmits light from 3SSF may be placed between sensor 25c and the dichroic mirror 62.

[0041] Referring again to Figure 2, the defect classification means 40 classifies a defect as a BPD if a bright, linear defect image is detected in the PL image. Because the filter 24 transmits PL light from the BPD, the BPD defect image in the PL image appears brighter against the background.

[0042] The defect classification means 40 analyzes whether the defects classified as SSF originated from BPD or were generated during epitaxial growth. In other words, the defect classification means 40 classifies the detected defects into either SSF originated from BPD or SSF generated during epitaxial growth. Specifically, the defect classification means 40 determines that a defect originated from BPD if one side of the triangle or quadrilateral representing the defect image of the SSF in the PL image is thicker than the other sides.

[0043] According to this embodiment, both defect images classified as BPD and defect images classified as SSF can be detected with high contrast. Referring to Figure 4, a bright rectangular (e.g., trapezoidal) defect image 71 is detected in the PL image, with one side 72 of the rectangle being thicker than the other sides. The rectangular (e.g., trapezoidal) shape is a defect image detected based on the PL wavelength light of the SSF, and the thick side 72 is a line-shaped defect image detected based on the detection of PL wavelength light of the BPD. Since the defect images overlap, it is thought that this SSF was formed when a BPD that was present on side 73 moved toward side 72. Thus, the defect classification means 40 determines that an SSF is caused by a BPD when one side of the shape of a defect image classified as an SSF is thicker than the other sides. In other words, the defect classification means 40 determines that an SSF is caused by a BPD when a defect image classified as a BPD is detected at a position corresponding to one side of the shape of a defect image classified as an SSF. Although the explanation used a rectangular defect image as an example, the same applies to a triangular defect image. That is, the defect classification means 40 may determine that an SSF caused by BPD is detected in the PL image when a bright triangular defect image is detected and one side of the triangle is thicker than the other sides.

[0044] Referring again to Figure 2, the defect classification means 40 may classify each defect based on the learning results obtained by deep learning or the like. The defect data memory 37 stores defect data, including the defect classification and address of each defect image.

[0045] The defect classification is supplied to the output means 41 along with address information. The operator inputs specified information, identifying the desired type and mode of occurrence of the defect, via an input device such as a keyboard.

[0046] Figure 5 is a flowchart showing the flow of the defect classification method according to Embodiment 1. First, the entire surface of the SiC substrate is scanned using an illumination beam (step S101). Then, the reflected light and PL light emitted from the SiC substrate are detected individually to form a reflected image and a PL image (step S102).

[0047] A reflected light inspection is performed on the formed reflected image to detect defects. In parallel, a PL light inspection is performed on the PL image to detect defects (step S103).

[0048] Next, the defects are classified based on the defect image (Step S104). For example, the defects are classified as 3C-SF based on the reflection image and as SSF based on the PL image.

[0049] In step S104, the cause of the SSF may be analyzed. For example, if the defect image of the SSF includes a defect image corresponding to the BPD (if a defect image classified as a BPD is detected at a position corresponding to one side of the shape of the defect image classified as an SSF), it may be determined that the SSF was caused by a BPD.

[0050] The defect classification method according to Embodiment 1 can improve the accuracy of defect classification. Furthermore, the defect classification method according to Embodiment 1 can analyze the causes of SSF occurrence.

[0051] Embodiment 2 Embodiment 2 does not require reflected light inspection and primarily performs PL light inspection. Figure 6 shows an example of an inspection apparatus for implementing the defect classification method according to Embodiment 2. Comparing Figure 3 and Figure 6, the inspection apparatus shown in Figure 6 does not have a first illumination system for forming a reflected image. Note that, as shown in Figure 1, the inspection apparatus may not have sensors 25a to 25d, and may only have sensor 25.

[0052] The dichroic mirror 59 transmits light with wavelengths of 700 nm or longer, for example, and reflects light of other wavelengths. The signal processing device 22 classifies a defect as a BPD when a bright, linear defect is detected in the PL image based on the image signal output from sensor 25d. Similar to Embodiment 1, the signal processing device 22 classifies a defect as 2SSF, 1SSF, 4SSF, or 3SSF, respectively, when a bright, triangular or square-shaped defect is detected in the PL image based on the image signals output from sensors 25, 25a, 25b, and 25c.

[0053] The signal processing device 22 determines that an SSF is caused by a BPD if a defect image classified as a BPD is detected at a position corresponding to one side of the shape of a defect image classified as one of 1SSF to 4SSF. Alternatively, the signal processing device 22 may determine that an SSF is caused by a BPD if one side of a defect image classified as an SSF is thicker than the other sides.

[0054] Figure 7 is a flowchart showing the flow of the defect classification method according to Embodiment 2. First, the entire surface of the SiC substrate is scanned using an illumination beam (step S201). Then, PL light emitted from the SiC substrate is detected and a PL image is formed (step S202). PL light inspection is performed from the PL image to detect a defect image (step S203). Subsequently, the cause of SSF is analyzed based on the defect image (step S204). Specifically, the signal processing device 22 determines whether or not the SSF was caused by BPD.

[0055] The defect classification method according to Embodiment 2 can analyze the causes of SSF occurrence.

[0056] While embodiments of this disclosure have been described above, this disclosure includes appropriate modifications that do not impair its purpose and advantages, and is not limited by the embodiments described above. Furthermore, combinations of the configurations of Embodiments 1 and 2 also fall within the scope of the technical concept of this disclosure. [Explanation of symbols]

[0057] 50 illumination light source 51 The first optical fiber 52 Fiber Optic Coupler 53. The second optical fiber 54 The third optical fiber 24, 55, 56 filters 4, 8, 10, 11, 23, 58 lenses 59, 60, 61, 62 Dichroic mirror 5 slits 6.57 Polarizer 7 Half Mirror 9. Vibrating Mirror 13 Nomarski prism 14 Objective lens 15 stages 16 SiC substrates 17 Position Sensor 18. Photometer 19 Positioner 20 First light detection means 21, 28 Amplifier 25 Second light detection means 30, 31 A / D converters 33 First Image Forming Means 34 Second image forming means 36 Defect image detection means 37. Defective data memory 40 Defect classification method 41 Output means 25a, 25b, 25c, 25d sensors 71 Defective Image 72, 73 sides

Claims

1. A defect classification method for classifying the types of defects present in a silicon carbide substrate, A step of projecting an illumination beam onto the silicon carbide substrate and scanning the position on the silicon carbide substrate to be illuminated by the illumination beam, A step of detecting reflected light and photoluminescent light including the visible range emitted from the silicon carbide substrate, The process includes classifying the defects based on the detection results of the reflected light and the detection results of the photoluminescence light including the visible range, The process of classifying the defects includes classifying the defects as 3C-SF (Stacking Fault) based on the detection of a first defect image by the reflected light, and classifying the defects as SSF (Shockley-type Stacking Fault) based on the detection of a second defect image by photoluminescence light including the visible range. Defect classification method.

2. The step of classifying the defects includes a step of classifying the SSFs into one of 1SSF to 4SSF based on the detection results for each wavelength of the photoluminescence light. The defect classification method according to claim 1.

3. The step of classifying the defects includes classifying the defects into BPD (Basal Plane Dislocation) based on the detection results of photoluminescence light in the infrared region. The defect classification method according to claim 1 or 2.

4. A method for analyzing defects present in a silicon carbide substrate, A step of projecting an illumination beam toward the silicon carbide substrate and scanning the position on the silicon carbide substrate to be irradiated by the illumination beam, A step of detecting photoluminescent light, including at least the visible and infrared regions, emitted from the silicon carbide substrate, A process to analyze the cause of SSF (Shockley-type Stacking Fault) based on the detection results of photoluminescence light in the visible range and photoluminescence light in the infrared range. Analytical methods including those mentioned.

5. The process of analyzing the cause of occurrence includes determining that the SSF was caused by BPD when one side of the shape of the defect image classified as SSF is thicker than the other sides. The analytical method according to claim 4.

6. The process of analyzing the cause of occurrence includes determining that the SSF was caused by a BPD when a defect image classified as a BPD is detected at a position corresponding to one side of the shape of the defect image classified as an SSF. The analytical method according to claim 4.

7. The process includes classifying defects in the silicon carbide substrate into BPDs (Basal Plane Dislocations) based on the detection results of photoluminescence light in the infrared region. The analytical method according to claim 4.

8. The process includes classifying the SSFs into 1SSF to 4SSF based on the detection results for each wavelength of the photoluminescence light in the visible range. The analytical method according to claim 4.

9. An inspection device for classifying the types of defects present in a silicon carbide substrate, An illumination means for projecting an illumination beam toward the silicon carbide substrate, A scanning means for scanning the position where the silicon carbide substrate is illuminated by the illumination beam, The system comprises: a photodetection means for detecting reflected light and photoluminescent light including the visible range emitted from the silicon carbide substrate; and a classification means for classifying the defects based on the detection results of the reflected light and the detection results of the photoluminescent light including the visible range. The classification means classifies the defect as a 3C-SF (Stacking Fault) based on the detection of a first defect image by the reflected light, and classifies the defect as an SSF (Shockley-type Stacking Fault) based on the detection of a second defect image by photoluminescence light including the visible range. Inspection device.

10. An inspection device for analyzing defects present in a silicon carbide substrate, An illumination means for projecting an illumination beam toward the silicon carbide substrate, A scanning means for scanning the position where the illumination beam irradiates the silicon carbide substrate, A detection means for detecting photoluminescent light, including at least the visible and infrared regions, emitted from the silicon carbide substrate, An analytical means for analyzing the cause of SSF (Shockley-type Stacking Fault) based on the detection results of photoluminescence light in the visible range and photoluminescence light in the infrared range. An inspection device equipped with the following features.

Images