WO2016002003A1 - Substrate inspection apparatus and substrate inspection method - Google Patents

Substrate inspection apparatus and substrate inspection method Download PDF

Info

Publication number
WO2016002003A1
WO2016002003A1 PCT/JP2014/067529 JP2014067529W WO2016002003A1 WO 2016002003 A1 WO2016002003 A1 WO 2016002003A1 JP 2014067529 W JP2014067529 W JP 2014067529W WO 2016002003 A1 WO2016002003 A1 WO 2016002003A1
Authority
WO
WIPO (PCT)
Prior art keywords
defect
sample
mirror
substrate
image
Prior art date
Application number
PCT/JP2014/067529
Other languages
French (fr)
Japanese (ja)
Inventor
野副 真理
長谷川 正樹
小林 健二
Original Assignee
株式会社日立ハイテクノロジーズ
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 株式会社日立ハイテクノロジーズ filed Critical 株式会社日立ハイテクノロジーズ
Priority to PCT/JP2014/067529 priority Critical patent/WO2016002003A1/en
Publication of WO2016002003A1 publication Critical patent/WO2016002003A1/en

Links

Images

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N23/00Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00
    • G01N23/20Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by using diffraction of the radiation by the materials, e.g. for investigating crystal structure; by using scattering of the radiation by the materials, e.g. for investigating non-crystalline materials; by using reflection of the radiation by the materials

Definitions

  • the present invention relates to a substrate inspection apparatus and method, for example, an apparatus and method for inspecting a substrate such as SiC, GaN, and Ga 2 O 3.
  • Semiconductor devices include what are called power semiconductors, as well as memory and logic products formed on Si substrates. Power semiconductors are mainly used in inverter / converter circuits, and are used for switching, conversion, motor control, etc.
  • the power semiconductor includes, for example, a power MOSFET (Metal Oxide Semiconductor Field Effect Transistor) / IPD (Intelligent Power Device), a diode, an IGBT (Insulated Gate Bipolar Transistor), a power module, and a bipolar transistor.
  • Si In power semiconductors, not only Si but also SiC, GaN, Ga2O3, etc. are used as substrate materials to ensure electrical performance.
  • a substrate formed of SiC, GaN, Ga2O3, or the like is likely to have defects on the surface and inside of the substrate. When a fatal defect occurs, the semiconductor at that location becomes defective. For this reason, reduction of defects is desired. In order to reduce defects, it is important to know how many defects are generated for each substrate material or each manufacturing process.
  • semiconductor devices formed with Si substrates also have an epitaxial growth film and other material layers stacked on the Si substrate in order to improve operating characteristics.
  • the number of processes is increasing, and defects are easily generated on the surface and inside of the substrate as in the case of the power semiconductor substrate.
  • a technique of imaging and inspecting electrons (mirror electrons) reflected near the surface of a sample to which a negative potential has been applied in advance is effective.
  • This inspection is called a mirror electron microscope method.
  • an image is formed by forming an image of mirror electrons reflected in the vicinity of the sample, so if the surface of the sample is unevenly charged or some electrons that are not reflected are irradiated onto the substrate.
  • Patent Document 1 an inspection method has been proposed in which ultraviolet light is irradiated simultaneously with electron beam irradiation to suppress the influence of charging.
  • Patent Document 1 can suppress the influence of non-uniform charging, it does not have sufficient sensitivity to detect internal defects, particularly defects called dislocations, which are the purpose of substrate inspection. Patent Document 1 does not disclose a method of revealing internal defects (dislocations) with high sensitivity in the mirror electron microscope system.
  • the present invention has been made paying attention to the above technical problems, and an epi growth film or a laminated film is formed on the surface thereof, like a power semiconductor substrate (SiC, GaN, Ga2O3, etc.) or a Si substrate.
  • the present invention adopts, for example, the configuration described in the claims.
  • the present specification includes a plurality of means for solving the above-described problems.
  • an electron optical system that irradiates an electron beam emitted from an electron source to a range including a visual field on a sample; and A voltage application unit that applies a voltage of intensity reflected before the electron beam reaches the sample to the sample, and a mirror that forms an image of the mirror electrons reflected by the application of the voltage to obtain a mirror electron image
  • a substrate inspection apparatus having an ultraviolet light irradiation unit that irradiates a range including the irradiation range of the electron beam during irradiation ”.
  • the quality of the substrate (including not only the surface of the sample but also internal defects) can be evaluated at high speed and with high sensitivity.
  • FIG. 1 is a diagram illustrating a configuration of a substrate inspection apparatus according to Embodiment 1.
  • FIG. The figure explaining the relationship between an ultraviolet light irradiation area
  • the schematic diagram which shows the defect part state at the time of irradiating the ultraviolet light from which a wavelength differs.
  • the figure which shows the relationship between the electric potential applied to a sample, and the brightness of a mirror electron image The figure which shows the relationship between the brightness of the mirror microscope image for every board
  • the detection principle proposed by the inventors will be described.
  • the inventors focused on the characteristics of dislocation defects as a technique for solving the above-described technical problems.
  • the detection principle of the present invention will be described by taking an inspection of a SiC substrate as an example.
  • the inventors have found that the SiC substrate can generate carriers on the surface of the sample or in a defective portion by irradiating ultraviolet light having a wavelength corresponding to energy larger than the band gap of the material. It was.
  • the inventors have found that when ultraviolet light having a wavelength corresponding to higher energy is irradiated, photoelectrons start to be generated, and when photoelectrons are generated, carriers accumulated in the defect portion are lost as photoelectrons. It was.
  • the inventors propose a method of generating carriers in the defect portion by irradiating ultraviolet light having an energy that is larger than the band gap but does not generate photoelectrons. Since the charge state of the carrier affects the trajectory of the mirror electrons, the defect portion can be made visible as the contrast in the image by the mirror electron microscope method, and the defect can be detected with high sensitivity. Irradiation with ultraviolet light not only generates carriers inside the defect, but also has the effect of stabilizing the non-uniform charging state as described above.
  • the inventors found 182 nm or more and 380 nm or less as a wavelength condition of ultraviolet light that satisfies the above-mentioned conditions. Note that the penetration depth of ultraviolet light into the substrate varies depending on the wavelength. Therefore, when inspecting an internal defect, it is desirable to have a function of changing the wavelength depending on the depth of the defect to be detected.
  • the inventors propose to shift the focus position of the mirror microscope to an under position or an over position with respect to the sample surface during inspection. By doing so, the change in the trajectory of the mirror electrons from the defective part can be increased, and as a result, the difference of the defective part in the acquired image can be enlarged.
  • the substrate inspection apparatus is equipped with an image processing unit for identifying dislocation species such as surface foreign matter, threading screw dislocation, threading edge dislocation, basal plane dislocation using the contrast and brightness profile of the defect. suggest. By mounting the image processing unit, it is possible to identify the defect type at the same time as the inspection and extract only the fatal defect.
  • dislocation defects not only the defect type but also the dislocation scale (depth and extent) can be determined from the profile.
  • the above-described technique can be applied to a review function for moving to the coordinates of a defect detected by an optical inspection apparatus or another inspection apparatus and capturing an image of the defect portion.
  • a review function for moving to the coordinates of a defect detected by an optical inspection apparatus or another inspection apparatus and capturing an image of the defect portion.
  • FIG. 1 shows the overall configuration of the substrate inspection apparatus according to the present embodiment. However, in FIG. 1, a vacuum exhaust pump, exhaust system piping, and the like are omitted.
  • the irradiation electron beam 100 a emitted from the electron gun 101 is deflected by the ExB deflector 103 while being converged by the condenser lens 102.
  • the irradiation electron beam 100a in which the crossover 100d is formed after the deflection is shaped into a substantially parallel bundle by the objective lens 107 and is irradiated to the sample 104 to be inspected.
  • FIG. 1 only one condenser lens 102 is illustrated, but a similar operation can be realized by a system in which a plurality of electron lenses are combined.
  • a Zr / O / W type Schottky electron source is usually used, but an electron source such as LaB6 may be used in order to increase the current.
  • Control voltage and current control
  • the electron gun control device 105 supplies control signals to the corresponding units, and controls the voltage and current described above.
  • the ExB deflector 103 is installed in the vicinity of the imaging surface of the irradiation electron beam 100a.
  • An aberration is generated in the irradiation electron beam 100 a by the ExB deflector 103.
  • an ExB deflector 106 for correcting aberration is further disposed between the condenser lens 102 and the ExB deflector 103.
  • the irradiated electron beam 100a deflected by the ExB deflector 103 along the axis perpendicular to the sample 104 to be inspected is incident in a direction substantially perpendicular to the surface of the sample 104 to be inspected by the objective lens 107. Formed.
  • the inspection substrate 104 is mounted on the sample stage 108.
  • the transport system up to the sample stage 108 is omitted in FIG.
  • a negative potential close to the acceleration voltage of the irradiation electron beam 100a is applied to the sample 104 to be inspected. Due to this negative potential, the irradiated electron beam 100a is reflected near the surface of the sample 104 to be inspected, and further accelerated upward by this negative potential.
  • the sample voltage controller 109 supplies and controls the voltage applied to the sample 104 to be inspected. In order to reflect the irradiation electron beam 100a in the very vicinity of the specimen 104 to be inspected, it is necessary to control the difference between the acceleration voltage of the irradiation electron beam 100a and the negative potential described above with high accuracy.
  • the sample voltage control unit 109 and the electron gun control unit 105 need to be controlled in conjunction with each other.
  • the negative potential is controlled so as not to be applied to the sample stage 108 when transporting the sample 104 to be inspected on the sample stage 108 or when the sample 104 to be inspected is carried out of the apparatus. Therefore, a negative potential is applied to the sample 104 to be inspected only when an image is captured, that is, when the irradiation electron beam 100a is irradiated.
  • the electron beam 100 c reflected in the vicinity of the sample 104 to be inspected is subjected to a convergence action by the objective lens 107.
  • the electron beam 100 c forms a crossover 100 b between the objective lens 107 and the ExB deflector 103.
  • the ExB deflector 103 is controlled so as not to deflect the electron beam 100c traveling from below. For this reason, the electron beam 100 c that has passed through the ExB deflector 103 rises vertically as it is, and is enlarged and projected onto the image detection unit 112 by the intermediate lens 110 and the projection lens 111.
  • FIG. 1 illustrates a case where the projection lens 111 is a single lens, a system in which a plurality of lenses are combined for the purpose of image correction may be used.
  • the image detection unit 112 converts the detected image signal into an electrical signal and transfers the electrical signal to the image processing unit 116.
  • the image detection unit 112 includes a fluorescent plate 112a for converting reflected electrons into an optical image, an optical image detection device 112b, and an optical image transmission system 112c.
  • An optical fiber bundle is used for the optical image transmission system 112c.
  • the optical image detection device 112b converts the optical image formed on the light receiving surface into an electrical signal and outputs it.
  • a device having an image acquisition mode for a normal stationary subject and a mode for time delay integration (TDI) for a moving subject is used as the optical image detection device 112b.
  • TDI time delay integration
  • the image processing unit 116 includes an image signal storage unit 116a and a defect determination unit 116b.
  • the image processing unit 116 receives electron beam irradiation position (deflection signal) data and stage position data from the electron optical system control device 113 and the stage control device 115, respectively, and receives the image data detected by the image detection unit 112 as the sample 104 to be inspected.
  • the image data is stored in the image signal storage unit 116a in association with the upper coordinate data.
  • the defect determination unit 116b includes (1) a comparison process between the image data with coordinates on the inspection sample 104 and a preset value, and (2) image data acquired or stored at an adjacent similar location. The presence / absence of a defect is determined by executing a comparison process or (3) a comparison process with image prediction data based on design data.
  • the defect determination unit 116b executes a storage process of image data having the corresponding coordinates, a defect type determination process based on detailed image characteristics of the defect part, and the like.
  • Information to be used in the above-described defect determination is selected from the inspection apparatus control unit 117. In any case, the determination value and the determination method set in advance in the inspection apparatus controller 117 are used for the defect determination process.
  • the operating conditions of each part constituting the substrate inspection apparatus are input / output through the inspection apparatus control unit 117.
  • the inspection device control unit 117 Through the inspection device control unit 117, (1) acceleration voltage at the time of electron beam generation, (2) current amount of the irradiated electron beam, (3) electron beam deflection width, (4) electron beam deflection speed, (5) Various conditions such as the moving speed of the sample stage and (6) image detection timing by the image detection apparatus are input in advance.
  • the inspection device control unit 117 comprehensively controls the control signals of each element and provides an interface with the user.
  • a monitor 118 is connected to the inspection apparatus control unit 117. The user can set a part of the conditions related to the above operation according to the contents of the inspection through the interface screen displayed on the monitor 118.
  • ultraviolet light is irradiated not only for the purpose of stabilizing the surface potential of the sample 104 to be inspected but also for the purpose of revealing internal defects in the substrate 104 to be inspected.
  • the ultraviolet ray irradiation system 126 is mounted on the substrate inspection apparatus of the present embodiment.
  • the ultraviolet light irradiation system 126 includes an ultraviolet light source 121, an aperture 123, and a reflection mirror 124.
  • the ultraviolet light source 121 is (1) in the case of a light source that irradiates a single wavelength, (2) in the case of a light source that generates a certain range of energy including necessary energy, and (3) irradiates with a specific energy selected by a spectroscope. It may be a light source, and any light source may be used. In this embodiment, it is assumed that the specimen 104 is irradiated with ultraviolet light having a plurality of wavelengths. For this reason, a band pass filter 122 is provided between the ultraviolet light source 121 and the aperture 123, and a structure capable of selecting light of a predetermined wavelength is adopted.
  • the timing of irradiating the ultraviolet light and the selection of the band pass filter 122 are performed by the ultraviolet light control unit 125.
  • the ultraviolet light irradiation conditions registered in advance for each specimen 104 to be inspected are input through the interface screen displayed on the monitor 118 and registered in the inspection apparatus control unit 117, as with other conditions.
  • FIG. 2 shows the relationship between the ultraviolet light irradiation region, the electron beam irradiation region, and the image pickup region (inspection region).
  • the ultraviolet light irradiation region 201 is wider than the electron beam irradiation region 202. The reason is to prevent the charged state of the surface from being changed by the electron beam irradiation.
  • the inspection area 203 is set inside the electron beam irradiation area 202.
  • the ultraviolet light irradiation region 201 and the electron beam irradiation region 202 are circular regions, and the inspection region 203 is a rectangular region, but if the relationship between the region sizes satisfies the relationship shown in FIG.
  • the shape of each region is arbitrary.
  • the substrate to be inspected 401 is a substrate used for a power semiconductor
  • SiC, GaN, Ga2O3 or the like is used in addition to Si.
  • surface foreign matter, micro scratches, crystal defects, etc. are inspected to identify areas where good products cannot be obtained, predict the yield rate (yield), and take measures against the cause of defects to reduce the defect rate. It is necessary to continue.
  • FIG. 3 shows characteristic examples of defects that are inspection targets.
  • Image (a) shows the foreign substance 301 adhering to the substrate surface.
  • Image (b) shows micro scratch 302 by polishing.
  • Image (c) shows minute pits 303.
  • Image (d) shows a stacking fault 304.
  • Image (e) shows a latent defect 305 that is not visible from the surface.
  • Image (f) shows a dislocation defect 306. All of these defects need to be detected.
  • a method of detecting scattered light generated from a substrate by laser light irradiation is used for inspection of these defects.
  • the conventional method can detect the foreign matter 301 and the micro scratch 302 accompanying the surface shape change, but cannot detect other defects with a small surface irregularity with high sensitivity. For example, in the case of a pit 303 having a very small defect, sufficient sensitivity cannot be obtained by the conventional method. Further, the conventional method cannot detect a part of the stacking fault 304, the latent defect 305, and the dislocation defect 306 in which the type of the defect is not accompanied by a shape abnormality on the surface.
  • dislocation defects 306 that are difficult to see from the surface can be classified mainly into three types: threading screw dislocation, threading edge dislocation, and basal plane dislocation.
  • the substrate inspection apparatus according to the present embodiment can inspect the dislocation defects 306 with high sensitivity.
  • FIG. 4 shows the image formation principle of the mirror electron microscope.
  • FIG. 4A shows the trajectory of mirror electrons when the surface has a concave shape
  • FIG. 4B shows the trajectory of mirror electrons when the surface has a convex shape.
  • a negative potential is applied to the sample 401 to be inspected with respect to the energy of the irradiation electron beam 100a. .
  • a potential distribution is formed immediately above the sample 401 to be inspected.
  • the potential distribution is substantially uniform and flat equipotential lines.
  • the equipotential line 403 that is the surface potential distribution is distorted in accordance with the concave shape 402. Since the mirror electrons have a property of taking a trajectory perpendicular to the equipotential line, the mirror electron 404a reflected at a flat portion becomes a trajectory reflecting upward as it is, but the equipotential line 403 is distorted downward. The mirror electrons 404b reflected at the locations change inward due to the distortion of the equipotential lines 403. On the other hand, as shown in FIG.
  • the equipotential line 403 is distorted upward and distorted in the opposite direction to the concave shape 402. Therefore, the trajectory of the mirror electrons 404 b reflected at the convex shape 405 changes to the outside due to the influence of the distortion of the equipotential line 403.
  • FIG. 4C shows the potential distribution on the surface and the trajectory of the mirror electrons when the positive charge 406 is locally present
  • FIG. 4D shows the case where the negative charge 407 is locally present.
  • the surface potential distribution and the mirror electron trajectory are shown. Since the sample 401 to be inspected is set to a relatively negative potential, the equipotential line 403 has a negative potential distribution.
  • the equipotential lines are distorted downward as shown in FIG.
  • the equipotential lines are distorted upward as shown in FIG. That is, even when local charge exists on the sample surface, the potential distribution is the same as when the physical unevenness exists on the sample surface.
  • FIG. 5 shows how the appearance of the contrast of the mirror microscope image changes due to the difference in trajectory change.
  • FIG. 5A shows how the contrast of the mirror microscope image appears when the concave shape 402 exists on the surface.
  • the upper image 501 is obtained when the in-focus position of the mirror electrons is set to the overfocus 409. In this case, the concave shape portion becomes dark.
  • the lower image 502 is obtained when the in-focus position of the mirror electrons is set to the under focus 408. In this case, the concave shape portion becomes bright.
  • the image processing unit 116 executes a defect determination process based on the contrast of the mirror microscope image.
  • Information about whether the surface of the electron optical system that is in focus is the position of the underfocus 408 or the position of the overfocus 109 with respect to the sample surface is given from the inspection apparatus control unit 117 to the image processing unit 116, for example. .
  • a specific method for classifying defective portions will be described later with reference to FIG. While the surface defect portion has a strong contrast with light and dark, the internal defect is characterized by an intermediate brightness. Therefore, the shape defect and the internal defect can be discriminated from the mirror microscope image.
  • a process 603 for growing an Epi growth film on the substrate surface is executed.
  • a substrate on which an Epi growth film is grown is referred to as an Epi formation substrate.
  • crystal defects are generated.
  • the first layer pattern formation processing 604 is executed.
  • processing 605 for forming an n-layer pattern on the upper layer sequentially is executed. The number of processing steps, pattern shapes, materials, and the like vary depending on the type of semiconductor.
  • a process 606 for evaluating the electrical characteristics of the completed semiconductor circuit is executed to discriminate between a good product and a defective product.
  • the semiconductor circuit is completed through this series of processing steps. If a substrate defect is inspected with high sensitivity at the stage of forming a polishing substrate before forming a pattern (process 602) or the stage of forming an epi growth film (process 603) in this series of processing steps, all patterns are obtained. It is possible to avoid a state in which the presence or absence of a defect is unknown until the electrical characteristic test stage (process 606) after the formation of. Therefore, in this embodiment, the substrate is inspected at the stage where the Epi growth film is grown on the substrate surface.
  • FIG. 1 An example of a mirror microscope image when the substrate (Epi forming substrate) after growing the Epi growth film is imaged by the mirror electron microscope shown in FIG. 1 is shown.
  • FIG. 1 An example of a mirror microscope image acquired without irradiating ultraviolet rays is shown in FIG. In this case, uneven brightness is seen on the entire substrate 701, and the contrast of the defective portions 702 and 703 assumed in two places in the screen is also weak.
  • FIG. 7B shows an example of a mirror microscope image acquired when the same part is observed with a mirror electron microscope in a state where the substrate surface is irradiated with ultraviolet light having a wavelength of 365 nm.
  • uneven brightness is eliminated from the entire substrate 701.
  • the contrast difference between the defective portions 702 and 703 also increased.
  • the defect portion 703 was difficult to recognize when it was not irradiated with ultraviolet light, but can be clearly recognized by irradiation with ultraviolet light having a wavelength of 365 nm.
  • FIG. 8 shows the evaluation results.
  • FIG. 8 shows the results of observing previously known stacking faults for 365 nm, 283 nm, 268 nm, 230 nm, and 180 nm in the wavelength range of 180 nm to 380 nm.
  • the energy of the excited band gap corresponds to 3.40 eV.
  • the penetration depth into the SiC film at this wavelength is equivalent to 120 ⁇ m.
  • the energy of the excited band gap corresponds to 4.38 eV.
  • the penetration depth into the SiC film at this wavelength is approximately 3 ⁇ m.
  • the energy of the excited band gap corresponds to 4.63 eV.
  • the penetration depth into the SiC film at this wavelength is approximately 1 ⁇ m.
  • the energy of the excited band gap corresponds to 5.39 eV.
  • the penetration depth into the SiC film at this wavelength is about 0.1 ⁇ m or less.
  • the energy of the excited band gap corresponds to 6.4 eV.
  • the wavelength range of ultraviolet light is in the range of 182 nm or more and 380 nm or less, but when a mirror microscope image is acquired while irradiating ultraviolet light having a wavelength outside this range of 180 nm, the whole is uniformly bright. It became an image, and it became difficult to recognize the defective portions 801 and 802 (FIG. 8E).
  • FIG. 9 is a diagram schematically illustrating a phenomenon that occurs in a defect when the individual mirror microscope images illustrated in FIG. 8 are acquired. That is, FIG. 9 is a diagram schematically showing the relationship between the irradiation of ultraviolet light of each wavelength described in FIG. 8 and the phenomenon occurring in the defect. As shown in FIG. 9 which is a cross-sectional view, the defect 901 was considered to extend to the depth 903 of about 10 ⁇ m from the substrate surface to the inside.
  • FIG. 9A is a schematic diagram for explaining the phenomenon when the wavelength of ultraviolet light is 365 nm. In the case of this wavelength, it was considered that ultraviolet light could reach deeper than 10 ⁇ m, which is the deepest part of the defect depth 903, and as a result, charges 902 as carriers were generated along the entire area of the defect and simultaneously trapped.
  • FIG. 9B is a schematic diagram for explaining the phenomenon when the wavelength of ultraviolet light is 283 nm.
  • the penetration depth of ultraviolet light is considered to be as shallow as about 3 ⁇ m.
  • the range in which carriers are generated and trapped was considered to be about 3 ⁇ m deep from the substrate surface.
  • FIG. 9C is a schematic diagram for explaining the phenomenon when the wavelength of ultraviolet light is 268 nm. In the case of this wavelength, it was considered that the penetration depth of ultraviolet light became as shallow as 1 ⁇ m, and the depth at which carriers were trapped was also reduced accordingly.
  • FIG. 9D is a schematic diagram for explaining the phenomenon when the wavelength of ultraviolet light is 230 nm. In the case of this wavelength, since carriers are generated only on the extreme surface of the substrate, it was considered that only defects near the surface were observed as potential changes.
  • FIG. 9E is a schematic diagram for explaining a phenomenon in the case where ultraviolet light having a wavelength of 180 nm, which is outside the wavelength condition of the present embodiment, is irradiated.
  • photoelectrons 904 were generated from the SiC substrate, brightened as a whole, the defect contrast was remarkably lowered, and defect recognition became difficult. This is presumably because the carriers trapped in the defect are emitted due to the generation of photoelectrons.
  • the inventors set the optimum inspection condition for ultraviolet light irradiation at a wavelength of 365 nm.
  • the wavelength of the optimum inspection condition was 365 nm, but the optimum wavelength condition depends on the defect type to be inspected and its expected depth. Is different. Therefore, the inventors set the wavelength range of ultraviolet rays to 182 nm or more and 380 nm or less.
  • the relationship between the defect type and / or the expected depth and the optimum wavelength condition of the ultraviolet ray is stored in a storage area (not shown) in a table format. Note that the defect type to be inspected and information on the expected depth thereof are input by the user through an interface screen, for example.
  • the inventors fine-tuned the voltage applied to the sample to be inspected while irradiating the ultraviolet light under the conditions after determining the optimum wavelength of the ultraviolet light irradiated at the time of image acquisition. This is because it is more sensitive to changes in the surface potential when mirror electrons are reflected as close to the surface of the specimen as possible. That is, it aims at improving the sensitivity with respect to the presence or absence of a defect.
  • the voltage application to the sample to be inspected is performed by the sample voltage control unit 109 in FIG.
  • the voltage applied by the sample voltage control unit 109 is set through the inspection apparatus control unit 117.
  • the user inputs a target value of the potential applied to the sample to be inspected through the monitor 118, and inputs a range in which the potential is changed in the plus direction and the minus direction with respect to the potential and a pitch of the change.
  • the target value is -3 kV
  • each image data is acquired while changing the range from -3.2 kV to -2.8 kV in steps of 0.02 kV, and an instruction is given to measure the change in brightness at that time. input.
  • the sample potential changes step by step from ⁇ 3.20 kV to ⁇ 3.18 kV, and a mirror microscope image is acquired each time.
  • the irradiation electron beam 100a emitted from the electron gun 101 is reflected near the surface of the sample 104 to be inspected, and the reflected electron beam 100c is detected by the image detection unit 112. It is stored in the image processing unit 116 as a mirror microscope image detected through the image detection unit 112. In the case of the present embodiment, a function for calculating the average brightness of a predetermined region portion of the stored mirror microscope image is mounted on the inspection apparatus control unit 117.
  • the electron beam 100a having an irradiation energy of 3 keV When the irradiation electron beam 100a having an irradiation energy of 3 keV is irradiated, the electron beam is negative, so that a repulsive force acts on the sample potential applied to ⁇ 3.20 kV and is reflected upward with respect to the sample surface. While all the electrons are mirror-reflected, the brightness is at a constant level. However, the energy of the irradiated electron beam may vary, and the energy of the electron and the sample potential may have a subtle error depending on the accuracy of the power source. For this reason, when the energy of the electron beam and the negative potential on the sample surface approach approximately the same level, most of the electrons are mirror-reflected, but some of the electrons are irradiated on the sample surface. When the sample surface is irradiated with an electron beam, secondary electrons are generated, but the amount is smaller than that of the mirror-reflected electrons. However, the brightness of the acquired mirror microscope image is lower than when all the
  • FIG. 10 shows a change in brightness of a mirror microscope image (mirror electron image) when the voltage applied to the sample to be inspected is changed.
  • the horizontal axis of the graph indicates the magnitude of the negative voltage applied to the sample to be inspected, and the vertical axis indicates the brightness of the mirror microscope image. Note that the absolute value of the negative voltage to be applied is larger on the left side of the horizontal axis and smaller on the right side of the horizontal axis.
  • the brightness of the mirror microscope image is stable at level 1001.
  • the negative applied voltage to the sample to be inspected is gradually changed to the plus side, the point 1002 at which the brightness starts to change is reached.
  • the brightness of the mirror microscope image gradually decreases and reaches a point 1003 where the change is reduced at a certain level. Above this point, the brightness of the mirror microscope image does not change even when the applied voltage is increased, and the region 1004 is where all the irradiated electrons irradiate the sample surface.
  • the voltage immediately before the point 1002 at which the brightness of the mirror microscope image starts changing is closest to the specimen surface.
  • the electrons are reflected near the equipotential line indicating the potential change of the sample surface, and as a result, it is easily affected by the change of the equipotential line, and the condition is highly sensitive to the change of the surface potential. Can be set.
  • the inventors set a voltage corresponding to the optimum range 1005 in FIG. 10 (a voltage within a certain range from the point 1002 to the minus side) as a condition that can achieve both sensitivity and stability. That is, the sample applied voltage is set to the optimum value.
  • the influence of brightness variation is suppressed by calculating the average of the brightness of the mirror microscope image before and after each applied voltage.
  • the point at which the change is obtained and the rate of change becomes equal to or higher than a certain value is defined as a point 1002 at which the change starts.
  • the optimum range 1005 is defined as a range of ⁇ 5 eV or less from the intersection of the tangent of the point 1002 and the tangent of the level 1001 where there is no change in brightness and ⁇ 100 eV or less from the intersection.
  • the optimum range 1005 can be arbitrarily changed according to the purpose, the state of the sample, and the required sensitivity when determining the sensitivity.
  • evaluation may be made with a certain amount of minus ⁇ side, such as ⁇ 100 eV with respect to the target value of ⁇ 3 kV.
  • sample application voltage optimization functions are mounted on the inspection apparatus control unit 117, for example.
  • the inventors optimized the conditions for actually acquiring a mirror microscope image and determining the presence or absence of defects.
  • a procedure for optimizing defect determination will be described.
  • the mirror microscope image captured for a predetermined area of the substrate to be inspected is transferred to the image processing unit 116 and stored in the storage unit.
  • FIG. 11A is an image of a first substrate
  • FIG. 11B is an image of a second substrate of the same type
  • FIG. 11C is an image of a third substrate of a different type from the first and second substrates described above.
  • FIGS. 11 (d), 11 (e), and 11 (f) show the brightness profiles extracted from these images at predetermined locations 1101 (indicated by dotted lines in the figure).
  • the allowable range 1105 in FIG. 11 (d) and the allowable range 1106 in FIG. 11 (f) are threshold values used when determining the presence or absence of defects.
  • the brightness variations 1102, 1103, and 1104 are assumed to vary from substrate to substrate or from place to place on the same substrate.
  • an allowable range 1105 ie, , Threshold value
  • the allowable range 1106 was set to a range larger than the variation 1104.
  • the allowable range can be set using another method. For example, a method of determining a permissible range by acquiring a mirror microscope image for each substrate to be inspected and comparing image signals between adjacent images as shown in FIG. First, as in the case of FIG. 11, the brightness is normalized to a certain level for each substrate to be inspected. Thereafter, the image (a) for the first region in the substrate to be inspected is compared with the image (b) for the adjacent second region, and a difference signal 1201 is calculated. By comparing with the adjacent region, changes in brightness and in-plane variation can be reduced. By setting an allowable level 1202 for the difference signal 1201 shown in the image (c), it is possible to determine a range of variation and a defect.
  • the threshold for defect determination was adjusted, a desired defect was detected, and an image processing condition that did not detect variation was selected.
  • a function for optimizing the condition for determining the presence / absence of a defect is mounted on the image processing unit 116, for example.
  • FIG. 13 shows an inspection processing procedure.
  • inspection conditions are input through an interface screen displayed on the monitor 118 (step 1301).
  • the inspection conditions are information for specifying a substrate to be inspected (for example, a slot in a cassette or a hoop for specifying a wafer, a wafer ID (for example, lot number, wafer number)), a recipe name (ultraviolet light irradiation condition, beam condition, Negative voltage value to be applied to the substrate to be inspected), inspection area, inspection result output contents and format, instruction contents regarding review and image storage after inspection, and the like.
  • optical condition A when detecting a defect with a depth of 20 ⁇ m existing on a 4-inch SiC substrate, the user selects “optical condition A” in which conditions for irradiating ultraviolet light with a wavelength of 350 nm are set when irradiating an electron beam.
  • the user sets the mode for instructing optimization of the sample applied voltage to “ON” as described above.
  • a relationship in which a depth to be inspected and an optimum ultraviolet light wavelength for inspecting the depth are associated with each other is stored in advance in a storage unit such as a hard disk, the user wants to make the inspection target.
  • These optical conditions can be set automatically simply by selecting or inputting the depth.
  • the user selects “area condition B” for setting a wafer size and a total of 6 areas of 20 mm ⁇ 20 mm in total, 5 in the center and 5 in the periphery of the wafer. Further, the user sets the inspection result output contents to automatically save the defect coordinate list of the inspected area, the defect map, and the image of the portion recognized as the defect, and to transfer to the external server. ”Is selected. A combination of these conditions is an inspection recipe. The user selects a recipe name “SiC substrate inspection recipe D” and inputs this recipe name from the monitor 118.
  • the user instructs the start of inspection through the interface screen (step 1302).
  • the wafer in the slot designated by the cassette or the hoop is automatically transferred to the sample exchange chamber and mounted on the holder.
  • the sample exchange chamber is evacuated, and when the vacuum reaches a predetermined level, the wafer along with the holder is transferred into the sample chamber (chamber) (step 1303).
  • the substrate inspection apparatus applies a negative potential to the sample stage 108 and irradiates ultraviolet light under the conditions specified in the inspection recipe D (step 1304).
  • the ultraviolet light can be irradiated continuously, or can be irradiated immediately before acquiring the electron beam image, and the irradiation can be turned off while the electron beam is not irradiated. In this inspection, as described above, ultraviolet light having a wavelength of 365 nm was irradiated.
  • the substrate inspection apparatus calibrates the state of the irradiated electron beam (step 1305).
  • the amount of electron beam is adjusted by current, and then the position and distortion state are corrected.
  • a standard pattern installed on the sample stage 108 is used.
  • the focus condition is set to be under focus.
  • the substrate inspection apparatus calibrates the position information (step 1306). In the stage of inspecting for the presence of defects, a pattern is usually not formed on the surface of the substrate. Therefore, at this stage, the position in the rotation direction is specified by using the center position of the substrate to be inspected, the notch and the orientation flat for specifying the crystal direction, and the coordinates of each inspection region and defect are specified.
  • the substrate inspection apparatus adjusts the gain and offset of the detection system so that the brightness level of the mirror microscope image becomes a predetermined level (step 1307).
  • determination thresholds are set for brightness and signal level. For this reason, adjustment processing is provided so that images with the same conditions can always be acquired.
  • the board inspection apparatus starts inspection (step 1308).
  • image acquisition and stage movement are repeated, and an image of a predetermined area is acquired.
  • an image is acquired in a state where the stage is stationary, and the next image is acquired when the stage is moved to the next inspection position and is stationary. That is, images are acquired sequentially.
  • processing for determining the presence or absence of defects is also executed.
  • a computer such as the image processing unit 116 sequentially determines the presence or absence of a defect based on the stored image, and outputs the result.
  • the presence / absence of a defect may be determined in real time using a dedicated image processing substrate.
  • the substrate inspection apparatus executes a review process for confirming the type and presence of defects (step 1309).
  • this review process in addition to the process of storing the acquired image, re-acquisition of an image under another imaging condition is performed as necessary. However, the review process itself may be skipped.
  • inspection result map output, image transfer, position information numerical data output, and the like are also performed.
  • the wafer is unloaded to the original cassette or hoop slot position (step 1310), and the inspection is terminated (step 1311).
  • FIG. 14 shows an output example of the inspection result.
  • positional information of detected defects is plotted as points 1402 with respect to a circle 1401 having a size indicating the shape and size of the substrate to be inspected.
  • Each point 1402 represents another defect.
  • FIG. 15 shows an example of a defect image.
  • the defect 1501 is detected brightly, but the other defects 1502, 1503, 1504, and 1505 are all dark. Accordingly, the brightness and darkness in the mirror electric microscope image enables separation of foreign matters and residues and other various defects (stacking defects, scratches, pits, dislocations, etc.) without teaching.
  • FIG. 16 shows the result of observing the characteristics of defects that appear dark.
  • a schematic diagram of the defect 1501 that appears bright in FIG. 15 is shown in FIG. 16A, and a brightness profile in the broken line direction in FIG. 15 is shown in FIG. Since this defect has a convex shape with respect to the substrate surface, the defect portion is brighter than the brightness of the substrate.
  • FIG. 16B shows a schematic diagram of the defect 1502 that appears as a dark spot in FIG. 15, and FIG. 16B ′ shows the brightness profile in the broken line direction in the figure. Since it looks like a dark circle, this defect is assumed to be concave with respect to the substrate surface. Furthermore, schematic diagrams of the defects 1503, 1504, and 1505 detected dark in FIG. 15 are shown in FIGS. 16C, 16D, and 16E, respectively, and the brightness profiles in the broken line direction in the figure are shown. 16 (c) ′, (d) ′, and (e) ′.
  • the defect 1503 (FIG. 16C) was identified as a crystal defect called threading screw dislocation because the periphery of the defect was etched away in addition to the pinhole similar to the pit.
  • the defect 1504 (FIG. 16D) has a very small crystal defect even inside, and was found to be a crystal defect called a basal plane dislocation.
  • the defect 1505 (FIG. 16E) is extremely fine and evenly etched, and from its characteristics, it was found that it is a threading edge dislocation. Thus, it has been found that there is a feature in the appearance of the defect in the mirror microscope image depending on the type of the defect.
  • the threading screw dislocation shown in FIG. Corresponds to the defect 1703 in FIG. 17, and it was found that a shadow occurred in the downstream direction of crystal formation. Further, it was found that the basal plane dislocation in FIG. 16D corresponds to the defect 1704 in FIG. 17 and has a shadow upstream of the crystal formation.
  • the image processing unit 116 of the present embodiment is equipped with a function that can automatically distinguish the characteristics of the profile of FIG. 16 based on these results. Specifically, the defects are classified based on the following conditions. First, the image processing unit 116 extracts brightness information for each pixel for a portion determined to be a defect (entire region where a predetermined brightness difference is detected).
  • the image processing unit 116 performs a smoothing process to suppress the influence of noise.
  • the image processing unit tentatively determines that the defective portion is a convex defect with respect to the substrate surface.
  • the image processing unit 116 sequentially executes the following processing.
  • the size LA1 in the longitudinal direction is divided into L11 and L12 starting from the darkest portion P1.
  • the defect is determined according to the brightness of the brightest or darkest point in the defect area and the shape of the defect area (more specifically, the length ratio in the direction perpendicular to the longitudinal direction).
  • Classify More specifically, defects are classified according to the following procedure. In the following description, the five types of defect classification methods shown in FIG. 16 will be described, but other methods can be classified by the same method. Further, a classification procedure other than that described in the present embodiment may be added, and the threshold values described below may be arbitrarily changed.
  • Bright defects may be classified as convex or foreign on the surface.
  • the image processing unit 116 identifies the darkest part from the mirror microscope image acquired for the defective part, and the length characteristic including the contrast and gradation part that looks like a shadow and the like.
  • the positional relationship is digitized, and defect types are automatically classified based on the numerical relationship and the length relationship with respect to the crystal growth direction.
  • the substrate inspection apparatus can detect latent scratches generated when the substrate is polished with high sensitivity, feedback for improving the polishing process can be performed at an early stage of the manufacturing process. Similarly, feedback for forming the laminated film and optimizing the temperature process can be performed at an early stage of the manufacturing process. In addition, it is possible to predict a non-defective product rate for a substrate with many defects. In addition, by removing a substrate with many defects, it is possible to reduce useless manufacturing costs. In addition, by applying a substrate whose quality is controlled by this technology, the quality of the semiconductor device can be further improved, and the time required to reduce the defect rate while improving the reliability of the semiconductor device and the like can be shortened. be able to.
  • Example 2 In Example 1 described above, the wavelength of ultraviolet light was set to 365 nm, and the substrate was inspected for defects. In contrast, in this embodiment, first, ultraviolet light having a first wavelength of 365 nm is irradiated to inspect the substrate for defects, and then the same substrate to be inspected is irradiated with ultraviolet light having the second wavelength of 230 nm. And inspecting the substrate for defects. As a result, as shown in FIGS. 18A and 18B, two inspection results were obtained.
  • FIG. 18 (a) 50 defects were detected, and in the example of FIG. 18 (b), 30 defects were detected.
  • the difference in the number of detections is that in the inspection using the ultraviolet light with the second wavelength of 230 nm (FIG. 18B), only defects near the surface of the sample to be inspected are detected, whereas the ultraviolet light with the first wavelength of 365 nm is detected.
  • FIG. 18A shows that in the inspection using light (FIG. 18A), defects up to about 10 ⁇ m are detected in the internal direction of the inspected substrate. That is, FIG. 18A shows that both surface defects and internal defects are detected, and FIG. 18B shows that only surface defects are detected.
  • the image processing unit 116 includes a process for comparing two inspection results, a process for determining a defect appearing at the same coordinates of the two inspection results as a defect near the surface, and FIG. ), A function for executing a process of determining a defect detected as an internal defect. By mounting this function, it becomes easier to discriminate defect types compared to the first embodiment. In addition, if the inspection is performed in a similar manner by combining a plurality of wavelengths other than the main inspection and a portion where the inspection results do not match is extracted, the depth of the defect can be identified.
  • the substrate inspection apparatus By using the substrate inspection apparatus according to the present embodiment, not only the surface but also internal defects can be detected with high sensitivity in the SiC substrate which is a laminated substrate. Further, since the depth direction of the defect can be identified without performing a cross-sectional analysis, the defect type can be easily specified. As a result, it becomes possible to identify the cause of failure and improve the substrate manufacturing process at an early stage, thereby contributing to the reduction of defects.
  • the substrate inspection apparatus that primarily detects the defect of the substrate has been described.
  • the above-described technique is a review in which the defect coordinates are acquired from another substrate inspection apparatus to inspect the defect. It can also be applied to devices.
  • the board inspection apparatus as a review apparatus is equipped with a communication unit that exchanges position information of an image capturing location with an external apparatus in advance.
  • the stage controller 115 may position the sample 104 to be inspected at the coordinate position designated as the review position and start the above-described inspection operation. According to the present embodiment, it is possible to increase the inspection accuracy of defects by the review device.
  • part or all of the functions executed by the processing unit and the control unit in the above-described embodiments may be realized as an integrated circuit or other hardware, for example.
  • Information such as programs, tables, and files used to implement the functions of the processing unit and control unit is a memory, hard disk, storage device such as SSD (Solid State Drive), storage medium such as IC card, SD card, and DVD. Can be stored.
  • SSD Solid State Drive
  • control lines and information lines indicate what is considered necessary for explanation, and do not represent all control lines and information lines necessary for the product. In practice, it can be considered that almost all components are connected to each other.
  • Optical image detection Device 112c: Optical image transmission system, 113: Electro-optical system control device, 115: Stage control device, 116: Image processing unit, 116a: Image signal storage unit, 116b: Defect determination unit, 117: Inspection device control unit, 118 ... Monitor, 121 ... UV light source, 122 ... Band pass filter, 123 ... Aperture, 124 ... Morphism mirror, 125 ... UV-light control unit, 126 ... UV light irradiation system.

Landscapes

  • Chemical & Material Sciences (AREA)
  • Crystallography & Structural Chemistry (AREA)
  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Analysing Materials By The Use Of Radiation (AREA)

Abstract

Provided is a technology of inspecting, at a higher speed and a higher sensitivity compared with conventional cases, surface defects and inner defects of substrates, such as power semiconductor substrates (SiC, GaN, Ga2O3 and the like) and Si substrates, which have epitaxial layers and laminated films formed thereon. In order to provide such technology, this substrate inspection apparatus is provided with: an electrooptical system that applies, to a sample, an electron beam in a range including a visual field, said electron beam being emitted from an electron source; a voltage applying unit that applies, to the sample, a voltage at a level at which the electron beam is reflected before the electron beam reaches the sample; a mirror electron image forming optical system that obtains a mirror electron image by image-forming mirror electrons reflected by means of the voltage application; a calculation unit that detects a defect of the sample using the mirror electron image; and an ultraviolet irradiation unit that applies, while the electron beam is being applied, ultraviolet having a wavelength arbitrarily set within a range of 182-380 nm, said ultraviolet being applied to the sample in the range of the region including the visual field.

Description

基板検査装置及び方法Substrate inspection apparatus and method
 本発明は、基板検査装置及び方法に関し、例えばSiC、GaN、Ga2O3等の基板を検査する装置及び方法に関する。 The present invention relates to a substrate inspection apparatus and method, for example, an apparatus and method for inspecting a substrate such as SiC, GaN, and Ga 2 O 3.
 半導体装置には、Si基板に形成されるメモリやロジック製品の他、パワー半導体と呼ばれるものがある。パワー半導体は、主にインバータ/コンバータ回路で使われており、スイッチング、変換、モータ制御等に使用される。パワー半導体には、例えばパワーMOSFET(Metal Oxide Semiconductor Field Effect Transistor)/IPD(Intelligent Power Device)、ダイオード、IGBT(Insulated Gate Bipolar Transistor)、パワーモジュール、バイポーラトランジスタが含まれる。 Semiconductor devices include what are called power semiconductors, as well as memory and logic products formed on Si substrates. Power semiconductors are mainly used in inverter / converter circuits, and are used for switching, conversion, motor control, etc. The power semiconductor includes, for example, a power MOSFET (Metal Oxide Semiconductor Field Effect Transistor) / IPD (Intelligent Power Device), a diode, an IGBT (Insulated Gate Bipolar Transistor), a power module, and a bipolar transistor.
 パワー半導体においては、電気的な性能を確保するために、基板材料にSiだけでなくSiC、GaN、Ga2O3等が用いられる。ところが、SiC、GaN、Ga2O3等で形成された基板は、基板表面及び内部に欠陥が発生し易い。致命欠陥が発生すると、当該箇所の半導体は不良になる。このため、欠陥の低減が望まれる。欠陥を低減するには、各基板の材料毎又は製造プロセス毎にどのような欠陥がどの程度発生しているかを把握することが重要となる。 In power semiconductors, not only Si but also SiC, GaN, Ga2O3, etc. are used as substrate materials to ensure electrical performance. However, a substrate formed of SiC, GaN, Ga2O3, or the like is likely to have defects on the surface and inside of the substrate. When a fatal defect occurs, the semiconductor at that location becomes defective. For this reason, reduction of defects is desired. In order to reduce defects, it is important to know how many defects are generated for each substrate material or each manufacturing process.
 また、近年における半導体装置を構成する回路パターンの微細化に伴い、Si基板で形成する半導体装置においても、動作特性を向上させるために、Si基板にEpi成長膜や他の材料の膜を積層する工程が増えており、パワー半導体の基板と同様に基板の表面や内部に欠陥が発生し易くなっている。 In addition, with the recent miniaturization of circuit patterns constituting semiconductor devices, semiconductor devices formed with Si substrates also have an epitaxial growth film and other material layers stacked on the Si substrate in order to improve operating characteristics. The number of processes is increasing, and defects are easily generated on the surface and inside of the substrate as in the case of the power semiconductor substrate.
 これら欠陥の検査技術として、表面のエッチング処理により欠陥部の周囲を削り、SEM(Scanning Electron Microscope)で観察する方法がある。本手法は、欠陥部の識別には有効であるが、ウエハ表面を削るため破壊検査となる。加えて、本手法は、被検査サンプルの前処理と検査に膨大な時間を要するという技術上の課題がある。基板を検査する別の技術として、レーザ光を照射し、欠陥部からの散乱光を検出する方法がある。本手法は、高速に基板全面を検査することができるが、半導体の電気特性にとって致命的な各種の結晶欠陥(例えば基板内部の転位と呼ばれる欠陥)に対して感度が低いという課題がある。 As an inspection technique for these defects, there is a method in which the periphery of the defective portion is shaved by etching the surface and observed with a scanning electron microscope (SEM). This method is effective for identifying defective portions, but it is a destructive inspection because the wafer surface is shaved. In addition, this technique has a technical problem that it takes a lot of time to pre-process and inspect the sample to be inspected. As another technique for inspecting a substrate, there is a method of irradiating a laser beam and detecting scattered light from a defective portion. Although this method can inspect the entire surface of the substrate at high speed, there is a problem that the sensitivity is low with respect to various crystal defects (for example, defects called dislocations inside the substrate) that are fatal to the electrical characteristics of the semiconductor.
 上記のように、従来の検査手法では、基板上の欠陥を非破壊で高感度に検査することが難しい。このため、予め負の電位を印加した試料の表面近傍で反射した電子(ミラー電子)を結像して検査する手法が有効である。本検査は、ミラー電子顕微鏡方式と呼ばれる。ミラー電子顕微鏡方式では、試料近傍で反射されるミラー電子を結像することによって画像を形成するため、試料表面に不均一な帯電が生じている場合や反射しない一部の電子が基板に照射されることで変化する帯電がある場合に、その帯電の電界によりミラー電子の軌道が曲げられ、本来の目的である試料の形状や電位の差異を反映した画像を得られなくなる。このため、特許文献1に開示されているように、電子線照射と同時に紫外光を照射し、帯電の影響を抑制する検査方法が提案されている。 As described above, it is difficult to inspect a defect on a substrate with high sensitivity and non-destructive by the conventional inspection method. For this reason, a technique of imaging and inspecting electrons (mirror electrons) reflected near the surface of a sample to which a negative potential has been applied in advance is effective. This inspection is called a mirror electron microscope method. In the mirror electron microscope method, an image is formed by forming an image of mirror electrons reflected in the vicinity of the sample, so if the surface of the sample is unevenly charged or some electrons that are not reflected are irradiated onto the substrate. When there is a charge that changes, the electric field of the charge causes the orbit of the mirror electrons to be bent, making it impossible to obtain an image that reflects the difference in shape and potential of the sample, which is the original purpose. For this reason, as disclosed in Patent Document 1, an inspection method has been proposed in which ultraviolet light is irradiated simultaneously with electron beam irradiation to suppress the influence of charging.
特開2007-192560号公報JP 2007-192560 A
 しかし、特許文献1に記載の技術は、不均一な帯電の影響こそ抑制できるものの、基板検査の目的である内部欠陥、特に転位と呼ばれる欠陥の検出感度が不十分である。そして、特許文献1には、ミラー電子顕微鏡方式において、内部欠陥(転位)を高感度に顕在化する方法については開示されていない。 However, although the technique described in Patent Document 1 can suppress the influence of non-uniform charging, it does not have sufficient sensitivity to detect internal defects, particularly defects called dislocations, which are the purpose of substrate inspection. Patent Document 1 does not disclose a method of revealing internal defects (dislocations) with high sensitivity in the mirror electron microscope system.
 本発明は、上記の技術課題に着目してなされたもので、パワー半導体用の基板(SiC、GaN、Ga2O3等)やSi基板のように、その表面にEpi成長膜や積層膜が形成された基板における表面欠陥や内部欠陥を、従来に比して、高速かつ高感度に検査する技術を提供する。 The present invention has been made paying attention to the above technical problems, and an epi growth film or a laminated film is formed on the surface thereof, like a power semiconductor substrate (SiC, GaN, Ga2O3, etc.) or a Si substrate. Provided is a technique for inspecting surface defects and internal defects in a substrate at a higher speed and with higher sensitivity than before.
 上記課題を解決するため、本発明は、例えば請求の範囲に記載の構成を採用する。本明細書は、上記課題を解決する手段を複数含んでいるが、その一例を挙げるならば、「電子源から放出された電子線を試料上の視野を含む範囲に照射する電子光学系と、前記電子線が前記試料に到達する前に反射される強度の電圧を前記試料に印加する電圧印加ユニットと、前記電圧の印加により反射されたミラー電子を結像してミラー電子像を取得するミラー電子結像光学系と、取得された前記ミラー電子像を用いて前記試料の欠陥を検出する演算ユニットと、182nm以上380nm以下の範囲で任意に設定された波長の紫外光を、前記電子線の照射中に、前記電子線の照射範囲を含む範囲に照射する紫外光照射ユニットと、を有する基板検査装置」を特徴とする。 In order to solve the above-described problems, the present invention adopts, for example, the configuration described in the claims. The present specification includes a plurality of means for solving the above-described problems. For example, "an electron optical system that irradiates an electron beam emitted from an electron source to a range including a visual field on a sample; and A voltage application unit that applies a voltage of intensity reflected before the electron beam reaches the sample to the sample, and a mirror that forms an image of the mirror electrons reflected by the application of the voltage to obtain a mirror electron image An electron imaging optical system, an arithmetic unit for detecting a defect of the sample using the acquired mirror electron image, and ultraviolet light having a wavelength arbitrarily set within a range of 182 nm to 380 nm. A substrate inspection apparatus having an ultraviolet light irradiation unit that irradiates a range including the irradiation range of the electron beam during irradiation ”.
 本発明により、高速かつ高感度に基板の品質(試料の表面だけでなく内部の欠陥を含む)を評価できる。前述した以外の課題、構成及び効果は、以下の実施の形態の説明により明らかにされる。 According to the present invention, the quality of the substrate (including not only the surface of the sample but also internal defects) can be evaluated at high speed and with high sensitivity. Problems, configurations, and effects other than those described above will become apparent from the following description of embodiments.
実施例1に係る基板検査装置の構成を示す図。1 is a diagram illustrating a configuration of a substrate inspection apparatus according to Embodiment 1. FIG. 紫外光照射領域、電子線照射領域、検査領域の関係を説明する図。The figure explaining the relationship between an ultraviolet light irradiation area | region, an electron beam irradiation area | region, and an inspection area | region. 検査対象の欠陥種を示す図。The figure which shows the defect kind of inspection object. ミラー電子の軌道を説明する図。The figure explaining the orbit of mirror electrons. ミラー顕微鏡画像における欠陥の見え方を示す図。The figure which shows the appearance of the defect in a mirror microscope image. パワー半導体の製造手順を示す図。The figure which shows the manufacturing procedure of a power semiconductor. 紫外光照射によるミラー顕微鏡画像の違いを説明する図。The figure explaining the difference in the mirror microscope image by ultraviolet light irradiation. 波長の異なる紫外光を照射した場合の欠陥部のミラー顕微鏡画像を示す図。The figure which shows the mirror microscope image of the defect part at the time of irradiating the ultraviolet light from which a wavelength differs. 波長の異なる紫外光を照射した場合の欠陥部状態を示す模式図。The schematic diagram which shows the defect part state at the time of irradiating the ultraviolet light from which a wavelength differs. 試料に印加する電位とミラー電子像の明るさの関係を示す図。The figure which shows the relationship between the electric potential applied to a sample, and the brightness of a mirror electron image. 基板毎のミラー顕微鏡画像の明るさと明るさばらつきの関係を示す図。The figure which shows the relationship between the brightness of the mirror microscope image for every board | substrate, and brightness dispersion | variation. 隣接領域を比較した場合の差信号を示す図。The figure which shows the difference signal at the time of comparing an adjacent area | region. 検査処理手順例を示す図。The figure which shows the example of an inspection process procedure. 検査結果例を示す図。The figure which shows an example of a test result. 欠陥箇所のミラー顕微鏡画像を示す図。The figure which shows the mirror microscope image of a defect location. 欠陥種毎の特徴とプロファイルを示す図。The figure which shows the characteristic and profile for every defect kind. 欠陥形状の特徴と結晶成長の方向の関係を示す図。The figure which shows the relationship between the characteristic of a defect shape, and the direction of crystal growth. 複数の紫外光波長を照射して検査した結果を示す図。The figure which shows the result examined by irradiating with a plurality of ultraviolet light wavelengths.
(1)検出原理
 最初に、発明者らが提案する検出原理を説明する。発明者らは、前述の技術課題を解決する手法として、転位欠陥の特性に着目した。以下では、SiC基板の検査を例に、本発明の検出原理を説明する。発明者らは、SiC基板においては、当該材料のバンドギャップより大きなエネルギーに相当する波長の紫外光を照射することにより、試料表面や内部の欠陥箇所にキャリアを発生させることができるという特性を見出した。また、発明者らは、さらに高いエネルギーに相当する波長の紫外光を照射すると、光電子が発生し始めること、光電子が発生すると、欠陥部に蓄積されたキャリアが光電子として損失してしまうことを見出した。
(1) Detection principle First, the detection principle proposed by the inventors will be described. The inventors focused on the characteristics of dislocation defects as a technique for solving the above-described technical problems. In the following, the detection principle of the present invention will be described by taking an inspection of a SiC substrate as an example. The inventors have found that the SiC substrate can generate carriers on the surface of the sample or in a defective portion by irradiating ultraviolet light having a wavelength corresponding to energy larger than the band gap of the material. It was. In addition, the inventors have found that when ultraviolet light having a wavelength corresponding to higher energy is irradiated, photoelectrons start to be generated, and when photoelectrons are generated, carriers accumulated in the defect portion are lost as photoelectrons. It was.
 そこで、発明者らは、バンドギャップより大きいが、光電子を発生しない範囲のエネルギーの紫外光を照射して、欠陥部にキャリアを発生させる手法を提案する。キャリアの電荷状態はミラー電子の軌道に影響を与えるので、ミラー電子顕微鏡方式により、欠陥箇所を画像内のコントラストとして顕在化でき、欠陥を高感度に検出することができる。また、紫外光の照射は、欠陥の内部にキャリアを発生させるだけでなく、前述したように、不均一な帯電状態を安定させる作用も同時に生じさせる効果もある。 Therefore, the inventors propose a method of generating carriers in the defect portion by irradiating ultraviolet light having an energy that is larger than the band gap but does not generate photoelectrons. Since the charge state of the carrier affects the trajectory of the mirror electrons, the defect portion can be made visible as the contrast in the image by the mirror electron microscope method, and the defect can be detected with high sensitivity. Irradiation with ultraviolet light not only generates carriers inside the defect, but also has the effect of stabilizing the non-uniform charging state as described above.
 前述の条件を満たす紫外光の波長条件として、発明者らは、182nm以上380nm以下を見出した。なお、紫外光の基板内部への侵入深さは、波長により変わる。そのため、内部欠陥を検査する際には、検出したい欠陥の深さにより波長を変える機能を具備することが望まれる。 The inventors found 182 nm or more and 380 nm or less as a wavelength condition of ultraviolet light that satisfies the above-mentioned conditions. Note that the penetration depth of ultraviolet light into the substrate varies depending on the wavelength. Therefore, when inspecting an internal defect, it is desirable to have a function of changing the wavelength depending on the depth of the defect to be detected.
 また、発明者らは、検査の際に、ミラー顕微鏡の焦点位置を、試料表面に対してアンダー位置又はオーバー位置にずらすことを提案する。こうすることで、欠陥部からのミラー電子の軌道変化を増大でき、結果として、取得画像における欠陥部の差異を拡大することができる。 In addition, the inventors propose to shift the focus position of the mirror microscope to an under position or an over position with respect to the sample surface during inspection. By doing so, the change in the trajectory of the mirror electrons from the defective part can be increased, and as a result, the difference of the defective part in the acquired image can be enlarged.
 ところで、前述した波長条件の紫外光を基板に照射して基板内部の欠陥部にキャリアを発生させ、ミラー電子顕微鏡方式で欠陥部のミラー電子像(ミラー顕微鏡画像)を撮像すると、画像の特徴により、欠陥種を識別できることが判明した。このため、基板検査装置には、欠陥部のコントラスト及び明暗のプロファイルを用いて表面異物、貫通らせん転位、貫通刃状転位、基底面転位等の転位種を識別する画像処理部を搭載することを提案する。当該画像処理部の搭載により、検査と同時に欠陥種を識別でき、かつ、致命欠陥のみを抽出することが可能になる。また、転位欠陥については、そのプロファイルから欠陥種だけでなく、転位の規模(深さや広がり程度)も判断することができる。 By the way, when the substrate is irradiated with the ultraviolet light having the wavelength condition described above to generate carriers in the defective portion inside the substrate, and a mirror electron image (mirror microscope image) of the defective portion is captured by the mirror electron microscope method, It was found that the defect type can be identified. For this reason, the substrate inspection apparatus is equipped with an image processing unit for identifying dislocation species such as surface foreign matter, threading screw dislocation, threading edge dislocation, basal plane dislocation using the contrast and brightness profile of the defect. suggest. By mounting the image processing unit, it is possible to identify the defect type at the same time as the inspection and extract only the fatal defect. As for dislocation defects, not only the defect type but also the dislocation scale (depth and extent) can be determined from the profile.
 また、前述の技術は、光学式検査装置や他の検査装置で検出された欠陥の座標に移動して欠陥箇所の画像を撮像するレビュー機能にも応用できる。当該レビュー機能の搭載により、非破壊で欠陥の種類を判別することができる。これらの技術の採用により、ミラー電子顕微鏡方式であっても、高速かつ高感度かつ高安定的に基板を検査することが可能になる。また、同技術の採用により、欠陥種に応じて最適な条件を設定することが可能になり、ミラー電子像(ミラー顕微鏡画像)から欠陥種を非破壊で識別・分類することが可能になる。 Also, the above-described technique can be applied to a review function for moving to the coordinates of a defect detected by an optical inspection apparatus or another inspection apparatus and capturing an image of the defect portion. By mounting the review function, it is possible to determine the type of defect nondestructively. By adopting these techniques, it becomes possible to inspect a substrate at high speed, high sensitivity and high stability even with the mirror electron microscope method. Further, by adopting this technique, it becomes possible to set an optimum condition according to the defect type, and it becomes possible to identify and classify the defect type non-destructively from the mirror electron image (mirror microscope image).
(2)実施例
 以下、図面に基づいて、本発明の実施例を説明する。なお、本発明の実施例は、後述する実施例に限定されるものではなく、その技術思想の範囲において、種々の変形が可能である。
(2) Examples Hereinafter, examples of the present invention will be described with reference to the drawings. In addition, the Example of this invention is not limited to the Example mentioned later, A various deformation | transformation is possible in the range of the technical idea.
(2-1)実施例1
(2-1-1)装置構成
 本実施例では、特定波長の紫外光の照射により、基板内部の欠陥を高感度に検査する技術について説明する。図1は、本実施例に係る基板検査装置の全体構成を示している。ただし、図1においては、真空排気ポンプや排気系配管等は省略されている。
(2-1) Example 1
(2-1-1) Apparatus Configuration In this example, a technique for inspecting defects inside a substrate with high sensitivity by irradiation with ultraviolet light having a specific wavelength will be described. FIG. 1 shows the overall configuration of the substrate inspection apparatus according to the present embodiment. However, in FIG. 1, a vacuum exhaust pump, exhaust system piping, and the like are omitted.
 まず、電子光学系の主な要素を説明する。電子銃101から放出された照射電子線100aは、コンデンサレンズ102によって収束されながら、ExB偏向器103により偏向される。偏向後にクロスオーバー100dを形成した照射電子線100aは、対物レンズ107によって略平行束に整形され、被検査試料104に照射される。図1では、コンデンサレンズ102は1つだけ描かれているが、複数の電子レンズを組み合わせたシステムによっても同様の作用を実現することができる。 First, the main elements of the electron optical system will be described. The irradiation electron beam 100 a emitted from the electron gun 101 is deflected by the ExB deflector 103 while being converged by the condenser lens 102. The irradiation electron beam 100a in which the crossover 100d is formed after the deflection is shaped into a substantially parallel bundle by the objective lens 107 and is irradiated to the sample 104 to be inspected. In FIG. 1, only one condenser lens 102 is illustrated, but a similar operation can be realized by a system in which a plurality of electron lenses are combined.
 電子銃101には、通常、Zr/O/W型のショットキー電子源が用いられるが、大電流にするためにLaB6などの電子源を用いても良い。電子銃101の引き出し電圧、引き出された電子線への加速電圧、電子源フィラメントへの加熱電流などの制御(電圧及び電流制御)は、電子銃制御装置105により実行される。電子銃制御装置105は、対応する各部に制御信号を供給し、前述した電圧や電流を制御する。 For the electron gun 101, a Zr / O / W type Schottky electron source is usually used, but an electron source such as LaB6 may be used in order to increase the current. Control (voltage and current control) such as the extraction voltage of the electron gun 101, the acceleration voltage to the extracted electron beam, and the heating current to the electron source filament is executed by the electron gun control device 105. The electron gun control device 105 supplies control signals to the corresponding units, and controls the voltage and current described above.
 ExB偏向器103は、照射電子線100aの結像面近傍に設置される。照射電子線100aには、ExB偏向器103により収差が発生する。この収差を補正する必要がある場合、コンデンサレンズ102とExB偏向器103の間に収差補正用のExB偏向器106を更に配置する。ExB偏向器103により、被検査試料104に対して垂直な軸に沿うように偏向された照射電子線100aは、対物レンズ107により被検査試料104表面に対してほぼ垂直な方向に入射する電子線に形成される。 The ExB deflector 103 is installed in the vicinity of the imaging surface of the irradiation electron beam 100a. An aberration is generated in the irradiation electron beam 100 a by the ExB deflector 103. When this aberration needs to be corrected, an ExB deflector 106 for correcting aberration is further disposed between the condenser lens 102 and the ExB deflector 103. The irradiated electron beam 100a deflected by the ExB deflector 103 along the axis perpendicular to the sample 104 to be inspected is incident in a direction substantially perpendicular to the surface of the sample 104 to be inspected by the objective lens 107. Formed.
 試料ステージ108には、被検査基板104が搭載される。試料ステージ108に至るまでの搬送系は、図1では省略している。被検査試料104には、照射電子線100aの加速電圧に近い負電位が印加されている。この負電位により、照射電子線100aは、被検査試料104の表面近傍で反射し、さらに、この負電位によって上方に加速される。被検査試料104に印加される電圧の供給と制御は、試料電圧制御部109が行う。被検査試料104の極近傍で照射電子線100aを反射させるには、照射電子線100aの加速電圧と前述の負電位との差を高精度に制御する必要がある。このため、試料電圧制御部109と電子銃制御部105とは互いに連動制御される必要がある。なお、被検査試料104を試料ステージ108に搭載するための搬送時や被検査試料104を装置の外に搬出する際には、この負電位は試料ステージ108に印加されないように制御される。従って、画像を撮像する際、すなわち照射電子線100aを照射する際だけに被検査試料104に負電位を印加している。 The inspection substrate 104 is mounted on the sample stage 108. The transport system up to the sample stage 108 is omitted in FIG. A negative potential close to the acceleration voltage of the irradiation electron beam 100a is applied to the sample 104 to be inspected. Due to this negative potential, the irradiated electron beam 100a is reflected near the surface of the sample 104 to be inspected, and further accelerated upward by this negative potential. The sample voltage controller 109 supplies and controls the voltage applied to the sample 104 to be inspected. In order to reflect the irradiation electron beam 100a in the very vicinity of the specimen 104 to be inspected, it is necessary to control the difference between the acceleration voltage of the irradiation electron beam 100a and the negative potential described above with high accuracy. Therefore, the sample voltage control unit 109 and the electron gun control unit 105 need to be controlled in conjunction with each other. Note that the negative potential is controlled so as not to be applied to the sample stage 108 when transporting the sample 104 to be inspected on the sample stage 108 or when the sample 104 to be inspected is carried out of the apparatus. Therefore, a negative potential is applied to the sample 104 to be inspected only when an image is captured, that is, when the irradiation electron beam 100a is irradiated.
 被検査試料104の近傍で反射した電子線100cは、対物レンズ107により収束作用を受ける。電子線100cは、対物レンズ107とExB偏向器103の間でクロスオーバー100bを形成する。ExB偏向器103は、下方から進行した電子線100cには偏向作用を持たないよう制御されている。このため、ExB偏向器103を通過した電子線100cは、そのまま垂直に上昇し、中間レンズ110、投影レンズ111により画像検出部112に拡大投影される。図1では、投影レンズ111が1つのレンズである場合を描いているが、像補正の目的で複数のレンズを組み合わせたシステムでも良い。画像検出部112は、検出された像信号を電気信号に変換し、画像処理部116に転送する。 The electron beam 100 c reflected in the vicinity of the sample 104 to be inspected is subjected to a convergence action by the objective lens 107. The electron beam 100 c forms a crossover 100 b between the objective lens 107 and the ExB deflector 103. The ExB deflector 103 is controlled so as not to deflect the electron beam 100c traveling from below. For this reason, the electron beam 100 c that has passed through the ExB deflector 103 rises vertically as it is, and is enlarged and projected onto the image detection unit 112 by the intermediate lens 110 and the projection lens 111. Although FIG. 1 illustrates a case where the projection lens 111 is a single lens, a system in which a plurality of lenses are combined for the purpose of image correction may be used. The image detection unit 112 converts the detected image signal into an electrical signal and transfers the electrical signal to the image processing unit 116.
 画像検出部112は、反射電子を光学像に変換するための蛍光板112aと、光学画像検出装置112bと、光学像伝送系112cとより構成される。光学像伝送系112cには、光ファイバー束が用いられる。光学画像検出装置112bは、その受光面上に結像された光学像を電気的な信号に変換して出力する。光学画像検出装置112bには、通常の静止した被写体に対する画像取得モードと、移動する被写体に対して時間遅延積分(TDI)するモードを備えたデバイスが用いられる。 The image detection unit 112 includes a fluorescent plate 112a for converting reflected electrons into an optical image, an optical image detection device 112b, and an optical image transmission system 112c. An optical fiber bundle is used for the optical image transmission system 112c. The optical image detection device 112b converts the optical image formed on the light receiving surface into an electrical signal and outputs it. As the optical image detection device 112b, a device having an image acquisition mode for a normal stationary subject and a mode for time delay integration (TDI) for a moving subject is used.
 画像処理部116は、画像信号記憶部116aと欠陥判定部116bより構成される。画像処理部116は、電子線照射位置(偏向信号)データとステージ位置データを電子光学系制御装置113、ステージ制御装置115から各々受け取り、画像検出部112で検出された画像データを被検査試料104上の座標データと関係づけて画像信号記憶部116aに記憶する。欠陥判定部116bは、被検査試料104上の座標が付された画像データと、(1)予め設定された値との比較処理、(2)隣接する類似箇所で取得若しくは記憶された画像データとの比較処理、又は、(3)設計データに基づく画像予測データとの比較処理などを実行することにより、欠陥の有無を判定する。 The image processing unit 116 includes an image signal storage unit 116a and a defect determination unit 116b. The image processing unit 116 receives electron beam irradiation position (deflection signal) data and stage position data from the electron optical system control device 113 and the stage control device 115, respectively, and receives the image data detected by the image detection unit 112 as the sample 104 to be inspected. The image data is stored in the image signal storage unit 116a in association with the upper coordinate data. The defect determination unit 116b includes (1) a comparison process between the image data with coordinates on the inspection sample 104 and a preset value, and (2) image data acquired or stored at an adjacent similar location. The presence / absence of a defect is determined by executing a comparison process or (3) a comparison process with image prediction data based on design data.
 「欠陥有」と判定された場合、欠陥判定部116bは、該当座標を有する画像データの保存処理、欠陥部の詳細な画像特徴に基づいた欠陥種の判定処理等を実行する。前述の欠陥判定で使用する情報(例えば被検査試料104に予め対応付けられている判断基準、ユーザによって設定された基準パラメータや数値等)を、検査装置制御部117から選択する。いずれの方法を用いる場合にも、欠陥判定処理には、検査装置制御部117に予め設定されている判定値と判定方法が使用される。 When it is determined that “there is a defect”, the defect determination unit 116b executes a storage process of image data having the corresponding coordinates, a defect type determination process based on detailed image characteristics of the defect part, and the like. Information to be used in the above-described defect determination (for example, determination criteria associated in advance with the sample 104 to be inspected, reference parameters and numerical values set by the user, etc.) is selected from the inspection apparatus control unit 117. In any case, the determination value and the determination method set in advance in the inspection apparatus controller 117 are used for the defect determination process.
 基板検査装置を構成する各部の動作条件は、検査装置制御部117を通じて入出力される。検査装置制御部117を通じ、電子線発生時の(1)加速電圧、(2)照射する電子ビームの電流量、(3)電子線の偏向幅、(4)電子線の偏向速度、(5)試料ステージの移動速度、(6)画像検出装置による画像検出タイミング等々の諸条件が予め入力される。検査装置制御部117は、各要素の制御信号を総括的に制御し、ユーザとのインターフェースを提供する。検査装置制御部117にはモニタ118が接続されている。ユーザは、モニタ118に表示されたインターフェース画面を通じ、検査の内容に応じた上記動作に係る条件の一部を設定できるようになっている。 The operating conditions of each part constituting the substrate inspection apparatus are input / output through the inspection apparatus control unit 117. Through the inspection device control unit 117, (1) acceleration voltage at the time of electron beam generation, (2) current amount of the irradiated electron beam, (3) electron beam deflection width, (4) electron beam deflection speed, (5) Various conditions such as the moving speed of the sample stage and (6) image detection timing by the image detection apparatus are input in advance. The inspection device control unit 117 comprehensively controls the control signals of each element and provides an interface with the user. A monitor 118 is connected to the inspection apparatus control unit 117. The user can set a part of the conditions related to the above operation according to the contents of the inspection through the interface screen displayed on the monitor 118.
 前述したように、本実施例に係る基板検査装置では、被検査試料104の表面電位を安定化する目的だけでなく、被検査基板104の内部欠陥を顕在化する目的でも、紫外光を照射する。すなわち、本実施例の基板検査装置には、紫外光照射系126が搭載されている。 As described above, in the substrate inspection apparatus according to this embodiment, ultraviolet light is irradiated not only for the purpose of stabilizing the surface potential of the sample 104 to be inspected but also for the purpose of revealing internal defects in the substrate 104 to be inspected. . That is, the ultraviolet ray irradiation system 126 is mounted on the substrate inspection apparatus of the present embodiment.
 紫外光照射系126は、紫外線光源121、アパーチャ123、反射ミラー124により構成される。紫外線光源121は、(1)単一波長を照射する光源の場合、(2)必要なエネルギーを含むある範囲のエネルギーを発生させる光源の場合、(3)分光器により特定エネルギーを選んで照射する光源の場合があり、いずれの光源でも良い。本実施例では、複数の波長の紫外光を被検査試料104に照射することを想定する。このため、紫外線光源121とアパーチャの123の間にバンドパスフィルタ122を設け、所定の波長の光を選択できる構造を採用している。紫外光を照射するタイミング及びバンドパスフィルタ122の選択は、紫外光制御部125に行われる。被検査試料104毎に予め登録する紫外光照射条件は、他の条件と同様、モニタ118に表示されるインターフェース画面を通じて入力され、検査装置制御部117に登録される。 The ultraviolet light irradiation system 126 includes an ultraviolet light source 121, an aperture 123, and a reflection mirror 124. The ultraviolet light source 121 is (1) in the case of a light source that irradiates a single wavelength, (2) in the case of a light source that generates a certain range of energy including necessary energy, and (3) irradiates with a specific energy selected by a spectroscope. It may be a light source, and any light source may be used. In this embodiment, it is assumed that the specimen 104 is irradiated with ultraviolet light having a plurality of wavelengths. For this reason, a band pass filter 122 is provided between the ultraviolet light source 121 and the aperture 123, and a structure capable of selecting light of a predetermined wavelength is adopted. The timing of irradiating the ultraviolet light and the selection of the band pass filter 122 are performed by the ultraviolet light control unit 125. The ultraviolet light irradiation conditions registered in advance for each specimen 104 to be inspected are input through the interface screen displayed on the monitor 118 and registered in the inspection apparatus control unit 117, as with other conditions.
 図2に、紫外光の照射領域と、電子線の照射領域と、画像撮像領域(検査領域)の関係を示す。紫外光照射領域201は、電子線照射領域202より広い。その理由は、電子線の照射により表面の帯電状態が変化しないようにするためである。検査領域203は、電子線照射領域202の内側に設定される。図2では、紫外光照射領域201と電子線照射領域202は円状の領域であり、検査領域203は矩形の領域であるが、領域の大きさの関係が図2に示す関係を満たせば、各領域の形状は任意である。 FIG. 2 shows the relationship between the ultraviolet light irradiation region, the electron beam irradiation region, and the image pickup region (inspection region). The ultraviolet light irradiation region 201 is wider than the electron beam irradiation region 202. The reason is to prevent the charged state of the surface from being changed by the electron beam irradiation. The inspection area 203 is set inside the electron beam irradiation area 202. In FIG. 2, the ultraviolet light irradiation region 201 and the electron beam irradiation region 202 are circular regions, and the inspection region 203 is a rectangular region, but if the relationship between the region sizes satisfies the relationship shown in FIG. The shape of each region is arbitrary.
 被検査基板401がパワー半導体に用いられる基板である場合、その材料には、Siの他にSiC、GaN、Ga2O3などが用いられる。これらの基板では、表面異物、マイクロスクラッチ、結晶欠陥等を検査し、良品が取得できない領域を特定したり、良品率(歩留り)を予想したり、欠陥原因に対する対策を施して不良率を低減していく必要がある。 When the substrate to be inspected 401 is a substrate used for a power semiconductor, SiC, GaN, Ga2O3 or the like is used in addition to Si. With these substrates, surface foreign matter, micro scratches, crystal defects, etc. are inspected to identify areas where good products cannot be obtained, predict the yield rate (yield), and take measures against the cause of defects to reduce the defect rate. It is necessary to continue.
(2-1-2)欠陥の種類
 図3に、検査対象である欠陥の特徴例を示す。画像(a)は、基板表面に付着した異物301を示す。画像(b)は、研磨によるマイクロスクラッチ302を示す。画像(c)は、微小なピット303を示す。画像(d)は、積層欠陥304を示す。画像(e)は、表面からは見えない潜傷欠陥305を示す。画像(f)は、転位欠陥306を示す。これらの欠陥は、すべて検出する必要がある。
(2-1-2) Types of Defects FIG. 3 shows characteristic examples of defects that are inspection targets. Image (a) shows the foreign substance 301 adhering to the substrate surface. Image (b) shows micro scratch 302 by polishing. Image (c) shows minute pits 303. Image (d) shows a stacking fault 304. Image (e) shows a latent defect 305 that is not visible from the surface. Image (f) shows a dislocation defect 306. All of these defects need to be detected.
 従来では、これらの欠陥の検査に、レーザ光の照射により基板から発生する散乱光を検出する手法が用いられている。従来手法は、表面の形状変化を伴う異物301やマイクロスクラッチ302を検出できるが、表面の凹凸量が少ない他の欠陥は、高感度に検出することができなかった。例えば欠陥が微小なピット303の場合、従来手法では、十分な感度が得られない。また、従来手法は、欠陥の種別が表面に形状異常を伴わない積層欠陥304の一部、潜傷欠陥305、転位欠陥306を検出できなかった。 Conventionally, a method of detecting scattered light generated from a substrate by laser light irradiation is used for inspection of these defects. The conventional method can detect the foreign matter 301 and the micro scratch 302 accompanying the surface shape change, but cannot detect other defects with a small surface irregularity with high sensitivity. For example, in the case of a pit 303 having a very small defect, sufficient sensitivity cannot be obtained by the conventional method. Further, the conventional method cannot detect a part of the stacking fault 304, the latent defect 305, and the dislocation defect 306 in which the type of the defect is not accompanied by a shape abnormality on the surface.
 ところで、表面から見え難い転位欠陥306は、主に、貫通らせん転位、貫通刃状転位、基底面転位の3種類に分類することができる。本実施例に係る基板検査装置は、この転位欠陥306についても高感度に検査することができる。以下では、従来手法では検査が困難な各種の欠陥を高感度に検査できる理由を説明する。 By the way, dislocation defects 306 that are difficult to see from the surface can be classified mainly into three types: threading screw dislocation, threading edge dislocation, and basal plane dislocation. The substrate inspection apparatus according to the present embodiment can inspect the dislocation defects 306 with high sensitivity. Hereinafter, the reason why various defects that are difficult to inspect with the conventional method can be inspected with high sensitivity will be described.
(2-1-3)ミラー顕微鏡画像とその形成原理
 図4に、ミラー電子顕微鏡の画像形成原理を示す。図4(a)は、表面に凹形状がある場合のミラー電子の軌道を示し、図4(b)は、表面に凸形状がある場合のミラー電子の軌道を示す。前述したように、ミラー電子顕微鏡では、被検査試料401に衝突する直前位置で照射電子線100aを反転させるため、照射電子線100aが有するエネルギーに対して負の電位を被検査試料401に印加する。このため、被検査試料401の直上には電位分布が形成されている。被検査試料401に凹凸、欠陥、局所的な電位差が無い場合、電位分布は、ほぼ均一で平坦な等電位線となる。
(2-1-3) Mirror Microscope Image and Its Formation Principle FIG. 4 shows the image formation principle of the mirror electron microscope. FIG. 4A shows the trajectory of mirror electrons when the surface has a concave shape, and FIG. 4B shows the trajectory of mirror electrons when the surface has a convex shape. As described above, in the mirror electron microscope, in order to invert the irradiation electron beam 100a immediately before the collision with the sample 401 to be inspected, a negative potential is applied to the sample 401 to be inspected with respect to the energy of the irradiation electron beam 100a. . For this reason, a potential distribution is formed immediately above the sample 401 to be inspected. When the sample 401 to be inspected does not have irregularities, defects, and local potential differences, the potential distribution is substantially uniform and flat equipotential lines.
 しかし、図4(a)に示すように、被検査試料401の表面に凹形状402が存在すると、表面の電位分布である等電位線403は凹形状402にあわせて歪む。ミラー電子は等電位線に対して垂直に軌道を取る性質があるため、平坦な箇所で反射するミラー電子404aは、そのまま上方に反射する軌道となるが、等電位線403が下方向に歪んだ箇所で反射するミラー電子404bは、等電位線403の歪みの影響で軌道が内側に変化する。一方、図4(b)に示すように、表面に凸形状405が存在する場合には、等電位線403は上方向に歪み、凹形状402の場合とは逆方向に歪む。そのため、凸形状405の箇所で反射するミラー電子404bは、等電位線403の歪みの影響で軌道が外側に変化する。 However, as shown in FIG. 4A, when the concave shape 402 exists on the surface of the sample 401 to be inspected, the equipotential line 403 that is the surface potential distribution is distorted in accordance with the concave shape 402. Since the mirror electrons have a property of taking a trajectory perpendicular to the equipotential line, the mirror electron 404a reflected at a flat portion becomes a trajectory reflecting upward as it is, but the equipotential line 403 is distorted downward. The mirror electrons 404b reflected at the locations change inward due to the distortion of the equipotential lines 403. On the other hand, as shown in FIG. 4B, when the convex shape 405 is present on the surface, the equipotential line 403 is distorted upward and distorted in the opposite direction to the concave shape 402. Therefore, the trajectory of the mirror electrons 404 b reflected at the convex shape 405 changes to the outside due to the influence of the distortion of the equipotential line 403.
 図4(c)及び(d)は、試料表面に局所的な帯電が存在する場合である。図4(c)は、局所的にプラスの帯電406が存在する場合における表面の電位分布とミラー電子の軌道を示し、図4(d)は、局所的にマイナスの帯電407が存在する場合の表面の電位分布とミラー電子の軌道を示す。なお、被検査試料401は、相対的に負電位に設定されているため、等電位線403はマイナス電位の電位分布となる。 4 (c) and 4 (d) show a case where local charge exists on the sample surface. FIG. 4C shows the potential distribution on the surface and the trajectory of the mirror electrons when the positive charge 406 is locally present, and FIG. 4D shows the case where the negative charge 407 is locally present. The surface potential distribution and the mirror electron trajectory are shown. Since the sample 401 to be inspected is set to a relatively negative potential, the equipotential line 403 has a negative potential distribution.
 試料表面に相対的にプラスの電位となる箇所が局所的に存在すると、等電位線は図4(c)に示すように下方向に歪む。逆に、試料表面に想定的にマイナスの電位となる箇所が存在すると、等電位線は図4(d)に示すように上方向に歪む。すなわち、試料表面に局所的な帯電が存在する場合も、試料表面に物理的な凹凸が存在する場合と同様の電位分布となる。等電位線403が下方向に歪む場合(すなわち、相対的にプラスの帯電406の場合)、反射するミラー電子404bの軌道は内側に変化し、等電位線403が上方向に歪む場合(すなわち、相対的にマイナスの帯電407の場合)、反射するミラー電子404bの軌道は外側に変化する。 When a portion having a relatively positive potential locally exists on the sample surface, the equipotential lines are distorted downward as shown in FIG. On the other hand, if there is a place that assumes a negative potential on the sample surface, the equipotential lines are distorted upward as shown in FIG. That is, even when local charge exists on the sample surface, the potential distribution is the same as when the physical unevenness exists on the sample surface. When the equipotential line 403 is distorted downward (ie, in the case of a relatively positive charge 406), the trajectory of the reflected mirror electrons 404b changes inward, and when the equipotential line 403 is distorted upward (ie, In the case of a relatively negative charge 407), the trajectory of the reflected mirror electrons 404b changes to the outside.
 図5に、軌道変化の違いにより、ミラー顕微鏡画像のコントラストの見え方がどのように変化するかを示す。図5(a)は、表面に凹形状402が存在する場合におけるミラー顕微鏡画像のコントラストの見え方を示している。上段の画像501は、ミラー電子の合焦点位置をオーバーフォーカス409に設定した場合である。この場合、凹形状部が暗くなる。一方、下段の画像502は、ミラー電子の合焦点位置をアンダーフォーカス408に設定した場合である。この場合、凹形状部は明るくなる。 Fig. 5 shows how the appearance of the contrast of the mirror microscope image changes due to the difference in trajectory change. FIG. 5A shows how the contrast of the mirror microscope image appears when the concave shape 402 exists on the surface. The upper image 501 is obtained when the in-focus position of the mirror electrons is set to the overfocus 409. In this case, the concave shape portion becomes dark. On the other hand, the lower image 502 is obtained when the in-focus position of the mirror electrons is set to the under focus 408. In this case, the concave shape portion becomes bright.
 試料表面に凸形状が存在する場合における明暗の見え方は、試料表面に凹形状が存在する場合における明暗の見え方と関係が逆転する。すなわち、ミラー電子の合焦点位置をオーバーフォーカス409に設定した上段の画像503では凸形状部が明るくなり、ミラー電子の合焦点位置をアンダーフォーカス408に設定した下段の画像504では凸形状部が暗くなる。図4の凸形状部を例に見ると、凸形状部の等電位線403は上方向に変化するため、反射電子は広がる。そのため、オーバーフォーカスでは明るく、アンダーフォーカスでは暗く観察されるからである。 The relationship between the appearance of light and darkness when a convex shape exists on the sample surface is reversed from the relationship between the appearance of light and darkness when a concave shape exists on the sample surface. That is, in the upper image 503 in which the focus position of the mirror electrons is set to the overfocus 409, the convex shape portion becomes bright, and in the lower image 504 in which the focus position of the mirror electrons is set to the under focus 408, the convex shape portion is dark. Become. Looking at the convex portion of FIG. 4 as an example, the equipotential line 403 of the convex portion changes upward, so that the reflected electrons spread. For this reason, it is observed bright in overfocus and dark in underfocus.
 前述したように、試料表面に物理的な凹凸が存在しなくても、局所的な電位変化が存在すれば、同様の現象を示す。従って、ミラー顕微鏡画像を観察すれば、形状欠陥の場合には凹と凸を画像のコントラストから区別でき、また、形状欠陥ではない内部欠陥の場合には、内部電位がプラスかマイナスかで区別することができる。本実施例の場合、画像処理部116(欠陥判定部116b)が、ミラー顕微鏡画像のコントラストに基づく欠陥判定処理を実行する。なお、電子光学系の焦点が合う面が試料面に対してアンダーフォーカス408の位置であるかオーバーフォーカス109の位置かの情報は、例えば検査装置制御部117から画像処理部116に与えられている。欠陥部の具体的な分類方法については、図16を用いて後述する。表面の欠陥部は明暗とも強いコントラストとなる一方、内部欠陥は中間的な明るさを伴うのが特徴なので、ミラー顕微鏡画像から、形状欠陥と内部欠陥は弁別可能である。 As described above, even if there is no physical unevenness on the sample surface, the same phenomenon is exhibited if there is a local potential change. Therefore, by observing the mirror microscope image, in the case of a shape defect, the concave and convex can be distinguished from the contrast of the image, and in the case of an internal defect that is not a shape defect, the internal potential is distinguished depending on whether it is positive or negative. be able to. In this embodiment, the image processing unit 116 (defect determination unit 116b) executes a defect determination process based on the contrast of the mirror microscope image. Information about whether the surface of the electron optical system that is in focus is the position of the underfocus 408 or the position of the overfocus 109 with respect to the sample surface is given from the inspection apparatus control unit 117 to the image processing unit 116, for example. . A specific method for classifying defective portions will be described later with reference to FIG. While the surface defect portion has a strong contrast with light and dark, the internal defect is characterized by an intermediate brightness. Therefore, the shape defect and the internal defect can be discriminated from the mirror microscope image.
(2-1-4)検査条件の設定
 ここでは、SiC基板を検査する場合について説明する。SiC基板では、図6に示す手順で基板上に半導体装置としての回路を形成する。まず、バルク基板の導入処理601が実行される。次に、バルク基板の研磨処理602が実行される。以下では、研磨後のバルク基板を「研磨基板」という。この段階では、研磨により発生する表面異物やスクラッチが検査の対象となる。なお、研磨処理602では、潜傷、積層欠陥、転位も発生し易い。
(2-1-4) Setting of inspection conditions Here, a case of inspecting a SiC substrate will be described. In the SiC substrate, a circuit as a semiconductor device is formed on the substrate by the procedure shown in FIG. First, a bulk substrate introduction process 601 is executed. Next, a bulk substrate polishing process 602 is performed. Hereinafter, the polished bulk substrate is referred to as “polishing substrate”. At this stage, the surface foreign matter and scratches generated by the polishing are to be inspected. In the polishing process 602, latent scratches, stacking faults, and dislocations are also likely to occur.
 ところで、電子線の照射に起因する有機物質汚染や金属汚染が基板の表面に存在すると、基板表面にEpi成長膜を正常に成長させることができない。このため、基板の検査時には、電子線を試料に照射させない必要がある。そこで、ミラー電子顕微鏡のように、電子を試料に直接当てることなく直上に反射させる方法が有効となる。 By the way, if organic substance contamination or metal contamination caused by electron beam irradiation exists on the surface of the substrate, the Epi growth film cannot be grown normally on the substrate surface. For this reason, it is necessary not to irradiate the sample with an electron beam when inspecting the substrate. Therefore, a method of reflecting the electrons directly without hitting the sample directly like a mirror electron microscope is effective.
 研磨処理602の後、基板表面にEpi成長膜を成長させる処理603が実行される。この明細書では、Epi成長膜を表面に成長させた基板をEpi形成基板という。なお、下層に潜傷や転位が存在していた場合、結晶欠陥が発生する。Epi形成基板が形成されると、1層目のパターン形成処理604を実行する。以後、順次上層にn層のパターンを形成する処理605を実行する。処理工程の数、パターンの形状、材料等は半導体の種類によって異なる。 After the polishing process 602, a process 603 for growing an Epi growth film on the substrate surface is executed. In this specification, a substrate on which an Epi growth film is grown is referred to as an Epi formation substrate. In addition, when there are latent scratches or dislocations in the lower layer, crystal defects are generated. When the Epi formation substrate is formed, the first layer pattern formation processing 604 is executed. Thereafter, processing 605 for forming an n-layer pattern on the upper layer sequentially is executed. The number of processing steps, pattern shapes, materials, and the like vary depending on the type of semiconductor.
 最後に、完成した半導体回路の電気特性を評価する処理606を実行し、良品と不良品を判別する。この一連の処理工程を経て、半導体回路が完成する。この一連の処理工程のうち、パターンを形成する前の研磨基板の形成段階(処理602)やEpi成長膜を形成した段階(処理603)で基板の欠陥を高感度に検査すれば、全てのパターンが形成された後の電気特性のテスト段階(処理606)まで不良の有無が不明という状態を回避できるようになる。そこで、本実施例では、Epi成長膜を基板表面に成長させた段階で基板の検査を実施する。 Finally, a process 606 for evaluating the electrical characteristics of the completed semiconductor circuit is executed to discriminate between a good product and a defective product. The semiconductor circuit is completed through this series of processing steps. If a substrate defect is inspected with high sensitivity at the stage of forming a polishing substrate before forming a pattern (process 602) or the stage of forming an epi growth film (process 603) in this series of processing steps, all patterns are obtained. It is possible to avoid a state in which the presence or absence of a defect is unknown until the electrical characteristic test stage (process 606) after the formation of. Therefore, in this embodiment, the substrate is inspected at the stage where the Epi growth film is grown on the substrate surface.
 以下、SiC基板の検査において、所望の欠陥を高感度に検査するための方法と最適パラメータの設定について説明する。まず、Epi成長膜を成長させた後の基板(Epi形成基板)を、図1に示すミラー電子顕微鏡によって撮像した場合のミラー顕微鏡画像の例を示す。まず、紫外線を照射しない状態で取得されるミラー顕微鏡画像の例を図7(a)に示す。この場合、基板全体701に明るさむらが見られ、画面内に2箇所想定されている欠陥部702及び703のコントラストも弱い。 Hereinafter, a method for inspecting a desired defect with high sensitivity and setting of optimum parameters in the inspection of the SiC substrate will be described. First, an example of a mirror microscope image when the substrate (Epi forming substrate) after growing the Epi growth film is imaged by the mirror electron microscope shown in FIG. 1 is shown. First, an example of a mirror microscope image acquired without irradiating ultraviolet rays is shown in FIG. In this case, uneven brightness is seen on the entire substrate 701, and the contrast of the defective portions 702 and 703 assumed in two places in the screen is also weak.
 次に、365nmの波長の紫外光を基板表面に照射した状態で、同一箇所をミラー電子顕微鏡で観察した場合に取得されるミラー顕微鏡画像の例を図7(b)に示す。この場合、基板全体701に明るさむらが無くなっている。さらに、欠陥部702及び703のコントラスト差も大きくなった。特に欠陥部703は、紫外光を照射しない場合には認識自体が困難であったが、365nmの波長の紫外光の照射により明確に認識できるようになった。 Next, FIG. 7B shows an example of a mirror microscope image acquired when the same part is observed with a mirror electron microscope in a state where the substrate surface is irradiated with ultraviolet light having a wavelength of 365 nm. In this case, uneven brightness is eliminated from the entire substrate 701. Furthermore, the contrast difference between the defective portions 702 and 703 also increased. In particular, the defect portion 703 was difficult to recognize when it was not irradiated with ultraviolet light, but can be clearly recognized by irradiation with ultraviolet light having a wavelength of 365 nm.
 発明者らが、図7(b)に示すミラー顕微鏡画像が得られている状態で紫外光の照射をオフしたところ、ミラー顕微鏡画像は図7(a)に示す画像に戻った。また、再び紫外光の照射をオンしたところ、ミラー顕微鏡画像は図7(b)に示す画像に戻った。この現象を通じ、紫外光を照射している間だけ、ミラー顕微鏡画像のコントラストが大きくなることが判明した。 When the inventors turned off the irradiation of the ultraviolet light in a state where the mirror microscope image shown in FIG. 7B was obtained, the mirror microscope image returned to the image shown in FIG. Further, when the ultraviolet light irradiation was turned on again, the mirror microscope image returned to the image shown in FIG. Through this phenomenon, it was found that the contrast of the mirror microscope image is increased only during the irradiation with ultraviolet light.
 ところで、紫外光の照射により表面の明るさむらが低減する現象は、不均一な帯電状態が均一になるためと解釈できるが、この解釈では、欠陥部のコントラスト差が大きくなる現象を説明できない。そこで、発明者らは、照射する紫外光の波長を数段階に亘って変化させることで、欠陥の見え方の違いを評価した。図8に、評価結果を示す。図8は、紫外光の波長を、180nmから380nmの範囲のうち、365nm、283nm、268nm、230nm、180nmについて、予め判明している積層欠陥部を観察した結果である。 By the way, the phenomenon in which the uneven brightness of the surface is reduced by the irradiation of ultraviolet light can be interpreted as a non-uniform charging state becoming uniform, but this interpretation cannot explain the phenomenon in which the contrast difference of the defective portion becomes large. Therefore, the inventors evaluated the difference in the appearance of defects by changing the wavelength of the ultraviolet light to be irradiated in several stages. FIG. 8 shows the evaluation results. FIG. 8 shows the results of observing previously known stacking faults for 365 nm, 283 nm, 268 nm, 230 nm, and 180 nm in the wavelength range of 180 nm to 380 nm.
・波長365nmは、励起させるバンドギャップのエネルギーが3.40eVに相当する。この波長でのSiC膜への侵入深さは120μm相当になる。 • At a wavelength of 365 nm, the energy of the excited band gap corresponds to 3.40 eV. The penetration depth into the SiC film at this wavelength is equivalent to 120 μm.
・波長283nmは、励起させるバンドギャップのエネルギーが4.38eVに相当する。この波長でのSiC膜への侵入深さは約3μm相当になる。 -At a wavelength of 283 nm, the energy of the excited band gap corresponds to 4.38 eV. The penetration depth into the SiC film at this wavelength is approximately 3 μm.
・波長263nmは、励起させるバンドギャップのエネルギーが4.63eVに相当する。この波長でのSiC膜への侵入深さは約1μm相当になる。 -At a wavelength of 263 nm, the energy of the excited band gap corresponds to 4.63 eV. The penetration depth into the SiC film at this wavelength is approximately 1 μm.
・波長230nmは、励起させるバンドギャップのエネルギーが5.39eVに相当する。この波長でのSiC膜への侵入深さは約0.1μm以下になる。 • At a wavelength of 230 nm, the energy of the excited band gap corresponds to 5.39 eV. The penetration depth into the SiC film at this wavelength is about 0.1 μm or less.
・波長180nmは、励起させるバンドギャップのエネルギーが6.4eVに相当する。 • At a wavelength of 180 nm, the energy of the excited band gap corresponds to 6.4 eV.
 これらの波長の紫外光を順次照射しながら、同一箇所についてミラー顕微鏡画像を取得し、基板と欠陥の見え方を比較した。その結果、波長が365nmの場合と283nmの場合においては、欠陥部801及び802のコントラストが高くなり、かつ、内部欠陥と思われる影803及び804も観察された(図8(a)、(b))。一方、波長が268nmの場合と230nmの場合においては、表面近傍の欠陥部801及び802は、図8(a)及び(b)の場合と同様に、高いコントラストを得ることができた。しかし、内部欠陥と思われる影803及び804は、だんだんコントラストが低下し、波長が230nmの条件で観察したミラー顕微鏡画像では影を確認できなかった(図8(c)(d))。 Mirror image of the same part was acquired while sequentially irradiating ultraviolet light of these wavelengths, and the appearance of the substrate and the defect were compared. As a result, when the wavelength is 365 nm and when the wavelength is 283 nm, the contrast of the defect portions 801 and 802 is high, and shadows 803 and 804 that are considered to be internal defects are also observed (FIGS. 8A and 8B). )). On the other hand, when the wavelength was 268 nm and 230 nm, the defect portions 801 and 802 in the vicinity of the surface were able to obtain a high contrast as in the case of FIGS. 8A and 8B. However, the shadows 803 and 804 that seemed to be internal defects gradually decreased in contrast, and the shadow could not be confirmed in the mirror microscope image observed under the condition where the wavelength was 230 nm (FIGS. 8C and 8D).
 なお、本実施例では、紫外光の波長範囲を182nm以上380nm以下の範囲とするが、この範囲外である波長が180nmの紫外光を照射しながらミラー顕微鏡画像を取得すると、全体が均一に明るい画像となり、欠陥部801及び802を認識することも困難となった(図8(e))。 In this example, the wavelength range of ultraviolet light is in the range of 182 nm or more and 380 nm or less, but when a mirror microscope image is acquired while irradiating ultraviolet light having a wavelength outside this range of 180 nm, the whole is uniformly bright. It became an image, and it became difficult to recognize the defective portions 801 and 802 (FIG. 8E).
 図9を用い、上記の実験結果に対する発明者らの考察結果を説明する。図9は、図8に示す個々のミラー顕微鏡画像の取得時に、欠陥に生じる現象を模式的に示す図である。すなわち、図9は、図8で説明した各波長の紫外光の照射と欠陥に生じる現象との関係を模式的に示す図である。断面図である図9に示すように、欠陥901は、基板表面から内部方向に、約10μm前後の深さ903まで及んでいると考察した。 9 will be used to explain the results of the inventors' consideration on the above experimental results. FIG. 9 is a diagram schematically illustrating a phenomenon that occurs in a defect when the individual mirror microscope images illustrated in FIG. 8 are acquired. That is, FIG. 9 is a diagram schematically showing the relationship between the irradiation of ultraviolet light of each wavelength described in FIG. 8 and the phenomenon occurring in the defect. As shown in FIG. 9 which is a cross-sectional view, the defect 901 was considered to extend to the depth 903 of about 10 μm from the substrate surface to the inside.
・図9(a)は、紫外光の波長が365nmの場合の現象を説明する模式図である。この波長の場合、欠陥の深さ903の最深部である10μmより深くまで紫外光が到達でき、その結果、欠陥の全域に沿ってキャリアである電荷902が発生し、同時にトラップされたと考えた。 FIG. 9A is a schematic diagram for explaining the phenomenon when the wavelength of ultraviolet light is 365 nm. In the case of this wavelength, it was considered that ultraviolet light could reach deeper than 10 μm, which is the deepest part of the defect depth 903, and as a result, charges 902 as carriers were generated along the entire area of the defect and simultaneously trapped.
・図9(b)は、紫外光の波長が283nmの場合の現象を説明する模式図である。この波長の場合、紫外光の侵入深さは約3μmまで浅くなると考えられる。このため、キャリアが発生してトラップされる範囲も、基板表面から約3μmの深さまでであったと考察した。 FIG. 9B is a schematic diagram for explaining the phenomenon when the wavelength of ultraviolet light is 283 nm. In the case of this wavelength, the penetration depth of ultraviolet light is considered to be as shallow as about 3 μm. For this reason, the range in which carriers are generated and trapped was considered to be about 3 μm deep from the substrate surface.
・図9(c)は、紫外光の波長が268nmの場合の現象を説明する模式図である。この波長の場合、紫外光の侵入深さは1μmまで浅くなり、その分、キャリアがトラップされる深さも浅くなったと考察した。 FIG. 9C is a schematic diagram for explaining the phenomenon when the wavelength of ultraviolet light is 268 nm. In the case of this wavelength, it was considered that the penetration depth of ultraviolet light became as shallow as 1 μm, and the depth at which carriers were trapped was also reduced accordingly.
・図9(d)は、紫外光の波長が230nmの場合の現象を説明する模式図である。この波長の場合、基板の極表面にしかキャリアが発生しないため、表面付近の欠陥のみが電位変化として観察されたと考察した。 FIG. 9D is a schematic diagram for explaining the phenomenon when the wavelength of ultraviolet light is 230 nm. In the case of this wavelength, since carriers are generated only on the extreme surface of the substrate, it was considered that only defects near the surface were observed as potential changes.
・図9(e)は、本実施例の波長条件外である180nmの波長の紫外光を照射した場合の現象を説明する模式図である。この場合、SiC基板より光電子904が発生し、全体に明るくなり、欠陥のコントラストが著しく低下し、欠陥の認識が困難となったと考察した。これは、光電子の発生により、欠陥にトラップされたキャリアが発散してしまったためと思われる。 FIG. 9E is a schematic diagram for explaining a phenomenon in the case where ultraviolet light having a wavelength of 180 nm, which is outside the wavelength condition of the present embodiment, is irradiated. In this case, it was considered that photoelectrons 904 were generated from the SiC substrate, brightened as a whole, the defect contrast was remarkably lowered, and defect recognition became difficult. This is presumably because the carriers trapped in the defect are emitted due to the generation of photoelectrons.
 以上の考察結果より、発明者らは、紫外光照射の最適検査条件を波長365nmと設定した。本実施例では、深さ10μm程度の積層欠陥をターゲットに最適条件を探索したため、最適検査条件の波長が365nmとなったが、検査の対象となる欠陥種及びその予想深さにより最適な波長条件は異なる。そこで、発明者らは、紫外線の波長範囲を、182nm以上380nm以下とした。本実施例の検査装置制御部117には、欠陥種及び/又は予想深さと最適な紫外線の波長条件との関係が、テーブル形式で不図示の記憶領域に格納されている。なお、検査の対象となる欠陥種及びその予想深さの情報は、例えばユーザがインターフェース画面を通じて入力する。 From the above consideration results, the inventors set the optimum inspection condition for ultraviolet light irradiation at a wavelength of 365 nm. In this example, since the optimum condition was searched for the stacking fault with a depth of about 10 μm as the target, the wavelength of the optimum inspection condition was 365 nm, but the optimum wavelength condition depends on the defect type to be inspected and its expected depth. Is different. Therefore, the inventors set the wavelength range of ultraviolet rays to 182 nm or more and 380 nm or less. In the inspection apparatus control unit 117 of this embodiment, the relationship between the defect type and / or the expected depth and the optimum wavelength condition of the ultraviolet ray is stored in a storage area (not shown) in a table format. Note that the defect type to be inspected and information on the expected depth thereof are input by the user through an interface screen, for example.
 さらに、発明者らは、画像取得時に照射する紫外光の最適な波長が決まった後に、当該条件で紫外光を照射しながら、被検査試料に印加する電圧を微調整した。これは、できるだけ被検査試料の表面近傍でミラー電子を反射させた方が表面の電位変化に対して感度が高くなるためである。すなわち、欠陥の有無に対する感度を向上させることを目的としている。 Furthermore, the inventors fine-tuned the voltage applied to the sample to be inspected while irradiating the ultraviolet light under the conditions after determining the optimum wavelength of the ultraviolet light irradiated at the time of image acquisition. This is because it is more sensitive to changes in the surface potential when mirror electrons are reflected as close to the surface of the specimen as possible. That is, it aims at improving the sensitivity with respect to the presence or absence of a defect.
 被検査試料に対する電圧の印加は、図1の試料電圧制御部109が行う。もっとも、試料電圧制御部109が印加する電圧は、検査装置制御部117を通じて設定される。ユーザは、モニタ118を通じて、被検査試料に印加する電位の目標値を入力し、当該電位に対してプラス方向とマイナス方向に電位を変化させる範囲と、その変化のピッチを入力する。例えば目標値を-3kVとする場合に、-3.2kVから-2.8kVの範囲を0.02kVきざみで変化させながら各々の画像データを取得し、その際の明るさの変化を測定するように指示を入力する。この場合、-3.20kV、-3.18kV、…と段階的に試料電位が変化し、その都度、ミラー顕微鏡画像を取得する。 The voltage application to the sample to be inspected is performed by the sample voltage control unit 109 in FIG. However, the voltage applied by the sample voltage control unit 109 is set through the inspection apparatus control unit 117. The user inputs a target value of the potential applied to the sample to be inspected through the monitor 118, and inputs a range in which the potential is changed in the plus direction and the minus direction with respect to the potential and a pitch of the change. For example, when the target value is -3 kV, each image data is acquired while changing the range from -3.2 kV to -2.8 kV in steps of 0.02 kV, and an instruction is given to measure the change in brightness at that time. input. In this case, the sample potential changes step by step from −3.20 kV to −3.18 kV, and a mirror microscope image is acquired each time.
 画像取得の際には、電子銃101から放出された照射電子線100aが被検査試料104の表面近傍で反射され、反射した電子線100cが画像検出部112で検出される。画像検出部112を通じて検出されたミラー顕微鏡画像として画像処理部116に保存される。本実施例の場合、保存されたミラー顕微鏡画像のうち所定の領域部分の平均明るさを算出する機能が検査装置制御部117に搭載されている。 At the time of image acquisition, the irradiation electron beam 100a emitted from the electron gun 101 is reflected near the surface of the sample 104 to be inspected, and the reflected electron beam 100c is detected by the image detection unit 112. It is stored in the image processing unit 116 as a mirror microscope image detected through the image detection unit 112. In the case of the present embodiment, a function for calculating the average brightness of a predetermined region portion of the stored mirror microscope image is mounted on the inspection apparatus control unit 117.
 照射エネルギーが3keVの照射電子線100aを照射した場合、当該電子線はマイナスのため、-3.20kVに印加された試料電位により斥力が作用し、試料面に対して上方に反射される。全ての電子がミラー反射されている間は、明るさは一定のレベルとなる。しかし、照射する電子線のエネルギーはばらつきを有する場合があり、電源の精度により、電子のエネルギーや試料電位が微妙な誤差を有する場合がある。そのため、電子線のエネルギーと試料表面の負電位がほぼ同じ程度に近づくと、大半の電子はミラー反射するが、一部の電子は試料表面に照射されるようになる。試料表面に電子線が照射されると二次電子が発生するが、その量はミラー反射する電子に比べて少ない。ただし、取得されるミラー顕微鏡画像は、全部の電子がミラー反射した場合に比べて明るさが低下する。 When the irradiation electron beam 100a having an irradiation energy of 3 keV is irradiated, the electron beam is negative, so that a repulsive force acts on the sample potential applied to −3.20 kV and is reflected upward with respect to the sample surface. While all the electrons are mirror-reflected, the brightness is at a constant level. However, the energy of the irradiated electron beam may vary, and the energy of the electron and the sample potential may have a subtle error depending on the accuracy of the power source. For this reason, when the energy of the electron beam and the negative potential on the sample surface approach approximately the same level, most of the electrons are mirror-reflected, but some of the electrons are irradiated on the sample surface. When the sample surface is irradiated with an electron beam, secondary electrons are generated, but the amount is smaller than that of the mirror-reflected electrons. However, the brightness of the acquired mirror microscope image is lower than when all the electrons are mirror-reflected.
 一方、試料電位をプラス側に変化させると、試料表面に照射される電子の量が増え、さらにミラー顕微鏡画像の明るさが低下する。この動作を繰り返した後、全ての電子が被検査試料に照射されるようになると、ミラー顕微鏡画像の明るさは変化しなくなる。この様子を図10に示す。図10に、被検査試料に印加する電圧を変化させた際のミラー顕微鏡画像(ミラー電子像)の明るさ変化を示す。グラフの横軸が被検査試料に印加する負の電圧の大きさ、縦軸がミラー顕微鏡画像の明るさを示す。なお、印加する負の電圧の絶対値は、横軸の左側ほど大きく、横軸の右側ほど小さいものとする。 On the other hand, when the sample potential is changed to the plus side, the amount of electrons irradiated on the sample surface increases and the brightness of the mirror microscope image further decreases. After repeating this operation, the brightness of the mirror microscope image does not change when all the electrons are irradiated onto the sample to be inspected. This is shown in FIG. FIG. 10 shows a change in brightness of a mirror microscope image (mirror electron image) when the voltage applied to the sample to be inspected is changed. The horizontal axis of the graph indicates the magnitude of the negative voltage applied to the sample to be inspected, and the vertical axis indicates the brightness of the mirror microscope image. Note that the absolute value of the negative voltage to be applied is larger on the left side of the horizontal axis and smaller on the right side of the horizontal axis.
 被検査試料の電位が照射電子線の電位よりも負側の場合には、電子はミラー反射する。このため、ミラー顕微鏡画像の明るさはレベル1001で安定している。被検査試料への負の印加電圧を徐々にプラス側に変化させると、やがて、明るさが変化し始めるポイント1002に達する。このポイントでは、前述したように、照射している電子の全てが反射せず、一部が試料表面に当たるようになり、その結果、反射するミラー電子の量が減り始める。さらに印加電圧をプラス側に変化させると、ミラー顕微鏡画像の明るさは徐々に暗くなり、ある一定のレベルで変化が少なくなるポイント1003に達する。このポイント以上は、印加電圧を大きくしてもミラー顕微鏡画像の明るさは変わらず、照射電子の全てが試料表面を照射する領域1004になる。 When the potential of the sample to be inspected is more negative than the potential of the irradiated electron beam, the electrons are mirror-reflected. For this reason, the brightness of the mirror microscope image is stable at level 1001. When the negative applied voltage to the sample to be inspected is gradually changed to the plus side, the point 1002 at which the brightness starts to change is reached. At this point, as described above, not all of the irradiated electrons are reflected and a part of the electrons hits the sample surface, and as a result, the amount of mirror electrons to be reflected starts to decrease. When the applied voltage is further changed to the plus side, the brightness of the mirror microscope image gradually decreases and reaches a point 1003 where the change is reduced at a certain level. Above this point, the brightness of the mirror microscope image does not change even when the applied voltage is increased, and the region 1004 is where all the irradiated electrons irradiate the sample surface.
 以上より、被検査試料に印加している負の電圧を徐々にプラス側に変化させた場合に、ミラー顕微鏡画像の明るさが変化を開始するポイント1002の直前の電圧が、最も試料表面の近傍で電子を反射させる。この場合、電子が、試料表面の電位変化を示す等電位線に近いところで反射するため、結果として等電位線の変化に対して影響を受け易くなり、表面電位の変化に対して高感度な条件を設定することができる。 As described above, when the negative voltage applied to the specimen to be inspected is gradually changed to the positive side, the voltage immediately before the point 1002 at which the brightness of the mirror microscope image starts changing is closest to the specimen surface. To reflect electrons. In this case, the electrons are reflected near the equipotential line indicating the potential change of the sample surface, and as a result, it is easily affected by the change of the equipotential line, and the condition is highly sensitive to the change of the surface potential. Can be set.
 また、発明者らは、図10の最適範囲1005に当たる電圧(ポイント1002からマイナス側に一定範囲内の電圧)を、感度と安定性を両立できる条件として設定する。すなわち、試料印加電圧の最適値とする。この感度と安定性を両立できる範囲を求めるために、各印加電圧とその前後でミラー顕微鏡画像の明るさの平均を算出することで明るさのばらつき影響を抑制し、さらに該明るさの平均の変化を求め、変化率が一定以上になる点を、変化が開始するポイント1002とした。本実施例では、ポイント1002の接線と明るさに変化がないレベル1001の接線との交点から-5eV以下と前記交点から-100eV以下の範囲を最適範囲1005とした。 Further, the inventors set a voltage corresponding to the optimum range 1005 in FIG. 10 (a voltage within a certain range from the point 1002 to the minus side) as a condition that can achieve both sensitivity and stability. That is, the sample applied voltage is set to the optimum value. In order to obtain a range in which both sensitivity and stability can be achieved, the influence of brightness variation is suppressed by calculating the average of the brightness of the mirror microscope image before and after each applied voltage. The point at which the change is obtained and the rate of change becomes equal to or higher than a certain value is defined as a point 1002 at which the change starts. In this embodiment, the optimum range 1005 is defined as a range of −5 eV or less from the intersection of the tangent of the point 1002 and the tangent of the level 1001 where there is no change in brightness and −100 eV or less from the intersection.
 この最適範囲1005は、感度を決める際にユーザが目的や試料の状態、必要な感度に応じて任意に変えることができる。最適条件を求めない場合には、目標値である-3kVに対し-100eVのように一定の量をマイナス側にして評価すれば良い。これらの試料印加電圧の最適化機能は、例えば検査装置制御部117に搭載する。 The optimum range 1005 can be arbitrarily changed according to the purpose, the state of the sample, and the required sensitivity when determining the sensitivity. When the optimum condition is not obtained, evaluation may be made with a certain amount of minus −side, such as −100 eV with respect to the target value of −3 kV. These sample application voltage optimization functions are mounted on the inspection apparatus control unit 117, for example.
 以上説明した画像取得条件が決定した後に、発明者らは、実際にミラー顕微鏡画像を取得し、欠陥の有無を判定する条件を最適化した。ここでは、欠陥判定を最適化する手順を説明する。まず、被検査基板の所定領域について撮像されたミラー顕微鏡画像を画像処理部116に転送し、記憶部に保存する。 After determining the image acquisition conditions described above, the inventors optimized the conditions for actually acquiring a mirror microscope image and determining the presence or absence of defects. Here, a procedure for optimizing defect determination will be described. First, the mirror microscope image captured for a predetermined area of the substrate to be inspected is transferred to the image processing unit 116 and stored in the storage unit.
 保存されたミラー顕微鏡画像の一例を図11に示す。図11(a)は、第1の基板の画像であり、図11(b)は、同一種類の第2の基板の画像である。さらに、図11(c)は、前述の第1及び第2の基板とは別種類の第3の基板の画像である。これらの画像のうち所定箇所1101(図中点線で示す)の明るさプロファイルを抜き出して描画したのが図11(d)、図11(e)、図11(f)である。 An example of the stored mirror microscope image is shown in FIG. FIG. 11A is an image of a first substrate, and FIG. 11B is an image of a second substrate of the same type. Further, FIG. 11C is an image of a third substrate of a different type from the first and second substrates described above. FIGS. 11 (d), 11 (e), and 11 (f) show the brightness profiles extracted from these images at predetermined locations 1101 (indicated by dotted lines in the figure).
 第1の基板に対応する図11(a)に示す画像と第2の基板に対応する図11(b)に示す画像の場合、明るさのレベルが多少異なるものの、明るさのばらつき1102及び1103は、ほぼ同レベルである。一方、膜種や加工条件が異なる第3の基板に対応する図11(c)に示す画像の明るさプロファイルは、明るさのレベルが異なるだけでなく、画像の面内におけるばらつき1104が大きくなっている。これらの多様な状態が想定されるミラー顕微鏡画像に基づいて、欠陥箇所と欠陥でない箇所を自動的に弁別可能するために、発明者らは、まず明るさの平均値を規格化することを提案する。次に、膜種毎に明るさのばらつきが異なるため、発明者らは、膜種毎にばらつきの許容範囲を決定することを提案する。 In the case of the image shown in FIG. 11A corresponding to the first substrate and the image shown in FIG. 11B corresponding to the second substrate, brightness variations 1102 and 1103, although the brightness levels are slightly different. Are almost at the same level. On the other hand, the brightness profile of the image shown in FIG. 11C corresponding to the third substrate having different film types and processing conditions not only has different brightness levels, but also has a large in-plane variation 1104. ing. In order to automatically discriminate between defective and non-defective parts based on the mirror microscope images in which these various states are assumed, the inventors first proposed normalizing the average value of brightness. To do. Next, since the variation in brightness differs for each film type, the inventors propose to determine an allowable range of variation for each film type.
 ここでは、図11(d)の許容範囲1105と図11(f)の許容範囲1106を、欠陥の有無を判定する際に使用するしきい値とした。明るさばらつき1102、1103及び1104は、基板毎に又は同一基板の場所毎に変動することを想定し、同一種の基板である第1の基板と第2の基板については、許容範囲1105(すなわち、しきい値)をばらつき1102及び1103よりも大きい範囲に設定した。同様に、第3の基板の膜種については、許容範囲1106(すなわち、しきい値)を、ばらつき1104よりも大きい範囲に設定した。 Here, the allowable range 1105 in FIG. 11 (d) and the allowable range 1106 in FIG. 11 (f) are threshold values used when determining the presence or absence of defects. The brightness variations 1102, 1103, and 1104 are assumed to vary from substrate to substrate or from place to place on the same substrate. For the first substrate and the second substrate, which are the same type of substrate, an allowable range 1105 (ie, , Threshold value) was set to a range larger than the variations 1102 and 1103. Similarly, for the film type of the third substrate, the allowable range 1106 (that is, the threshold value) was set to a range larger than the variation 1104.
 なお、許容範囲は、別の方法を使用して設定することもできる。例えば被検査基板毎にミラー顕微鏡画像を取得し、図12に示すように、隣接画像間で画像信号を比較することにより、許容範囲を決定する方法も考えられる。まず、図11の場合と同様、被検査基板毎に明るさをある一定のレベルに規格化する。その後、被検査基板内の第1の領域についての画像(a)と、隣接する第2の領域についての画像(b)とを比較し、差信号1201を計算する。隣接領域と比較することで、明るさや面内ばらつきの変化を小さくすることができる。画像(c)に示す差信号1201に対し、許容レベル1202を設定することで、ばらつきの範囲と欠陥を判定することができる。これらの方法により、欠陥判定のしきい値を調整し、所望の欠陥を検出し、ばらつきは検出しない画像処理条件を選択した。この欠陥有無を判定する条件を最適化する機能は、例えば画像処理部116に搭載する。 Note that the allowable range can be set using another method. For example, a method of determining a permissible range by acquiring a mirror microscope image for each substrate to be inspected and comparing image signals between adjacent images as shown in FIG. First, as in the case of FIG. 11, the brightness is normalized to a certain level for each substrate to be inspected. Thereafter, the image (a) for the first region in the substrate to be inspected is compared with the image (b) for the adjacent second region, and a difference signal 1201 is calculated. By comparing with the adjacent region, changes in brightness and in-plane variation can be reduced. By setting an allowable level 1202 for the difference signal 1201 shown in the image (c), it is possible to determine a range of variation and a defect. By these methods, the threshold for defect determination was adjusted, a desired defect was detected, and an image processing condition that did not detect variation was selected. A function for optimizing the condition for determining the presence / absence of a defect is mounted on the image processing unit 116, for example.
(2-1-5)検査処理
 前述の検査条件の設定の後、基板検査装置による被検査基板の検査を開始する。図13に、検査処理手順を示す。まず、モニタ118に表示されるインターフェース画面を通じ、検査条件を入力する(ステップ1301)。検査条件とは、被検査基板を特定する情報(例えばウエハを特定するカセット又はフープ内のスロット、ウエハのID(例えばロット番号、ウエハ番号))、レシピ名(紫外光の照射条件、ビーム条件、被検査基板に印加する負の電圧値等)、被検査領域、検査結果出力内容や形式、検査実施後のレビューや画像保存に関する指示内容等である。
(2-1-5) Inspection Processing After setting the above inspection conditions, inspection of the substrate to be inspected by the substrate inspection apparatus is started. FIG. 13 shows an inspection processing procedure. First, inspection conditions are input through an interface screen displayed on the monitor 118 (step 1301). The inspection conditions are information for specifying a substrate to be inspected (for example, a slot in a cassette or a hoop for specifying a wafer, a wafer ID (for example, lot number, wafer number)), a recipe name (ultraviolet light irradiation condition, beam condition, Negative voltage value to be applied to the substrate to be inspected), inspection area, inspection result output contents and format, instruction contents regarding review and image storage after inspection, and the like.
 例えば4インチのSiC基板に存在する深さ20μmまでの欠陥を検出したい場合、ユーザは、電子線の照射時に波長350nmの紫外光を照射する条件が設定された「光学条件A」を選択する。また、ユーザは、高感度に欠陥の電位を測定するために、前述したように試料印加電圧の最適化を指示するモードを「オン(ON)」に設定する。もっとも、検査対象となる深さと当該深さを検査するために最適な紫外光の波長を対応づけた関係が予めハードディスクなどの記憶部に記憶されている場合には、ユーザは、検査対象としたい深さを選択又は入力するだけで、これらの光学条件を自動的に設定することができる。 For example, when detecting a defect with a depth of 20 μm existing on a 4-inch SiC substrate, the user selects “optical condition A” in which conditions for irradiating ultraviolet light with a wavelength of 350 nm are set when irradiating an electron beam. In addition, in order to measure the defect potential with high sensitivity, the user sets the mode for instructing optimization of the sample applied voltage to “ON” as described above. However, when a relationship in which a depth to be inspected and an optimum ultraviolet light wavelength for inspecting the depth are associated with each other is stored in advance in a storage unit such as a hard disk, the user wants to make the inspection target. These optical conditions can be set automatically simply by selecting or inputting the depth.
 次に、ユーザは、ウエハサイズと、ウエハの中央と周辺5箇所の計6箇所の20mm x 20mmの領域を設定する「領域条件B」を選択する。さらに、ユーザは、検査結果出力内容として、被検査領域の欠陥座標リスト、欠陥マップ、欠陥と認識された箇所の画像を自動的に保存し、外部サーバに転送することを設定する「出力様式C」を選択する。これらの条件を組み合わせたものが検査レシピである。ユーザは、SiC基板の検査レシピDというレシピ名を選択し、このレシピ名をモニタ118から入力する。 Next, the user selects “area condition B” for setting a wafer size and a total of 6 areas of 20 mm × 20 mm in total, 5 in the center and 5 in the periphery of the wafer. Further, the user sets the inspection result output contents to automatically save the defect coordinate list of the inspected area, the defect map, and the image of the portion recognized as the defect, and to transfer to the external server. ”Is selected. A combination of these conditions is an inspection recipe. The user selects a recipe name “SiC substrate inspection recipe D” and inputs this recipe name from the monitor 118.
 これら条件の入力後、ユーザは、インターフェース画面を通じ、検査スタートを指示する(ステップ1302)。検査スタートが指示されると、カセット又はフープより指定されたスロットのウエハが試料交換室に自動的に搬送され、ホルダに搭載される。この後、試料交換室が真空引きされ、真空が所定のレベルに達したら試料室(チャンバ)内にホルダごとウエハが搬送される(ステップ1303)。試料室内にウエハがロードされると、基板検査装置は、試料ステージ108に負電位を印加し、検査レシピDで指定された条件で紫外光を照射する(ステップ1304)。紫外光は、連続的に照射することも、電子線画像を取得する直前に照射し、電子線を照射しない間は照射をオフすることもできる。本検査では、前記のとおり、波長365nmの紫外光を照射することとした。 After inputting these conditions, the user instructs the start of inspection through the interface screen (step 1302). When the inspection start is instructed, the wafer in the slot designated by the cassette or the hoop is automatically transferred to the sample exchange chamber and mounted on the holder. Thereafter, the sample exchange chamber is evacuated, and when the vacuum reaches a predetermined level, the wafer along with the holder is transferred into the sample chamber (chamber) (step 1303). When the wafer is loaded into the sample chamber, the substrate inspection apparatus applies a negative potential to the sample stage 108 and irradiates ultraviolet light under the conditions specified in the inspection recipe D (step 1304). The ultraviolet light can be irradiated continuously, or can be irradiated immediately before acquiring the electron beam image, and the irradiation can be turned off while the electron beam is not irradiated. In this inspection, as described above, ultraviolet light having a wavelength of 365 nm was irradiated.
 次に、基板検査装置は、照射する電子線の状態を校正する(ステップ1305)。電子線の量を、電流により調整し、次に位置や歪みの状態を補正する。この補正には、試料ステージ108に設置されている標準パターンなどを使用する。本実施例では、焦点条件をアンダーフォーカスになるように設定した。ビーム校正の終了後、基板検査装置は、位置情報を校正する(ステップ1306)。欠陥の有無を検査する段階では、通常、基板の表面にパターンが形成されていない。そこで、この段階では、被検査基板の中心位置と、結晶方向を特定するためのノッチやオリエンテーションフラットを用いて回転方向の位置を特定し、一つ一つの検査領域や欠陥の座標を特定する。 Next, the substrate inspection apparatus calibrates the state of the irradiated electron beam (step 1305). The amount of electron beam is adjusted by current, and then the position and distortion state are corrected. For this correction, a standard pattern installed on the sample stage 108 is used. In this embodiment, the focus condition is set to be under focus. After the beam calibration is completed, the substrate inspection apparatus calibrates the position information (step 1306). In the stage of inspecting for the presence of defects, a pattern is usually not formed on the surface of the substrate. Therefore, at this stage, the position in the rotation direction is specified by using the center position of the substrate to be inspected, the notch and the orientation flat for specifying the crystal direction, and the coordinates of each inspection region and defect are specified.
 次に、基板検査装置は、ミラー顕微鏡画像の明るさレベルが所定のレベルになるように検出系のゲインやオフセットを調整する(ステップ1307)。前述したように、明るさや信号レベルに対して判定用のしきい値を設定している。このため、常に同じ条件の画像を取得できるようにするための調整処理を設ける。 Next, the substrate inspection apparatus adjusts the gain and offset of the detection system so that the brightness level of the mirror microscope image becomes a predetermined level (step 1307). As described above, determination thresholds are set for brightness and signal level. For this reason, adjustment processing is provided so that images with the same conditions can always be acquired.
 これらの事前準備の完了後、基板検査装置は、検査を開始する(ステップ1308)。この処理ステップでは、画像取得とステージ移動が繰り返され、所定の領域の画像が取得される。本実施例では、ステージが静止した状態で画像を取得し、次の検査位置にステージが移動して静止すると、次の画像を取得する。すなわち、順次画像を取得する。もっとも、ステージを連続移動させながらリアルタイムで画像を撮像し、撮像されたミラー顕微鏡画像を保存する方法もある。 After completion of these advance preparations, the board inspection apparatus starts inspection (step 1308). In this processing step, image acquisition and stage movement are repeated, and an image of a predetermined area is acquired. In the present embodiment, an image is acquired in a state where the stage is stationary, and the next image is acquired when the stage is moved to the next inspection position and is stationary. That is, images are acquired sequentially. However, there is also a method of capturing an image in real time while continuously moving the stage and storing the captured mirror microscope image.
 また、この処理ステップでは、欠陥の有無を判定する処理も実行される。本実施例の場合、保存された画像に基づいて、画像処理部116等の計算機が欠陥の有無を順次判定し、結果を出力する。もっとも、専用の画像処理基板を用いて、リアルタイムで欠陥の有無を判定しても良い。 In this processing step, processing for determining the presence or absence of defects is also executed. In the case of the present embodiment, a computer such as the image processing unit 116 sequentially determines the presence or absence of a defect based on the stored image, and outputs the result. However, the presence / absence of a defect may be determined in real time using a dedicated image processing substrate.
 検査が終了すると、基板検査装置は、欠陥の種類や有無を確認するレビュー処理を実行する(ステップ1309)。このレビュー処理では、取得した画像の保存処理に加え、別の撮像条件による画像の再取得等も必要に応じて実施される。もっとも、レビュー処理自体は、その実施をスキップしても良い。また、この処理ステップでは、検査結果のマップ出力、画像転送、位置情報の数値データ出力等も実施される。最後に、ウエハを元のカセット又はフープのスロット位置にアンロードし(ステップ1310)、検査を終了する(ステップ1311)。 When the inspection is completed, the substrate inspection apparatus executes a review process for confirming the type and presence of defects (step 1309). In this review process, in addition to the process of storing the acquired image, re-acquisition of an image under another imaging condition is performed as necessary. However, the review process itself may be skipped. In this processing step, inspection result map output, image transfer, position information numerical data output, and the like are also performed. Finally, the wafer is unloaded to the original cassette or hoop slot position (step 1310), and the inspection is terminated (step 1311).
 図14に、検査結果の出力例を示す。図14では、被検査基板の形状や大きさを示すサイズの円1401に対し、検出された欠陥の位置情報が点1402としてプロットされている。点1402の1つ1つが別の欠陥を示している。この欠陥の座標を出力することにより、当該座標の欠陥について、別の検査装置を用いたさらに詳細な検査の実施が可能になる。 FIG. 14 shows an output example of the inspection result. In FIG. 14, positional information of detected defects is plotted as points 1402 with respect to a circle 1401 having a size indicating the shape and size of the substrate to be inspected. Each point 1402 represents another defect. By outputting the coordinates of the defect, it becomes possible to carry out a more detailed inspection using another inspection apparatus for the defect of the coordinate.
 検査で欠陥と判定された箇所をミラー電子顕微鏡で観察すると、異物や残渣は凸形状のため明るくなり、他は暗くなった。図15に、欠陥の画像例を示す。例えば欠陥1501は明るく検出されているが、他の欠陥1502、1503、1504、1505はいずれも暗くなっている。従って、ミラー電気顕微鏡画像内の明暗により、教示無しでも、異物及び残渣と、他の各種欠陥(積層欠陥、スクラッチ、ピット、転位等)を分離できるようになった。 When the part determined to be defective in the inspection was observed with a mirror electron microscope, the foreign matter and residue became bright due to the convex shape, and the others became dark. FIG. 15 shows an example of a defect image. For example, the defect 1501 is detected brightly, but the other defects 1502, 1503, 1504, and 1505 are all dark. Accordingly, the brightness and darkness in the mirror electric microscope image enables separation of foreign matters and residues and other various defects (stacking defects, scratches, pits, dislocations, etc.) without teaching.
 次に、暗く見える欠陥の特徴を観察した結果を図16に示す。図15で明るく見えた欠陥1501の模式図を図16(a)に示し、図中の破線方向の明るさプロファイルを図16(a)’に示す。この欠陥は、基板表面に対して凸形状であるため、基板の明るさに対して欠陥部が明るくなっている。 Next, FIG. 16 shows the result of observing the characteristics of defects that appear dark. A schematic diagram of the defect 1501 that appears bright in FIG. 15 is shown in FIG. 16A, and a brightness profile in the broken line direction in FIG. 15 is shown in FIG. Since this defect has a convex shape with respect to the substrate surface, the defect portion is brighter than the brightness of the substrate.
 図15で暗い点として見えた欠陥1502の模式図を図16(b)に示し、図中の破線方向の明るさプロファイルを図16(b)’に示す。暗い円形に見えているので、この欠陥は、基板表面に対して凹形状であることが推測される。さらに、図15で暗く検出された欠陥1503、1504、1505の模式図を、それぞれ図16(c)、(d)、(e)に示し、図中の破線方向におけるそれぞれの明るさプロファイルを図16(c)’、(d)’、(e)’に示す。 FIG. 16B shows a schematic diagram of the defect 1502 that appears as a dark spot in FIG. 15, and FIG. 16B ′ shows the brightness profile in the broken line direction in the figure. Since it looks like a dark circle, this defect is assumed to be concave with respect to the substrate surface. Furthermore, schematic diagrams of the defects 1503, 1504, and 1505 detected dark in FIG. 15 are shown in FIGS. 16C, 16D, and 16E, respectively, and the brightness profiles in the broken line direction in the figure are shown. 16 (c) ′, (d) ′, and (e) ′.
 これらの欠陥は、いずれも暗く見える点で共通しているが、例えば欠陥1503(図16(c)、(c)’)では、暗い円形の周りに楕円状の影のようなグラデーションが見えている。欠陥1504(図16(d)、(d)’)では、暗い楕円にライン状の影のようなグラデーションが見えている。このように、欠陥毎に見え方が異なるため、被検査基板をエッチング処理し、表面を電子顕微鏡で観察することにより、内部欠陥が発生しているかどうかを調べた。欠陥1501(図16(a))は、エッチングにより除去されてしまったため、表面に付着していた異物であったことが分かった。これに対し、欠陥1502(図16(b))は、基板内部に凹み形状が発生しているピットとよばれる欠陥であることが分かった。他方、欠陥1503(図16(c))は、ピットと同様のピンホールに加え、その周囲もエッチングで削れたことから、貫通らせん転位と呼ばれる結晶欠陥であると同定された。欠陥1504(図16(d))は、ごく微小な結晶欠陥が内部まで発生しており、基底面転位と呼ばれる結晶欠陥であることが分かった。さらに欠陥1505(図16(e))は、極微細で均等にエッチングされており、その特徴から貫通刃状転位であることが分かった。このように、欠陥の種類により、ミラー顕微鏡画像での欠陥の見え方に特徴があることが分かった。 These defects are common in that they both appear dark. For example, in the defect 1503 (FIGS. 16C and 16C ′), a gradation like an elliptical shadow appears around a dark circle. Yes. In the defect 1504 (FIGS. 16D and 16D), a gradation like a line-like shadow is seen on a dark ellipse. In this way, since the appearance differs for each defect, the substrate to be inspected was etched and the surface was observed with an electron microscope to examine whether or not an internal defect had occurred. It was found that the defect 1501 (FIG. 16A) was a foreign matter adhering to the surface because it was removed by etching. On the other hand, it was found that the defect 1502 (FIG. 16B) is a defect called a pit in which a concave shape is generated inside the substrate. On the other hand, the defect 1503 (FIG. 16C) was identified as a crystal defect called threading screw dislocation because the periphery of the defect was etched away in addition to the pinhole similar to the pit. The defect 1504 (FIG. 16D) has a very small crystal defect even inside, and was found to be a crystal defect called a basal plane dislocation. Further, the defect 1505 (FIG. 16E) is extremely fine and evenly etched, and from its characteristics, it was found that it is a threading edge dislocation. Thus, it has been found that there is a feature in the appearance of the defect in the mirror microscope image depending on the type of the defect.
 さらに、楕円やライン状に影があるように見えた欠陥1503、1504の向きとSiC基板の結晶の方向の関係について調べたところ、図17に示すように、図16(c)の貫通らせん転位は、図17の欠陥1703に相当し、結晶形成の下流方向に影が発生していることが分かった。また、図16(d)の基底面転位は、図17の欠陥1704に相当し、結晶形成の上流に影があることが分かった。 Further, when the relationship between the orientation of the defects 1503 and 1504 that appear to have shadows in the ellipse or line shape and the crystal direction of the SiC substrate was investigated, as shown in FIG. 17, the threading screw dislocation shown in FIG. Corresponds to the defect 1703 in FIG. 17, and it was found that a shadow occurred in the downstream direction of crystal formation. Further, it was found that the basal plane dislocation in FIG. 16D corresponds to the defect 1704 in FIG. 17 and has a shadow upstream of the crystal formation.
 本実施例の画像処理部116は、これらの結果に基づいて、図16のプロファイルの特徴を自動的に区別できる機能を搭載する。具体的には、以下に示す条件に基づいて、欠陥を分類する。画像処理部116は、まず、欠陥と判定された箇所(所定の明るさの差異が検出された領域の全体)について、画素毎に明るさ情報を抽出する。 The image processing unit 116 of the present embodiment is equipped with a function that can automatically distinguish the characteristics of the profile of FIG. 16 based on these results. Specifically, the defects are classified based on the following conditions. First, the image processing unit 116 extracts brightness information for each pixel for a portion determined to be a defect (entire region where a predetermined brightness difference is detected).
 次に、画像処理部116は、ノイズの影響を抑制するためにスムージング処理を実施する。周囲の平均明るさに対して欠陥部の明るさが明るい場合、画像処理部116は、当該欠陥部は、基板表面に対して凸形状の欠陥と仮判定する。欠陥部の明るさが、周囲の平均明るさAに対して暗い場合、画像処理部116は、以下の処理を順番に実行する。 Next, the image processing unit 116 performs a smoothing process to suppress the influence of noise. When the brightness of the defective portion is brighter than the surrounding average brightness, the image processing unit tentatively determines that the defective portion is a convex defect with respect to the substrate surface. When the brightness of the defective portion is dark with respect to the surrounding average brightness A, the image processing unit 116 sequentially executes the following processing.
(1)欠陥と思われる領域の最も明るい/暗い箇所P1を特定する。 (1) Identify the brightest / darkest part P1 of the area that appears to be defective.
(2)最も明るい/暗い箇所P1の明るさ差分D1を求める。 (2) Find the brightness difference D1 of the brightest / darkest part P1.
(3)欠陥領域の長手方向のサイズLA1を測定する。 (3) The size LA1 in the longitudinal direction of the defect area is measured.
(4)長手方向に対して直交する方向のサイズLA2を測定する。 (4) The size LA2 in the direction orthogonal to the longitudinal direction is measured.
(5)前記最も暗い箇所P1を起点に長手方向のサイズLA1をL11とL12に分割する。 (5) The size LA1 in the longitudinal direction is divided into L11 and L12 starting from the darkest portion P1.
(6)前記最も暗い箇所P1を起点に、長手方向に対して直交する方向のサイズLA2をL21とL22に分割する。 (6) The size LA2 in the direction orthogonal to the longitudinal direction is divided into L21 and L22 starting from the darkest portion P1.
 以上のステップを経た後に、欠陥領域中の最も明るい又は暗い点の明るさと、欠陥領域の形状(より具体的には、長手方向と長手方向に直交する方向の長さ比)とにより、欠陥を分類する。より具体的には、以下の手順で欠陥を分類する。なお、以下では、図16に示す5種類の欠陥の分類法について記載するが、それ以外についても同様の手法にて分類可能である。また、本実施例に記載以外の分類手順を追加しても良いし、以下に記載するしきい値は任意に変更されてもよい。 After the above steps, the defect is determined according to the brightness of the brightest or darkest point in the defect area and the shape of the defect area (more specifically, the length ratio in the direction perpendicular to the longitudinal direction). Classify. More specifically, defects are classified according to the following procedure. In the following description, the five types of defect classification methods shown in FIG. 16 will be described, but other methods can be classified by the same method. Further, a classification procedure other than that described in the present embodiment may be added, and the threshold values described below may be arbitrarily changed.
(1)平均明るさAに対する最も明るい/暗い箇所P1の明るさ差分D1がプラスの場合には明欠陥、マイナスの場合には暗欠陥に分類する。 (1) When the brightness difference D1 of the brightest / darkest part P1 with respect to the average brightness A is positive, it is classified as a bright defect, and when it is negative, it is classified as a dark defect.
(2)明欠陥は、表面に凸形状又は異物と分類しても良い。 (2) Bright defects may be classified as convex or foreign on the surface.
(3)暗欠陥であって、さらに、(平均明るさAに対する最も明るい/暗い箇所P1の明るさ差分D1)-(所定の明るさB)がプラス、かつ、LA2/LA1の比が70%以上の場合には、図16(e)の貫通刃状転位に分類する。 (3) It is a dark defect, and (the brightness difference D1 of the brightest / darkest part P1 with respect to the average brightness A)-(predetermined brightness B) is positive and the ratio of LA2 / LA1 is 70% In the above case, it is classified into the threading edge dislocation shown in FIG.
(4)暗欠陥であって、さらに、(最も明るい/暗い箇所P1の明るさ差分D1)-(所定の明るさB)がマイナス、かつ、所定の明るさCにおける欠陥の長手方向サイズLC1と所定の明るさBにおける欠陥の長手方向サイズLB1の差分のLA1に対する比率が50%以下であり、かつ、LA2/LA1の比が70%以上であり、かつ、L11/L12の比が70%以上130%以下の場合には、図16(b)に示す凹み、すなわちピット欠陥と分類する。 (4) It is a dark defect, and (the brightness difference D1 of the brightest / darkest part P1) − (predetermined brightness B) is negative and the longitudinal size LC1 of the defect at the predetermined brightness C is The ratio of the difference in the longitudinal size LB1 of the defect at a predetermined brightness B to LA1 is 50% or less, the ratio of LA2 / LA1 is 70% or more, and the ratio of L11 / L12 is 70% or more If it is 130% or less, it is classified as a dent shown in FIG.
(5)暗欠陥であって、さらに、上記の(3)又は(4)に分類されない欠陥であり、かつ、(平均明るさAに対する最も明るい/暗い箇所P1の明るさ差分D1)-(所定の明るさB)がマイナス、かつ、所定の明るさCにおける欠陥の長手方向サイズLC1と所定の明るさBにおける欠陥の長手方向サイズLB1の差分のLA1に対する比率が30%以上であり、かつ、L11/L12の比が3以上の場合には、図16(d)に示す基底面転位であると分類する。 (5) A dark defect that is not classified into the above (3) or (4), and (the brightness difference D1 of the brightest / darkest part P1 with respect to the average brightness A) − (predetermined And the ratio of the difference between the longitudinal size LC1 of the defect at the predetermined brightness C and the longitudinal size LB1 of the defect at the predetermined brightness B to LA1 is 30% or more, and When the ratio of L11 / L12 is 3 or more, it is classified as the basal plane dislocation shown in FIG.
(6)暗欠陥であって、さらに、上記の(3)又は(4)に分類されない欠陥であり、かつ、(平均明るさAに対する最も明るい/暗い箇所P1の明るさ差分D1)-(所定の明るさB)がマイナス、かつ、所定の明るさCにおける欠陥の長手方向サイズLC1と所定の明るさBにおける欠陥の長手方向サイズLB1の差分のLA1に対する比率が30%以上であり、かつ、L11/L12の比が3未満の場合には、図16(c)に示す貫通らせん転位と分類する。 (6) A dark defect that is not classified into the above (3) or (4), and (the brightness difference D1 of the brightest / darkest part P1 with respect to the average brightness A) − (predetermined And the ratio of the difference between the longitudinal size LC1 of the defect at the predetermined brightness C and the longitudinal size LB1 of the defect at the predetermined brightness B to LA1 is 30% or more, and When the ratio of L11 / L12 is less than 3, it is classified as a threading screw dislocation shown in FIG.
 このように、本実施例の画像処理部116は、欠陥部について取得されたミラー顕微鏡画像の中から最も暗い箇所を特定し、影のように見えるコントラストやグラデーション部分を含めた長さの特徴及び位置関係を数値化し、数値の関係や結晶成長の向きに対する長さの関係より欠陥種を自動的に分類する。 As described above, the image processing unit 116 according to the present exemplary embodiment identifies the darkest part from the mirror microscope image acquired for the defective part, and the length characteristic including the contrast and gradation part that looks like a shadow and the like. The positional relationship is digitized, and defect types are automatically classified based on the numerical relationship and the length relationship with respect to the crystal growth direction.
(2-1-6)実施例により得られる効果
 本実施例に係る基板検査装置では、積層基板であるSiC基板において、その表面の欠陥に加え、その内部の欠陥についても高感度に検出することができる。特に、ピットや転位の検出感度を向上でき、従来の検査手法では検出できない欠陥を顕在化できる。このように、本実施例に係る基板検査装置では、各種の欠陥を自動検査できるため、結晶製造プロセスを改善することに寄与することができる。
(2-1-6) Effects Obtained by the Example In the substrate inspection apparatus according to this example, in the SiC substrate which is a laminated substrate, in addition to defects on the surface, defects inside the substrate can be detected with high sensitivity. Can do. In particular, the detection sensitivity of pits and dislocations can be improved, and defects that cannot be detected by conventional inspection methods can be made obvious. Thus, since the substrate inspection apparatus according to the present embodiment can automatically inspect various defects, it can contribute to improving the crystal manufacturing process.
 また、本実施例に係る基板検査装置は、基板を研磨した際に生じる潜傷を高感度に検知できるため、研磨プロセスを改善するためのフィードバックを製造工程の早い段階で実施することができる。同様に、積層膜形成や温度プロセスの最適化のためのフィードバックも製造工程の早い段階で実施することができる。また、不良が多い基板に対し、良品率を予測することができるようになる。また、不良が多い基板を除くことで、無駄な製造コストを削減することができるようになる。また、この技術によって品質が管理された基板を適用することで、半導体装置の品質を一段と向上することができ、半導体装置等の信頼性を高めると共に不良率を低減するまでに要する期間を短縮することができる。 In addition, since the substrate inspection apparatus according to the present embodiment can detect latent scratches generated when the substrate is polished with high sensitivity, feedback for improving the polishing process can be performed at an early stage of the manufacturing process. Similarly, feedback for forming the laminated film and optimizing the temperature process can be performed at an early stage of the manufacturing process. In addition, it is possible to predict a non-defective product rate for a substrate with many defects. In addition, by removing a substrate with many defects, it is possible to reduce useless manufacturing costs. In addition, by applying a substrate whose quality is controlled by this technology, the quality of the semiconductor device can be further improved, and the time required to reduce the defect rate while improving the reliability of the semiconductor device and the like can be shortened. be able to.
(2-2)実施例2
 前述の実施例1においては、紫外光の波長365nmに設定し、基板の欠陥を検査した。これに対し、本実施例では、まず、第1の波長365nmの紫外光を照射して基板の欠陥を検査し、続いて同一の被検査基板に対して第2の波長230nmの紫外光を照射して基板の欠陥を検査する。その結果、図18(a)及び(b)に示すように、2つの検査結果が得られた。
(2-2) Example 2
In Example 1 described above, the wavelength of ultraviolet light was set to 365 nm, and the substrate was inspected for defects. In contrast, in this embodiment, first, ultraviolet light having a first wavelength of 365 nm is irradiated to inspect the substrate for defects, and then the same substrate to be inspected is irradiated with ultraviolet light having the second wavelength of 230 nm. And inspecting the substrate for defects. As a result, as shown in FIGS. 18A and 18B, two inspection results were obtained.
 図18(a)の例では50個の欠陥が検出され、図18(b)の例では30個の欠陥が検出された。検出個数の違いは、第2の波長230nmの紫外光を用いた検査(図18(b))では被検査試料の表面近傍の欠陥のみが検出されるのに対し、第1の波長365nmの紫外光を用いた検査(図18(a))では被検査基板の内部方向に10μm程度までの欠陥が検出されるからである。すなわち、図18(a)は表面の欠陥と内部欠陥の両方が検出され、図18(b)は表面の欠陥だけが検出されているからである。 In the example of FIG. 18 (a), 50 defects were detected, and in the example of FIG. 18 (b), 30 defects were detected. The difference in the number of detections is that in the inspection using the ultraviolet light with the second wavelength of 230 nm (FIG. 18B), only defects near the surface of the sample to be inspected are detected, whereas the ultraviolet light with the first wavelength of 365 nm is detected. This is because in the inspection using light (FIG. 18A), defects up to about 10 μm are detected in the internal direction of the inspected substrate. That is, FIG. 18A shows that both surface defects and internal defects are detected, and FIG. 18B shows that only surface defects are detected.
 そこで、本実施例に係る画像処理部116には、2つの検査結果を比較する処理と、2つの検査結果の同一座標に出現する欠陥を表面付近の欠陥と判定する処理と、図18(a)だけで検出された欠陥を内部欠陥と判定する処理とを実行する機能を搭載する。この機能の搭載により、欠陥種別の弁別が実施例1に比べて容易になる。また、本検査以外の波長を複数組み合わせて同様に検査を行い、検査結果の一致しない箇所を抽出すれば、欠陥の深さも同定することができる。 Therefore, the image processing unit 116 according to the present embodiment includes a process for comparing two inspection results, a process for determining a defect appearing at the same coordinates of the two inspection results as a defect near the surface, and FIG. ), A function for executing a process of determining a defect detected as an internal defect. By mounting this function, it becomes easier to discriminate defect types compared to the first embodiment. In addition, if the inspection is performed in a similar manner by combining a plurality of wavelengths other than the main inspection and a portion where the inspection results do not match is extracted, the depth of the defect can be identified.
 本実施例に係る基板検査装置の使用により、積層基板であるSiC基板において、表面だけでなく内部の欠陥についても高感度に検出することができる。また、欠陥の深さ方向を、断面解析することなく同定することができるため、欠陥種の特定が容易になる。これにより、不良要因の特定や基板製造プロセスの改善を早い段階で実施することが可能になり、不良低減に寄与できるようになる。 By using the substrate inspection apparatus according to the present embodiment, not only the surface but also internal defects can be detected with high sensitivity in the SiC substrate which is a laminated substrate. Further, since the depth direction of the defect can be identified without performing a cross-sectional analysis, the defect type can be easily specified. As a result, it becomes possible to identify the cause of failure and improve the substrate manufacturing process at an early stage, thereby contributing to the reduction of defects.
(2-3)実施例3
 前述の実施例1及び2においては、専ら一次的に基板の欠陥を検出する基板検査装置について説明したが、前述の技術は、他の基板検査装置から欠陥座標を取得して欠陥を検査するレビュー装置にも適用することができる。このため、レビュー装置としての基板検査装置には、図1に示す装置構成に加え、予め画像を撮像する箇所の位置情報を外部装置との間で授受する通信ユニットを搭載する。なお、本実施例の場合、ステージ制御部115は、レビュー位置に指定された座標位置に被検査試料104を位置決めし、前述の検査動作を開始すれば良い。本実施例によれば、レビュー装置による欠陥の検査精度を高めることができる。
(2-3) Example 3
In the above-described first and second embodiments, the substrate inspection apparatus that primarily detects the defect of the substrate has been described. However, the above-described technique is a review in which the defect coordinates are acquired from another substrate inspection apparatus to inspect the defect. It can also be applied to devices. For this reason, in addition to the apparatus configuration shown in FIG. 1, the board inspection apparatus as a review apparatus is equipped with a communication unit that exchanges position information of an image capturing location with an external apparatus in advance. In the case of the present embodiment, the stage controller 115 may position the sample 104 to be inspected at the coordinate position designated as the review position and start the above-described inspection operation. According to the present embodiment, it is possible to increase the inspection accuracy of defects by the review device.
(2-4)他の実施例
 本発明は、上述した実施例の構成に限定されるものでなく、様々な変形例を含んでいる。例えば前述の実施例では、本発明を分かりやすく説明するために、一部の実施例について詳細に説明したものであり、必ずしも説明した全ての構成を備える必要は無い。また、前述の実施例の構成に他の構成を追加し、他の構成で置換し、又は、実施例の一部構成を削除することも可能である。
(2-4) Other Embodiments The present invention is not limited to the configuration of the embodiment described above, and includes various modifications. For example, in the above-described embodiments, some of the embodiments have been described in detail in order to explain the present invention in an easy-to-understand manner. Further, it is possible to add another configuration to the configuration of the above-described embodiment, replace it with another configuration, or delete a part of the configuration of the embodiment.
 また、前述の実施例における処理部や制御部が実行する機能の一部又は全部を、例えば集積回路その他のハードウェアとして実現しても良い。なお、処理部や制御部の機能の実現に使用するプログラム、テーブル、ファイル等の情報は、メモリ、ハードディスク、SSD(Solid State Drive)等の記憶装置、ICカード、SDカード、DVD等の記憶媒体に格納することができる。 Further, part or all of the functions executed by the processing unit and the control unit in the above-described embodiments may be realized as an integrated circuit or other hardware, for example. Information such as programs, tables, and files used to implement the functions of the processing unit and control unit is a memory, hard disk, storage device such as SSD (Solid State Drive), storage medium such as IC card, SD card, and DVD. Can be stored.
 また、制御線や情報線は、説明上必要と考えられるものを示すものであり、製品上必要な全ての制御線や情報線を表すものでない。実際にはほとんど全ての構成が相互に接続されていると考えて良い。 Also, the control lines and information lines indicate what is considered necessary for explanation, and do not represent all control lines and information lines necessary for the product. In practice, it can be considered that almost all components are connected to each other.
 100a…照射電子線、100b…クロスオーバー、100c…反射した電子線、100d…クロスオーバー、101…電子銃、102…コンデンサレンズ、103…ExB偏向器、104…被検査試料、105…電子銃制御装置、106…ExB偏向器、107…対物レンズ、108…試料ステージ、109…試料電圧制御部、110…中間レンズ、111…投影レンズ、112…画像検出部、112a…蛍光板、112b…光学画像検出装置、112c…光学像伝送系、113…電子光学系制御装置、115…ステージ制御装置、116…画像処理部、116a…画像信号記憶部、116b…欠陥判定部、117…検査装置制御部、118…モニタ、121…紫外線光源、122…バンドパスフィルタ、123…アパーチャ、124…反射ミラー、125…紫外光制御部、126…紫外光照射系。 DESCRIPTION OF SYMBOLS 100a ... Irradiated electron beam, 100b ... Crossover, 100c ... Reflected electron beam, 100d ... Crossover, 101 ... Electron gun, 102 ... Condenser lens, 103 ... ExB deflector, 104 ... Sample to be inspected, 105 ... Electron gun control Device: 106 ... ExB deflector, 107 ... Objective lens, 108 ... Sample stage, 109 ... Sample voltage control unit, 110 ... Intermediate lens, 111 ... Projection lens, 112 ... Image detection unit, 112a ... Fluorescent plate, 112b ... Optical image detection Device: 112c: Optical image transmission system, 113: Electro-optical system control device, 115: Stage control device, 116: Image processing unit, 116a: Image signal storage unit, 116b: Defect determination unit, 117: Inspection device control unit, 118 ... Monitor, 121 ... UV light source, 122 ... Band pass filter, 123 ... Aperture, 124 ... Morphism mirror, 125 ... UV-light control unit, 126 ... UV light irradiation system.

Claims (15)

  1.  電子源から放出された電子線を、試料上の視野を含む範囲に照射する電子光学系と、
     前記電子線が前記試料に到達する前に反射される強度の電圧を前記試料に印加する電圧印加ユニットと、
     前記電圧の印加により反射されたミラー電子を結像してミラー電子像を取得するミラー電子結像光学系と、
     取得された前記ミラー電子像を用いて前記試料の欠陥を検出する演算ユニットと、
     182nm以上380nm以下の範囲で任意に設定された波長の紫外光を、前記電子線の照射中に、前記電子線の照射範囲を含む範囲に照射する紫外光照射ユニットと、
     を有する基板検査装置。
    An electron optical system that irradiates an electron beam emitted from an electron source to a range including a visual field on the sample; and
    A voltage application unit for applying a voltage having an intensity reflected before the electron beam reaches the sample to the sample;
    A mirror electron imaging optical system for imaging a mirror electron reflected by the application of the voltage to obtain a mirror electron image;
    An arithmetic unit for detecting defects in the sample using the acquired mirror electron image;
    An ultraviolet light irradiation unit that irradiates ultraviolet light of a wavelength arbitrarily set in a range of 182 nm or more and 380 nm or less during irradiation of the electron beam to a range including the irradiation range of the electron beam,
    A substrate inspection apparatus.
  2.  請求項1に記載の基板検査装置において、
     前記電子光学系は、前記電子線を面状に形成して前記試料に照射する
     ことを特徴とする基板検査装置。
    The board inspection apparatus according to claim 1,
    The electron optical system forms the electron beam in a planar shape and irradiates the sample.
  3.  請求項1に記載の基板検査装置において、
     前記紫外光照射ユニットは、前記紫外光の波長を200nm以上300nm以下の範囲で任意に設定できる
     ことを特徴とする基板検査装置。
    The board inspection apparatus according to claim 1,
    The substrate inspection apparatus, wherein the ultraviolet light irradiation unit can arbitrarily set the wavelength of the ultraviolet light within a range of 200 nm to 300 nm.
  4.  請求項1に記載の基板検査装置において、
     前記電子光学系は、前記電子光学系の焦点面を、前記試料に対してアンダーフォーカス条件又はオーバーフォーカス条件に設定して前記ミラー電子像を取得する
     ことを特徴とする基板検査装置。
    The board inspection apparatus according to claim 1,
    The substrate inspection apparatus, wherein the electron optical system acquires the mirror electron image by setting a focal plane of the electron optical system to an underfocus condition or an overfocus condition for the sample.
  5.  請求項4に記載の基板検査装置において、
     前記演算ユニットは、前記アンダーフォーカス条件で取得した前記ミラー顕微鏡画像における欠陥部のコントラスト、又は、前記オーバーフォーカス条件で取得した前記ミラー電子画像とにおける欠陥部のコントラストにより、表面の異物と内部欠陥を区別する
     ことを特徴とする基板検査装置。
    The board inspection apparatus according to claim 4,
    The arithmetic unit removes foreign matter and internal defects on the surface based on the contrast of the defect portion in the mirror microscope image acquired under the underfocus condition or the contrast of the defect portion in the mirror electronic image acquired under the overfocus condition. A board inspection device characterized by distinguishing.
  6.  請求項1に記載の基板検査装置において、
     前記試料は、SiC基板、GaN基板、又はGa2O3基板のいずれかである
     ことを特徴とする基板検査装置。
    The board inspection apparatus according to claim 1,
    The substrate inspection apparatus, wherein the sample is any one of a SiC substrate, a GaN substrate, and a Ga2O3 substrate.
  7.  請求項1に記載の基板検査装置において、
     前記試料は、Epi成長膜が形成された基板である
     ことを特徴とする基板検査装置。
    The board inspection apparatus according to claim 1,
    The substrate inspection apparatus, wherein the sample is a substrate on which an Epi growth film is formed.
  8.  請求項1に記載の基板検査装置において、
     前記演算ユニットは、貫通らせん転位、貫通刃状転位、基底面転位を区別する欠陥分類部を有する
     ことを特徴とする基板検査装置。
    The board inspection apparatus according to claim 1,
    The arithmetic unit has a defect classification unit that distinguishes between threading screw dislocations, threading edge dislocations, and basal plane dislocations.
  9.  請求項1に記載の基板検査装置において、
     前記演算ユニットは、第1の波長の紫外光を照射しながら取得した第1のミラー電子像により検出された欠陥と、前記第1の波長とは異なる第2の波長の紫外光を照射しながら取得した第2のミラー電子像により検出された欠陥とを比較して、欠陥の種類を判別する
     ことを特徴とする基板検査装置。
    The board inspection apparatus according to claim 1,
    The arithmetic unit irradiates the defect detected by the first mirror electron image acquired while irradiating the ultraviolet light of the first wavelength and the ultraviolet light of the second wavelength different from the first wavelength. A substrate inspection apparatus characterized by comparing the defect detected by the acquired second mirror electron image and determining the type of the defect.
  10.  請求項1に記載の基板検査装置において、
     前記演算ユニットは、前記ミラー電子像における前記欠陥の領域のうち明暗の情報に基づいて決定した最も暗い箇所から、前記欠陥の領域の最外周までの距離を少なくとも2つの方向に測定し、当該測定された各方向の距離の比に基づいて前記欠陥の種類を判別する
     ことを特徴とする基板検査装置。
    The board inspection apparatus according to claim 1,
    The arithmetic unit measures in at least two directions the distance from the darkest part determined based on light and dark information among the defect areas in the mirror electron image to the outermost periphery of the defect area, and performs the measurement. The type of the defect is determined based on the ratio of the distances in each direction.
  11.  請求項1に記載の基板検査装置において、
     前記電圧印加ユニットは、前記試料に印加する電圧を複数の条件に変更可能であり、
     前記演算ユニットは、前記複数の条件の各々における前記ミラー電子像の明るさを監視して、前記ミラー電子像の明るさが変化し始めるときの電圧を前記試料に印加される電圧として決定する
     ことを特徴とする基板検査装置。
    The board inspection apparatus according to claim 1,
    The voltage application unit can change the voltage applied to the sample into a plurality of conditions,
    The arithmetic unit monitors the brightness of the mirror electronic image under each of the plurality of conditions, and determines a voltage when the brightness of the mirror electronic image starts to change as a voltage applied to the sample. A board inspection apparatus characterized by the above.
  12.  請求項1の基板検査装置において、
     予め画像を撮像する箇所の位置情報を外部装置との間で授受するユニットと、
     前記位置情報に基づいて特定される座標位置が前記視野に含まれるように、前記試料を位置決めするステージ制御部と
     を更に有することを特徴とする基板検査装置。
    The substrate inspection apparatus according to claim 1,
    A unit for exchanging position information of an image capturing location with an external device;
    A substrate inspection apparatus further comprising: a stage control unit for positioning the sample so that a coordinate position specified based on the position information is included in the visual field.
  13.  電子源から放出された電子線を、試料上の視野を含む範囲に照射するステップと、
     前記電子線が前記試料に到達する前に反射される強度の電圧を前記試料に印加するステップと、
     182nm以上380nm以下の範囲で任意に設定された波長の紫外光を、前記電子線の照射中に、前記電子線の照射範囲を含む範囲に照射するステップと、
     前記電圧の印加により反射されたミラー電子を結像してミラー電子像を取得するステップと、
     取得された前記ミラー電子像を用いて前記試料の欠陥を検出するステップと、
     を有する基板検査方法。
    Irradiating an electron beam emitted from an electron source to a range including a visual field on the sample;
    Applying to the sample a voltage of an intensity that is reflected before the electron beam reaches the sample;
    Irradiating ultraviolet light of a wavelength arbitrarily set in the range of 182 nm or more and 380 nm or less during irradiation of the electron beam to a range including the irradiation range of the electron beam;
    Imaging a mirror electron reflected by the application of the voltage to obtain a mirror electron image;
    Detecting a defect of the sample using the acquired mirror electron image;
    A substrate inspection method.
  14.  請求項13に記載の基板検査方法において、
     前記試料の欠陥を検出するステップは、第1の波長の紫外光を照射しながら取得した第1のミラー電子像により検出された欠陥と、前記第1の波長とは異なる第2の波長の紫外光を照射しながら取得した第2のミラー電子像により検出された欠陥とを比較して、欠陥の種類を判別する
     ことを特徴とする基板検査方法。
    The substrate inspection method according to claim 13,
    The step of detecting the defect of the sample includes the defect detected by the first mirror electron image acquired while irradiating the ultraviolet light having the first wavelength, and the ultraviolet having the second wavelength different from the first wavelength. A substrate inspection method comprising: comparing a defect detected by a second mirror electron image acquired while irradiating light to determine the type of defect.
  15.  請求項13に記載の基板検査方法において、
     前記試料の欠陥を検出するステップは、前記ミラー電子像における前記欠陥の領域のうち明暗の情報に基づいて決定した最も暗い箇所から、前記欠陥の領域の最外周までの距離を少なくとも2つの方向に測定し、当該測定された各方向の距離の比に基づいて前記欠陥の種類を判別する
     ことを特徴とする基板検査方法。
    The substrate inspection method according to claim 13,
    The step of detecting a defect of the sample includes, in at least two directions, the distance from the darkest portion determined based on light and dark information among the defect regions in the mirror electron image to the outermost periphery of the defect region. A substrate inspection method comprising: measuring and determining the type of the defect based on the measured distance ratio in each direction.
PCT/JP2014/067529 2014-07-01 2014-07-01 Substrate inspection apparatus and substrate inspection method WO2016002003A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/JP2014/067529 WO2016002003A1 (en) 2014-07-01 2014-07-01 Substrate inspection apparatus and substrate inspection method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2014/067529 WO2016002003A1 (en) 2014-07-01 2014-07-01 Substrate inspection apparatus and substrate inspection method

Publications (1)

Publication Number Publication Date
WO2016002003A1 true WO2016002003A1 (en) 2016-01-07

Family

ID=55018603

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2014/067529 WO2016002003A1 (en) 2014-07-01 2014-07-01 Substrate inspection apparatus and substrate inspection method

Country Status (1)

Country Link
WO (1) WO2016002003A1 (en)

Cited By (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2017158742A1 (en) * 2016-03-16 2017-09-21 株式会社 日立ハイテクノロジーズ Defect inspection device
WO2017158744A1 (en) * 2016-03-16 2017-09-21 株式会社 日立ハイテクノロジーズ Defect inspection method and defect inspection device
CN108604522A (en) * 2016-03-28 2018-09-28 株式会社日立高新技术 The method of adjustment of charged particle beam apparatus and charged particle beam apparatus
WO2019058441A1 (en) * 2017-09-20 2019-03-28 株式会社日立ハイテクノロジーズ Charged particle beam device
WO2019058440A1 (en) * 2017-09-20 2019-03-28 株式会社日立ハイテクノロジーズ Charged particle beam device
DE112017008019T5 (en) 2017-10-30 2020-07-16 Hitachi High-Tech Corporation SEMICONDUCTOR SUBSTRATE FOR EVALUATION AND METHOD FOR USING THIS TO DETECT THE ERROR DETECTION SENSITIVITY OF A TEST DEVICE
TWI701448B (en) * 2016-06-14 2020-08-11 台灣積體電路製造股份有限公司 Metho and apparatus for estimating electrical property of semiconductor device
WO2020166049A1 (en) * 2019-02-15 2020-08-20 株式会社日立ハイテク Defect inspection device and defect inspection method
TWI716901B (en) * 2018-06-28 2021-01-21 日商日立全球先端科技股份有限公司 Semiconductor inspection equipment
WO2021106128A1 (en) * 2019-11-28 2021-06-03 株式会社日立ハイテク Defect inspection device and method
CN112899788A (en) * 2021-01-14 2021-06-04 山东天岳先进科技股份有限公司 Preliminary screening method and device for silicon carbide crystal ingot
CN113284150A (en) * 2021-07-26 2021-08-20 常州微亿智造科技有限公司 Industrial quality inspection method and industrial quality inspection device based on unpaired industrial data
DE102019217080B4 (en) 2018-11-08 2022-09-29 Hitachi High-Tech Corporation electron beam device
US11460497B2 (en) 2017-06-29 2022-10-04 Hamamatsu Photonics K.K. Device analysis apparatus and device analysis method
WO2022219667A1 (en) * 2021-04-12 2022-10-20 株式会社日立ハイテク Defect inspection device
WO2023096908A1 (en) * 2021-11-23 2023-06-01 Trustees Of Tufts College Detection and identification of defects using artificial intelligence analysis of multi-dimensional information data

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2007207688A (en) * 2006-02-06 2007-08-16 Hitachi High-Technologies Corp Mirror electron microscope, and inspection device using mirror electron microscope
JP2008515239A (en) * 2004-09-30 2008-05-08 ケイエルエイ−テンコー・テクノロジーズ・コーポレーション Flexible and hybrid defect classification for semiconductor manufacturing

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2008515239A (en) * 2004-09-30 2008-05-08 ケイエルエイ−テンコー・テクノロジーズ・コーポレーション Flexible and hybrid defect classification for semiconductor manufacturing
JP2007207688A (en) * 2006-02-06 2007-08-16 Hitachi High-Technologies Corp Mirror electron microscope, and inspection device using mirror electron microscope

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
HASEGAWA,M. ET AL.: "Non-destructive observation of in-grown stacking faults in 4H- SiC epitaxial layer using mirror electron microscope", JOURNAL OF APPLIED PHYSICS, vol. 110, no. Issue 7, pages II - IV, XP012154382 *
HASEGAWA,M. ET AL.: "Observation of electron trapping along scratches on SiO2 surface in mirror electron microscope images under ultraviolet light irradiation", JOURNAL OF APPLIED PHYSICS, vol. 107, no. Issue 8, 23 April 2010 (2010-04-23), pages 6, XP012133966 *
HIROYUKI SHINADA ET AL.: "Mirror Electron Microscope Technology Having Possibilities of High-speed and Highly Sensitive Inspection", HITACHI HYORON, vol. 94, 1 February 2012 (2012-02-01), pages 46 - 51, XP055248704 *

Cited By (34)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108603851B (en) * 2016-03-16 2021-01-01 株式会社日立高新技术 Defect inspection device
WO2017158742A1 (en) * 2016-03-16 2017-09-21 株式会社 日立ハイテクノロジーズ Defect inspection device
US11002687B2 (en) 2016-03-16 2021-05-11 Hitachi High-Tech Corporation Defect inspection method and defect inspection device
WO2017158744A1 (en) * 2016-03-16 2017-09-21 株式会社 日立ハイテクノロジーズ Defect inspection method and defect inspection device
CN108603850B (en) * 2016-03-16 2020-10-13 株式会社日立高新技术 Defect inspection method and defect inspection apparatus
JPWO2017158744A1 (en) * 2016-03-16 2019-02-14 株式会社日立ハイテクノロジーズ Defect inspection method and defect inspection apparatus
JPWO2017158742A1 (en) * 2016-03-16 2019-02-28 株式会社日立ハイテクノロジーズ Defect inspection equipment
CN108603851A (en) * 2016-03-16 2018-09-28 株式会社日立高新技术 Flaw detection apparatus
CN108603850A (en) * 2016-03-16 2018-09-28 株式会社日立高新技术 Defect detecting method and flaw detection apparatus
CN108604522B (en) * 2016-03-28 2020-07-14 株式会社日立高新技术 Charged particle beam device and method for adjusting charged particle beam device
CN108604522A (en) * 2016-03-28 2018-09-28 株式会社日立高新技术 The method of adjustment of charged particle beam apparatus and charged particle beam apparatus
TWI701448B (en) * 2016-06-14 2020-08-11 台灣積體電路製造股份有限公司 Metho and apparatus for estimating electrical property of semiconductor device
US11480606B2 (en) 2016-06-14 2022-10-25 Taiwan Semiconductor Manufacturing Co., Ltd. In-line device electrical property estimating method and test structure of the same
TWI783997B (en) * 2017-06-29 2022-11-21 日商濱松赫德尼古斯股份有限公司 Component analysis device and component analysis method
US11460497B2 (en) 2017-06-29 2022-10-04 Hamamatsu Photonics K.K. Device analysis apparatus and device analysis method
WO2019058440A1 (en) * 2017-09-20 2019-03-28 株式会社日立ハイテクノロジーズ Charged particle beam device
WO2019058441A1 (en) * 2017-09-20 2019-03-28 株式会社日立ハイテクノロジーズ Charged particle beam device
JPWO2019058440A1 (en) * 2017-09-20 2020-10-22 株式会社日立ハイテク Charged particle beam device
US11107655B2 (en) 2017-09-20 2021-08-31 Hitachi High-Technologies Corporation Charged particle beam device
DE112017008019T5 (en) 2017-10-30 2020-07-16 Hitachi High-Tech Corporation SEMICONDUCTOR SUBSTRATE FOR EVALUATION AND METHOD FOR USING THIS TO DETECT THE ERROR DETECTION SENSITIVITY OF A TEST DEVICE
US11193895B2 (en) 2017-10-30 2021-12-07 Hitachi High-Tech Corporation Semiconductor substrate for evaluation and method using same to evaluate defect detection sensitivity of inspection device
TWI716901B (en) * 2018-06-28 2021-01-21 日商日立全球先端科技股份有限公司 Semiconductor inspection equipment
US11719746B2 (en) 2018-06-28 2023-08-08 Hitachi High-Tech Corporation Semiconductor inspection device
DE102019217080B4 (en) 2018-11-08 2022-09-29 Hitachi High-Tech Corporation electron beam device
US11515121B2 (en) 2018-11-08 2022-11-29 Hitachi High-Tech Corporation Electron beam device
DE112019006527T5 (en) 2019-02-15 2021-09-16 Hitachi High-Tech Corporation Defect inspection device and defect inspection method
JPWO2020166049A1 (en) * 2019-02-15 2021-10-14 株式会社日立ハイテク Defect inspection equipment and defect inspection method
WO2020166049A1 (en) * 2019-02-15 2020-08-20 株式会社日立ハイテク Defect inspection device and defect inspection method
WO2021106128A1 (en) * 2019-11-28 2021-06-03 株式会社日立ハイテク Defect inspection device and method
CN112899788A (en) * 2021-01-14 2021-06-04 山东天岳先进科技股份有限公司 Preliminary screening method and device for silicon carbide crystal ingot
WO2022219667A1 (en) * 2021-04-12 2022-10-20 株式会社日立ハイテク Defect inspection device
CN113284150B (en) * 2021-07-26 2021-10-19 常州微亿智造科技有限公司 Industrial quality inspection method and industrial quality inspection device based on unpaired industrial data
CN113284150A (en) * 2021-07-26 2021-08-20 常州微亿智造科技有限公司 Industrial quality inspection method and industrial quality inspection device based on unpaired industrial data
WO2023096908A1 (en) * 2021-11-23 2023-06-01 Trustees Of Tufts College Detection and identification of defects using artificial intelligence analysis of multi-dimensional information data

Similar Documents

Publication Publication Date Title
WO2016002003A1 (en) Substrate inspection apparatus and substrate inspection method
JP4248382B2 (en) Inspection method and inspection apparatus using charged particle beam
JP3973372B2 (en) Substrate inspection apparatus and substrate inspection method using charged particle beam
US6583634B1 (en) Method of inspecting circuit pattern and inspecting instrument
JP4528014B2 (en) Sample inspection method
US7351968B1 (en) Multi-pixel electron emission die-to-die inspection
JP5164317B2 (en) Inspection / measurement method and inspection / measurement device using electron beam
JP2007257969A (en) Charged particle beam device
KR101685274B1 (en) Charged particle beam device
WO2017158742A1 (en) Defect inspection device
JP2006332296A (en) Focus correction method in electronic beam applied circuit pattern inspection
WO2010055610A1 (en) Pattern check device and pattern check method
US9177757B2 (en) Charged particle beam apparatus
US20100225905A1 (en) Inspection method and inspection apparatus for semiconductor substrate
US20110293167A1 (en) Defect inspecting method, defect inspecting apparatus, and recording medium
JP4728361B2 (en) Substrate inspection apparatus and substrate inspection method using charged particle beam
JP2002353279A (en) Circuit-pattern inspection method and its apparatus
JP6033325B2 (en) Semiconductor inspection apparatus and inspection method using charged particle beam
US8884223B2 (en) Methods and apparatus for measurement of relative critical dimensions
JP2012058206A (en) Method and device for inspecting defect in mask
JP4147233B2 (en) Electron beam equipment
JP5470194B2 (en) Charged particle beam equipment
JP2004157135A (en) Method of and apparatus for inspecting circuit pattern
JP6230831B2 (en) Charged particle beam apparatus and image acquisition method
JP2004347483A (en) Pattern inspection device using electron beam and pattern inspection method using electron beam

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 14896301

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 14896301

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: JP