Method for manufacturing nitride semiconductor substrate, nitride semiconductor substrate, and laminated structure

Abstract

The present application relates to a manufacturing method of a nitride semiconductor substrate, a nitride semiconductor substrate, and a laminated structure. The manufacturing method has a process of preparing a base substrate; a first process of directly epitaxially growing a single crystal of a Group III nitride semiconductor having a top surface with a (0001) surface exposed on a main surface of the base substrate, causing the top surface to have a plurality of recesses composed of inclined interfaces other than the (0001) surface, causing the inclined interfaces to gradually expand as they go upward from the main surface of the base substrate, and causing the (0001) surface to disappear from the top surface, thereby growing a first layer having a surface composed only of the inclined interfaces; and a second process of epitaxially growing a single crystal of the Group III nitride semiconductor on the first layer, causing the inclined interfaces to disappear, and growing a second layer having a mirror-finished surface.

Description

[0001] This application is a divisional application of an application with the application date of August 22, 2019, the application number of 201980056303.6, and the invention name of "Method for manufacturing nitride semiconductor substrate, nitride semiconductor substrate, and laminated structure". TECHNICAL FIELD

[0002] The present application relates to a method for manufacturing a nitride semiconductor substrate, a nitride semiconductor substrate, and a laminated structure. BACKGROUND

[0003] A method is known in which a substrate formed of a single crystal of a Group III nitride semiconductor is used as a base substrate (seed substrate), and a crystal layer formed of a single crystal of a Group III nitride semiconductor is further grown on a main surface of which the nearest low-index crystal plane is a (0001) plane in the base substrate. According to this method, by slicing the crystal layer grown in a prescribed thickness, at least one nitride semiconductor substrate can be obtained (for example, Patent Document 1).

[0004] PRIOR ART DOCUMENTS

[0005] PATENT DOCUMENTS

[0006] Patent Document 1: Japanese Patent Application Publication No. 2013-60349 SUMMARY

[0007] Problem to be solved by the invention

[0008] An object of the present application is to improve the crystal quality of a nitride semiconductor substrate.

[0009] Solution to solve the problem

[0010] According to one embodiment of the present application, there is provided a method for manufacturing a nitride semiconductor substrate,

[0011] which is a method for manufacturing a nitride semiconductor substrate using a vapor phase growth method, and has the following steps:

[0012] a step of preparing a base substrate formed of a single crystal of a Group III nitride semiconductor, having a mirror-finished main surface, and having a (0001) plane as the nearest low-index crystal plane with respect to the main surface;

[0013] A first step of epitaxially growing a single crystal of a Group III nitride semiconductor having a top surface with a (0001) surface directly on the main surface of the aforementioned base substrate, causing the aforementioned top surface to have a plurality of recesses composed of inclined interfaces other than the aforementioned (0001) surface, causing the aforementioned inclined interfaces to gradually expand as they go upward from the aforementioned main surface of the aforementioned base substrate, and causing the aforementioned (0001) surface to disappear from the aforementioned top surface, thereby growing a first layer whose surface is composed only of the aforementioned inclined interfaces; and

[0014] A second step of epitaxially growing a single crystal of a Group III nitride semiconductor on the aforementioned first layer, causing the aforementioned inclined interfaces to disappear, and growing a second layer having a mirror-finished surface,

[0015] In the aforementioned first step,

[0016] The aforementioned (0001) surface is caused to disappear by causing the aforementioned top surface of the aforementioned single crystal to have the aforementioned plurality of recesses, thereby forming a plurality of valley portions and a plurality of top portions on the surface of the aforementioned first layer,

[0017] When an arbitrary cross section perpendicular to the aforementioned main surface is observed, the average distance between the closest pair of top portions among the aforementioned plurality of top portions sandwiching one valley portion among the aforementioned plurality of valley portions in the direction along the aforementioned main surface is set to be more than 100 μm.

[0018] According to another aspect of the present application, there is provided a nitride semiconductor substrate,

[0019] which has a diameter of 2 inches or more and has a main surface whose nearest low-index crystal plane is a (0001) surface,

[0020] When an X-ray rocking curve measurement of (0002) surface diffraction is performed by irradiating an X-ray of Kα1 of Cu to the aforementioned main surface through a double-crystal monochromator and a slit of a Ge (220) surface,

[0021] The half-value width FWHMb of the aforementioned (0002) surface diffraction when the width of the ω direction of the aforementioned slit is set to 0.1 mm is 32 arcsec or less,

[0022] The difference FWHMa-FWHMb obtained by subtracting FWHMb from the half-value width FWHMa of the aforementioned (0002) surface diffraction when the width of the ω direction of the aforementioned slit is set to 1 mm is 30% or less of FWHMa,

[0023] The diffraction pattern when the width of the ω direction of the aforementioned slit is set to 1 mm has a single peak.

[0024] According to another aspect of the present application, there is provided a nitride semiconductor substrate,

[0025] has a diameter of 2 inches or more and has a main surface with a nearest low-index crystal plane of (0001) surface,

[0026] When a main surface of the nitride semiconductor substrate is observed with a field of view of 250 μm square by a multi-photon excitation microscope, and a dislocation density is calculated from a density of dark spots, there is no region in the main surface where the dislocation density exceeds 3 x 10 6 cm -2 -2, and a region where the dislocation density is less than 1 x 10 6 cm -2 -2 is present in 80% or more of the main surface,

[0027] the main surface has non-overlapping dislocation-free regions of 50 μm square at a density of 100 / cm 2 or more.

[0028] According to another aspect of the present application, there is provided a laminated structure including:

[0029] a base substrate formed of a single crystal of a Group III nitride semiconductor, having a mirror-finished main surface with a nearest low-index crystal plane of (0001) surface;

[0030] a first low-oxygen-concentration region provided directly on the main surface of the base substrate and formed of a single crystal of a Group III nitride semiconductor;

[0031] a high-oxygen-concentration region provided on the first low-oxygen-concentration region and formed of a single crystal of a Group III nitride semiconductor; and

[0032] a second low-oxygen-concentration region provided on the high-oxygen-concentration region and formed of a single crystal of a Group III nitride semiconductor,

[0033] the oxygen concentration of the high-oxygen-concentration region is higher than the oxygen concentration of each of the first low-oxygen-concentration region and the second low-oxygen-concentration region,

[0034] when any cross section perpendicular to the main surface is observed,

[0035] the upper surface of the first low-oxygen-concentration region has a plurality of valleys and a plurality of hills,

[0036] the nearest pair of hills among the plurality of hills sandwiching one valley among the plurality of valleys are spaced apart from each other by an average distance of more than 100 μm in a direction along the main surface.

[0037] Effects of the invention

[0038] According to the present application, the crystal quality of a nitride semiconductor substrate can be improved. BRIEF DESCRIPTION OF DRAWINGS

[0039] FIG. 1 is a flowchart showing the manufacturing method of a nitride semiconductor substrate according to an embodiment of the present application.

[0040] FIG. 2 (a) to (g) of FIG. 1 are schematic cross-sectional views showing a part of the manufacturing method of a nitride semiconductor substrate according to an embodiment of the present application.

[0041] FIG. 3 (a) to (c) of FIG. 2 are schematic cross-sectional views showing a part of the manufacturing method of a nitride semiconductor substrate according to an embodiment of the present application.

[0042] FIG. 4 is a schematic perspective view showing a part of the manufacturing method of a nitride semiconductor substrate according to an embodiment of the present application.

[0043] FIG. 5 (a) to (b) of FIG. 3 are schematic cross-sectional views showing a part of the manufacturing method of a nitride semiconductor substrate according to an embodiment of the present application.

[0044] FIG. 6 (a) to (b) of FIG. 4 are schematic cross-sectional views showing a part of the manufacturing method of a nitride semiconductor substrate according to an embodiment of the present application.

[0045] FIG. 7 (a) of FIG. 5 is a schematic cross-sectional view showing a growth process under a reference growth condition in which neither the inclined interface nor the c-plane is enlarged or reduced, FIG. 7 (b) of FIG. 5 is a schematic cross-sectional view showing a growth process under a first growth condition in which the inclined interface is enlarged and the c-plane is reduced.

[0046] FIG. 8 is a schematic cross-sectional view showing a growth process under a second growth condition in which the inclined interface is reduced and the c-plane is enlarged.

[0047] FIG. 9 (a) of FIG. 6 is a schematic top surface view of a nitride semiconductor substrate according to an embodiment of the present application, FIG. 9 (b) of FIG. 6 is a schematic cross-sectional view along the m-axis of a nitride semiconductor substrate according to an embodiment of the present application, FIG. 9 (c) of FIG. 6 is a schematic cross-sectional view along the a-axis of a nitride semiconductor substrate according to an embodiment of the present application.

[0048] FIG. 10 (a) of FIG. 7 is a schematic cross-sectional view showing X-ray diffraction for a curved c-plane, FIG. 10(b) and (c) of FIG. 2 are graphs showing fluctuations in diffraction angles of the (0002) plane, which are represented by the radius of curvature of the c-plane.

[0049] FIG. 11 is a graph showing an observation image obtained by observing a cross section of the layered structure of the example using a fluorescence microscope.

[0050] FIG. 12 (a) of FIG. 1 is a graph showing a normalized X-ray diffraction pattern when a slit is changed and X-ray rocking curve measurement is performed for the nitride semiconductor substrate of the example, FIG. 12 (b) of FIG. 1 is a graph showing a normalized X-ray diffraction pattern when the same measurement as the example is performed for the base substrate.

[0051] FIG. 13 (a) of FIG. 4 is a graph showing an observation image obtained by observing a surface of the layered structure of Experiment 2 using an optical microscope, FIG. 13 (b) of FIG. 4 is a graph showing an observation image obtained by observing a surface of the layered structure of Experiment 2 using a scanning electron microscope.

[0052] FIG. 14 (a) of FIG. 5 is a graph showing an observation image obtained by observing an M cross section of the layered structure of Experiment 2 using an optical microscope, FIG. 14 (b) of FIG. 5 is a graph showing an observation image obtained by observing an M cross section of the layered structure of Experiment 2 using a scanning electron microscope.

[0053] FIG. 15 (a) of FIG. 6 is a graph showing an observation image obtained by observing an a cross section of the layered structure of Experiment 2 using an optical microscope, FIG. 15 (b) of FIG. 6 is a graph showing an observation image obtained by observing an a cross section of the layered structure of Experiment 2 using a scanning electron microscope.

[0054] FIG. 16 is a graph showing observation of a main surface of the nitride semiconductor substrate of Sample 1 using a multiphoton excitation microscope.

[0055] FIG. 17 is a graph showing observation of a main surface of the nitride semiconductor substrate of Sample 1 using a multiphoton excitation microscope.

[0056] FIG. 18 is a graph showing observation of a main surface of the nitride semiconductor substrate of Sample 1 using a multiphoton excitation microscope.

[0057] FIG. 19 is a graph showing observation of a main surface of the nitride semiconductor substrate of Sample 1 using a multiphoton excitation microscope.

[0058] FIG. 20is a view of the main surface of the nitride semiconductor substrate of Sample 1 observed using a multiphoton excitation microscope.

[0059] FIG. 21 is a view of the main surface of the nitride semiconductor substrate of Sample 1 observed using a multiphoton excitation microscope.

[0060] FIG. 22 is a view of the main surface of the nitride semiconductor substrate of Sample 1 observed using a multiphoton excitation microscope.

[0061] FIG. 23 is a view of the main surface of the nitride semiconductor substrate of Sample 1 observed using a multiphoton excitation microscope.

[0062] FIG. 24 is a view of the main surface of the nitride semiconductor substrate of Sample 1 observed using a multiphoton excitation microscope.

[0063] FIG. 25 is a view of the main surface of the nitride semiconductor substrate of Sample 1 observed using a multiphoton excitation microscope.

[0064] FIG. 26 is a view of the main surface of the nitride semiconductor substrate of Sample 1 observed using a multiphoton excitation microscope.

[0065] FIG. 27 is a view of the main surface of the nitride semiconductor substrate of Sample 1 observed using a multiphoton excitation microscope.

[0066] FIG. 28 is a view of the main surface of the nitride semiconductor substrate of Sample 1 observed using a multiphoton excitation microscope.

[0067] FIG. 29 is a view of the main surface of the nitride semiconductor substrate of Sample 1 observed using a multiphoton excitation microscope.

[0068] FIG. 30 is a view of the main surface of the nitride semiconductor substrate of Sample 1 observed using a multiphoton excitation microscope.

[0069] FIG. 31 is a view of the main surface of the nitride semiconductor substrate of Sample 1 observed using a multiphoton excitation microscope.

[0070] FIG. 32 is a view of the main surface of the nitride semiconductor substrate of Sample 1 observed using a multiphoton excitation microscope.

[0071] FIG. 33 is a view of the main surface of the nitride semiconductor substrate of Sample 2 observed using a multiphoton excitation microscope.

[0072] Explanation of reference numerals

[0073] 10 base substrate

[0074] 30 first layer

[0075] 40 second layer

[0076] 50 nitride semiconductor substrate (substrate) DETAILED DESCRIPTION

[0077] <Insights obtained by the inventors>

[0078] First, the insights obtained by the inventors will be described.

[0079] (i) Regarding dislocation density

[0080] Conventionally, as described above, in the case where a crystal layer is further epitaxially grown on a base substrate formed of a single crystal of a Group III nitride semiconductor, for example, the crystal layer on the base substrate is grown so as not to expose a tilted interface other than a c-plane, and only the c-plane is used as a growth surface. In this case, there is a tendency that the dislocation density in the surface of the crystal layer is inversely proportional to the thickness of the crystal layer.

[0081] However, when the crystal layer is grown using only the c-plane as a growth surface, if the crystal layer is not grown to be very thick, the dislocation density in the surface of the crystal layer cannot be sufficiently reduced. Therefore, the productivity for obtaining a nitride semiconductor substrate having a desired dislocation density in a main surface is reduced.

[0082] Therefore, a technique capable of efficiently obtaining a nitride semiconductor substrate having a low dislocation density is desired.

[0083] (ii) Regarding deviation of off-angle

[0084] In a nitride semiconductor substrate, sometimes a (0001) plane is curved in a concave spherical shape with respect to a main surface. If the (0001) plane is curved with respect to the main surface, the angle formed by the <0001> axis with respect to the normal line of the main surface, that is, the off-angle, becomes uneven in the main surface.

[0085] The off-angle of a nitride semiconductor substrate, for example, can affect the surface morphology of a semiconductor functional layer grown on the substrate. For example, in the case where the radius of curvature of the (0001) plane of the substrate is small and the deviation of the off-angle of the substrate is large, sometimes the surface morphology of the semiconductor functional layer is deteriorated at a part of the substrate due to the off-angle. Therefore, when a semiconductor device as a Schottky barrier diode (SBD) is manufactured using the substrate, in the semiconductor device cut out from the part where the surface morphology of the semiconductor functional layer has been deteriorated, there is a possibility that the withstand voltage and the reliability are reduced.

[0086] In addition, in a case where, for example, indium (In) is doped on the substrate to form a light emitting layer, the off-angle of the nitride semiconductor substrate affects the In content in the light emitting layer. For example, in a case where the radius of curvature of the (0001) surface of the substrate is small and the deviation of the off-angle of the substrate is large, depending on the deviation of the off-angle of the substrate, the In content in the light emitting layer deviates. Therefore, in a light emitting element having the light emitting layer, it is possible that the deviation of the light emitting wavelength, light emitting unevenness, etc. occur.

[0087] Therefore, in order to not cause the deterioration of the surface morphology, light emitting unevenness, etc. practical problems, it is desirable to be able to reduce the deviation of the off-angle in the nitride semiconductor substrate.

[0088] The present application is based on the above insights (i) and (ii) discovered by the inventors.

[0089] <One embodiment of the present application>

[0090] Hereinafter, one embodiment of the present application will be described with reference to the drawings.

[0091] (1) Method for manufacturing nitride semiconductor substrate

[0092] The method for manufacturing a nitride semiconductor substrate described in the present embodiment will be described using FIGS. 1 to 6

[0093] FIG. 1 is a flowchart showing the method for manufacturing a nitride semiconductor substrate described in the present embodiment. FIG. 2 (a) to (g) of FIG. 3 (a) to (c) of FIG. 5 (a) to FIG. 6 (b) of is a schematic cross-sectional view showing a part of the method for manufacturing a nitride semiconductor substrate described in the present embodiment. FIG. 4 is a schematic perspective view showing a part of the method for manufacturing a nitride semiconductor substrate described in the present embodiment. Note that FIG. 4 corresponds to FIG. 3 (b) of this timing, a part of the first layer 30 grown on the base substrate 10 is shown. In addition, in FIG. 5 (b) of, the solid line shows a crystal surface in the middle of growth, and in FIG. 3 (c) to FIG. 6 (b) of, the broken line shows a dislocation.

[0094] As FIG. 1 ​As shown, the method for manufacturing a nitride semiconductor substrate according to this embodiment includes, for example, a substrate preparation process S100, a first process S200, a second process S300, a slicing process S400, and a polishing process S500.

[0095] (S100: Substrate preparation process)

[0096] First, in the substrate preparation process S100, a substrate 10 formed from a single crystal of a group III nitride semiconductor is prepared. In this embodiment, a gallium nitride (GaN) self-supporting substrate is prepared as the substrate 10, for example.

[0097] It should be noted that, in the following, in the crystal of a group III nitride semiconductor with a wurtzite structure, <0001> The axis (e.g., the

[0001] axis) is called the "c-axis", and the (0001) plane is called the "c-plane". It should be noted that sometimes the (0001) plane is called the "+c-plane (group III element polar plane)", and the (000-1) plane is called the "-c-plane (nitrogen (N) polar plane)". Additionally, the <1-100> axis (e.g., the [1-100] axis) is called the "m-axis", and the {1-100} plane is called the "m-plane". It should be noted that the m-axis can be denoted as the <10-10> axis. Furthermore, the <11-20> axis (e.g., the [11-20] axis) is called the "a-axis", and the {11-20} plane is called the "a-plane".

[0098] In the substrate preparation process S100 of this embodiment, for example, the substrate 10 is fabricated by the VAS (Void-Assisted Separation) method.

[0099] Specifically, the substrate preparation process S100 includes, for example, a substrate preparation process for crystal growth S110, a first crystal layer formation process S120, a metal layer formation process S130, a via formation process S140, a second crystal layer formation process S150, a peeling process S160, a slicing process S170, and a polishing process S180.

[0100] (S110: Preparation process for substrate for crystal growth)

[0101] First, such as FIG. 2 As shown in (a), a substrate 1 (hereinafter sometimes abbreviated as "substrate 1") for crystal growth is prepared. Substrate 1 is, for example, a sapphire substrate. It should be noted that substrate 1 may also be, for example, a Si substrate or a gallium arsenide (GaAs) substrate. Substrate 1 has, for example, a main surface 1s that serves as the growth surface. The low-index crystal surface closest to the main surface 1s is, for example, the c-surface 1c.

[0102] In this embodiment, the c-plane 1c of the substrate 1 is inclined with respect to the main surface Is. The c-axis 1ca of the substrate 1 is inclined at a prescribed off-angle θ0 with respect to the normal line of the main surface Is. The off-angle θ0 in the main surface Is of the substrate 1 is uniform throughout the main surface Is. The off-angle θ0 in the main surface Is of the substrate 1 affects the off-angle θ3 at the center of the main surface 10s of the base substrate 10 described later.

[0103] (S120: First Crystal Layer Formation Step)

[0104] Next, as shown in (b) of FIG. 1, for example, using a metal organic vapor phase epitaxy (MOVPE) method, the substrate 1 heated to a prescribed growth temperature is supplied with a trimethyl gallium (TMG) gas as a group III raw material gas, an ammonia (NH3) gas as a nitrogen raw material gas, and a silane (SiH4) gas as an n-type dopant gas, thereby causing a low-temperature growth GaN buffer layer and a Si-doped GaN layer as a first crystal layer (base growth layer) 2 to grow in this order on the main surface Is of the substrate 1. At this time, the thickness of the low-temperature growth GaN buffer layer and the thickness of the Si-doped GaN layer are set to, for example, 20 nm and 0.5 μm, respectively. FIG. 2

[0105] (S130: Metal Layer Formation Step)

[0106] Next, as shown in (c) of FIG. 1, a metal layer 3 is vapor-deposited on the first crystal layer 2. As the metal layer 3, for example, a titanium (Ti) layer is used. In addition, the thickness of the metal layer 3 is set to, for example, 20 nm. FIG. 2

[0107] (S140: Pore Formation Step)

[0108] Next, the above-described substrate 1 is put into an electric furnace, and the substrate 1 is placed on a susceptor having a prescribed heater. After the substrate 1 is placed on the susceptor, the substrate 1 is heated using the heater, and heat treatment is performed in an atmosphere containing a hydrogen gas or a hydride gas. Specifically, for example, heat treatment is performed at a prescribed temperature for 20 minutes in a hydrogen (H2) gas stream containing 20% of an NH3 gas. Note that the heat treatment temperature is set to, for example, 850°C or higher and 1100°C or lower. By performing this heat treatment, the metal layer 3 is nitrided, and a metal nitride layer 5 having a high density of fine pores on the surface is formed. In addition, by performing the above-described heat treatment, a portion of the first crystal layer 2 is etched by the pores of the metal nitride layer 5, and a high density of pores is formed in the first crystal layer 2.

[0109] Thus, as shown in (d) of FIG. 1, a first crystal layer 4 containing pores is formed. FIG. 2

[0110] (S150: Second Crystal Layer Formation Step)​​​

[0111] Next, a substrate 1 heated to a prescribed growth temperature is supplied with gallium chloride (GaCl) gas, NH3 gas, and dichlorosilane (SiH2Cl2) gas as an n-type dopant gas by a hydride vapor phase epitaxy (HVPE) method, for example, whereby a Si-doped GaN layer as a second crystal layer (formal growth layer) 6 is epitaxially grown on the first crystal layer 4 containing voids and the metal nitride layer 5. Note that, as the n-type dopant gas, instead of the SiH2Cl2 gas, germanium tetrachloride (GeCl4) gas or the like can be supplied to epitaxially grow a Ge-doped GaN layer as the second crystal layer 6.

[0112] At this time, the second crystal layer 6 grows on the first crystal layer 4 containing voids and the metal nitride layer 5 through the voids of the metal nitride layer 5 from the first crystal layer 4 containing voids. Some of the voids in the first crystal layer 4 containing voids are filled with the second crystal layer 6, but other voids in the first crystal layer 4 containing voids remain. Due to the voids remaining in the first crystal layer 4 containing voids, a flat gap is formed between the second crystal layer 6 and the metal nitride layer 5. This gap causes the second crystal layer 6 to be peeled off in the peeling process S160 described later.

[0113] In addition, at this time, the second crystal layer 6 grows while inheriting the orientation of the substrate 1. That is, the misorientation angle θ1 in the main surface of the second crystal layer 6 is uniform in the entire main surface, like the misorientation angle θ0 in the main surface 1s of the substrate 1.

[0114] In addition, at this time, the thickness of the second crystal layer 6 is set to, for example, 600 μm or more, and preferably 1 mm or more. Note that, the upper limit of the thickness of the second crystal layer is not particularly limited, and from the viewpoint of improving productivity, the thickness of the second crystal layer 6 is preferably set to 50 mm or less.

[0115] (S160: Peeling Process)

[0116] After the growth of the second crystal layer 6 is completed, during the process of cooling the HVPE device used to grow the second crystal layer 6, the second crystal layer 6 naturally peels off from the substrate 1 with the first crystal layer 4 containing voids and the metal nitride layer 5 as boundaries.

[0117] At this time, in the second crystal layer 6, tensile stress is introduced due to the initial nuclei generated during the growth thereof attracting each other. Thus, due to the tensile stress generated in the second crystal layer 6, the internal stress acts to cause the surface side of the second crystal layer 6 to be recessed. In addition, the dislocation density on the main surface (surface) side of the second crystal layer 6 becomes low, whereas the dislocation density on the back surface side of the second crystal layer 6 becomes high. Thus, due to the difference in the dislocation density in the thickness direction of the second crystal layer 6 as well, the internal stress acts to cause the surface side of the second crystal layer 6 to be recessed.

[0118] As a result, as shown in (f) of FIG. 6, the second crystal layer 6 warps with the surface side thereof being recessed after being peeled from the substrate 1. Thus, the c-plane 6c of the second crystal layer 6 is curved in a concave spherical shape with respect to a plane perpendicular to the normal direction of the center of the main surface 6s of the second crystal layer 6. The deviation angle Θ2 of the c-axis 6ca with respect to the normal of the center of the main surface 6s of the second crystal layer 6 has a prescribed distribution. FIG. 2

[0119] (S170: Slicing Step)

[0120] Next, as shown in (f) of FIG. 6, the second crystal layer 6 is sliced using a wire saw, for example, along a slicing surface SS that is substantially perpendicular to the normal direction of the center of the main surface 6s of the second crystal layer 6. FIG. 2 Thus, as shown in (g) of FIG. 6, the base substrate 10 is formed as an in-situ sliced substrate. At this time, the thickness of the base substrate 10 is set to be, for example, 450 μm. Note that there is a possibility that the deviation angle Θ3 of the base substrate 10 changes from the deviation angle Θ2 of the second crystal layer 6 depending on the slicing direction.

[0121] FIG. 2 (S180: Polishing Step)

[0122] Next, both surfaces of the base substrate 10 are polished using a polishing device. Thus, the main surface 10s of the base substrate 10 is mirror-finished.

[0123] Through the above base substrate preparation step S100, the base substrate 10 formed of a single crystal of GaN is obtained.

[0124] The diameter of the base substrate 10 is, for example, 2 inches or more. In addition, the thickness of the base substrate 10 is, for example, 300 μm or more and 1 mm or less.

[0125] The main surface 10s of the base substrate 10 has, for example, a main surface (base surface) 10s that becomes an epitaxial growth surface. In the present embodiment, the low-index crystal plane closest to the main surface 10s is, for example, the c-plane (+c-plane) 10c.

[0126] The main surface 10s of the base substrate 10 has, for example, a main surface (base surface) 10s that becomes an epitaxial growth surface. In the present embodiment, the low-index crystal plane closest to the main surface 10s is, for example, the c-plane (+c-plane) 10c. ​​

[0127] The c-plane 10c in the base substrate 10 is curved into a concave spherical shape with respect to the main surface 10s. The "spherical shape" referred to herein means a curved surface shape that approximates a sphere. In addition, the "approximate sphere" referred to herein means approximation with respect to a true sphere or an ellipsoid within a prescribed error range.

[0128] In the present embodiment, the c-plane 10f of the base substrate 10 exhibits an approximate spherical curved surface shape, for example, in a cross section along the m-axis and a cross section along the a-axis, respectively. The radius of curvature of the c-plane 10c in the base substrate 10 is, for example, 1 m or more and less than 10 m.

[0129] The deviation angle θ3 of the c-axis 10ca with respect to the normal line of the center of the main surface 10s of the base substrate 10 has a prescribed distribution.

[0130] In the present embodiment, the magnitude of the deviation angle θ3 at the center of the main surface 10s of the base substrate 10 is, for example, 1° or less, and preferably 0.4° or less. If the magnitude of the deviation angle θ3 at the center of the main surface 10s exceeds 1°, it is sometimes difficult to exhibit three-dimensional growth of the first layer 30 due to the first growth conditions in the first process S200 described later. Thus, it is difficult to make the c-plane 30c disappear. In contrast, in the present embodiment, by making the magnitude of the deviation angle θ3 at the center of the main surface 10s 1° or less, it is possible to make the first layer 30 easily grow three-dimensionally in the first process S200 described later. Thus, it is possible to make the c-plane 30c easily disappear. Furthermore, by making the magnitude of the deviation angle θ3 at the center of the main surface 10s 0.4° or less, it is possible to make the first layer 30 grow three-dimensionally under a wider range of growth conditions, and to make the c-plane 30c stably disappear.

[0131] Note that, from the viewpoint of three-dimensional growth of the first layer 30, the smaller the magnitude of the deviation angle θ3 at the center of the main surface 10s is, the better. However, if the magnitude of the deviation angle θ3 at the center of the main surface 10s is too close to 0°, there is a possibility that the surface of the first layer 30 becomes too rough. Thus, the magnitude of the deviation angle θ3 at the center of the main surface 10s is, for example, preferably 0.1° or more.

[0132] Note that the magnitude and direction of the deviation angle θ3 at the center of the main surface 10s of the base substrate 10 can be adjusted by, for example, the magnitude and direction of the deviation angle θ0 of the crystal growth substrate 1 used in the VAS method described above, and the slicing angle and slicing direction in the slicing process S170.

[0133] In addition, in the present embodiment, for example, the main surface 10s of the base substrate 10 is held in a so-called as-received state in which a single crystal of a Group III nitride semiconductor can be epitaxially grown, and is roughly polished.

[0134] Specifically, the root mean square roughness RMs of the main surface 10s of the base substrate 10 is, for example, 1 nm or more and 10 nm or less. By setting the RMs of the main surface 10s of the base substrate 10 within the above range, the occurrence of the inclined interface 30i other than the c-plane in the surface of the first layer 30 can be promoted when the first layer 30 is grown on the base substrate 10 in the first process S200 described later. In addition, by setting the RMs of the main surface 10s of the base substrate 10 within the above range, the surface of the first layer 30 can be prevented from being excessively roughened, and the most proximal top-to-top average distance L in the first layer 30 described later can be prevented from being shortened.

[0135] In addition, in the present embodiment, for example, the crystal quality of the bulk portion in the base substrate 10 can be favorably maintained, and the crystal strain introduced by the processing of the slicing process S170, the polishing process S180, and the like of the base substrate 10 can be left on the main surface 10s side of the base substrate 10. Specifically, the full width at half maximum (FWHM) of the (10-10) plane diffraction when X-ray rocking curve measurement is performed with an incident angle of 2° with respect to the main surface 10s of the base substrate 10 after processing is set to be greater than, for example, the FWHM of the base substrate 10 before processing, and is set to be 60 arcsec or more and 200 arcsec or less. By setting the FWHM of the (10-10) plane diffraction within the above range, the stable crystal plane appearing on the surface of the first layer 30 described later can be changed due to the crystal strain on the main surface 10s side of the base substrate 10. As a result, the inclined interface 30i other than the c-plane can be generated on the surface of the first layer 30. In addition, by setting the FWHM of the (10-10) plane diffraction within the above range, the excessive dislocations can be suppressed from being generated in the first layer 30 described later due to the crystal strain on the main surface 10s side of the base substrate 10.

[0136] In addition, in the present embodiment, the base substrate 10 is produced by the VAS method described above, and thus the dislocation density in the main surface 10s of the base substrate 10 is low. Specifically, the dislocation density in the main surface 10s of the base substrate 10 is, for example, 3 x 10 6 cm -2 1 x 10 7 cm -2 .

[0137] (S200: First Process (First Layer Growth Process))

[0138] After the base substrate 10 is prepared, the first layer 30 is grown on the main surface 10s of the base substrate 10 in the first process S200. FIG. 3As shown in (a), the first process S200 is performed using the base substrate 10 in a state where any of the processes of forming a mask layer on the main surface 10s and forming a concavo-convex pattern on the main surface 10s has not been performed. Note that the "mask layer" referred to here means a mask layer having a prescribed opening, for example, used in a so-called ELO (Epitaxial Lateral Overgrowth) method. Further, the "concavo-convex pattern" referred to here means at least either of a groove and a ridge obtained by directly patterning the main surface of a base substrate, for example, used in a so-called suspended epitaxial method. The concavo-convex pattern referred to here has a height difference of, for example, 100 nm or more. The base substrate 10 of the present embodiment is used in the first process S200 in a state where it does not have the above-described structure.

[0139] First, as shown in (a), the first process S200 is performed using the base substrate 10 in a state where any of the processes of forming a mask layer on the main surface 10s and forming a concavo-convex pattern on the main surface 10s has not been performed. Note that the "mask layer" referred to here means a mask layer having a prescribed opening, for example, used in a so-called ELO (Epitaxial Lateral Overgrowth) method. Further, the "concavo-convex pattern" referred to here means at least either of a groove and a ridge obtained by directly patterning the main surface of a base substrate, for example, used in a so-called suspended epitaxial method. The concavo-convex pattern referred to here has a height difference of, for example, 100 nm or more. The base substrate 10 of the present embodiment is used in the first process S200 in a state where it does not have the above-described structure. FIG. 3 of (b), FIG. 3 of (c), and FIG. 4 As shown in (b), the III-V nitride semiconductor single crystal having the top surface 30u exposing the c-plane 30c is directly epitaxially grown on the main surface 10s of the base substrate 10. Thus, the first layer (three-dimensionally grown layer) 30 is grown.

[0140] At this time, the top surface 30u of the single crystal is made to have a plurality of recesses 30p surrounded by the inclined interface 30i other than the c-plane, and as it goes up from the main surface 10s of the base substrate 10, the inclined interface 30i is made to gradually expand and the c-plane 30c is made to gradually shrink. Thus, the c-plane 30c disappears from the top surface 30u. As a result, the first layer 30 having a surface composed of only the inclined interface 30i is grown.

[0141] That is, in the first process S200, the first layer 30 is made to three-dimensionally grow in such a manner that the main surface 10s of the base substrate 10 is intentionally roughened. Note that the first layer 30 is grown as a single crystal even if it is formed in this growth form. In this regard, the first layer 30 is different from a so-called low-temperature growth buffer layer formed in an amorphous or polycrystalline form on a foreign substrate before the III-V nitride semiconductor is epitaxially grown on the foreign substrate such as sapphire.

[0142] In the present embodiment, as the first layer 30, a layer formed of the same III-V nitride semiconductor as the III-V nitride semiconductor constituting the base substrate 10 is epitaxially grown, for example. Specifically, for example, the base substrate 10 is heated, and GaCl gas and NH3 gas are supplied to the heated base substrate 10, whereby a GaN layer is epitaxially grown as the first layer 30, using the HVPE method.

[0143] Here, in the first process S200, in order to exhibit the above-described growth process, the first layer 30 is grown under prescribed first growth conditions, for example.

[0144] First, use FIG. 7 (a) describes the baseline growth conditions under which the inclined interfaces 30i and c-face 30c neither expand nor shrink. FIG. 7 (a) is a schematic cross-sectional view showing the growth process under the baseline growth conditions where the inclined interface and the c-plane neither expand nor shrink.

[0145] exist FIG. 7 In (a), the thick solid line represents the surface of the first layer 30 every unit time. FIG. 7 The inclined interface 30i shown in (a) is set as the most inclined interface relative to the c-plane 30c. Additionally, in FIG. 7 In (a), the growth rate of surface 30c in the first layer 30 is denoted as G. c0 Let G be the growth rate of the inclined interface 30i in the first layer 30. i Let α be the angle between surface c 30c in the first layer 30 and the inclined interface 30i. Additionally, in FIG. 7 In (a), the first layer 30 grows while maintaining the angle α between the c-surface 30c and the inclined interface 30i. It should be noted that the deviation angle of the c-surface 30c of the first layer 30 is negligible compared to the angle α between the c-surface 30c and the inclined interface 30i.

[0146] like FIG. 7 As shown in (a), when the inclined interface 30i and the c-plane 30c neither expand nor shrink, the trajectory of the intersection point of the inclined interface 30i and the c-plane 30c is perpendicular to the c-plane 30c. Therefore, the reference growth condition that the inclined interface 30i and the c-plane 30c neither expand nor shrink satisfies the following equation (a).

[0147] G c0 =G i / cosα···(a)

[0148] Next, use FIG. 7 (b) describes the first growth condition in which the inclined interface 30i expands and the c-surface 30c shrinks. FIG. 7 (b) is a schematic cross-sectional view showing the growth process under the first growth condition where the inclined interface expands and the c-plane shrinks.

[0149] exist FIG. 7 In (b), it is also related to FIG. 7 Similarly, in (a), the thick solid line represents the surface of the first layer (30) every unit time. Additionally, FIG. 7 The inclined interface 30i shown in (b) is also set as the most inclined interface relative to surface c 30c. Additionally, in FIG. 3In (b), the growth rate of the c-plane 30c in the first layer 30 is denoted as G c1 The advancing rate of the locus of the intersection of the inclined interface 30i and the c-plane 30c in the first layer 30 is denoted as Rl. Further, the angle between the locus of the intersection of the inclined interface 30i and the c-plane 30c and the c-plane 30c is denoted as α R1 When the angle between the Rl direction and the G i direction is denoted as α', α' = α + 90 - α R1 It should be noted that the deviation angle of the c-plane 30c of the first layer 30 is negligible compared to the angle α between the c-plane 30c and the inclined interface 30i.

[0150] As shown in (b), the advancing rate Rl of the locus of the intersection of the inclined interface 30i and the c-plane 30c is expressed by the following equation (b). FIG. 4

[0151] Rl = G i / cos α' ··· (b)

[0152] Further, the growth rate G c1 of the c-plane 30c in the first layer 30 is expressed by the following equation (c).

[0153] G c1 = Rl sin α R1 ··· (c)

[0154] By substituting equation (b) into equation (c), G c1 can be expressed using G i and the following equation (d).

[0155] G c1 = G i sin α R1 / cos (α + 90 - α R1 ) ··· (d)

[0156] In order to expand the inclined interface 30i and shrink the c-plane 30c, it is preferable that α R1 < 90°. Therefore, the first growth condition in which the inclined interface 30i is expanded and the c-plane 30c is shrunk is preferably satisfied by equation (1) below and α R1 < 90°.

[0157] G c1 > G i / cos α ··· (1)

[0158] where G i ​G is the growth rate of the most inclined inclined surface 30i with respect to the c-plane 30c, and α is the angle of the most inclined inclined surface 30i with respect to the c-plane 30c.

[0159] Alternatively, it can also be considered that G c1 is preferably greater than G c0 under the reference growth conditions. c1 Thus, by substituting formula (a) into G c0 , formula (1) can also be derived.

[0160] Note that the growth conditions under which the most inclined inclined surface 30i with respect to the c-plane 30c is enlarged are the most severe conditions, and therefore, if the first growth conditions satisfy formula (1), the other inclined surfaces 30i can also be enlarged.

[0161] Specifically, for example, when the most inclined inclined surface 30i with respect to the c-plane 30c is the {10-11} plane, α = 61.95°. Therefore, the first growth conditions preferably satisfy, for example, formula (1') below.

[0162] G c1 > 2.13G i ···(1')

[0163] Alternatively, as described later, for example, when the inclined surface 30i is the {11-2m} plane with m ≥ 3, the most inclined inclined surface 30i with respect to the c-plane 30c is the {11-23} plane, and therefore, α = 47.3°. Therefore, the first growth conditions preferably satisfy, for example, formula (1") below.

[0164] G c1 > 1.47G i ···(1")

[0165] As the first growth conditions of the present embodiment, for example, the growth temperature in the first process S200 is made lower than the growth temperature in the second process S300 described later. Specifically, the growth temperature in the first process S200 is set to, for example, 980°C or higher and 1020°C or lower, and preferably 1000°C or higher and 1020°C or lower.

[0166] In addition, as the first growth conditions of the present embodiment, for example, the ratio of the partial pressure of the flow rate of the NH3 gas as the nitrogen source gas with respect to the partial pressure of the GaCl gas as the Group III source gas (hereinafter also referred to as "V / III ratio") in the first process S200 can be made greater than the V / III ratio in the second process S300 described later. Specifically, the V / III ratio in the first process S200 is set to, for example, 2 or higher and 20 or lower, and preferably 2 or higher and 15 or lower.

[0167] In fact, as the first growth condition, at least one of the growth temperature and the V / III ratio is adjusted within the above range in a manner that satisfies Equation (1).

[0168] It should be noted that other conditions in the first growth conditions of this embodiment are as follows, for example.

[0169] Growth pressure: 90–105 kPa, preferably 90–95 kPa

[0170] Partial pressure of GaCl gas: 1.5~15kPa

[0171] N2 gas flow rate / H2 gas flow rate: 0~1

[0172] Here, the first step S200 of this embodiment is classified into two steps, for example, according to the morphology of the first layer 30 during growth. Specifically, the first step S200 of this embodiment includes, for example, a tilted interface expansion step S220 and a tilted interface maintenance step S240. Through these steps, the first layer 30 includes, for example, a tilted interface expansion layer 32 and a tilted interface maintenance layer 34.

[0173] (S220: Inclined interface expansion process)

[0174] First, such as FIG. 3 (b) and FIG. 4 As shown, the tilted interface expansion layer 32 of the first layer 30, formed from a single crystal of a group III nitride semiconductor, is directly epitaxially grown on the main surface 10s of the substrate 10 under the aforementioned first growth conditions.

[0175] In the initial stage of the growth of the inclined interface expansion layer 32, the inclined interface expansion layer 32 is grown with the c-plane 30c as the growth surface in the normal direction (along the c-axis direction) of the main surface 10s of the substrate 10.

[0176] By allowing the inclined interface expansion layer 32 to grow slowly under the first growth conditions, thus... FIG. 3 (b) and FIG. 3 As shown, the top surface 30u of the inclined interface expansion layer 32, which exposes the c-surface 30c, is formed with a plurality of recesses 30p composed of inclined interfaces 30i other than the c-surface. These recesses 30p, composed of inclined interfaces 30i other than the c-surface, are randomly formed on the top surface 30u. Thus, an inclined interface expansion layer 32 is formed with the c-surface 30c and inclined interfaces 30i other than the c-surface mixed on its surface.

[0177] It should be noted that the "tilted interface 30i" mentioned here refers to the growth interface that is tilted relative to the c-plane 30c, including facets with low exponents other than the c-plane, facets with high exponents other than the c-plane, or tilted surfaces that cannot be represented by facet exponents. It should be noted that facets other than the c-plane are, for example, {11-2m}, {1-10n}, etc., where m and n are integers other than 0.

[0178] In this embodiment, by using the aforementioned substrate 10 and adjusting the first growth conditions in a manner that satisfies equation (1), it is possible to generate a {11-2m} surface as an inclined interface 30i with, for example, m≥3. This reduces the tilt angle of the {11-2m} surface relative to the c-surface 30c. Specifically, this tilt angle can be reduced to 47.3° or less.

[0179] By further growing the inclined interface expansion layer 32 under the first growth conditions, thus as FIG. 3 As shown in (b) and (c), as the substrate 10 moves upward, the inclined interface 30i (excluding the c-plane) gradually expands in the inclined interface expansion layer 32, while the c-plane 30c gradually shrinks. It should be noted that, at this time, as the substrate 10 moves upward, the angle of inclination of the inclined interface 30i relative to the main surface 10s of the substrate 10 gradually decreases. Thus, the majority of the inclined interface 30i ultimately forms the aforementioned {11-2m} plane with m≥3.

[0180] Furthermore, if the inclined interface expansion layer 32 is allowed to grow gradually, the c-face 30c of the inclined interface expansion layer 32 disappears from the top surface 30u, and the surface of the inclined interface expansion layer 32 is composed only of the inclined interface 30i. Thus, a mountain-shaped inclined interface expansion layer 32 is formed by continuously connecting the cones.

[0181] In this way, by generating multiple recesses 30p on the top surface 30u of the inclined interface expansion layer 32, which are composed of inclined interfaces 30i other than the c-surface, the c-surface 30c disappears, thereby achieving the desired effect. FIG. 5 As shown in (c), a plurality of valleys 30v and a plurality of tops 30t are formed on the surface of the inclined interface expansion layer 32. The plurality of valleys 30v are downwardly protruding inflection points on the surface of the inclined interface expansion layer 320, and are formed above the respective locations where the inclined interfaces 30i appear, except for the c-surface. On the other hand, the plurality of tops 30t are upwardly protruding inflection points on the surface of the inclined interface expansion layer 320, and are formed between a pair of inclined interfaces 30i that expand in opposite directions, at or above the location where the c-surface 30c (last) disappears and ends. The valleys 30v and tops 30t are formed alternately in the direction along the main surface 10s of the substrate 10.

[0182] In this embodiment, during the initial stage of growing the inclined interface expansion layer 32, the inclined interface expansion layer 32 is grown to a predetermined thickness on the main surface 10s of the substrate 10 without creating an inclined interface 30i, using the c-surface 30c as the growth surface. Then, an inclined interface 30i other than the c-surface is created on the surface of the inclined interface expansion layer 32. As a result, multiple valleys 30v are formed at a distance above the main surface 10s of the substrate 10.

[0183] Through the growth process of layer 32, which is expanded via a tilted interface as described above, dislocations bend and propagate as follows. Specifically, as... FIG. 5 As shown in (c), multiple dislocations extending along the c-axis within the substrate 10 propagate from the substrate 10 toward the c-axis of the inclined interface expansion layer 32. In the region of the inclined interface expansion layer 32 where the c-plane 30c is grown, dislocations propagate from the substrate 10 toward the c-axis of the inclined interface expansion layer 32. However, if a dislocation propagating along the c-axis of the inclined interface expansion layer 32 is exposed at the inclined interface 30i, the dislocation bends and propagates in a direction substantially perpendicular to the inclined interface 30i at the exposed location. That is, the dislocation bends and propagates in a direction inclined relative to the c-axis. Therefore, in the inclined interface expansion process S220 and subsequent processes, dislocations locally accumulate above approximately the center between the pair of tops 30t. As a result, the dislocation density on the surface of the second layer 40 (described later) can be reduced.

[0184] At this time, in the present embodiment, when observing an arbitrary cross section perpendicular to the main surface 10s of the base substrate 10, the average distance (also referred to as "closest apex-to-apex average distance") L at which the closest pair of apexes 30t among the plurality of apexes 30t sandwiching one of the plurality of valleys 30v is spaced apart from each other in the direction along the main surface 10s of the base substrate 10 is set to, for example, more than 100 μm. As in the case where the fine hexagonal pyramid-shaped crystal nuclei are generated on the main surface 10s of the base substrate 10 from the initial stage of the inclined interface expansion process S220, or the like, when the closest apex-to-apex average distance L is 100 μm or less, the distance over which the dislocations propagate by bending is shortened in the inclined interface expansion process S220 and the processes thereafter. Thus, the dislocations do not sufficiently converge to the upper side of the substantially center between the pair of apexes 30t in the inclined interface expansion layer 32. As a result, the dislocation density in the surface of the second layer 40 described later can not be sufficiently reduced. In contrast, in the present embodiment, by making the closest apex-to-apex average distance L more than 100 μm, it is possible to ensure that the distance over which the dislocations propagate by bending is at least more than 50 μm in the inclined interface expansion process S220 and the processes thereafter. Thus, it is possible to sufficiently converge the dislocations to the upper side of the substantially center between the pair of apexes 30t in the inclined interface expansion layer 32. As a result, it is possible to sufficiently reduce the dislocation density in the surface of the second layer 40 described later.

[0185] On the other hand, in the present embodiment, the closest apex-to-apex average distance L is made less than 800 μm. If the closest apex-to-apex average distance L is 800 μm or more, the height from the valley 30v to the apex 30t of the inclined interface expansion layer 32 from the main surface 10s of the base substrate 10 is excessively high. Thus, in the second process S300 described later, the thickness until the second layer 40 is mirror-finished is thick. In contrast, in the present embodiment, by making the closest apex-to-apex average distance L less than 800 μm, it is possible to reduce the height from the valley 30v to the apex 30t of the inclined interface expansion layer 32 from the main surface 10s of the base substrate 10. Thus, in the second process S300 described later, it is possible to quickly mirror-finish the second layer 40.

[0186] In addition, at this time, in the inclined interface expansion layer 320, depending on the difference in growth surface during growth, a first c-surface growth region 60 grown with the c-surface 30c as the growth surface, and an inclined interface growth region 70 (gray portion in the figure) grown with the inclined interface 30i other than the c-surface as the growth surface are formed.

[0187] In addition, at this time, in the first c-plane growth region 60, a valley 60a is formed at a position where the inclined interface 30i is generated, and a hill 60b is formed at a position where the c-plane 30c disappears. In addition, in the first c-plane growth region 60, on both sides sandwiching the hill 60b, a pair of inclined portions 60i are formed as traces of the intersection points of the c-plane 30c and the inclined interface 30i.

[0188] In addition, at this time, by causing the first growth conditions to satisfy the formula (1), the angle β formed by the pair of inclined portions 60i is, for example, 70° or less.

[0189] Details of these regions will be described later.

[0190] (S240: Inclined Interface Maintaining Step)

[0191] After causing the c-planes 30c to disappear from the surface of the inclined interface expansion layer 32, as shown in (a) of FIG. 6, the state in which the inclined interface 30i occupies more of the surface than the c-planes 30c, preferably the state in which the surface is composed only of the inclined interface 30i, is maintained, and the growth of the first layer 30 is continued over a prescribed thickness. Thus, the inclined interface maintaining layer 34 having a surface in which the inclined interface 30i occupies more than the c-planes 30c is formed on the inclined interface expansion layer 32. By forming the inclined interface maintaining layer 34, the c-planes 30c can be caused to certainly disappear in the entire surface of the first layer 30. FIG. 6

[0192] At this time, the c-planes 30c can reappear on a part of the surface of the inclined interface maintaining layer 34, but it is preferable that the inclined interface 30i mainly be exposed in the surface of the inclined interface maintaining layer 34 so that the area ratio of the inclined interface growth region 70 in the cross section along the main surface 10s of the base substrate 10 becomes 80% or more. Note that the higher the area ratio of the inclined interface growth region 70 in the cross section, the better, and 100% is preferable.

[0193] At this time, the growth conditions in the inclined interface maintaining step S240 are maintained as the above-described first growth conditions as in the inclined interface expansion step S220. Thus, the inclined interface maintaining layer 34 can be grown with only the inclined interface 30i as the growth surface.

[0194] In addition, at this time, by growing the inclined interface maintaining layer 34 with the inclined interface 30i as the growth surface under the first growth conditions, as described above, the dislocations that propagate in the direction in which the c-axis is inclined also continue to propagate in the same direction in the inclined interface maintaining layer 34 at the position where the inclined interface 30i is exposed in the inclined interface expansion layer 32.

[0195] ​Further, at this time, with respect to the inclined interface maintaining layer 34, by growing with the inclined interface 30i as a growth surface, the entirety of the inclined interface maintaining layer 34 becomes a part of the inclined interface growth region 70.

[0196] By the above first process S200, the first layer 30 having the inclined interface expanding layer 32 and the inclined interface maintaining layer 34 is formed.

[0197] In the first process S200 of the present embodiment, the height from the main surface 10s of the base substrate 10 to the top 30t of the first layer 30 (the maximum height in the thickness direction of the first layer 30) is set to, for example, more than 100 μm and less than 1.5 mm.

[0198] (S300: Second Process (Second Layer Growth Process))

[0199] After the first layer 30 in which the c-plane 30c has been made to disappear has been grown, as shown in (b) of FIG. 2 and (a) of FIG. 3, a single crystal of a Group III nitride semiconductor is further epitaxially grown on the first layer 30. FIG. 8 FIG. 8 As shown in (b) of FIG. 2 and (a) of FIG. 3, a single crystal of a Group III nitride semiconductor is further epitaxially grown on the first layer 30.

[0200] At this time, as the inclined interface 40i is made to gradually decrease toward the upper side of the main surface 10s of the base substrate 10, the c-plane 40c is made to gradually expand. Thereby, the inclined interface 30i formed on the surface of the first layer 30 is made to disappear. As a result, the second layer (planarization layer) 40 having a mirror-finished surface is grown. Note that the "mirror surface" referred to here means a surface in which the maximum difference in height of surface irregularities is below the wavelength of visible light.

[0201] In the present embodiment, as the second layer 40, a layer having the same Group III nitride semiconductor as that constituting the first layer 30 as a main component is epitaxially grown, for example. Note that in the second process S300, with respect to the base substrate 10 heated to a prescribed growth temperature, a GaCl gas, an NH3 gas, and a dichlorosilane (SiH2Cl2) gas as an n-type dopant gas are supplied, whereby a silicon (Si)-doped GaN layer is epitaxially grown as the second layer 40. Note that as the n-type dopant gas, a GeCl4 gas or the like can be supplied instead of the SiH2Cl2 gas.

[0202] Here, in the second process S300, in order to exhibit the above-described growth process, the second layer 40 is grown under prescribed second growth conditions, for example.

[0203] Using the above-described second growth conditions, the inclined interface 40i is made to decrease and the c-plane 40c is made to expand. FIG. 8 FIG. 8 ​​is a schematic cross-sectional view showing a growth process under a second growth condition in which the inclined interface is reduced and the c-plane is enlarged. FIG. 7 shows a process in which the second layer 40 grows on the first layer 30 exposing the inclined interface 30i most inclined with respect to the c-plane 30c.

[0204] In FIG. 8 , also as in (a) of FIG. 8 , the thick solid line indicates the surface of the second layer 40 every unit time. In addition, in FIG. 8 , the growth rate of the c-plane 40c among the second layer 40 is denoted as G c2 , the growth rate of the inclined interface 40i among the second layer 40 is denoted as G i , and the advancing rate of the locus of the intersection of the inclined interface 40i and the c-plane 40c among the second layer 40 is denoted as R2. In addition, the angle narrower than the angle formed between the locus of the intersection of the inclined interface 40i and the c-plane 40c and the c-plane 30c is denoted as a R2 . When the angle formed between the direction of R2 and the direction of G i is denoted as a", a" = a - (90 - a R2 ). In addition, in FIG. 5 , the second layer 40 grows while maintaining the angle a formed between the c-plane 30c and the inclined interface 30i in the first layer 30. Note that the deviation angle of the c-plane 40c of the second layer 40 can be ignored compared to the angle a formed between the c-plane 30c and the inclined interface 30i.

[0205] As shown in FIG. 6 , the advancing rate R2 of the locus of the intersection of the inclined interface 40i and the c-plane 40c is expressed by the following equation (e).

[0206] R2 = G i / cos a"... (e)

[0207] In addition, the growth rate G c2 of the c-plane 40c among the second layer 40 is expressed by the following equation (f).

[0208] G c2 = R2 sin a R2 ... (f)

[0209] By substituting equation (e) into equation (f), G c2 can be expressed using G i and the following equation (g).

[0210] G c2 = G i sin a R2 / cos (a + a R2 - 90)... (g)

[0211] To reduce the size of the inclined interface 40i and increase the size of the c-plane 40c, α is preferred. R2 <90°. Therefore, the second growth condition where the inclined interface 40i shrinks and the c-plane 40c expands is preferably achieved by formula (g) and α. R2 <90° to satisfy the following equation (2).

[0212] G c2 <G i / cosα···(2)

[0213] Among them, as mentioned above, G i α represents the growth rate of the inclined interface 40i, which is most inclined relative to the c-plane 40c, and α is the angle between the inclined interface 40i and the c-plane 40c.

[0214] Alternatively, the growth rate of surface 30c in the second layer 40 under the baseline growth conditions can be denoted as G. c0 At that time, G can also be considered to be under the second growth condition. c2 Preferred G is smaller than that under the baseline growth conditions. c0 Therefore, by substituting equation (a) into G... c2 <G c0 In this context, equation (2) can also be derived.

[0215] It should be noted that the growth condition that shrinks the inclined interface 40i that is most inclined relative to the c-plane 40c is the most stringent condition. Therefore, if the second growth condition satisfies equation (2), then other inclined interfaces 40i can also be shrunk.

[0216] Specifically, when the inclined interface 40i, which is most inclined relative to the c-plane 40c, is the {10-11} plane, the second growth condition preferably satisfies the following equation (2').

[0217] G c2 <2.13G i ···(2')

[0218] Alternatively, for example, when the inclined interface 30i is a {11-2m} surface with m≥3, since the inclined interface 30i that is most inclined relative to the c surface 30c is the {11-23} surface, the second growth condition preferably satisfies, for example, the following equation (2”).

[0219] G c2 <1.47G i ···(2”)

[0220] As a second growth condition in this embodiment, the growth temperature in the second process S300 is, for example, higher than the growth temperature in the first process S200. Specifically, the growth temperature in the second process S300 is set to, for example, 990°C or higher and 1120°C or lower, preferably 1020°C or higher and 1100°C or lower.

[0221] Furthermore, as a second growth condition in this embodiment, the V / III ratio in the second step S300 can be adjusted. For example, the V / III ratio in the second step S300 can be made smaller than the V / III ratio in the first step S200. Specifically, the V / III ratio in the second step S300 can be set to, for example, 1 or more and 10 or less, preferably 1 or more and 5 or less.

[0222] In fact, as a second growth condition, at least one of the growth temperature and the V / III ratio is adjusted within the above range in a manner that satisfies Equation (2).

[0223] It should be noted that other conditions in the second growth conditions of this embodiment are as follows, for example.

[0224] Growth pressure: 90–105 kPa, preferably 90–95 kPa

[0225] Partial pressure of GaCl gas: 1.5~15kPa

[0226] N2 gas flow rate / H2 gas flow rate: 1~20

[0227] Here, the second process S300 of this embodiment is classified into two processes, for example, based on the morphology of the second layer 40 during growth. Specifically, the second process S300 of this embodiment includes, for example, a c-plane enlargement process S320 and a main growth process S340. Through these processes, the second layer 40 includes, for example, a c-plane enlargement layer 42 and a main growth layer 44.

[0228] (S320: c-surface enlargement process)

[0229] like FIG. 6 As shown in (b), the second growth conditions described above are used to epitaxially grow the c-plane extension layer 42 of the second layer 40 formed from a single crystal of a group III nitride semiconductor on the first layer 30.

[0230] At this point, as it moves upwards from the first layer 30, the c-surface 40c expands while the inclined interface 40i, excluding the c-surface, shrinks.

[0231] Specifically, by growth under the second growth conditions, the c-plane expansion layer 42 grows from the tilted interface 30i of the tilted interface maintaining layer 34 with the tilted interface 40i as a growth surface and in a direction along a direction perpendicular to the c-axis (i.e., a planar direction or a lateral direction). If the c-plane expansion layer 42 is gradually grown in the lateral direction, the c-plane 40c of the c-plane expansion layer 42 starts to be exposed again above the top portion 30t of the tilted interface maintaining layer 34. Thus, the c-plane expansion layer 42 in which the c-plane 40c and the tilted interface 40i other than the c-plane are mixed on the surface is formed.

[0232] Further, if the c-plane expansion layer 42 is gradually grown in the lateral direction, the c-plane 40c is slowly expanded and the tilted interface 40i of the c-plane expansion layer 42 is slowly reduced. Thus, in the surface of the first layer 30, the plurality of recesses 30p composed of the tilted interface 30i are slowly filled.

[0233] Thereafter, if the c-plane expansion layer 42 is further grown, the tilted interface 40i of the c-plane expansion layer 42 completely disappears and in the surface of the first layer 30, the plurality of recesses 30p composed of the tilted interface 30i are completely filled. Thus, the surface of the c-plane expansion layer 42 forms a mirror surface (a flat surface) composed of only the c-plane 40c.

[0234] At this time, during the growth of the first layer 30 and the c-plane expansion layer 42, dislocations are locally gathered, and thus, the dislocation density can be reduced. Specifically, the dislocations that propagate in the first layer 30 while bending toward a direction in which the c-axis is tilted continue to propagate in the same direction in the c-plane expansion layer 42. Thus, in the c-plane expansion layer 42, above the substantially central portion between the pair of top portions 30t, at the meeting portion of the adjoining tilted interfaces 40i, the dislocations are locally gathered. Among the plurality of dislocations gathered at the meeting portion of the adjoining tilted interfaces 40i in the c-plane expansion layer 42, dislocations having opposite Burgers vectors cancel each other out when meeting. In addition, a part of the plurality of dislocations gathered at the meeting portion of the adjoining tilted interfaces 40i forms a closed loop, and propagation in the direction along the c-axis (i.e., the surface side of the c-plane expansion layer 42) is suppressed. Note that the propagation direction of the other part of the plurality of dislocations gathered at the meeting portion of the adjoining tilted interfaces 40i in the c-plane expansion layer 42 is changed from the direction in which the c-axis is tilted to the direction along the c-axis again, and propagates to the surface side of the second layer 40. In this way, by canceling a part of the plurality of dislocations or suppressing the propagation of a part of the plurality of dislocations to the surface side of the c-plane expansion layer 42, the dislocation density in the surface of the second layer 40 can be reduced. In addition, by gathering the dislocations locally, a low dislocation density region can be formed in the second layer 40 above the portion in which the dislocations propagate toward the direction in which the c-axis is tilted.

[0235] Further, at this time, in the c-plane expansion layer 42, the second c-plane growth region 80 to be described later, which grows with the c-plane 40c as a growth plane, expands gradually and is formed as it goes upward in the thickness direction.

[0236] On the other hand, in the c-plane expansion layer 42, the inclined interface growth region 70 gradually shrinks as it goes upward in the thickness direction by gradually reducing the inclined interface 40i, and ends at a prescribed position in the thickness direction. By this growth process of the c-plane expansion layer 42, a valley portion 70a of the inclined interface growth region 70 is formed at a position where the c-plane 40c is generated again in a cross section. Further, in a process in which the recess portion composed of the inclined interface 40i is gradually filled, a hill portion 70b of the inclined interface growth region 70 is formed at a position where the inclined interface 40i disappears in a cross section.

[0237] In the c-plane expansion process S320, the surface of the c-plane expansion layer 42 is composed of only the mirror surface of the c-plane 40c, and thus the height in the thickness direction (the maximum height in the thickness direction) of the c-plane expansion layer 42 becomes, for example, a height from the valley portion 30v to the top portion 30t of the inclined interface maintaining layer 34.

[0238] (S340: Main growth process (c-plane growth process))

[0239] In the c-plane expansion layer 42, the inclined interface 40i disappears and the surface is mirrorized, and then, as shown in (a) of FIG. 4, the main growth layer 44 is formed on the c-plane expansion layer 42 over a prescribed thickness with the c-plane 40c as a growth plane. Thus, the main growth layer 44 is formed in which the surface has only the c-plane 40c and does not have the inclined interface 40i. FIG. 6

[0240] At this time, the growth conditions in the main growth process S340 are maintained as the second growth conditions described above as in the c-plane expansion process S320. Thus, the main growth layer 44 can be grown in a laminar flow with the c-plane 40c as a growth plane.

[0241] Further, at this time, the curvature radius of the c-plane 40c of the main growth layer 44 can be made larger than the curvature radius of the c-plane 10c of the base substrate 10. Thus, the deviation of the off angle of the c-axis with respect to the normal line of the surface in the main growth layer 44 can be made smaller than the deviation of the off angle of the c-axis 10ca with respect to the normal line of the main surface 10s in the base substrate 10.

[0242] Further, at this time, the main growth layer 44 is grown with only the c-plane 40c as a growth plane by not exposing the inclined interface 40i, and thus the entire main growth layer 44 becomes the second c-plane growth region 80 to be described later.

[0243] ​In the main growth process S340, the thickness of the main growth layer 44 is set to, for example, 300 μm or more and 10 mm or less. By setting the thickness of the main growth layer 44 to 300 μm or more, at least one substrate 50 can be cut from the main growth layer 44 in the slicing process S400 described later. On the other hand, by setting the thickness of the main growth layer 44 to 10 mm, when a 700 μm thick substrate 50 is cut from the main growth layer 44 with a final thickness of 650 μm, at least 10 substrates 50 can be obtained even considering a cutting loss of about 200 μm.

[0244] Through the second process S300 described above, a second layer 40 having a c-plane enlargement layer 42 and a main growth layer 44 is formed. As a result, the laminated structure 90 of this embodiment is formed.

[0245] It should be noted that the first process S200 to the second process S300 are performed continuously in the same chamber without exposing the substrate 10 to the atmosphere. As a result, it is possible to suppress the formation of an unexpected high oxygen concentration region (a region with an excessively high oxygen concentration compared to the inclined interface growth region 70) at the interface between the first layer 30 and the second layer 40.

[0246] (S400: Slicing process)

[0247] Next, as FIG. 9 As shown in (b), for example, the main growth layer 44 is sliced ​​using a wire saw along a cutting surface that is substantially parallel to the surface of the main growth layer 44. This forms at least one nitride semiconductor substrate 50 (also called substrate 50) serving as an in-situ slicing substrate. At this time, the thickness of the substrate 50 is set to, for example, 300 μm or more and 700 μm or less.

[0248] At this point, the radius of curvature of the c-plane 50c of the substrate 50 can be made greater than the radius of curvature of the c-plane 10c of the substrate 10. It should be noted that, at this point, the radius of curvature of the c-plane 50c of the substrate 50 can be made greater than the radius of curvature of the c-plane 40c of the main growth layer 44 before slicing. Therefore, the deviation angle θ of the c-axis 50ca of the substrate 50 relative to the normal of the main plane 50s can be made smaller than the deviation angle of the c-axis 10ca of the substrate 10.

[0249] (S500: Grinding process)

[0250] Next, both sides of the substrate 50 are polished using a polishing apparatus. It should be noted that, at this time, the final thickness of the substrate 50 is set to, for example, 250 μm or more and 650 μm or less.

[0251] The substrate 50 described in this embodiment is manufactured through the above processes S100 to S500.

[0252] (Manufacturing process of semiconductor laminate and manufacturing process of semiconductor device)

[0253] After the substrate 50 is manufactured, a semiconductor functional layer formed of a Group III nitride semiconductor is epitaxially grown on the substrate 50, for example, to manufacture a semiconductor laminate. After the semiconductor laminate is manufactured, an electrode or the like is formed using the semiconductor laminate, and the semiconductor laminate is cut to cut out a chip of a prescribed size. Thus, a semiconductor device is manufactured.

[0254] (2) Laminated structure

[0255] Next, the laminated structure 90 according to the present embodiment will be described with reference to (a) of FIG. 1. FIG. 9

[0256] The laminated structure 90 according to the present embodiment has, for example, a base substrate 10, a first layer 30, and a second layer 40.

[0257] The first layer 30 is grown on the main surface 10s of the base substrate 10, for example.

[0258] The first layer 30 has, for example, a plurality of valley portions 30v and a plurality of top portions 30t formed by causing the top surface 30u of a single crystal of a Group III nitride semiconductor to have a plurality of recesses 30p composed of inclined interfaces 30i other than a c-plane and to lose the c-plane 30c. When any cross section perpendicular to the main surface of the base substrate 10 is observed, the average distance between the nearest top portions exceeds 100 μm, for example.

[0259] In addition, the first layer 30 has a first c-plane growth region (first low oxygen concentration region) 60 and an inclined interface growth region (high oxygen concentration region) 70 based on, for example, a difference in growth surface during growth.

[0260] ​The 1st c-plane growth region 60 is a region grown with the c-plane 30c as a growth plane. The 1st c-plane growth region 60 has, for example, a plurality of valleys 60a and a plurality of hills 60b in a cross section. Note that the valleys 60a and the hills 60b referred to here are portions of shapes observed based on differences in luminous intensity when a cross section of the layered structure 90 is observed with a fluorescence microscope or the like, and are not portions of the most surface shapes generated in the middle of growth of the first layer 30. The plurality of valleys 60a are each a turning point that protrudes downward in a cross section in the 1st c-plane growth region 60, and is formed at a position where the tilted interface 30i is generated. At least one of the plurality of valleys 60a is disposed at a position above the main surface 10s of the base substrate 10 by a distance. On the other hand, the plurality of hills 60b are each a turning point that protrudes upward in a cross section in the 1st c-plane growth region 60, and is formed at a position where the c-plane 30c (finally) disappears and ends, sandwiching a pair of tilted interfaces 30i that expand toward opposite directions. The valleys 60a and the hills 60b are alternately formed in a direction along the main surface 10s of the base substrate 10.

[0261] When an arbitrary cross section perpendicular to the main surface 10s of the base substrate 10 is observed, a pair of hills 60b closest to each other among the plurality of hills 60b sandwiching one valley among the plurality of valleys 60a are spaced apart from each other in a direction along the main surface 10s of the base substrate 10 by an average distance equivalent to the above-described average distance L between the closest top portions of the first layer 30, for example, more than 100 μm.

[0262] The 1st c-plane growth region 60 has a pair of tilted portions 60i disposed as trajectories of intersections of the c-plane 30c and the tilted interface 30i on both sides of one hill among the plurality of hills 60b. Note that the tilted portions 60i referred to here are portions of shapes observed based on differences in luminous intensity when a cross section of the layered structure 90 is observed with a fluorescence microscope or the like, and are not the tilted interface 30i of the most surface generated in the middle of growth of the first layer 30.

[0263] In a cross section, an angle β formed by the pair of tilted portions 60i is, for example, 70° or less, and preferably 20° or more and 65° or less. The angle β formed by the pair of tilted portions 60i being 70° or less means that, under the first growth conditions, the growth rate G c1 of the tilted interface 30i most inclined with respect to the c-plane 30c in the first layer 30 is G i / G c1 . ihigh. Thus, the inclined interface 30i other than the c-plane can be easily generated. As a result, the dislocation can be easily bent at a position where the inclined interface 30i is exposed. In addition, by setting the angle β of the pair of inclined portions 60i to 70° or less, the plurality of valley portions 30v and the plurality of top portions 30t can be easily generated above the main surface 10s of the base substrate 10. Further, by setting the angle β of the pair of inclined portions 60i to 65° or less, the inclined interface 30i other than the c-plane can be further easily generated, and the plurality of valley portions 30v and the plurality of top portions 30t can be more easily generated above the main surface 10s of the base substrate 10. Note that, by setting the angle β of the pair of inclined portions 60i to 20° or more, the height from the valley portion 30v to the top portion 30t of the first layer 30 can be suppressed from becoming high, and the thickness until the mirror surface of the second layer 40 can be suppressed from becoming thick.

[0264] On the other hand, the inclined interface growth region 70 is a region grown with the inclined interface 30i other than the c-plane as a growth surface. The lower surface of the inclined interface growth region 70 is formed, for example, along the shape of the 1st c-plane growth region 60. The inclined interface growth region 70 is continuously provided along the main surface of the base substrate 10.

[0265] In the inclined interface growth region 70, oxygen is easily mixed in, compared to the 1st c-plane growth region 60. Thus, the oxygen concentration in the inclined interface growth region 70 is higher than the oxygen concentration in the 1st c-plane growth region 60. Note that, the oxygen mixed into the inclined interface growth region 70 is, for example, oxygen accidentally mixed in within the vapor phase growth apparatus, or oxygen released from a member (quartz member or the like) constituting the vapor phase growth apparatus, or the like.

[0266] Note that, the oxygen concentration in the 1st c-plane growth region 60 is, for example, 5 x 10 16 cm -3 or more, and 3 x 10 16 cm -3 or less. On the other hand, the oxygen concentration in the inclined interface growth region 70 is, for example, 9 x 10 17 cm -3 or more, and 5 x 10 19 cm -3 or less.

[0267] The second layer 40 has, for example, the inclined interface growth region (high oxygen concentration region) 70 and the 2nd c-plane growth region (second low oxygen concentration region) 80 based on a difference in growth surface during growth.

[0268] The upper surface of the inclined interface growth region 70 in the second layer 40 has, for example, a plurality of valleys 70a and a plurality of hills 70b in cross section. Note that the valleys 70a and the hills 70b referred to here each refer to a portion of the shape observed based on a difference in light emission intensity when the cross section of the layered structure 90 is observed with a fluorescence microscope or the like, and do not refer to a portion of the shape of the top surface generated during the growth of the second layer 40. The plurality of valleys 70a of the inclined interface growth region 70 are formed at positions where the c-plane 40c is generated again, as described above, in cross section. In addition, the plurality of valleys 70a of the inclined interface growth region 70 are each formed above the plurality of hills 60b of the first c-plane growth region 60 in cross section. On the other hand, the plurality of hills 70b of the inclined interface growth region 70 are formed at positions where the inclined interface 40i disappears and ends, as described above, in cross section. In addition, the plurality of hills 70b of the inclined interface growth region 70 are each formed above the plurality of valleys 60a of the first c-plane growth region 60 in cross section.

[0269] In addition, the face of the upper end of the inclined interface growth region 70 in the second layer 40 and substantially parallel to the main surface 10s of the base substrate 10 is a boundary face 40b of the position where the inclined interface 40i in the second layer 40 disappears and ends.

[0270] The second c-plane growth region 80 is a region grown with the c-plane 40c as a growth plane. In the second c-plane growth region 80, the incorporation of oxygen is suppressed as compared with the inclined interface growth region 70. Thus, the oxygen concentration in the second c-plane growth region 80 is lower than the oxygen concentration in the inclined interface growth region 70. The oxygen concentration in the second c-plane growth region 80 is, for example, 5 x 10 16 cm -3 preferably 3 x 10 16 cm -3 or less.

[0271] In this embodiment, during the growth of the first layer 30, at a position where the inclined interface 30i other than the c-plane is exposed, dislocations are bent and propagate in a direction substantially perpendicular to the inclined interface 30i, whereby in the second layer 40, a part of the plurality of dislocations disappears or a part of the plurality of dislocations is suppressed from propagating to the surface side of the c-plane spreading layer 42. Thus, the dislocation density in the surface of the second layer 40 is reduced as compared with the dislocation density in the main surface 10s of the base substrate 10.

[0272] In addition, in this embodiment, the dislocation density in the surface of the second layer 40 is sharply reduced in the thickness direction.

[0273] Here, the dislocation density in the main surface 10s of the base substrate 10 is denoted as N0, and the dislocation density in the boundary surface 40b of the position where the inclined interface 40i in the second layer 40 disappears is denoted as N. Note that the average dislocation density in the boundary surface 40b is denoted as N. On the other hand, the dislocation density in the surface of the crystal layer when the crystal layer of the Group III nitride semiconductor is epitaxially grown on the main surface 10s of the base substrate 10 with only the c-plane as the growth surface, with the thickness of the crystal layer being equal to the thickness from the main surface of the base substrate 10 to the boundary surface 40b (hereinafter also referred to as "the case of c-plane limited growth") is denoted as N'.

[0274] In the case of c-plane limited growth, there is a tendency that the dislocation density in the surface of the crystal layer is inversely proportional to the thickness of the crystal layer. Specifically, in the case of c-plane limited growth, when the thickness of the crystal layer is 1.5 mm, the reduction rate of the dislocation density calculated from N' / N0 is about 0.6.

[0275] In contrast to this, in the present embodiment, the reduction rate of the dislocation density calculated from N / N0 is, for example, less than the reduction rate of the dislocation density calculated from N' / N0 in the case of c-plane limited growth.

[0276] Specifically, in the present embodiment, the thickness of the boundary surface 40b of the position where the inclined interface 40i in the second layer 40 disappears from the main surface 10s of the base substrate 10 is, for example, 1.5 mm or less, and preferably 1.2 mm or less. In addition, in the present embodiment, the above-mentioned reduction rate of the dislocation density calculated from N / N0 is, for example, 0.3 or less, and preferably 0.23 or less, and more preferably 0.15 or less.

[0277] Note that, in the present embodiment, the lower limit value of the thickness from the main surface 10s of the base substrate 10 to the boundary surface 40b is better to be thin, and thus is not limited. However, in the first process S200 and the second process S300, if the process from the generation of the inclined interface 30i to the disappearance of the inclined interface 40i is taken into consideration, the thickness from the main surface 10s of the base substrate 10 to the boundary surface 40b is, for example, more than 200 μm.

[0278] In addition, in the present embodiment, the lower limit value of the reduction rate of the dislocation density is better to be small, and thus is not limited. However, if the thickness from the main surface 10s of the base substrate 10 to the boundary surface 40b is 1.5 mm or less, the reduction rate of the dislocation density is, for example, 0.01 or more.

[0279] In addition, in the present embodiment, the surface of the second layer 40 is entirely composed of a +c plane, and the first layer 30 and the second layer 40 each do not include an inversion domain. In this regard, the layered structure 90 of the present embodiment is different from a layered structure formed by a so-called DEEP (Dislocation Elimination by the Epitaxial-growth with inverse-pyramidal Pits) method, that is, a layered structure in which a core located at the center of a pit includes an inversion domain.

[0280] (3) Nitride semiconductor substrate (nitride semiconductor free-standing substrate, nitride crystal substrate)

[0281] Next, the nitride semiconductor substrate 50 according to the present embodiment will be described using FIG. 9 FIG. 10 (a) is a schematic top view of the nitride semiconductor substrate according to the present embodiment, (b) is a schematic cross-sectional view of the nitride semiconductor substrate according to the present embodiment along the m axis, and (c) is a schematic cross-sectional view of the nitride semiconductor substrate according to the present embodiment along the a axis. Note that the direction along the m axis is set as the x direction, and the direction along the a axis is set as the y direction.

[0282] In the present embodiment, the substrate 50 obtained by slicing the second layer 40 using the above-described manufacturing method is, for example, a free-standing substrate formed of a single crystal of a Group III nitride semiconductor. In the present embodiment, the substrate 50 is, for example, a GaN free-standing substrate.

[0283] The diameter of the substrate 50 is, for example, 2 inches or more. In addition, the thickness of the substrate 50 is, for example, 300 μm or more and 1 mm or less.

[0284] The conductivity of the substrate 50 is not particularly limited, and when the substrate 50 is used to manufacture a semiconductor device that is a vertical Schottky barrier diode (SBD), the substrate 50 is, for example, n-type, the n-type impurity in the substrate 50 is, for example, Si or germanium (Ge), and the n-type impurity concentration in the substrate 50 is, for example, 1.0 x 1018cm-3or more and 1.0 x 1020cm-3or less. 18 cm -3 20 cm -3 or less.

[0285] The substrate 50 has, for example, a main surface 50s that is an epitaxial growth surface. In the present embodiment, the lowest-index crystal plane closest to the main surface 50s is, for example, a c plane 50c.

[0286] ​​It should be noted that the main surface 50s of the substrate 50 is mirror-finished, and the root mean square roughness (RMS) of the main surface 50s of the substrate 50 is, for example, less than 1 nm.

[0287] Furthermore, in this embodiment, the impurity concentration in the substrate 50 obtained by the above manufacturing method is lower than that of the substrate obtained by the flux method or the ammonothermal method.

[0288] Specifically, the hydrogen concentration in substrate 50 is, for example, less than 1 × 10⁻⁶. 17 cm -3 Preferably 5×10 16 cm -3 the following.

[0289] Furthermore, in this embodiment, the substrate 50 is formed by slicing the main growth layer 44 grown with the c-plane 40c as the growth surface. Therefore, it does not include the inclined interface growth region 70 grown with the inclined interface 30i or the inclined interface 40i as the growth surface. That is, the substrate 50 is entirely composed of a low oxygen concentration region.

[0290] Specifically, the oxygen concentration in substrate 50 is, for example, 5 × 10⁻⁶. 16 cm -3 The following, preferably 3×10 16 cm -3 the following.

[0291] In addition, in this embodiment, the substrate 50 does not include an inversion domain, for example, as described above.

[0292] (The curvature and deviation angle of surface c)

[0293] like FIG. 10 As shown in (b) and (c), in this embodiment, the c-plane 50c, which is the low index crystal plane closest to the main plane 50s of the substrate 50, is bent into a concave spherical shape relative to the main plane 50s due to the above-described manufacturing method of the substrate 50.

[0294] In this embodiment, the c-surface 50c of the substrate 50 presents an approximately spherical curved surface in cross-sections along the m-axis and along the a-axis, respectively.

[0295] In this embodiment, since the c-plane 50f of the substrate 50 is bent into a concave spherical shape as described above, at least a portion of the c-axis 50ca is tilted relative to the normal of the main surface 50s. The angle between the c-axis 50ca and the normal of the main surface 50s, i.e., the deviation angle θ, has a predetermined distribution within the main surface 50s.

[0296] Note that, among the deviation angle Θ of the c-axis 50ca with respect to the normal line of the main surface 50s, the component in the direction along the m-axis is denoted as "Θ m ", and the component in the direction along the a-axis is denoted as "Θ a ". Note that, Θ 2 = Θ m 2 + Θ a 2 .

[0297] In this embodiment, since the c-plane 50c of the substrate 50 is curved in a concave spherical shape as described above, the deviation angle m-axis component Θ m and the deviation angle a-axis component Θ a can be approximately expressed by a linear function of x and a linear function of y, respectively.

[0298] In this embodiment, the radius of curvature of the c-plane 50c of the substrate 50 is larger than the radius of curvature of the c-plane 10c of the base substrate 10 used in the manufacturing method of the substrate 50 described above, for example.

[0299] Specifically, the radius of curvature of the c-plane 50c of the substrate 50 is, for example, 23 m or more, preferably 30 m or more, and further preferably 40 m or more.

[0300] Note that, as a reference, even in the case of c-plane limited growth, the radius of curvature of the c-plane in a substrate cut from a crystal layer having the same thickness as the total thickness of the first layer 30 and the second layer 40 of this embodiment is sometimes larger than the radius of curvature of the c-plane 10c of the base substrate 10. However, in the case of c-plane limited growth, the radius of curvature of the c-plane in a substrate cut from a crystal layer having a thickness of 2 mm is about 11 m, which is about 1.4 times or so the radius of curvature of the c-plane 10c of the base substrate 10.

[0301] In this embodiment, the larger the upper limit value of the radius of curvature of the c-plane 50c of the substrate 50 is, the better, and thus is not particularly limited. In the case where the c-plane 50c of the substrate 50 is substantially flat, the radius of curvature of the c-plane 50c can be considered to be infinite.

[0302] In addition, in this embodiment, by making the radius of curvature of the c-plane 50c of the substrate 50 large, it is possible to make the deviation of the deviation angle Θ of the c-axis 50ca with respect to the normal line of the main surface 50s smaller than the deviation of the deviation angle of the c-axis 10ca of the base substrate 10.

[0303] Specifically, when the deviation of the angle of the c-axis 50ca from the normal line of the main surface 50s is calculated from the maximum-minimum difference of the magnitude of the angle of the deviation θ within 29.6 mm from the center of the main surface 50s, the deviation is, for example, 0.075° or less, preferably 0.057° or less, and further preferably 0.043° or less, based on the angle of the diffraction peak of the (0002) plane.

[0304] Note that, as a reference, the deviation of the angle of the c-axis 10ca calculated by the above-described measurement method is about 0.22° in the base substrate 10 produced by the above-described VAS method. In addition, in the case of c-plane limited growth, when the thickness of the crystal layer is set to the same thickness (for example, 2 mm) as the total thickness of the first layer 30 and the second layer 40 of the present embodiment, the deviation of the angle of the c-axis calculated by the above-described measurement method is about 0.15° in the nitride semiconductor substrate obtained from the crystal layer.

[0305] In the present embodiment, the lower the lower limit value of the deviation of the angle θ of the c-axis 50ca of the substrate 50 is, the better, and thus is not particularly limited. In the case where the c-plane 50c of the substrate 50 is substantially flat, it is considered that the deviation of the angle θ of the c-axis 50ca of the substrate 50 is 0°.

[0306] In addition, in the present embodiment, the curvature of the c-plane 50c is isotropically reduced with respect to the main surface 50s of the substrate 50, and thus the direction dependence of the radius of curvature of the c-plane 50c is small.

[0307] Specifically, the difference between the radius of curvature of the c-plane 50c in the direction along the a-axis and the radius of curvature of the c-plane 50c in the direction along the m-axis calculated by the above-described measurement method is, for example, 50% or less, preferably 20% or less, of the larger one of them.

[0308] (Dark spots)

[0309] Next, the dark spots in the main surface 50s of the substrate 50 of the present embodiment will be described. Note that the "dark spots" referred to here mean points having low light emission intensity observed in an observation image of the main surface 50s in a multiphoton excitation microscope, a cathodoluminescence image of the main surface 50s, or the like, and include not only dislocations but also non-luminescent centers caused by foreign matter or point defects. Note that the "multiphoton excitation microscope" is sometimes referred to as a two-photon excitation fluorescence microscope.

[0310] In the present embodiment, the substrate 50 is manufactured using the base substrate 10 formed of a GaN single crystal with high purity, which is produced by the VAS method, and thus, in the substrate 50, non-radiative centers caused by foreign substances or point defects are less. Therefore, more than 95%, preferably more than 99% of dark spots when the main surface of the substrate 50 is observed by a multiphoton excitation microscope or the like are not non-radiative centers caused by foreign substances or point defects, but dislocations.

[0311] In addition, in the present embodiment, by the manufacturing method described above, the dislocation density in the surface of the second layer 40 is reduced compared to the dislocation density in the main surface 10s of the base substrate 10. Thus, in the main surface 50s of the substrate 50 formed by slicing the second layer 40, dislocations are also reduced.

[0312] In addition, in the present embodiment, by the manufacturing method described above, the first process S200 and the second process S300 are performed using the base substrate 10 in a state where no processing has been performed, and thus, in the main surface 50s of the substrate 50 formed by slicing the second layer 40, a region with a high dislocation density caused by concentration of dislocations is not formed, but a region with a low dislocation density is uniformly formed.

[0313] Specifically, in the present embodiment, when the main surface 50s of the substrate 50 is observed with a field of view of 250 μm square by a multiphoton excitation microscope, and the dislocation density is calculated from the dark spot density, there is no region where the dislocation density exceeds 3 x 10 6 cm -2 , and a region where the dislocation density is less than 1 x 10 6 cm -2 is present in more than 80%, preferably more than 90%, and more preferably more than 95% of the main surface 50s.

[0314] Note that, when the manufacturing method of the present embodiment is used, it is desirable that the upper limit value of the proportion of the region where the dislocation density is less than 1 x 10 6 cm -2 be close to 100%, and for example, it sometimes becomes 99% of the main surface 50s.

[0315] In addition, in the present embodiment, the dislocation density obtained by averaging the entire main surface 50s of the substrate 50 is, for example, less than 1 x 10 6 cm -2 , preferably less than 5.5 x 10 5 cm -2 , and more preferably 3 x 10 5 cm -2 or less.

[0316] In addition, the main surface 50s of the substrate 50 of the present embodiment contains dislocation-free regions of 50 μm square, for example, based on the closest top inter-averaged distance L in the first process S200 described above. In addition, the dislocation-free regions of 50 μm square are dispersed throughout the main surface 50s of the substrate 50, for example. In addition, the main surface 50s of the substrate 50 of the present embodiment contains, for example, 100 dislocation-free regions of 50 μm square per 1 cm2. 2 The above, preferably 800 per 1 cm2 2 The above, more preferably 1600 per 1 cm2 2 The above density has dislocation-free regions of 50 μm square that do not overlap. Note that the density of dislocation-free regions of 50 μm square that do not overlap is 1600 per 1 cm2 2 The above case corresponds to a case in which the main surface 50s has at least one dislocation-free region of 50 μm square within any field of view of 250 μm square, for example.

[0317] Note that the upper limit of the density of dislocation-free regions of 50 μm square that do not overlap is 40000 per 1 cm2 2 .

[0318] Note that, as a reference, in a substrate obtained by a conventional manufacturing method that does not perform a special process for gathering dislocations, the size of a dislocation-free region is smaller than 50 μm square, or the density of dislocation-free regions of 50 μm square is less than 100 per 1 cm2 2 .

[0319] Next, the Burgers vector of dislocations in the substrate 50 of the present embodiment will be described.

[0320] In the present embodiment, the dislocation density in the main surface 10s of the base substrate 10 used in the manufacturing method described above is low, and thus the occurrence of combination (mixing) of a plurality of dislocations when the first layer 30 and the second layer 40 are grown on the base substrate 10 is less. Thus, in the substrate 50 obtained from the second layer 40, the generation of dislocations having a large Burgers vector can be suppressed.

[0321] Specifically, in the substrate 50 of the present embodiment, for example, dislocations having any of a Burgers vector of <11-20> / 3, <0001>, or <11-23> / 3 are more. Note that the "Burgers vector" here can be measured by, for example, a large angle convergent beam electron diffraction method (LACBED method) using a transmission electron microscope (TEM). In addition, a dislocation having a Burgers vector of <11-20> / 3 is an edge-type dislocation, a dislocation having a Burgers vector of <0001> is a screw dislocation, and a dislocation having a Burgers vector of <11-23> / 3 is a mixed dislocation that is a mixture of an edge-type dislocation and a screw dislocation.

[0322] In this embodiment, when 100 dislocations are randomly selected from the main surface 50s of the substrate 50, the Burgers vector is <11-20> / 3. <0001> The proportion of the number of dislocations of either <11-23> / 3 is, for example, 50% or more, preferably 70% or more, and more preferably 90% or more. It should be noted that at least a portion of the main surface 50s of the substrate 50 may also contain dislocations with Burgers vectors of 2<11-20> / 3 or <11-20>, etc.

[0323] (Determination of X-ray rocking curves to change slit width)

[0324] Here, the inventors discovered that by changing the slit width on the incident side to perform X-ray rocking curve measurement, it is possible to simultaneously evaluate both the crystal quality factors of the substrate 50 constituting this embodiment and the bending (warping) of the c-surface 50c.

[0325] First, the influence of crystal quality factors on X-ray rocking curve determination is explained.

[0326] The half-width at half-maximum (WWHM) of the diffraction pattern in X-ray rocking curve measurements is significantly affected by crystal quality factors such as dislocation density, mosaicity, stacking defect density, basal dislocation density, point defect (void, etc.) density, in-plane fluctuation of lattice constant, and impurity concentration distribution. When these crystal quality factors are poor, the fluctuation of the diffraction angle in X-ray rocking curve measurements increases, and the WWHM of the diffraction pattern becomes larger.

[0327] Next, use FIG. 10 (a) explains the effect of the curvature of the c-plane 50c in X-ray rocking curve measurement. FIG. 10 (a) is a schematic cross-sectional view showing X-ray diffraction over a curved c-plane.

[0328] Let a denote the slit width on the incident side of the X-rays, let b denote the irradiation width (imprint) of the X-rays irradiating the main surface of the substrate, and let θ denote the Bragg angle of the crystal. B When the X-ray irradiation width b in the main surface of the substrate is determined using the following formula (h), the irradiation width b is calculated.

[0329] b = a / sinθ B ···(h)

[0330] like FIG. 10As shown in (a), when the c-plane of the substrate is bent, the radius of curvature of the c-plane is denoted as R, and half of the central angle formed by the c-plane bending within the X-ray irradiation width b is denoted as γ. In this case, the radius of curvature R of the c-plane is very large relative to the X-ray irradiation width b. Therefore, the angle γ is obtained using the following equation (i).

[0331] γ=sin -1 (b / 2R)≈b / 2R···(i)

[0332] At this point, the diffraction angle θ between the incident end (right end in the figure) of the region irradiated with X-rays on plane c of the substrate and the main surface of the substrate. B +γ=θ B +b / 2R.

[0333] On the other hand, the end of the X-ray-irradiated region on the c-plane of the substrate (the left end in the figure) has a diffraction angle of θ relative to the main surface of the substrate. B -γ=θ B -b / 2R.

[0334] Therefore, based on the difference between the diffraction angle of the incident side end of the c-plane of the substrate relative to the main surface of the substrate and the diffraction angle of the light-receiving side end of the c-plane of the substrate relative to the main surface of the substrate, the fluctuation of the X-ray diffraction angle for the curved c-plane is b / R.

[0335] FIG. 10 Figures (b) and (c) are diagrams representing the fluctuations in the diffraction angle of the (0002) plane relative to the radius of curvature of the c-plane. It should be noted that... Embodiment The vertical axis of (b) becomes a logarithmic scale. Comparative example The vertical axis of (c) becomes a linear scale.

[0336] like Base substrate As shown in (b) and (c), when the width a of the slit on the incident side of the X-ray is increased, i.e., the X-ray irradiation width b is increased, the fluctuation of the diffraction angle of the (0002) plane increases accordingly to the X-ray irradiation width b. Furthermore, as the radius of curvature R of the c plane decreases, the fluctuation of the diffraction angle of the (0002) plane gradually increases. Additionally, the difference in the fluctuation of the diffraction angle of the (0002) plane when the X-ray irradiation width b is changed increases as the radius of curvature R of the c plane decreases.

[0337] When the width a of the slit on the incident side is narrow, the effect of the curvature of the c-plane on the fluctuation of the diffraction angle of the (0002) plane is small, and the effect of the crystal quality factors described above becomes a dominant factor. However, when the width a of the slit on the incident side is wide, the effects of both the crystal quality factors and the curvature of the c-plane overlap with respect to the fluctuation of the diffraction angle of the (0002) plane. Therefore, if the width a of the slit on the incident side is changed to perform X-ray rocking curve measurement, both the crystal quality factors and the curvature (warpage) of the c-plane can be evaluated simultaneously over the entire region where X-rays are irradiated.

[0338] Here, the characteristics at the time of performing X-ray rocking curve measurement on the substrate 50 according to the present embodiment are described.

[0339] Hereinafter, when performing X-ray rocking curve measurement of the (0002) plane diffraction by irradiating the main surface 50s of the substrate 50 with X-rays of Cu Kα1 through the double monochromator of the Ge (220) plane and the slit, the half-value width of the (0002) plane diffraction when the width of the slit in the ω direction is 1 mm is denoted as "FWHMa", and the half-value width of the (0002) plane diffraction when the width of the slit in the ω direction is 0.1 mm is denoted as "FWHMb". Note that the "ω direction" refers to the direction of rotation when the substrate 50 is rotated with the central axis passing through the center of the substrate 50 and parallel to the main surface of the substrate 50 as the central axis in X-ray rocking curve measurement.

[0340] In the substrate 50 according to the present embodiment, all of the crystal quality factors, such as the level of dislocation density, the level of mosaicity, the size of stacking fault density, the size of basal plane dislocation density, the size of point defect (void, etc.) density, the size of in-plane deviation amount of lattice constant, the distribution of impurity concentration, and the like, are all good.

[0341] As a result, in the substrate 50 according to the present embodiment, the half-value width FWHMb of the (0002) plane diffraction when performing X-ray rocking curve measurement of the (0002) plane diffraction with the width of the slit in the ω direction set to 0.1 mm is, for example, 80 arcsec or less, preferably 50 arcsec or less, and more preferably 32 arcsec or less.

[0342] In addition, in the substrate 50 according to the present embodiment, as described above, all of the crystal quality factors are good over a wide range of the main surface 50s.

[0343] As a result, when the X-ray rocking curve measurement of the (0002) plane diffraction is performed with the slit width in the ω direction set to 0.1 mm at the plurality of measurement points set at intervals of 5 mm within the main surface 50s (between the center and the outer edge) of the substrate 50 of the present embodiment, the half width FWHMbof the (0002) plane diffraction is, for example, 80 arcsec or less, preferably 50 arcsec or less, and more preferably 32 arcsec or less at 90% or more of all the measurement points.

[0344] In addition, in the substrate 50 of the present embodiment, the in-plane deviation of the above-described crystal quality factor is small. Therefore, there is a tendency that the diffraction pattern of the (0002) plane when the X-ray rocking curve measurement is performed with the slit width on the incident side widened is difficult to narrow compared to the diffraction pattern of the (0002) plane when the X-ray rocking curve measurement is performed with the slit width on the incident side narrowed.

[0345] As a result, in the substrate 50 of the present embodiment, the half width FWHMaof the (0002) plane diffraction when the slit width in the ω direction is set to 1 mm can be, for example, the half width FWHMbof the (0002) plane diffraction when the slit width in the ω direction is set to 0.1 mm or more.

[0346] Note that even in the case where the crystal quality factor of the substrate 50 is good, in a state where FWHMbis very small, FWHMa< FWHMmay sometimes occur.

[0347] In addition, in the substrate 50 of the present embodiment, as described above, the dislocation is small and the above-described crystal quality factor is all well balanced over a wide range of the main surface 50s. Furthermore, the curvature radius of the c-plane 50c of the substrate 50 is large. Thus, in the substrate 50 of the present embodiment, even when the X-ray rocking curve measurement is performed with the slit width on the incident side widened, the above-described crystal quality factor is well balanced over the entire region where X-rays are irradiated, and by making the curvature radius of the c-plane large, the fluctuation of the diffraction angle of the (0002) plane does not become too large. Therefore, even when the X-ray rocking curve measurement is performed with the slit width on the incident side changed, the difference in the fluctuation of the diffraction angle of the (0002) plane becomes small.

[0348] As a result, at the prescribed measurement point (for example, the center of the main surface) of the substrate 50 of the present embodiment, the difference FWHMa-FWHMbobtained by subtracting the half width FWHMbof the (0002) plane diffraction when the slit width in the ω direction is set to 0.1 mm from the half width FWHMaof the (0002) plane diffraction when the slit width in the ω direction is set to 1 mm is, for example, 30% or less, and preferably 22% or less of FWHMa.

[0349] Note that in the substrate 50 of the present embodiment, even if FWHMa< FWHMb, | FWHMa- FWHMb| / FWHMabecomes 30% or less. Also, in the substrate 50 of the present embodiment, FWHMaand FWHMbare approximately equal, and sometimes | FWHMa- FWHMb| / FWHMa becomes 0%.

[0350] Also, in the substrate 50 of the present embodiment, even if the slit width on the incident side is widened to perform X-ray rocking curve measurement, the deviation of the above-described crystal quality elements is small in the entire region irradiated with X-rays, and thus the diffraction pattern has a single peak.

[0351] Note that, as a reference, a substrate produced by a conventional production method (hereinafter also referred to as a conventional substrate) is described. The conventional production method referred to here means, for example, a conventional VAS method, a method of performing thick film growth with a c-plane as a growth surface, the above-described DEEP method, a THVPE (Tri-halide vapor phase epitaxy) method, an ammonothermal method, a flux method, and the like.

[0352] For the conventional substrate, at least any one of the above-described crystal quality elements is not as good as the substrate 50 of the present embodiment. Therefore, FWHMb of the conventional substrate becomes larger than FWHMb of the substrate 50 of the present embodiment.

[0353] For the conventional substrate, at least any one of the above-described crystal quality elements can have an in-plane deviation. Therefore, the diffraction pattern of the (0002) plane when the slit width on the incident side is widened to perform X-ray rocking curve measurement is sometimes wider than the diffraction pattern of the (0002) plane when the slit width on the incident side is narrowed to perform X-ray rocking curve measurement. As a result, for the conventional substrate, sometimes FWHMa< FWHMb.

[0354] For the conventional substrate, the radius of curvature of the c-plane is smaller than the substrate 50 of the present embodiment. When the slit width is widened, in at least a part of the region irradiated with X-rays, a site where at least any one of the crystal quality elements is not as good as the substrate 50 of the present embodiment is inevitably included. Therefore, the difference FWHMa- FWHMb in the base substrate 10 becomes larger than the difference in the substrate 50 of the present embodiment.

[0355] For the conventional substrate, at least any one of the above-described crystal quality elements can have an in-plane deviation. When the slit width is widened, in at least a part of the region irradiated with X-rays, a site where the fluctuation of the diffraction angle is different is likely to be generated. Therefore, the diffraction pattern when the slit width is widened sometimes has multiple peaks.

[0356] As above, the existing substrate can not satisfy the above conditions prescribed for the substrate 50 of the present embodiment.

[0357] (4) Effects obtained by the present embodiment

[0358] According to the present embodiment, one or more of the effects shown below are obtained.

[0359] (a) In the first process S200, by causing the surface of the single crystal constituting the first layer 30 to have a tilted interface 30i other than the c-plane, at a position where the tilted interface 30i is exposed, dislocations can be caused to propagate while bending in a direction substantially perpendicular to the tilted interface 30i. Thus, dislocations can be caused to locally converge. By causing dislocations to locally converge, dislocations having opposite Burgers vectors to each other can be caused to cancel each other out. Alternatively, by causing locally converged dislocations to form a closed loop, the propagation of dislocations to the surface side of the second layer 40 can be suppressed. By so doing, the dislocation density in the surface of the second layer 40 can be reduced. As a result, a substrate 50 having a reduced dislocation density compared to the base substrate 10 can be obtained.

[0360] (b) As described above, during the growth of the second layer 40, by causing a portion of the plurality of dislocations to cancel out or suppressing a portion of the plurality of dislocations from propagating to the surface side of the second layer 40, the dislocation density can be reduced extremely quickly compared to the case of c-plane restricted growth. That is, the reduction rate of the dislocation density in the present embodiment, calculated by N / N0, can be made smaller than the reduction rate of the dislocation density in the case of c-plane restricted growth, calculated by N' / N0. As a result, a substrate 50 having a reduced dislocation density compared to the base substrate 10 can be efficiently obtained, and the productivity thereof can be improved.

[0361] (c) In the first process S200, the c-plane 30c is caused to cancel out from the top surface 30u of the first layer 30. Thus, a plurality of valleys 30v and a plurality of peaks 30t can be formed in the surface of the first layer 30. As a result, dislocations propagating from the base substrate 10 can be reliably caused to bend at the position where the tilted interface 30i is exposed in the first layer 30.

[0362] Here, in the first process, the case where the c-plane remains is considered. In this case, in the portion where the c-plane remains, dislocations propagating from the base substrate propagate without bending in a substantially vertical upward direction and reach the surface of the second layer. Therefore, above the portion where the c-plane remains, dislocations do not decrease but form a high dislocation density region.

[0363] On the other hand, according to the present embodiment, in the first process S200, by causing the c-plane 30c to disappear from the top surface 30u of the first layer 30, it is possible to form the surface of the first layer 30 only with the inclined interface 30i other than the c-plane, and to form a plurality of valley portions 30v and a plurality of top portions 30t on the surface of the first layer 30. Thus, it is possible to reliably bend the dislocations propagated from the base substrate 10 over the entire surface of the first layer 30. By reliably bending the dislocations, it is easy to cause a part of the plurality of dislocations to disappear, or difficult to propagate a part of the plurality of dislocations to the surface side of the second layer 40. As a result, it is possible to reduce the dislocation density over the entire main surface Is of the substrate 50 obtained by the second layer 40.

[0364] (d) In the present embodiment, by causing the RMS of the main surface 10s of the base substrate 10 to be 1 nm or more, it is possible to promote the occurrence of the inclined interface 30i other than the c-plane in the surface of the first layer 30 when the first layer 30 is grown on the base substrate 10 in the first process S200.

[0365] In addition, in the present embodiment, the crystal strain introduced by the processing of the base substrate 10 is left on the main surface 10s side of the base substrate 10. At this time, the half value width (FWHM) of the (10-10) plane diffraction when the X-ray rocking curve measurement is performed with the incident angle with respect to the main surface 10s of the processed base substrate 10 being set to 2° is larger than the half value width of the base substrate 10 before processing, and is set to 60 arcsec or more. Thus, due to the crystal strain on the main surface 10s side of the base substrate 10, it is possible to change the stable crystal plane that appears on the surface of the first layer 30. As a result, it is possible to cause the inclined interface 30i other than the c-plane in the surface of the first layer 30.

[0366] (e) In the present embodiment, by using the above-described base substrate 10, and adjusting the first growth conditions in a manner satisfying formula (1), it is possible to generate the {11-2m} plane with m ≥ 3 as the inclined interface 30i in the first process S200. Thus, it is possible to relax the inclination angle of the {11-2m} plane with respect to the c-plane 30c. Specifically, it is possible to cause the inclination angle to be 47.3° or less. By relaxing the inclination angle of the {11-2m} plane with respect to the c-plane 30c, it is possible to extend the period of the plurality of top portions 30t. Specifically, when observing an arbitrary cross section perpendicular to the main surface 10s of the base substrate 10, it is possible to cause the average distance L between the closest top portions to exceed 100 μm.

[0367] Note that, as a reference, generally, when an etch pit is generated in a nitride semiconductor substrate using a prescribed etchant, an etch pit composed of {1-10n} planes is formed on the surface of the substrate. In contrast, in the present embodiment, the {11-2m} plane with m ≥ 3 can be generated on the surface of the first layer 30 grown under prescribed conditions. It is thus considered that, in the present embodiment, the inclined interface 30i unique to the production method is formed, as compared with the general etch pit.

[0368] (f) In the present embodiment, when observing an arbitrary cross section perpendicular to the main surface 10s of the base substrate 10, by making the average distance L between the nearest top portions exceed 100 μm, it is possible to ensure that the distance of the dislocation bend propagation exceeds at least 50 μm. Thereby, it is possible to sufficiently concentrate the dislocations above the substantially central portion between the pair of top portions 30t in the first layer 30. As a result, it is possible to sufficiently reduce the dislocation density in the surface of the second layer 40.

[0369] (g) In the first step S200, after the c-plane 30c is made to disappear from the surface of the first layer 30, the state in which the surface is composed only of the inclined interface 30i is maintained, and the growth of the first layer 30 is continued over a prescribed thickness. Thereby, it is possible to surely make the c-plane 30c disappear over the entire surface of the first layer 30. For example, even if the c-plane 30c is shifted at the time of disappearance from the surface of the first layer 30 in the inclined interface expansion step S220, and the c-plane 30c remains in a portion of the inclined interface expansion layer 32, it is possible to surely make the c-plane 30c disappear.

[0370] In addition, after the c-plane 30c disappears, by continuing the growth based on the inclined interface 30i of the first layer 30, it is possible to sufficiently ensure the time for making the dislocations bend at the position where the inclined interface 30i is exposed. Here, if the c-plane is grown immediately after the c-plane disappears, the dislocations can not sufficiently bend, but can propagate in the substantially vertical direction toward the surface of the second layer. In contrast, in the present embodiment, by sufficiently ensuring the time for making the dislocations bend at the position where the inclined interface 30i other than the c-plane is exposed, particularly, it is possible to surely bend the dislocations in the vicinity of the top portion 30t of the first layer 30, and it is possible to suppress the dislocations from the base substrate 10 from propagating in the substantially vertical direction toward the surface of the second layer 40. Thereby, it is possible to suppress the concentration of the dislocations above the top portion 30t of the first layer 30.

[0371] (h) By the production method of the present embodiment, it is possible to make the radius of curvature of the c-plane 50c of the substrate 50 larger than the radius of curvature of the c-plane 10c of the base substrate 10. Thereby, it is possible to make the deviation of the deviation angle Θ of the c-axis 50ca of the substrate 50 with respect to the normal line of the main surface 50s smaller than the deviation of the deviation angle of the c-axis 10ca of the base substrate 10.

[0372] As one reason why the radius of curvature of the c-plane 50c of the substrate 50 can be increased, the following reason can be considered, for example.

[0373] As described above, in the first process S200, the first layer 30 is caused to grow three-dimensionally with the inclined interface 30i other than the c-plane as a growth surface, thereby forming the inclined interface growth region 70. In the inclined interface growth region 70, oxygen is easily mixed compared to the first c-plane growth region 60. Therefore, the oxygen concentration in the inclined interface growth region 70 is higher than the oxygen concentration in the first c-plane growth region 60. In other words, the inclined interface growth region 70 can be regarded as a high oxygen concentration region.

[0374] As such, by mixing oxygen into the high oxygen concentration region, the lattice constant of the high oxygen concentration region can be made larger than the lattice constant of other regions other than the high oxygen concentration region (see: Chris G. Van de Walle, Physical Review B vol. 68, 165209 (2003)). With respect to the first c-plane growth region 60 in the base substrate 10 or the first layer 30 that is grown with the c-plane 30c as a growth surface, stress is applied toward the center of curvature of the c-plane by the bending of the c-plane 10c of the base substrate 10. In contrast, by relatively increasing the lattice constant of the high oxygen concentration region, stress that expands the c-plane 30c of the high oxygen concentration region to the outside in the in-plane direction can be generated in the high oxygen concentration region. Thus, the stress that concentrates toward the center of curvature of the c-plane 30c on the lower side of the high oxygen concentration region and the stress that expands the c-plane 30c of the high oxygen concentration region to the outside in the in-plane direction can be canceled out.

[0375] As such, by obtaining the stress canceling effect based on the first layer 30, the radius of curvature of the c-plane 50c of the substrate 50 obtained by the second layer 40 can be made larger than the radius of curvature of the c-plane 10c of the base substrate 10 obtained by the conventional VAS method.

[0376] (i) In the substrate 50 obtained by the manufacturing method of the present embodiment, not only can the dislocation density be reduced, but also the misalignment angle deviation can be reduced, and all of the above-described crystal quality elements that determine the half-value width of the X-ray rocking curve measurement can be balanced and good. Thus, in the substrate 50 of the present embodiment, FWHMb can be 32 arcsec or less. Furthermore, in the substrate 50 of the present embodiment, even when the slit width is set to 1 mm, the radius of curvature of the c-plane is large in the entire region where X-rays are irradiated, and the above-described crystal quality elements are balanced and good, and thus (FWHMa-FWHMb) / FWHMa can be 30% or less.

[0377] <Other Embodiments>

[0378] The above describes embodiments of the present application. However, the present application is not limited to the above-described embodiments, and various changes can be made without departing from the gist thereof.

[0379] In the above-described embodiments, the case where the base substrate 10 is a GaN free-standing substrate has been described, but the base substrate 10 is not limited to a GaN free-standing substrate, and can be, for example, a free-standing substrate formed of a Group III nitride semiconductor such as AlN, AlGaN, InN, InGaN, AlInGaN, or the like, i.e., Al x In y Ga 1-x-y N (0≤x≤1, 0≤y≤1, 0≤x+y≤1) free-standing substrate.

[0380] In the above-described embodiments, the case where the substrate 50 is a GaN free-standing substrate has been described, but the substrate 50 is not limited to a GaN free-standing substrate, and can be, for example, a free-standing substrate formed of a Group III nitride semiconductor such as AlN, AlGaN, InN, InGaN, AlInGaN, or the like, i.e., Al x In y Ga 1-x-y N (0≤x≤1, 0≤y≤1, 0≤x+y≤1) free-standing substrate.

[0381] In the above-described embodiments, the case where the substrate 50 is n-type has been described, but the substrate 50 can be p-type or have semi-insulating properties. For example, when the substrate 50 is used to manufacture a semiconductor device that is a high electron mobility transistor (HEMT), the substrate 50 preferably has semi-insulating properties.

[0382] In the above-described embodiments, the case where the growth temperature is mainly adjusted as the first growth condition in the first process S200 has been described, but as long as the first growth condition satisfies the formula (1), the growth condition other than the growth temperature or the growth condition other than the growth temperature in combination with the growth temperature can be adjusted as the first growth condition.

[0383] In the above-described embodiments, the case where the growth temperature is mainly adjusted as the second growth condition in the second process S300 has been described, but as long as the second growth condition satisfies the formula (2), the growth condition other than the growth temperature or the growth condition other than the growth temperature in combination with the growth temperature can be adjusted as the second growth condition.

[0384] In the above embodiment, the case where the growth conditions in the inclined interface maintaining step S240 are maintained as the first growth conditions as in the inclined interface enlarging step S220 is described, but as long as the growth conditions in the inclined interface maintaining step S240 satisfy the first growth conditions, the growth conditions in the inclined interface maintaining step S240 can be made different from the growth conditions in the inclined interface enlarging step S220.

[0385] In the above embodiment, the case where the growth conditions in the main growth step S340 are maintained as the second growth conditions as in the c-plane enlarging step S320 is described, but as long as the growth conditions in the main growth step S340 satisfy the second growth conditions, the growth conditions in the main growth step S340 can be made different from the growth conditions in the c-plane enlarging step S320.

[0386] In the above embodiment, the case where the second crystal layer 6 or the main growth layer 44 is sliced using a wire saw in the slicing step S170 and the slicing step S400 is described, but a peripheral blade slicer, an inner peripheral blade slicer, an electric discharge machine, or the like can be used.

[0387] In the above embodiment, the case where the substrate 50 is obtained by slicing the main growth layer 44 in the stacked structure 90 is described, but is not limited to this case. For example, the stacked structure 90 can be directly used to manufacture a semiconductor stack for manufacturing a semiconductor device. Specifically, after the stacked structure 90 is manufactured, in a semiconductor stack manufacturing step, a semiconductor functional layer is epitaxially grown on the stacked structure 90, thereby manufacturing a semiconductor stack. After the semiconductor stack is manufactured, the back surface side of the stacked structure 90 is polished, and the base substrate 10, the first layer 30, and the c-plane enlarging layer 42 in the stacked structure 90 are removed. Thus, as in the above embodiment, a semiconductor stack having the main growth layer 44 and the semiconductor functional layer is obtained. According to this case, the slicing step S400 and the polishing step S500 for obtaining the substrate 50 can be omitted.

[0388] In the above-described embodiments, the case where the manufacturing process ends after the substrate 50 is manufactured is described, but the substrate 50 can be used as the base substrate 10, and the processes S200 to S500 can be performed again. Thus, a substrate 50 in which the dislocation density is further reduced can be obtained. In addition, a substrate 50 in which the deviation of the misorientation angle Θ of the c-axis 50ca is further reduced can be obtained. In addition, the processes S200 to S500 using the substrate 50 as the base substrate 10 can be regarded as one cycle, and the cycle can be repeated multiple times. Thus, the dislocation density of the substrate 50 can be gradually reduced in accordance with the number of times the cycle is repeated. In addition, the deviation of the misorientation angle Θ of the c-axis 50ca in the substrate 50 can also be gradually reduced in accordance with the number of times the cycle is repeated.

[0389] Embodiments

[0390] Hereinafter, various experimental results for verifying the effects of the present application will be described. Note that hereinafter, a "nitride semiconductor substrate" will be sometimes abbreviated as a "substrate".

[0391] (1) Experiment 1

[0392] (1-1) Production of Nitride Semiconductor Substrate

[0393] The substrates of the embodiments and the comparative examples were produced by the following operations. Note that for the embodiments, a laminated structure before the substrate is sliced was also produced.

[0394] [Production Conditions of Nitride Semiconductor Substrate of Embodiments]

[0395] (Base Substrate)

[0396] Material: GaN

[0397] Production Method: VAS method

[0398] Diameter: 2 inches

[0399] Thickness: 400 μm

[0400] Low-index crystal plane closest to the main surface: c-plane

[0401] The main surface was not subjected to pattern processing such as a mask layer.

[0402] Root mean square roughness RMS of the main surface: 2 nm

[0403] Misorientation angle of the main surface: 0.4° in the m direction

[0404] FWHM of (10-10) plane diffraction in XRC measurement: 100 arcsec

[0405] (First Layer)

[0406] Material: GaN

[0407] Growth method: HVPE method

[0408] First growth condition:

[0409] The growth temperature is set to 980°C or higher and 1020°C or lower, and the V / III ratio is set to 2 or higher and 20 or lower. At this time, at least one of the growth temperature and the V / III ratio is adjusted within the above range so that the first growth condition satisfies Equation (1).

[0410] (Second layer)

[0411] Material: GaN

[0412] Growth method: HVPE method

[0413] Growth temperature: 1050℃

[0414] V / III ratio: 2

[0415] It should be noted that the second growth condition mentioned above satisfies equation (2).

[0416] Thickness from the main surface of the substrate to the surface of the second layer: 2mm

[0417] (Slicing conditions)

[0418] Substrate thickness: 400μm

[0419] Cutting loss: 200μm

[0420] It should be noted that, in this embodiment, two substrates were manufactured in slightly different states.

[0421] [Fabrication conditions of nitride semiconductor substrates in comparative examples]

[0422] (Substrate)

[0423] Material: GaN

[0424] Production method: VAS method

[0425] Diameter: 2 inches

[0426] Thickness: 400μm

[0427] The nearest low-index crystal plane relative to the principal plane: the c-plane

[0428] The main surface was not patterned using a mask layer or other similar process.

[0429] The root mean square roughness (RMS) of the principal surface is 0.7 nm.

[0430] Misalignment angle of main surface: 0.4° in m direction

[0431] FWHM of (10-10) plane diffraction in XRC measurement: 50 arcsec

[0432] (Crystalline layer)

[0433] Material: GaN

[0434] Growth method: HVPE method

[0435] Growth temperature: 1050°C (same as the second layer of the example)

[0436] V / III ratio: 2 (same as the second layer of the example)

[0437] Note that the above growth conditions satisfy formula (2).

[0438] Thickness from the main surface of the base substrate to the surface of the crystalline layer: 2 mm

[0439] (Slicing conditions)

[0440] The same as the example.

[0441] (1-2) Evaluation

[0442] (Observation based on fluorescence microscope)

[0443] Using a fluorescence microscope, the cross section of the layered structure before slicing the substrate of the example was observed.

[0444] (Observation based on multiphoton excitation microscope)

[0445] Using a multiphoton excitation microscope, the main surface of each of the base substrate, the substrate of the example, and the substrate of the comparative example was observed. At this time, the dislocation density was measured by measuring the dark spot density over the entire main surface with a 250 μm field of view. Note that the dark spots in these substrates were confirmed to be dislocations by measuring with the focal point shifted in the thickness direction. Also at this time, the proportion of the number of regions (low dislocation density regions) in which the dislocation density under a 250 μm square field of view was less than 1 x 10 6 cm -2 , with respect to the total number of measurement regions was found. Note that the "low dislocation density region" referred to here means a region in the comparative example in which the crystalline layer was grown without the first process, and which had a lower dislocation density than the average dislocation density in the main surface of the crystalline layer, as shown in the results described later.

[0446] (X-ray rocking curve measurement)

[0447] The following two kinds of X-ray rocking curve measurement were performed for the base substrate, the substrate of the embodiment, and the substrate of the comparative example, respectively.

[0448] In the X-ray rocking curve measurement, "X'Pert-PRO MRD" manufactured by SPECTRIS was used, and a "hybrid monochromator" manufactured by the same company was used as a monochromator on the incident side. The hybrid monochromator has an X-ray mirror and a double crystal of a Ge (220) plane in this order from the X-ray source side. In this measurement, first, the X-rays emitted from the X-ray source were made into parallel light by the X-ray mirror. By this, the number of photons of the X-rays used (that is, the intensity of the X-rays) can be increased. Next, the parallel light from the X-ray mirror was made into monochromatic light of Cu Kα1 by the double crystal of the Ge (220) plane. Next, the monochromatic light from the double crystal of the Ge (220) plane was narrowed to a prescribed width by a slit, and was made incident on the substrate. Note that, if the half-value width when the rocking curve of the (0002) plane of GaN of a complete crystal is measured using this hybrid monochromator is found by a simulation experiment, it is 25.7 arcsec. That is, this half-value width is the theoretical measurement limit when the measurement is performed by the above optical system.

[0449] Note that, in this measurement, the X-rays incident on the substrate are made into parallel light toward the substrate side in a cross section along the ω direction, but do not become parallel light in a cross section along a direction orthogonal to the ω direction (the direction of the rotation axis of the substrate). Therefore, the width of the X-rays in the ω direction is substantially constant during the period from when the X-rays exit the slit to when they reach the substrate, but the width of the X-rays in the direction orthogonal to the ω direction widens. Therefore, in the X-ray rocking curve measurement, the half-value width of the X-rays that diffract on a prescribed crystal plane depends on the width of the X-rays that become parallel light in the ω direction among the slits on the incident side.

[0450] On the other hand, the light-receiving side was made open. The window width of the detector on the light-receiving side was set to 14.025 mm. In the above optical system, since the angle radius was 420 mm, the variation of the Bragg angle of ±0.95° could be measured.

[0451] (X-ray rocking curve measurement 1)

[0452] The width of the ω direction of the entrance side slit was set to 0.1 mm, and X-ray rocking curve measurement was performed on the (0002) plane of each of the base substrate, the substrate of the example, and the substrate of the comparative example. At this time, the measurement was performed at a plurality of measurement points set at intervals of 5 mm in the m axis direction and the a axis direction among the main surfaces of each substrate, respectively. As a result of the measurement, based on the diffraction peak angle of the (0002) plane at each measurement point, the radius of curvature of the c plane and the angle of the c axis with respect to the normal line of the main surface, that is, the misorientation angle, were calculated. In addition, as the maximum-minimum difference in the misorientation angle from the center of the main surface to a diameter of 29.6 mm, the deviation of the misorientation angle was calculated. In addition, at each measurement point, the half-value width FWHMb of the (0002) plane diffraction when the width of the ω direction of the entrance side slit was set to 0.1 mm was calculated.

[0453] (X-ray rocking curve measurement 2)

[0454] The width of the ω direction of the entrance side slit was set to 1 mm, and X-ray rocking curve measurement was performed on each of the base substrate and the substrate of the example. Note that the measurement was performed at the center of the main surface of each substrate. As a result of the measurement, the half-value width FWHMa of the (0002) plane diffraction when the width of the ω direction of the entrance side slit was set to 1 mm was calculated. Furthermore, at the center of the main surface in each substrate, the ratio of FWHMa-FWHMb to FWHMa was calculated.

[0455] Note that in the X-ray rocking curve measurements 1 and 2, for the main surface of each substrate, when the Bragg angle of the (0002) plane was 17.28°, the width of the ω direction of the slit was 0.1 mm, the footprint of the X-ray was about 0.337 mm, and the width of the ω direction of the slit was 1 mm, the footprint of the X-ray was about 3.37 mm.

[0456] (1-3) Results

[0457] The results are shown in Table 1.

[0458] [Table 1]

[0459] Proportion of low dislocation density region (%) Curvature radius of c-plane (m) Deviation of off-angle (°) (within diameter 29.6 mm) average dislocation density (cm -2 ) 4.5 x 10 5 ]]> 1.5 x 10 6 ]]> 3.0 x 10 6 ]]> FWHMb (arcsec) (in-plane) 95 30 0 (FWHMa-FWHMb) / FWHMa (%) 33.0～68.2 11.3 7.64 FIG. 11 0.025～0.052 0.15 0.22 FIG. 11 26.2～31.5 38.5～66.2 40.1～77.8 FIG. 12 4.5～28.1 - 54.2～79.6

[0460] FIG. 12 is a view showing an observation image obtained by observing the cross section of the layered structure of the example using a fluorescence microscope. As shown in FIG. 12As shown, in the layered structure of the example, based on the difference in the growth surface during growth (i.e., the difference in oxygen concentration), the first layer has a first c-plane growth region grown with the c-plane as the growth surface and a tilted interface growth region grown with the tilted interface as the growth surface. The first c-plane growth region has a plurality of recesses and a plurality of protrusions. The average of the angle formed by a pair of tilted portions in the first c-plane growth region is about 52°. In addition, the average distance to the nearest top is about 234 μm. In addition, the height from the main surface of the base substrate to the top of the first c-plane growth region is about 298 to 866 μm. In addition, the tilted interface growth region is continuously formed along the main surface of the base substrate. In addition, the thickness of the boundary surface from the main surface of the base substrate at the position where the tilted interface disappears in the second layer is about 1 mm.

[0461] As shown in Table 1, in the substrate of the example, the average dislocation density in the main surface is greatly reduced compared to the base substrate and the substrate of the comparative example, and is less than 5.5 x cm -2 Even when the crystal layer is grown thicker as in the comparative example, the dislocation density of the substrate is reduced compared to the base substrate, but in the substrate of the example, the dislocation density is further reduced compared to the comparative example.

[0462] In addition, when the dislocation density of the substrate of the example is denoted as N, the reduction rate of the dislocation density calculated from N / N0 above is 0.15.

[0463] In addition, in the substrate of the example, there is no region where the dislocation density exceeds 3 x 10 6 cm -2 It should be noted that even in the region with the highest dislocation density, the dislocation density is less than 1.5 x 10 6 cm -2 In addition, in the substrate of the example, a region (low dislocation density region) where the dislocation density is less than 1 x 10 6 cm -2 exists in more than 90% of the main surface 50s. The dislocation density in this low dislocation density region is 1.7 x 10 5 to 8.1 x 10 5 cm -2 .

[0464] In addition, as shown in Table 1, in the substrate of the example, the radius of curvature of the c-plane is larger than that of the base substrate and the substrate of the comparative example, and is 22 m or more. In addition, in the substrate of the example, the deviation of the misorientation angle of the c-axis within a diameter of 29.6 mm is reduced compared to the base substrate and the substrate of the comparative example, and is 0.075° or less. Even when the crystal layer is grown thicker as in the comparative example, the deviation of the misorientation angle of the c-axis in the substrate is smaller than that of the base substrate, but in the substrate of the example, the deviation of the misorientation angle of the c-axis is further reduced compared to the comparative example.

[0465] In addition, as shown in Table 1, in the substrate of the example, the half value width FWHMb of the (0002) plane diffraction when the width of the slit in the ω direction is set to 0.1 mm was 32 arcsec or less at all measurement points (i.e., 100%).

[0466] FIG. 12 (a) of FIG. 10 is a graph showing the normalized X-ray diffraction pattern when the slit was changed and X-ray rocking curve measurement was performed on the substrate of the example, and (b) is a graph showing the normalized X-ray diffraction pattern when the same measurement as the example was performed on the base substrate. Note that, FIG. 13 (a) and (b) of FIG. 10 show the measurement results in the direction along the m axis. In addition, in the graph, "line width" refers to the footprint of the X-ray.

[0467] As shown in (b) of FIG. 10, in the base substrate, the X-ray diffraction pattern was narrow when the width of the slit in the ω direction was set to 0.1 mm, but the X-ray diffraction pattern was broadened when the width of the slit in the ω direction was set to 1 mm. FIG. 14

[0468] Therefore, as shown in Table 1, in the base substrate, FWHMa-FWHMb was 50% or more of FWHMa.

[0469] In contrast, as shown in (a) of FIG. 10, in the substrate of the example, even when the width of the slit in the ω direction was widened from 0.1 mm to 1 mm, the X-ray diffraction pattern was slightly broadened, but the broadening was small. FIG. 15

[0470] Thus, as shown in Table 1, in the substrate of the example, FWHMa-FWHMb was 0% or more and 30% or less of FWHMa.

[0471] According to the above example, the root mean square roughness RMs of the main surface of the base substrate was set to 1 nm or more, and the misorientation of the main surface of the base substrate was set to 0.4° or less. In addition, the crystal strain introduced by processing of the base substrate was left on the main surface side of the base substrate, and the FWHM of the (10-10) plane diffraction in XRC measurement of the processed base substrate was 60 arcsec or more. Thus, the generation of the tilted interface other than the c-plane in the surface of the first layer was sufficiently promoted. In addition, in the first process, the first growth conditions were adjusted in such a manner as to satisfy Expression (1). Thus, during the growth of the first layer, the c-plane was surely disappeared. By surely disappearing the c-plane, dislocations were surely bent at the position where the tilted interface was exposed in the first layer. As a result, it was confirmed that the dislocation density in the main surface of the substrate was efficiently reduced. ​​

[0472] Further, according to the embodiment, it is possible to make the radius of curvature of the c-plane of the substrate larger than the radius of curvature of the c-plane of the base substrate, and it is possible to make the deviation of the misorientation angle of the c-axis in the substrate smaller than the deviation of the misorientation angle of the c-axis in the base substrate.

[0473] Further, according to the embodiment, as described above, dislocations are few throughout a wide range of the main surface of the substrate, and the crystal quality elements are all well balanced. It is thereby confirmed that, in the substrate of the embodiment, the FWHMb is 32 arcsec or less throughout a wide range of the main surface.

[0474] Further, according to the embodiment, as described above, the crystal quality elements are all well balanced, and the radius of curvature of the c-plane of the substrate is large. It is thereby confirmed that, in the embodiment, the difference FWHMa-FWHMb of the half-value width at the time of X-ray rocking curve measurement by changing the width of the slit on the incident side is 30% or less of FWHMa.

[0475] (2) Experiment 2

[0476] (2-1) Production of the Laminated Structure

[0477] In order to investigate the tilted interface generated on the surface of the first layer, a laminated structure having a base substrate and a first layer but not having a second layer was produced. Note that the conditions of the base substrate and the first layer were set to be substantially the same as those of the embodiment of Experiment 1.

[0478] (2-2) Evaluation

[0479] (Observation based on an optical microscope)

[0480] The surface of the first layer of the laminated structure was observed using an optical microscope.

[0481] (Observation based on a fluorescence microscope)

[0482] The cross section of the laminated structure was observed using a fluorescence microscope.

[0483] (2-3) Results

[0484] FIG. 13 (a) of FIG. 10 is a view showing an observation image obtained by observing the surface of the laminated structure of Experiment 2 using an optical microscope, and (b) is a view showing an observation image obtained by observing the surface of the laminated structure of Experiment 2 using a scanning electron microscope. FIG. 15 (a) of FIG. 11 is a view showing an observation image obtained by observing the M cross section of the laminated structure of Experiment 2 using an optical microscope, and (b) is a view showing an observation image obtained by observing the M cross section of the laminated structure of Experiment 2 using a scanning electron microscope. FIG. 13(a) is a view showing an observation image obtained by observing the a cross section of the layered structure of Experiment 2 using an optical microscope, and (b) is a view showing an observation image obtained by observing the a cross section of the layered structure of Experiment 2 using a scanning electron microscope.

[0485] As shown in (a) and (b) of FIG. 6, the angle of the inclined surface in the first layer with respect to the main surface of the base substrate was about 47° or less in the M cross section. FIG. 13 As shown in (a) and (b) of FIG. 6, the angle of the inclined surface in the first layer with respect to the main surface of the base substrate was about 47° or less in the M cross section. FIG. 14 As shown in (a) and (b) of FIG. 6, the angle of the inclined surface in the first layer with respect to the main surface of the base substrate was about 47° or less in the M cross section.

[0486] As shown in (a) of FIG. 6, six light-emitting visible surfaces were formed in the recesses produced on the top surface of the first layer, that is, the recesses had six inclined surfaces. FIG. 14 As shown in (a) of FIG. 6, six light-emitting visible surfaces were formed in the recesses produced on the top surface of the first layer, that is, the recesses had six inclined surfaces.

[0487] As shown in (a) of FIG. 6, six light-emitting visible surfaces were formed in the recesses produced on the top surface of the first layer, that is, the recesses had six inclined surfaces. FIG. 14 As shown in (a) of FIG. 6, six light-emitting visible surfaces were formed in the recesses produced on the top surface of the first layer, that is, the recesses had six inclined surfaces.

[0488] FIG. 15 As shown in (a) and (b) of FIG. 6, the M cross section (a cross section in the direction along the <11-20> axis) roughly vertically cut the inclined surfaces that constituted the recesses of the inverse regular hexagonal pyramid.

[0489] As shown in (a) and (b) of FIG. 6, the M cross section (a cross section in the direction along the <11-20> axis) roughly vertically cut the inclined surfaces that constituted the recesses of the inverse regular hexagonal pyramid. FIG. 15 As shown in (a) and (b) of FIG. 6, the angle of the inclined surface in the first layer with respect to the main surface of the base substrate was about 47° or less in the M cross section. FIG. 15 As shown in (a) of FIG. 6, six light-emitting visible surfaces were formed in the recesses produced on the top surface of the first layer, that is, the recesses had six inclined surfaces.

[0490] On the other hand, as shown in (a) and (b) of FIG. 7, the a cross section (a cross section in the direction along the <1-100> axis) was cut along the ridge lines of the recesses that constituted the inverse regular hexagonal pyramid. FIGS. 16 to 33 As shown in (a) and (b) of FIG. 7, the a cross section (a cross section in the direction along the <1-100> axis) was cut along the ridge lines of the recesses that constituted the inverse regular hexagonal pyramid.

[0491] As shown in (a) and (b) of FIG. 7, the angle of the ridge line of the recess that constituted the inverse regular hexagonal pyramid with respect to the main surface of the base substrate was mostly about 43° in the a cross section. FIGS. 16 to 32 As shown in (a) and (b) of FIG. 7, the angle of the ridge line of the recess that constituted the inverse regular hexagonal pyramid with respect to the main surface of the base substrate was mostly about 43° in the a cross section. FIGS. 16 to 32 As shown in (a) and (b) of FIG. 7, the angle of the ridge line of the recess that constituted the inverse regular hexagonal pyramid with respect to the main surface of the base substrate was mostly about 43° in the a cross section.

[0492] Here, the angle of the {11-2m} plane with respect to the {0001} plane is as follows.

[0493] {11-21} plane: 72.9°

[0494] {11-22} plane: 58.4°

[0495] {11-23} plane: 47.3°

[0496] {11-24} plane: 39.1°

[0497] From the above, it was confirmed that the inclined interface generated in the surface of the first layer grown under the conditions of Experiment 2 was the {11-2m} plane with m > 3. It was also confirmed that the inclined interface was mostly the {11-23} plane.

[0498] According to Experiment 2, as with Experiment 1, by using the above-described base substrate and adjusting the first growth conditions so as to satisfy Equation (1), it was possible to generate the {11-2m} plane with m > 3 as the inclined interface. From this, it was confirmed that in the first layer, it was possible to make the nearest top-to-top average distance exceed 100 μm.

[0499] (3) Experiment 3

[0500] (3-1) Production of Nitride Semiconductor Substrate

[0501] In order to compare the in-plane distribution of dislocation-free regions and the in-plane distribution in X-ray rocking curve measurement, the following Samples 1 to 3 were produced. The substrate of Sample 1 was a substrate equivalent to the substrate of the example of Experiment 1. The substrate of Sample 2 was a substrate obtained from a crystal layer grown as a thick film with the c-plane as the growth plane. In addition, the substrate of Sample 3 was a substrate produced by the conventional VAS method, and was equivalent to the base substrate.

[0502] [Method of Producing Nitride Semiconductor Substrate of Sample 1]

[0503] The substrate of Sample 1 was produced by the same method as the example of Experiment 1. Note that, regarding Sample 1, the curvature radius of the c-plane, the dislocation density, and the like were set to be equivalent to those of the substrate of the example of Experiment 1, except that the absolute value of the misorientation angle and the misorientation direction were different from those of the substrate of the example of Experiment 1.

[0504] [Production Conditions of Nitride Semiconductor Substrate of Sample 2]

[0505] (Base Substrate)

[0506] Material: GaN

[0507] Production Method: VAS method

[0508] Diameter: 62 mm

[0509] Thickness: 400 μm

[0510] Low-index crystal plane closest to the main surface: c-plane

[0511] Deviation angle: 0.5° in the m-axis direction

[0512] The main surface was not subjected to pattern processing such as masking layer or the like.

[0513] (Crystal layer)

[0514] Material: GaN

[0515] Growth method: HVPE method

[0516] Growth temperature: 1050°C

[0517] V / III ratio: 2.8

[0518] Growth time: 15 hours

[0519] (Processing)

[0520] Lapping: The cylindrical region was removed, and a cylindrical region having a diameter of 56 mm was obtained.

[0521] Slicing: 5 pieces each having a thickness of 630 μm

[0522] Beveling: The diameter was set to 50.8 mm.

[0523] Grinding processing: The thickness was set to 400 to 450 μm.

[0524] [Production conditions of nitride semiconductor substrate of sample 3]

[0525] The substrate of sample 3 was produced by the same existing VAS method as the base substrate. Note that, regarding sample 3, the curvature radius of the c-plane, the dislocation density, and the like were set to be the same as the base substrate except that the absolute value and the deviation direction of the deviation angle were different from those of the base substrate.

[0526] (3-2) Evaluation

[0527] (Observation based on multiphoton excitation microscope)

[0528] The main surface of each of the substrates of samples 1 to 3 was observed using a multiphoton excitation microscope under the same conditions as in Experiment 1.

[0529] (X-ray rocking curve measurement)

[0530] The same two kinds of X-ray rocking curve measurement as in Experiment 1 were performed on the substrates of Samples 1 to 3. At this time, the measurement was performed at a plurality of measurement points each set at an interval of 5 mm in the m-axis direction and the a-axis direction among the main surfaces. Thus, the proportion of FWHMa-FWHMb to FWHMa was found at the plurality of measurement points of each sample.

[0531] (3-3) Results

[0532] (3-3-1) In-Plane Distribution of Dislocation-Free Regions

[0533] Using FIG. 33 , the distribution of dislocation-free regions in the substrates of Samples 1 and 2 will be described. FIGS. 16 to 33 is a view of the main surface of the substrate of Sample 1 observed using a multiphoton excitation microscope. Note that in FIG. 33 , (x, y) represents the coordinates in the m-axis direction and the coordinates in the a-axis direction. FIG. 33 is a view of the main surface of the substrate of Sample 2 observed using a multiphoton excitation microscope. Note that in FIGS. 16 to 33 , the thick-line frame represents a 50-μm square dislocation-free region.

[0534] [Sample 2]

[0535] In the substrate of Sample 2 obtained from the crystal layer subjected to thick-film growth with the c-plane as the growth surface, the dislocation density decreased in inverse proportion to the thickness of the crystal layer, and thus the average dislocation density was 6.3 x 10 5 cm -2 .

[0536] However, as shown in FIGS. 16 to 33 , in the substrate of Sample 2, the dislocations were present in a state of being uniformly dispersed in the plane. Note that the distribution of dislocations in the region not shown in the figure was also the same as ​ . Thus, throughout the entire substrate of Sample 2, the size of the dislocation-free regions was smaller than 50 μm square, and no 50-μm square dislocation-free region was formed.

[0537] As such, even with the method of Sample 2, which is a method capable of obtaining a high-quality substrate as a conventional method, the obtained substrate did not form a 50-μm square dislocation-free region. It is thus considered that even a substrate produced by a conventional other manufacturing method that does not perform a special process of collecting dislocations does not form a 50-μm square dislocation-free region.

[0538] [Sample 1]

[0539] In contrast, as shown in ​As shown, the main surface of the substrate of Sample 1 contained dislocation-free regions of at least 50 μm square. In addition, in the substrate of Sample 1, dislocation-free regions of 50 μm square were dispersed throughout the entire main surface.

[0540] In addition, in the substrate of Sample 1, dislocation-free regions of 50 μm square were dispersed throughout the entire main surface. ​ As shown, within the entire field of view of 250 μm square, there were at least 1 dislocation-free region of 50 μm square. The main surface of the substrate of Sample 1 had a density of 1600 / cm 2 The above density had dislocation-free regions of 50 μm square that did not overlap. In detail, the density of dislocation-free regions of 50 μm square that did not overlap in the main surface of the substrate of Sample 1 was about 5200 / cm 2 .

[0541] In summary, according to Sample 1, by using the above-described base substrate, and adjusting the first growth conditions in a manner satisfying Equation (1), it was possible to make the average distance between the nearest top portions exceed 100 μm. It was confirmed that this made it possible to sufficiently reduce the dislocation density in the main surface of the substrate. In addition, it was confirmed that by making the average distance between the nearest top portions exceed 100 μm, it was possible to form dislocation-free regions of at least 50 μm square, and to disperse these dislocation-free regions throughout the entire main surface. In addition, it was confirmed that it was possible to make the density of dislocation-free regions of 50 μm square that did not overlap in the main surface be 1600 / cm 2 or more.

[0542] (3-3-2) In-Plane Distribution in X-Ray Rocking Curve Measurement

[0543] The results of Samples 1 to 3 are shown in Tables 2, 3, and 4, respectively. In the following tables, “difference” means (FWHMa-FWHMb) / FWHMa (%).

[0544] [Table 2]

[0545] [Sample 1]

[0546]

[0547] [Table 3]

[0548] [Sample 2]

[0549]

[0550] [Table 4]

[0551] [Sample 3]: Corresponds to Base Substrate

[0552]

[0553] [Sample 3]

[0554] As shown in Table 4, in the substrate of Sample 3 produced by the conventional VAS method, the deviation of the misalignment angle of the c-axis within a diameter of 40 mm was about ±0.24°. In addition, in the substrate of Sample 3, the FWHMb exceeded 32 arcsec at all measurement points. In addition, in the substrate of Sample 3, the (FWHMa-FWHMb) / FWHMa exceeded 30% at all measurement points.

[0555] [Sample 2]

[0556] As shown in Table 3, in the substrate of Sample 2 obtained from the crystal layer produced by thick film growth with the c-plane as the growth surface, the deviation of the misalignment angle of the c-axis within a diameter of 40 mm was improved to about ±0.074° compared with the substrate of Sample 3. In addition, the FWHMb of the substrate of Sample 2 was improved compared with the FWHMb of the substrate of Sample 3.

[0557] However, in the substrate of Sample 2, a plurality of portions where the FWHMb exceeded 32 arcsec were observed. In addition, in the substrate of Sample 2, the (FWHMa-FWHMb) / FWHMa greatly exceeded 30% at all measurement points.

[0558] As such, in the substrate of Sample 2 which is a high-quality substrate of the related art, the dislocation density and the misalignment angle deviation were improved compared with the base substrate, but there was not a single point in the substrate of Sample 2 that satisfied the condition of the half-value width of FWHMb < 32 arcsec and (FWHMa-FWHMb) / FWHMa ≤ 30%. It is considered that this is because at least any one of the above-described crystal quality elements was not as good as the substrate of Sample 1.

[0559] Therefore, even the substrate of Sample 2 which is a high-quality substrate of the related art did not satisfy the condition of the above-described half-value width, and it is considered that the substrates produced by other conventional production methods also do not satisfy the condition of the above-described half-value width.

[0560] [Sample 1]

[0561] In contrast, as shown in Table 2, in the substrate of Sample 1, the deviation of the misalignment angle of the c-axis within a diameter of 40 mm was about ±0.03°, which was smaller than the substrates of Samples 2 and 3.

[0562] In addition, in the substrate of Sample 1, the FWHMb was 32 arcsec or less at all measurement points. In addition, in the substrate of Sample 1, the (FWHMa-FWHMb) / FWHMa was 30% or less at all measurement points.

[0563] As described above, in the substrate of Sample 1 obtained by the above production method, not only the dislocation density can be reduced, but also the misorientation deviation can be reduced, and all of the above-mentioned crystal quality factors that determine the half-value width can be balanced and good. Thus, it is confirmed that in the substrate of Sample 1, FWHMb can be 32 arcsec or less. Furthermore, it is confirmed that in Sample 1, even when the slit width is set to 1 mm, in the entire region where X-rays are irradiated, the radius of curvature of the c-plane is large, and the above-mentioned crystal quality factors are balanced and good, and thus (FWHMa-FWHMb) / FWHMa can be 30% or less.

[0564] <Preferred Embodiment of the Invention>

[0565] Hereinafter, a preferred embodiment of the present invention will be described.

[0566] (Note 1)

[0567] A production method of a nitride semiconductor substrate, which is a method of producing a nitride semiconductor substrate using a vapor phase growth method, includes the following steps:

[0568] A step of preparing a base substrate formed of a single crystal of a Group III nitride semiconductor, having a mirror-finished main surface, and having a (0001) plane as a lowest index crystal plane closest to the main surface;

[0569] A first step of epitaxially growing a single crystal of a Group III nitride semiconductor having a top surface exposing a (0001) plane directly on the main surface of the base substrate, causing the top surface to have a plurality of recesses composed of tilted interfaces other than the (0001) plane, causing the tilted interfaces to gradually expand as they go upward from the main surface of the base substrate, and causing the (0001) plane to disappear from the top surface, thereby growing a first layer having a surface composed only of the tilted interfaces; and

[0570] A second step of epitaxially growing a single crystal of a Group III nitride semiconductor on the first layer, causing the tilted interfaces to disappear, and growing a second layer having a mirror-finished surface,

[0571] In the first step,

[0572] The plurality of recesses are caused to be formed in the top surface of the single crystal, and the (0001) plane is caused to disappear, thereby forming a plurality of valley portions and a plurality of top portions on the surface of the first layer.

[0573] When an arbitrary cross section perpendicular to the main surface is observed, the average distance between a pair of top portions closest to each other among the plurality of top portions sandwiching one valley portion among the plurality of valley portions in a direction along the main surface is set to be more than 100 μm.

[0574] (Note 2)

[0575] The method for manufacturing a nitride semiconductor substrate according to Note 1, in the process of preparing a base substrate,

[0576] The root mean square roughness of the main surface of the base substrate is 1 nm or more.

[0577] (Note 3)

[0578] The method for manufacturing a nitride semiconductor substrate according to Note 1 or 2, in the process of preparing a base substrate,

[0579] The crystal strain introduced by processing the base substrate remains on the main surface side of the base substrate.

[0580] The half-value width of (10-10) plane diffraction when X-ray rocking curve measurement is performed with an incident angle of 2° with respect to the main surface of the base substrate after processing is greater than the half-value width of the base substrate before processing, and is 60 arcsec or more and 200 arcsec or less.

[0581] (Note 4)

[0582] The method for manufacturing a nitride semiconductor substrate according to any one of Notes 1 to 3, in the first process,

[0583] The average distance of the closest pair of top portions from each other is less than 800 μm.

[0584] (Note 5)

[0585] The method for manufacturing a nitride semiconductor substrate according to any one of Notes 1 to 4, in the first process,

[0586] After the (0001) plane disappears from the surface, the state in which the surface is composed only of the inclined interface is maintained, and the growth of the first layer is continued over a predetermined thickness.

[0587] (Note 6)

[0588] The method for manufacturing a nitride semiconductor substrate according to any one of Notes 1 to 5, after the second process, has a process of cutting at least one nitride semiconductor substrate from the second layer.

[0589] (Note 7)

[0590] The method for manufacturing a nitride semiconductor substrate according to any one of Notes 1 to 6, in the process of preparing a base substrate,

[0591] the aforementioned base substrate is bent into a concave spherical shape with respect to the aforementioned main surface,

[0592] in a process of cutting out the nitride semiconductor substrate,

[0593] the deviation of the angle of the <0001> axis in the nitride semiconductor substrate with respect to the normal line of the main surface, that is, the deviation of the off angle, is smaller than the deviation of the angle of the <0001> axis in the base substrate with respect to the normal line of the main surface, that is, the deviation of the off angle.

[0594] (Note 8)

[0595] The method of manufacturing a nitride semiconductor substrate according to any one of Notes 1 to 7, wherein

[0596] in the first process, the first layer is grown under first growth conditions that satisfy Equation (1),

[0597] in the second process, the second layer is grown under second growth conditions that satisfy Equation (2).

[0598] G c1 G i / cos α... (1)

[0599] G c2 G i / cos α... (2)

[0600] (where G is the growth rate of the (0001) plane in the first layer, c1 G is the growth rate of the (0001) plane in the second layer, c2 G is the growth rate of the inclined interface that is most inclined with respect to the (0001) plane in each of the first layer and the second layer, i α is the angle of the inclined interface that is most inclined with respect to the (0001) plane in each of the first layer and the second layer, with respect to the (0001) plane.)

[0601] (Note 9)

[0602] The method of manufacturing a nitride semiconductor substrate according to any one of Notes 1 to 8, wherein

[0603] a 1c plane growth region in which a 1c plane that is grown with the (0001) plane as a growth surface is formed in the first layer,

[0604] A convex portion is formed at a position where the aforementioned (0001) plane disappears and ends as an inflection point protruding upward in the aforementioned first c-plane growth region, and a pair of inclined portions are formed as traces of intersections of the aforementioned (0001) plane and the aforementioned inclined interface on both sides of the aforementioned convex portion in the aforementioned first c-plane growth region,

[0605] The angle of the pair of inclined portions is 70° or less.

[0606] (Addendum 10)

[0607] The method of manufacturing a nitride semiconductor substrate according to any one of Addenda 1 to 9, wherein the first process has the following process:

[0608] a process of forming an inclined interface expansion layer by gradually expanding the inclined interface toward the upper side of the base substrate; and

[0609] a process of forming an inclined interface maintenance layer over the inclined interface expansion layer in which the (0001) plane disappears from the surface, in a state in which the surface is composed only of inclined interfaces other than the (0001) plane, and over a prescribed thickness.

[0610] (Addendum 11)

[0611] The method of manufacturing a nitride semiconductor substrate according to any one of Addenda 1 to 10, wherein the second process has the following process:

[0612] a process of forming a c-plane expansion layer by expanding the (0001) plane toward the upper side of the first layer and reducing inclined interfaces other than the (0001) plane; and

[0613] a process of forming a main growth layer over the c-plane expansion layer in which the surface is mirror-finished, using the (0001) plane as a growth surface, over a prescribed thickness.

[0614] (Addendum 12)

[0615] The method of manufacturing a nitride semiconductor substrate according to any one of Addenda 1 to 11, wherein in the first process,

[0616] as the inclined interface, a {11-2m} plane where m ≥ 3 is generated.

[0617] (Addendum 13)

[0618] A nitride semiconductor substrate having a diameter of 2 inches or more and having a main surface in which the nearest low-index crystal plane is a (0001) plane,

[0619] When an X-ray rocking curve measurement of the (0002) plane diffraction is performed on the X-ray irradiating the Cu's Kα1 through the double monochromator and the slit of the Ge (220) face, the half value width FWHM a of the (0002) plane diffraction when the width of the slit in the ω direction is set to 1 mm is 30% or less of FWHM a,

[0620] the half value width FWHM b of the (0002) plane diffraction when the width of the slit in the ω direction is set to 0.1 mm is 32 arcsec or less,

[0621] the difference FWHM a - FWHM b obtained by subtracting FWHM b from FWHM a is 30% or less of FWHM a,

[0622] the diffraction pattern when the width of the slit in the ω direction is set to 1 mm has a single peak.

[0623] (Paragraph 14)

[0624] The nitride semiconductor substrate according to Paragraph 13, wherein when an X-ray rocking curve measurement of the (0002) plane diffraction is performed on the X-ray irradiating the Cu's Kα1 through the double monochromator and the slit of the Ge (220) face, the half value width FWHM b of the (0002) plane diffraction when the width of the slit in the ω direction is set to 0.1 mm is 32 arcsec or less at 90% or more of all the measurement points set at intervals of 5 mm within the main face.

[0625] (Paragraph 15)

[0626] The nitride semiconductor substrate according to Paragraph 13 or 14, wherein when the main face is observed with a field of view of 250 μm square by a multiphoton excitation microscope, and a dislocation density is calculated from a dark point density, there is no region in the main face where the dislocation density exceeds 3 x 10 6 cm -2 and a region where the dislocation density is less than 1 x 10 6 cm -2 exists in 80% or more of the main face.

[0627] (Paragraph 16)

[0628] The nitride semiconductor substrate according to any one of Paragraphs 13 to 15, wherein the main face has non-overlapping dislocation-free regions of 50 μm square at a density of 100 / cm 2 or more.

[0629] (Paragraph 17)

[0630] A nitride semiconductor substrate having a diameter of 2 inches or more and having a main face whose nearest low-index crystal plane is a (0001) plane,

[0631] When the main surface of the aforementioned nitride semiconductor substrate was observed using a multiphoton excitation microscope with a field of view of 250 μm square, and the dislocation density was calculated from the dark spot density, no dislocation density exceeding 3 × 10⁻⁶ was found in the aforementioned main surface. 6 cm -2 In regions where the aforementioned dislocation density is less than 1×10⁻⁶, 6 cm -2 The area exists in more than 80% of the aforementioned main surface.

[0632] The aforementioned main surface is 100 pieces / cm 2 The above densities have non-overlapping 50μm square dislocation-free regions.

[0633] (Postscript 18)

[0634] The nitride semiconductor substrate according to any one of Appendices 13 to 17 has an oxygen concentration of 5 × 10⁻⁶. 16 cm -3 the following.

[0635] (Postscript 19)

[0636] The nitride semiconductor substrate according to any one of Appendices 13 to 18 has a hydrogen concentration of less than 1 × 10⁻⁶. 17 cm -3 .

[0637] (Postscript 20)

[0638] According to any one of Appendices 13 to 19, in the nitride semiconductor substrate, when 100 dislocations are randomly selected from the aforementioned main surface, the Burgers vector is <11-20> / 3. <0001> The proportion of dislocations in either <11-23> / 3 is more than 50%.

[0639] (Postscript 21)

[0640] A layered structure having:

[0641] The substrate is formed from a single crystal of a group III nitride semiconductor and has a mirrored main surface, with the nearest low index crystal surface to the main surface being the (0001) surface.

[0642] The first low oxygen concentration region is directly disposed on the aforementioned main surface of the aforementioned substrate and is formed by a single crystal of a group III nitride semiconductor.

[0643] A high-oxygen concentration region, disposed on the aforementioned first low-oxygen concentration region, and formed of a single crystal of a group III nitride semiconductor; and

[0644] The second low-oxygen concentration region is disposed on top of the aforementioned high-oxygen concentration region and is formed of a single crystal of a group III nitride semiconductor.

[0645] the oxygen concentration of the aforementioned high oxygen concentration region is higher than the oxygen concentration of each of the aforementioned first low oxygen concentration region and the aforementioned second low oxygen concentration region,

[0646] when an arbitrary cross section perpendicular to the aforementioned main surface is observed,

[0647] the upper surface of the aforementioned first low oxygen concentration region has a plurality of valleys and a plurality of hills,

[0648] the closest pair of hills among the aforementioned plurality of hills sandwiching one valley among the aforementioned plurality of valleys are spaced apart from each other by an average distance exceeding 100 μm in a direction along the aforementioned main surface.

[0649] (Paragraph 22)

[0650] The layered structure according to Paragraph 21, wherein the aforementioned high oxygen concentration region is continuously provided along the aforementioned main surface of the aforementioned base substrate.

[0651] (Paragraph 23)

[0652] The layered structure according to Paragraph 21 or 22, wherein the aforementioned first low oxygen concentration region further has a pair of inclined portions provided on both sides sandwiching the aforementioned hill,

[0653] the angle formed by the aforementioned pair of inclined portions is 70° or less.

[0654] (Paragraph 24)

[0655] The layered structure according to any one of Paragraphs 21 to 23, wherein the dislocation density in the aforementioned main surface of the aforementioned base substrate is denoted by N0, and the dislocation density in the boundary surface at the upper end of the aforementioned high oxygen concentration region and along the aforementioned main surface is denoted by N, and the reduction rate of the dislocation density obtained from N / N0 is less than the reduction rate of the dislocation density obtained from N' / N0 when the surface of the crystal layer of the Ill-nitride semiconductor is denoted by N' when the crystal layer is epitaxially grown on the aforementioned main surface of the aforementioned base substrate with only the aforementioned (0001) plane as a growth surface with a thickness equal to the thickness from the aforementioned main surface of the aforementioned base substrate to the aforementioned boundary surface.

[0656] (Paragraph 25)

[0657] The layered structure according to any one of Paragraphs 21 to 24, wherein the thickness from the aforementioned main surface of the aforementioned base substrate to the boundary surface at the upper end of the aforementioned high oxygen concentration region and along the aforementioned main surface is 1.5 mm or less,

[0658] The dislocation density in the aforementioned main surface of the aforementioned base substrate is denoted as N0, and the dislocation density in the aforementioned boundary surface of the aforementioned second layer is denoted as N, and the reduction rate of the dislocation density, which is obtained from N / N0, is 0.3 or less.

Claims

1. A nitride semiconductor substrate having a diameter of 2 inches or more and a thickness of 300 μm or more and 1 mm or less, formed only of a single crystal of a Group III nitride semiconductor in the entire thickness, and having a main surface of which the nearest low-index crystal face is a (0001) face, more than 95% of dark spots detected when the main surface is observed with a field of view of 250 μm square using a multiphoton excitation microscope correspond to dislocations, rather than to non-luminescent centers due to foreign matter or point defects, The main face has 100 / cm 2 The above density has dislocation-free regions of 50 μm square that do not overlap.

2. The nitride semiconductor substrate of claim 1, wherein When the main surface is observed with a field of view of 250 μm square by a multiphoton excitation microscope, and the dislocation density is calculated from the density of dark spots, the dislocation density is less than 1 x 10 6 cm -2 -2 in the region of 80% or more of the main surface.

3. The nitride semiconductor substrate of claim 1 or 2, wherein When the main surface is observed with a field of view of 250 μm square by a multiphoton excitation microscope, and the dislocation density is calculated from the density of dark spots, there is no region in the main surface where the dislocation density exceeds 3 x 1010cm 6 cm -2 -2.

4. The nitride semiconductor substrate of claim 1 or 2, having an oxygen concentration of 5 x 1016 cm"2 or less. 16 cm -3 below.

5. The nitride semiconductor substrate of claim 1 or 2, having a hydrogen concentration of less than 1 x 1016 cm"3. 17 cm -3 .

Images