JP7164946B2 - Gallium nitride crystal substrate and method for evaluating crystal quality thereof - Google Patents

Description

The present invention relates to a gallium nitride crystal substrate and a method for evaluating the crystal quality thereof.

Japanese Patent Application Laid-Open No. 2013-203653 (Patent Document 1) describes a method for producing a high-quality Group III nitride crystal having a small warpage and suppressed cracking, in which a single crystal made of a Group III nitride crystal is heated at 1000° C. or higher. Manufacture of Group III nitride crystals comprising the steps of obtaining a Group III nitride seed substrate by performing a heat treatment, and forming a Group III nitride crystal film on the Group III nitride seed substrate to obtain Group III nitride crystals. Disclose how.

In addition, Japanese Patent Application Laid-Open No. 2006-071354 (Patent Document 2) describes a method for evaluating the crystal quality in the surface layer of a group III nitride crystal, using an X-ray diffraction method, from the surface of the crystal to the depth direction. A method for evaluating the crystallinity of a crystal surface layer by evaluating a change in the crystallinity, wherein X By irradiating the crystal with X-rays while changing the line penetration depth, the amount of change in at least one of the interplanar spacing in the diffraction profile of the crystal lattice plane, the half-value width of the diffraction peak, and the half-value width in the rocking curve is measured. Disclosed is a crystallinity evaluation method for a crystal surface layer to be evaluated.

In addition, Katsuhiko Inaba, "Evaluation of GaN materials by high-resolution X-ray diffraction method", Rigaku Journal, 44 (2), 2013, pp. 7-15 (Non-Patent Document 1) describes GaN grown on a sapphire substrate. We disclose that forbidden reflections are observed in high-resolution X-ray diffraction ω-2θ scans of thin films, which are believed to originate from multiple reflections of lattice planes that are not parallel to the surface. Further, J. Blasing and A. Krost, “X-ray multiple diffraction (Umweganregung) in wurtzite-type GaN and ZnO epitaxial layers”, phys. stat. sol., (a)201, No.4, 2004, pp. R17-R20 disclose that diffraction peaks are observed in φ scans at ω-2θ of forbidden (0001) reflections of X-ray diffraction for GaN thin films grown on silicon substrates.

JP 2013-203653 A JP 2006-071354 A

Katsuhiko Inaba, "Evaluation of GaN materials by high-resolution X-ray diffraction method", Rigaku Journal, 44 (2), 2013, pp. 7-15 J. Blasing and A. Krost, “X-ray multiple diffraction (Umweganregung) in wurtzite-type GaN and ZnO epitaxial layers”, phys. stat. sol., (a)201, No.4, 2004, pp.R17- R20

Even with gallium nitride crystals obtained by the method for producing group III nitride crystals disclosed in Japanese Patent Application Laid-Open No. 2013-203653 (Patent Document 1), the crystal quality is reduced when processed into gallium nitride crystal substrates. I have a problem getting it to work. In addition, the industry is demanding the development of gallium nitride crystal substrates with higher crystal quality in order to manufacture semiconductor devices with higher characteristics.

Further, the method for evaluating the crystallinity of a crystal surface layer disclosed in Japanese Patent Application Laid-Open No. 2006-071354 (Patent Document 2) is a method for evaluating crystal quality such as uniform strain, non-uniform strain, and misalignment of plane orientation in the surface layer of a gallium nitride crystal. is useful as an index for evaluating However, in the industry, development of an evaluation method for more detailed evaluation of the crystal quality of gallium nitride crystal substrates with high crystal quality is desired.

Also, Katsuhiko Inaba, "Evaluation of GaN materials by high-resolution X-ray diffraction method", Rigaku Journal, 44 (2), 2013, pp. 7-15 (Non-Patent Document 1) and J. Blasing and A. Krost, " X-ray multiple diffraction (Umweganregung) in wurtzite-type GaN and ZnO epitaxial layers”, phys. stat. Diffraction peaks appearing in the ω-2θ scan and φ scan in the X-ray diffraction obtained have not been linked to the detailed evaluation of the crystal quality of gallium nitride crystal substrates.

Therefore, in view of the above situation, it is an object of the present invention to provide a gallium nitride crystal substrate having high crystal quality and a method for evaluating the crystal quality of the gallium nitride crystal substrate capable of evaluating the crystal quality in detail.

A gallium nitride crystal substrate according to an aspect of the present invention has a {0001} plane as a principal surface, and X-ray diffraction measurement using a small divergence angle incident X-ray parallel beam shows that the incidence angles ω and ω of 0001 symmetric forbidden Bragg reflections In the φ scan pattern measurement with the [0001] axis at the diffraction angle 2θ as the rotation axis, when the [1-100] direction is φ = 0 °, φ is from -180 ° to 180 ° In the measurement, The X-ray multiple diffraction pattern that appears by X-ray multiple diffraction has a three-fold rotational symmetry that repeats every 120° from -180° to -60°, from -60° to 60°, and from 60° to 180°. It has three 120° rotationally symmetrical peak regions with polarities. Each 120° rotationally symmetric peak region has mirror symmetry with respect to the plane containing the line indicating the angle of φ at the center of the range of φ of the region and the [0001] axis as the mirror plane. Prepare. Each −region comprises a −L region and a −R region having rotational symmetry with respect to each other about the intersection of the [0001] axis and the line indicating the φ angle in the middle of the φ range of that region. Each + region comprises a +L region and a +R region that are rotationally symmetrical to each other about the intersection of the [0001] axis and the line indicating the φ angle in the middle of the φ range of that region. The maximum peak intensity ratio obtained by dividing the maximum peak intensity of X-ray multiple diffraction peaks in at least one of the -L region, each -R region, each +L region, and each +R region by the background intensity is 500 or more. be.

A method for evaluating the crystal quality of a gallium nitride crystal substrate according to one aspect of the present invention is to evaluate the crystal quality of a gallium nitride crystal substrate whose main surface is the {0001} plane by X-rays using a small divergence angle incident X-ray parallel beam. A method of evaluating by diffraction measurement, comprising a step of placing a gallium nitride crystal substrate on a sample stage so that its [1-100] direction coincides with the direction of φ = 0 °; Calibrating the position of the gallium nitride crystal substrate by ω-scan and z-scan at the folded angle 2θ, and aligning the χ-axis, and rotating the [0001] axis at the incident angle ω and the diffraction angle 2θ of the 0001 symmetrical forbidden Bragg reflection. and C. performing an axial φ scan pattern measurement. From the high symmetry of the X-ray multiple diffraction peak obtained from the φ scan pattern measurement and the maximum peak intensity ratio obtained by dividing the maximum peak intensity of the X-ray multiple diffraction peak by the background intensity, gallium nitride crystal Evaluate the crystalline quality of the substrate.

According to the above, it is possible to provide a gallium nitride crystal substrate having a high crystal quality and a method for evaluating the crystal quality of the gallium nitride crystal substrate capable of evaluating the crystal quality in detail.

FIG. 1 is a schematic diagram showing an example of a parallel X-ray beam method used in a method for evaluating the crystal quality of a gallium nitride crystal substrate according to one aspect of the present invention. FIG. 2 is a schematic diagram showing an example of 0002 symmetrical Bragg reflection in the method for evaluating crystal quality of a gallium nitride crystal substrate according to one aspect of the present invention. FIG. 3 is a schematic diagram showing an example of 0001 symmetrical forbidden Bragg reflection in the method for evaluating crystal quality of a gallium nitride crystal substrate according to one aspect of the present invention. FIG. 4 is a schematic diagram showing an example of {0001} planes of a gallium nitride crystal substrate. FIG. 5 is a schematic diagram showing an example of {0002} planes of a gallium nitride crystal substrate. FIG. 6 is a schematic diagram showing an example of {01-10} planes of a gallium nitride crystal substrate. FIG. 7 is a schematic diagram showing an example of {11-20} planes of a gallium nitride crystal substrate. FIG. 8 is a schematic diagram showing the arrangement of gallium atoms and nitrogen atoms in a gallium nitride crystal substrate. FIG. 9 is a schematic plan view showing crystal orientations and regions of a gallium nitride crystal in a method for evaluating crystal quality of a gallium nitride crystal substrate according to one aspect of the present invention. FIG. 10 is a schematic diagram showing an example of ω scan by {0002} symmetrical Bragg reflection when the crystal substrate is set in the crystal quality evaluation method of the gallium nitride crystal substrate according to one aspect of the present invention. FIG. 11 is a schematic diagram showing an example of z-scan by {0002} symmetrical Bragg reflection when the crystal substrate is set in the crystal quality evaluation method of the gallium nitride crystal substrate according to one aspect of the present invention. FIG. 12 shows an example of a 360° X-ray multiple diffraction pattern in the region where φ is from −180° to 180° in φ scan pattern measurement of a gallium nitride crystal substrate according to one aspect of the present invention. 1 is a schematic diagram showing FIG. FIG. 13 shows an example of a 360° X-ray multiple diffraction pattern in a region where φ is from −180° to 180° in φ scan pattern measurement of a gallium nitride crystal substrate according to one aspect of the present invention, with the vertical axis on a logarithmic scale. 1 is a schematic diagram showing FIG. FIG. 14 shows an example of a 120° X-ray multiple diffraction pattern in a 120° rotationally symmetrical peak region with φ from −60° to 60° in φ scan pattern measurement of a gallium nitride crystal substrate according to one aspect of the present invention. FIG. 4 is a schematic diagram showing the axes on a square root scale; FIG. 15 shows an example of a 120° X-ray multiple diffraction pattern in a 120° rotationally symmetrical peak region with φ from −60° to 60° in φ scan pattern measurement of a gallium nitride crystal substrate according to one aspect of the present invention. FIG. 4 is a schematic diagram showing the axes on a logarithmic scale; FIG. 16 is a schematic diagram showing a comparison between the pattern of the A ₊ region of the 120° rotationally symmetrical peak region shown in FIG. 14 and the pattern of the A ₋ region inverted around φ=0°. FIG. 17 is a schematic diagram showing a comparison between the pattern of the A ₊ region of the 120° rotationally symmetrical peak region shown in FIG. 14 and the pattern in which the A ₊ region is inverted around φ=30°. FIG. 18 is a schematic diagram showing the pattern of the A _+R region of the 120° rotationally symmetrical peak region shown in FIG. 19 is a schematic diagram showing the pattern of the A _+L region of the 120° rotationally symmetrical peak region shown in FIG. 14; FIG. FIG. 20 is a schematic diagram showing the A- _R region pattern of the 120° rotationally symmetrical peak region shown in FIG. 21 is a schematic diagram showing the pattern of the A- _L region of the 120° rotationally symmetrical peak region shown in FIG. 14; FIG.

[Description of an embodiment of the present invention]
First, embodiments of the present invention will be listed and described.

[1] A gallium nitride crystal substrate according to an embodiment of the present invention has a {0001} plane as the main surface, and uses an X-ray lens on the incident side to point focus with a cross slit of 0.3 mm × 0.3 mm. A 0.27° collimator with a 0.27° slit for reflectance measurement on the receiving side and a graphite plate monochromator for selection of CuKα1 monochromatic X _- rays were used. Incidence of 0001 symmetrical forbidden Bragg reflection performed by X-ray diffraction measurement under the measurement conditions of 0.02° in the measurement φ angle step and measurement time at each step ranging from 1 second to 50 seconds according to the measurement φ angle range. In the φ scan pattern measurement with the [0001] axis as the rotation axis at the angle ω and the diffraction angle 2θ (that is, the incident angle ω: 8.5428 ° and the diffraction angle 2θ: 17.0856 °), the [1-100] direction is φ = 0°, in the measurement from -180° to 180°, the X-ray multiple diffraction pattern that appears due to the multiple diffraction of X-rays in the crystal is -60° from -180° to -60°. It comprises three 120° rotationally symmetrical peak regions with 3-fold rotational symmetry repeating every 120° from ° to 60° and from 60° to 180°. Each 120° rotationally symmetric peak region has mirror symmetry with respect to the plane containing the line indicating the angle of φ at the center of the range of φ of the region and the [0001] axis as the mirror plane. Prepare. Each −region comprises a −L region and a −R region having rotational symmetry with respect to each other about the intersection of the [0001] axis and the line indicating the φ angle in the middle of the φ range of that region. Each + region comprises a +L region and a +R region that are rotationally symmetrical to each other about the intersection of the [0001] axis and the line indicating the φ angle in the middle of the φ range of that region. The maximum peak intensity ratio obtained by dividing the maximum peak intensity of X-ray multiple diffraction peaks in at least one of the -L region, each -R region, each +L region, and each +R region by the background intensity is 500 or more. be. The gallium nitride crystal substrate of the present embodiment has high symmetry and a large maximum peak intensity ratio compared to conventional gallium nitride crystal substrates, and thus has high crystal quality.

[2] In the gallium nitride crystal substrate, at least one of the X-ray multiple diffraction peaks exhibiting the maximum peak intensity ratio in each −L region, each −R region, each +L region, and each +R region is at least (1 −100) and (−1101). Such a gallium nitride crystal substrate has higher symmetry and thus higher crystal quality.

[3] The gallium nitride crystal substrate may be a self-supporting substrate. Such gallium nitride crystal substrates are self-supporting substrates with high crystal quality, so they are suitably used for manufacturing semiconductor devices.

[4] The gallium nitride crystal substrate may have a thickness of 250 μm or more. Such a gallium nitride crystal substrate has a high crystal quality and a thickness of 250 μm or more, and is therefore suitably used for manufacturing semiconductor devices.

[5] A method for evaluating the crystal quality of a gallium nitride crystal substrate according to another embodiment of the present invention uses an X-ray lens on the incident side to evaluate the crystal quality of a gallium nitride crystal substrate whose main surface is the {0001} plane. A 0.3 mm × 0.3 mm cross slit point-focused CuKα ₁ monochromatic X-ray incident X-ray parallel beam with a small divergence angle was used. This is a method of evaluation by X-ray diffraction measurement using a .27° collimator and a graphite plate monochromator. placing the gallium nitride crystal substrate on the sample stage so that its [1-100] direction coincides with the direction of φ=0°; A step of calibrating the position of the gallium nitride crystal substrate by aligning the χ-axis, a measurement φ angle step of 0.02°, and a measurement time of 1 second to 50 seconds at each step depending on the measurement φ angle range. In a range of measurement conditions, the [0001] axis was the rotation axis at the incident angle ω and the diffraction angle 2θ of the 0001 symmetrical forbidden Bragg reflection (i.e., the incident angle ω: 8.5428° and the diffraction angle 2θ: 17.0856°). and C. performing a φ scan pattern measurement. From the high symmetry of the X-ray multiple diffraction peak obtained from the φ scan pattern measurement and the maximum peak intensity ratio obtained by dividing the maximum peak intensity of the X-ray multiple diffraction peak by the background intensity, gallium nitride crystal Evaluate the crystalline quality of the substrate. In the method for evaluating the crystal quality of a gallium nitride crystal substrate of the present embodiment, in X-ray diffraction measurement using a small divergence angle incident X-ray parallel beam, the symmetry of the X-ray multiple diffraction peaks obtained from the above φ scan pattern measurement is The higher the peak intensity ratio of the X-ray multiple diffraction peaks, the higher the crystal quality of the gallium nitride crystal substrate, and the crystal quality of the gallium nitride crystal substrate can be evaluated in detail.

[6] In the φ scan pattern measurement in the method for evaluating the crystal quality of the gallium nitride crystal substrate, when φ is measured from −180° to 180°, φ is from −180° to −150° and from −150°. -120° to -120° to -90°, -90° to -60°, -60° to -30°, -30° to 0°, 0° to 30°, 30° to 60°, 60° to 90°, 90° to 120°, 120° to 150°, and 150° to 180°. In the X-ray multiple diffraction peaks of the two regions, the reference peak intensity ratio obtained by dividing the peak intensity of the reference peak derived from a plurality of arbitrarily specified predetermined crystal planes by the background intensity From the small variation of the contrast peak relative intensity ratio, which is the relative ratio of the contrast peak intensity ratio obtained by dividing the peak intensity of the contrast peak derived from another predetermined multiple crystal planes specified by the background intensity , the crystal quality of the gallium nitride crystal substrate can be evaluated. In this gallium nitride crystal substrate crystal quality evaluation method, the smaller the variation in the relative peak relative intensity ratio, the higher the crystal quality of the gallium nitride crystal substrate, and the more detailed the crystal quality of the gallium nitride crystal substrate can be evaluated. can.

[7] In the φ scan pattern measurement in the method for evaluating the crystal quality of the gallium nitride crystal substrate, when φ is measured from −180° to 180°, φ is from −180° to −150° and from −150°. -120° to -120° to -90°, -90° to -60°, -60° to -30°, -30° to 0°, 0° to 30°, 30° to 60°, 60° to 90°, 90° to 120°, 120° to 150°, and 150° to 180°. The φ-axis corresponding to each segmented region can be aligned for measurement, and the angle at which the φ-axis corresponding to each segmented segmented region can be set to the center φ angle of the φ range of the segmented region. This method for evaluating the crystal quality of a gallium nitride crystal substrate can obtain more precise X-ray multiple diffraction peaks, so that the crystal quality of the gallium nitride crystal substrate can be evaluated in more detail.

[Details of the embodiment of the present invention]
Embodiments of the invention are explained in detail below on the basis of the drawings. First, a method for evaluating the crystal quality of a GaN (gallium nitride) crystal substrate will be described, and then a GaN crystal substrate with high crystal quality evaluated by this evaluation method will be described. In the specification, claims and drawings of the present application, a specific crystal plane of a hexagonal GaN crystal is represented by (hkil), and a set of equivalent crystal planes is represented by {hkil} from crystal symmetry. Represented by Also, a specific direction is represented by [hkil], and a set of equivalent directions based on crystal symmetry is represented by <hkil>. where h, k, i, and l are the Miller indices. Here, there is a relationship i=-(h+k). Miller indices are usually represented by 0, natural numbers, and numbers with a bar above them, but in the specification and drawings, natural numbers with a bar are represented with a minus sign in front of the natural number. . For example, the notation of (11-20) is read as the crystal plane of "ichi-ichi-ni/bar-zero", and the notation of [1-100] is the direction of "ichi-ichi/bar-zero-zero". and read.

<Evaluation Method for Crystal Quality of GaN Crystal Substrate>
The method for evaluating the crystal quality of a GaN (gallium nitride) crystal substrate according to the present embodiment is to measure the crystal quality of a GaN crystal substrate whose main surface is the {0001} plane by using an X-ray lens on the incident side and measuring the crystal quality by 0.3 mm × 0.3 mm. A 0.27° collimator and a 0.27° collimator with a 0.27° slit for reflectance measurement on the receiving side were used with a small divergence angle incident X-ray beam of CuKα ₁ monochromatic X-rays point-focused by a 0.3 mm cross-slit. This is a method of evaluation by X-ray diffraction measurement using a graphite plate monochromator. A step of placing a GaN crystal substrate on a sample stage so that its [1-100] direction coincides with the direction of φ=0°; °) and z-scan and χ-axis alignment, a step of calibrating the position of the GaN crystal substrate, a measurement φ angle step of 0.02°, and depending on the measurement φ angle range, Under the measurement conditions where the measurement time is in the range of 1 second to 50 seconds, at the incident angle ω and the diffraction angle 2θ of the 0001 symmetrical forbidden Bragg reflection (that is, the incident angle ω: 8.5428° and the diffraction angle 2θ: 17.0856°), and a step of performing φ scan (also referred to as Reinninger scan, hereinafter the same) pattern measurement with the [0001] axis as the rotation axis. From the high symmetry of the X-ray multiple diffraction peak obtained from the φ scan pattern measurement and the maximum peak intensity ratio obtained by dividing the maximum peak intensity of the X-ray multiple diffraction peak by the background intensity, gallium nitride crystal Evaluate the crystalline quality of the substrate. In the method for evaluating the crystal quality of a gallium nitride crystal substrate of the present embodiment, in X-ray diffraction measurement using a small divergence angle incident X-ray parallel beam, the symmetry of the X-ray multiple diffraction peaks obtained from the above φ scan pattern measurement is The higher the peak intensity ratio of the X-ray multiple diffraction peaks, the higher the crystal quality of the gallium nitride crystal substrate, and the crystal quality of the gallium nitride crystal substrate can be evaluated in detail.

In the method for evaluating the crystal quality of a GaN crystal substrate according to the present embodiment, in X-ray diffraction measurement of a GaN crystal substrate having {0001} plane as a specific crystal plane of a single crystal, in addition to a large diffraction peak due to 0002 symmetrical Bragg reflection, annihilation The diffraction peak observed at the position of the 0001 symmetrical forbidden Bragg reflection determined from the structure factor indicated by the law is utilized. Such diffraction is multiple diffraction at two or more crystal planes with high diffraction intensity inside the crystal that are not parallel to the crystal surface, and is also called Umweganregung (long-distance) diffraction. Such multiple diffraction is often observed as the single crystal approaches a perfect crystal. In addition, since the atoms of the crystal structure of the single crystal material are not the rigid sphere model in the theoretical calculation, but have an atomic cloud with a shape extending in the direction of the atomic bond, the fact that the structure factor is not zero is the background. This is considered to be the cause of the observation. Such a background region is a region where the multiple diffraction of X-rays does not occur. The inventors of the present invention have conducted φ scan pattern measurement with the [0001] axis as the rotation axis at the incident angle ω and the diffraction angle 2θ of the 0001 symmetric forbidden Bragg reflection using a small divergence incident X-ray parallel beam. Non-destructively evaluating the crystal quality inside the vicinity of the surface of the GaN crystal substrate by utilizing X-ray multiple diffraction peaks observed by X-ray multiple diffraction on crystal planes not parallel to the crystal surface of the present embodiment. A method for evaluating the crystal quality of GaN crystal substrates has been completed. In the method for evaluating the crystal quality of the GaN crystal substrate of the present embodiment, multiple diffraction of X-rays on the inner crystal plane near the crystal surface occurs depending on the rotation angle φ of the φ scan with the [0001] axis as the rotation axis. Therefore, it is suitable for evaluation of the uniformity within the crystal plane near the surface of the crystal substrate.

In the evaluation method using the half width of the diffraction peak due to 0002 symmetrical Bragg reflection of the ω rocking curve in the ω-2θ measurement, which is a conventional evaluation method of the crystal quality of a GaN crystal substrate, the crystal in the depth direction from the crystal surface Focusing on the structure, the incident angle ω of the 0002 symmetric Bragg reflection is 17.2833°. On the other hand, in the method for evaluating the crystal quality of the GaN crystal substrate of the present embodiment, the incident angle ω of the 0001 symmetrical forbidden Bragg reflection is 8.5428°, which is approximately half or less of the incident angle of the 0002 symmetrical Bragg reflection. Since the crystal surface is a square, the X-ray transmission thickness of the sample on the crystal surface is approximately half, and it is possible to evaluate the crystal quality close to the crystal surface. be done. Furthermore, since the incident X-ray is a point-focused parallel beam and has a dot shape of 0.3 mm × 0.3 mm, the in-plane uniformity of the crystal structure in the plane near the crystal surface can be obtained by moving the measurement location within the plane. etc. can be evaluated non-destructively.

(Setting of incident X-ray side of X-ray diffractometer)
The X-ray diffractometer can detect monochromatic X-rays with a monochromator, and there are no particular restrictions other than using a parallel beam with a small divergence angle. Adoption of the goniometer method is preferable. The optical axis of the X-ray is fixed, and the incident angle ω, the diffraction angle (reflection angle) 2θ, the coordinates x, y, z of the GaN crystal substrate position, the rotation angle φ of the GaN crystal substrate, the tilt angle χ of the GaN crystal substrate, etc. set.

(Use of small divergence angle incident X-ray parallel beam)
In the method for evaluating the crystal quality of a GaN crystal substrate according to the present embodiment, the incident X-ray divergence angle of the X-ray diffractometer is is required to be small, a parallel beam method as shown in FIG. 1 is used. In the parallel beam method, an X-ray lens and/or a monocapillary is used to obtain a parallel beam. A beam diameter of 0.3 mm×0.3 mm is obtained by the parallel beam method using an X-ray lens, and a beam diameter of 0.1 mm is obtained by the parallel beam method using a monocapillary. The X-ray divergence angle is the same whether the X-ray lens is used or the monocapillary is used, and the intensity of X-rays decreases when the monocapillary is used. Therefore, the parallel beam method using the X-ray lens is preferable. In this embodiment, an X-ray lens is used to point-focus a 0.3 mm × 0.3 mm cross-slit X-ray parallel beam with _a small divergence angle. A spectroscope for selecting X-rays is arranged to generate CuKα1 monochromatic X-rays, but it is also possible to arrange _a spectroscope for selecting CuKα1 monochromatic X _- rays on the incident side.

(Setting the sample stage of the X-ray diffractometer)
0002 Unlike the ω-2θ measurement that measures symmetrical Bragg reflection, in the case of φ scanning, the rotation axis φ of the sample table shown in FIG. First, the observed value of the Bragg reflection intensity fluctuates due to the φ scan, and the GaN crystal substrate, which is the sample, has an off angle. A two-axis tilted sample table is prepared for this correction, but in the case of this φ scan measurement, the measurement range is narrow (for example, 30°). Since the measurement is performed with the axis and the rotation axis aligned, the [0001] direction of the GaN crystal substrate and the rotation axis φ of the sample stage will coincide within the measurement range, and a biaxially inclined sample stage must not be used. good too. In addition, in this φ scan measurement, as described later, when measuring by dividing into 12 segmented regions of 30° each, the φ axis corresponding to the segmented region is set up and measured, so that the φ scan May not be used as the intensity can be measured closer to the maximum.

(Setting on the light receiving part side of the X-ray diffraction device)
In order to detect diffracted X-rays with a small divergence angle of incident X-rays, a 0.27° collimator and a 0.27° slit for reflectance measurement are also used on the light receiving side. Also, since it is necessary to select CuKα ₁ monochromatic X-rays, a parallel plate graphite monochromator for CuKα ₁ monochromatic X-ray selection is used. When _a spectroscope for selecting CuKα1 monochromatic X _- rays is arranged on the incident side, the parallel plate graphite monochromator for selecting CuKα1 monochromatic X-rays may be omitted in some cases.

(Step of placing GaN crystal substrate on sample table)
In the step of placing the GaN crystal substrate on the sample table, the gallium nitride crystal substrate is placed on the sample table of the X-ray diffractometer so that its [1-100] direction coincides with the direction of φ=0°. Specifically, a φ scan of the 01-11 symmetrical Bragg reflection by the {01-11} plane having the same normal to the [1-100] direction and the φ angle is performed, and the position of the peak is set to φ=0 °. The value of χ at this time is 61.96°.

(Step of Calibrating Position of GaN Crystal Substrate)
Referring to FIG. 2, in the step of calibrating the position of the GaN crystal substrate, ω scanning is performed at and near the incident angle ω of 17.2889° when 2θ of {0002} symmetrical reflection is 34.5667°. , the ω rocking curve is measured, and the offset calibration of ω is performed by setting the value of ω at which the diffraction peak appears to be 17.2889°. Next, in the ω-2θ measurement using ω calibrated above, z is calibrated by performing a z scan and determining the value of z at which the peak appears. At ω-2θ and z calibrated and determined above, by setting the χ-axis by adjusting the angle of χ, χ=0° and the φ-axis and the {0001}-axis are aligned (that is, the crystal coincide with the axis of rotation of the crystal). Here, the setting of the χ axis by adjusting the angle of χ means that the angle of χ is changed in the range of 2° to 3° in steps of 0.3°, and ω scanning is repeatedly performed at the position of the 0002 symmetrical Bragg reflection, This means executing an optimization program that finds the value of χ that maximizes the peak value in each scan, and setting the offset value of the value of χ so that χ=0°.

(Measurement of forbidden Bragg reflection)
Referring to FIG. 3, in the step of measuring the φ scan pattern with the [0001] axis of 0001 symmetrical forbidden Bragg reflection as the axis of rotation, when the rotation angle φ of the GaN crystal substrate is scanned from −180° to 180°, A highly symmetrical diffraction pattern appears (see FIGS. 12-13). 2 and 3, the incident angle ω in the 0002 symmetrical Bragg reflection measurement is 17.2833°, whereas the incident angle ω in the 0001 symmetrical forbidden Bragg reflection measurement is 8.5428°, which is approximately half. Therefore, the crystal quality in the vicinity of the crystal surface can be known in detail.

(Evaluation of crystal quality)
The crystal quality near the crystal surface is evaluated from the symmetry of the X-ray multiple diffraction peaks obtained from the φ scan pattern measurement and the maximum peak intensity ratio obtained by dividing the maximum peak intensity of the X-ray multiple diffraction peaks by the background intensity. . Here, the higher the crystal quality of the GaN crystal substrate, the higher the symmetry of the X-ray multiple diffraction peaks and the higher the maximum peak intensity ratio. It is evaluated that the crystal quality of the GaN crystal substrate is higher as the value increases.

(Relationship Between Measurement of Forbidden Bragg Reflection and Crystal Structure of GaN Crystal Substrate)
The relationship between the φ scan pattern measurement with the [0001] axis of 0001 symmetrical forbidden Bragg reflection as the axis of rotation and the crystal structure of the GaN crystal substrate will be described below.

(1) Crystal Structure of GaN Crystal Substrate The GaN crystal forming the GaN crystal substrate has a hexagonal wurtzite crystal structure. The lattice constants are a (a ₁ =a ₂ =a ₃ ) of 0.31891 nm and c of 0.51855 nm. As crystal planes, the c plane which is the {0001} plane shown in FIG. 4, the (1/2) c plane which is the {0002} plane shown in FIG. 5, the m plane which is the {10-10} plane shown in FIG. There is the a-plane which is the {11-20} plane shown in FIG. FIG. 8 shows the arrangement of Ga and N atoms in the crystal lattice.

(2) Relationship between X-ray Incidence Direction and Crystal Orientation in GaN Crystal Substrate The GaN crystal forming the GaN crystal substrate has a hexagonal wurtzite crystal structure, and space group No. It has a 6-fold symmetry of 186 P6 ₃ mc, but a 3-fold rotational symmetry in the plane perpendicular to the c-axis. Therefore, the GaN crystal substrate has the crystal orientation shown in FIG. There are three regions: A region (φ from −60° to 60°), B region (φ from 60° to 180°) and C region (φ from −180° to −60°). Divide into areas.

The A region has a mirror symmetry with respect to the plane containing the line indicating φ=0° and the [0001] axis, and the A ₋ region (the region where φ is from −60° to 0°) and A ₊ It is divided into regions (regions where φ is from 0° to 60°). The B region includes the B ₋ region (the region where φ is from 60° to 120°) and the B ₊ region, which have mirror symmetry with respect to the plane containing the line indicating φ=120° and the [0001] axis. (the area where φ is from 120° to 180°). Region C has mirror symmetry with respect to the plane containing the line indicating φ=−120° and the [0001] axis, and the region C ₋ (region where φ is from −180° to −120°) and has mirror symmetry with each other. It is divided into C ₊ regions (regions where φ is from −120° to −60°).

The A ₋ region further has rotational _symmetry with respect to the intersection of the line indicating φ=−30° and the [0001] axis. and A- _R region (φ from -30° to 0°). The A ₊ region consists of the A _{+ L} region (the region where φ is 0° to 30°) and the A _{+ R} region, which have rotational symmetry with respect to the intersection of the line indicating φ=30° and the [0001] axis. (φ is from 30° to 60°). The B _- region is further divided into the B- _L region (the region where φ is from 60° to 90°) _and the B- It is divided into _R region (φ is from 90° to 120°). The B ₊ region further includes the B _{+ L} region (the region where φ is from 120° to 150°) and the B _{+ R} region (φ is from 150° to 180°). The C _- region further has rotational symmetry around the intersection of the line indicating φ=−150° and the [0001] axis, and the C- _L region (the region where φ is from −180° to −150°). and a C- _R region (φ from -150° to -120°). The C ₊ region further has rotational symmetry with respect to the intersection of the line indicating φ = -90° and the [0001] axis, and the C _{+ L} region (the region where φ is from -120° to -90°). and the C _+R region (φ from -90° to -30°). In this way, the crystal planes perpendicular to the c-axis are φ from −180° to 180°, C _−L region, C _−R region, C _+L region, C _+R region, A −L region, A _−L region, A It is divided into 12 segmented areas: _-R area, A _+L area, A _+R area, B- _L area, B- _R area, B _+L area and B _+L area.

In the φ scan pattern measurement, the C _-L region from -180° to -150°, the C _-R region from -150° to -120°, the C _+L region from -120° to -90°, C _+R area from -90° to -60°, A- _L area from -60° to -30°, A- _R area from -30° to 0°, A ₊ from 0° to 30° _L region, A _+R region from 30° to 60°, B- _L region from 60° to 90°, B- _R region from 90° to 120°, B _+L region from 120° to 150° , and 150 ° to 180 ° B _{+ L} region divided into 12 segmented regions every 30 °, in the X-ray multiple diffraction peaks of at least two regions of each segmented region, an arbitrarily specified predetermined plurality The contrast peak derived from another predetermined crystal plane arbitrarily specified other than the reference peak to the reference peak intensity ratio obtained by dividing the peak intensity of the reference peak derived from the crystal plane by the background intensity. It is preferable to evaluate the crystal quality of the gallium nitride crystal substrate from the small variation of the relative peak intensity ratio obtained by dividing the intensity by the background intensity. Here, the variation of the contrasting peak relative intensity ratio is any of the variance, standard deviation, and range (difference between the maximum value and the minimum value) of the corresponding contrasting peak relative intensity ratio in at least two regions of each segmented region. can be a statistic of In this gallium nitride crystal substrate crystal quality evaluation method, the smaller the variation in the relative peak relative intensity ratio, the higher the crystal quality of the gallium nitride crystal substrate, and the more detailed the crystal quality of the gallium nitride crystal substrate can be evaluated. can.

In the φ scan pattern measurement, the C _-L region from -180° to -150°, the C _-R region from -150° to -120°, the C _+L region from -120° to -90°, C _+R area from -90° to -60°, A- _L area from -60° to -30°, A- _R area from -30° to 0°, A ₊ from 0° to 30° _L region, A _+R region from 30° to 60°, B- _L region from 60° to 90°, B- _R region from 90° to 120°, B _+L region from 120° to 150° , and from 150 ° to 180 °, the B _{+ L} region is divided into 12 segmented regions every 30 °, and each segmented region is measured by setting the φ axis corresponding to each segmented region, and each segment It is preferable that the angle at which the φ axis corresponding to is set is the central φ angle of the φ range of the segmented region. That is, as shown in Table 1, the angles at which the φ axis is set in each segmented region are −165° for the C _−L region, −135° for the C _−R region, −105° for the C _+L region, and −105° for the C +L region. C _+R area -75°, A _-L area -45°, A _-R area -15°, A _+L area 15°, A _+R area 45°, B _-L area 75° , the _BR region is 105°, the B _+L region is 135°, and the B _+L region is 165°. This method for evaluating the crystal quality of a gallium nitride crystal substrate can obtain more precise X-ray multiple diffraction peaks, so that the crystal quality of the gallium nitride crystal substrate can be evaluated in more detail.

12 to 15, the specific X-ray multiple diffraction patterns appearing in the measurement of the GaN crystal substrate when φ in the X-ray incident direction ranges from −180° to 180° are: Three 120° rotationally symmetrical peak regions (A region, B region, and C region) with 3-fold rotational symmetry repeating every 120° from -60° to 60°, and from 60° to 180° Prepare. Each 120° rotationally symmetric peak region (A region, B region, or C region) has a φ angle (φ = 0°, φ = 120°, or φ = -120°) at the center of the φ range for that region. and the plane containing the [0001] axis as the reflection plane, the − area (A ₋ area, B ₋ area, or C ₋ area) and the + area (A ₊ area, B ₊ area , or C ₊ region). Each − region (A ₋ region, B ₋ region, or C ₋ region) has a φ angle (φ=−30°, φ=90°, or φ=−150°) at the center of the φ range of that region. -L region (A _-L region, B _-L region, or C _-L region) and -R region (A _-R region, B- _R region or C- _R region). Each + region (A ₊ region, B ₊ region, or C ₊ region) has a φ angle (φ = 30°, φ = 150°, or φ = -90°) at the center of the φ range for that region. +L region (A _+L region, B _+L region, or C _+L region) and +R region (A _+R region, B _+R region, or C _+R region). Here, the symmetry of the X-ray multiple diffraction pattern and the “symmetry” in the mirror symmetry relationship means that the positions of the background region (the region where X-ray multiple diffraction does not occur) and the X-ray multiple diffraction peak region are symmetrical. and does not mean that the number and intensity of each X-ray multiple diffraction peak in the X-ray multiple diffraction peak region are symmetrical.

As shown in FIG. 14, if the A _{+ R} region is the basic segmental region, the A _{+ L} region is the rotational basic segmental region that has a rotationally symmetrical relationship with the A _{+ R} region, and the A _−R region is the A ₊ It is a reflection basic segmented area having a relationship of reflection symmetry with the _L area, and the A- _L area is a reflection basic segmented area having a relationship of reflection symmetry with the A _+R area. Considering such symmetry, when the two X-ray multiplex diffraction peak groups present in the A _{+ R} region are denoted by mp1 and mp2, the two X-ray multiplex diffraction peak groups present in the A _{+ L} region are mp1 and mp2 can be denoted as mp1′ and mp2′ by adding a “′” symbol to indicate rotational symmetry. The two multiple peak groups present in the A- _R region are denoted as (mp1′) _m and (mp2′) _m by adding the “() _m ” symbol indicating reflection symmetry to mp1′ and mp2′. can be done. The two X-ray multiple diffraction peak groups present in the A- _L region can be expressed as (mp1) _m and (mp2) _m by attaching the "() _m " symbol indicating reflection symmetry to mp1 and mp2. can.

<GaN crystal substrate>
12 to 15, the GaN (gallium nitride) crystal substrate of this embodiment has a {0001} plane as the main surface, and a 0.3 mm×0.3 mm crystal substrate on the incident side using an X-ray lens. A 0.27° collimator with a 0.27° slit for reflectance measurement on the receiving side and graphite for CuKα1 monochromatic X _- ray selection using a small divergence incident X-ray parallel beam point-focused by a cross-slit of X-ray diffraction measurement using a flat plate monochromator is performed under the measurement conditions where the measurement φ angle step is 0.02° and the measurement time at each step is in the range of 1 to 50 seconds depending on the measurement φ angle range. In the φ scan pattern measurement with the [0001] axis as the rotation axis at the incident angle ω and the diffraction angle 2θ (that is, the incident angle ω: 8.5428 ° and the diffraction angle 2θ: 17.0856 °) of the symmetrical forbidden Bragg reflection, [1 -100] direction is φ = 0°, the X-ray multiple diffraction pattern that appears due to the multiple diffraction of X-rays in the crystal in the measurement from -180° to 180° when φ is from -180° to - Three 120° rotationally symmetrical peak regions (A region, B region and C ). Each 120° rotationally symmetric peak region (A region, B region, or C region) has a φ angle (φ = 0°, φ = 120°, or φ = -120°) at the center of the φ range for that region. and the plane containing the [0001] axis as the reflection plane, the − area (A ₋ area, B ₋ area, or C ₋ area) and the + area (A ₊ area, B ₊ area , or C ₊ region). Each − region (A ₋ region, B ₋ region, or C ₋ region) has a φ angle (φ=−30°, φ=90°, or φ=−150°) at the center of the φ range of that region. -L region (A _-L region, B _-L region, or C _-L region) and -R region (A _-R region, B- _R region or C- _R region). Each + region (A ₊ region, B ₊ region, or C ₊ region) has a φ angle (φ = 30°, φ = 150°, or φ = -90°) at the center of the φ range for that region. +L region (A _+L region, B _+L region, or C _+L region) and +R region (A _+R region, B _+R region, or C _+R region). Each -L area (A- _L area, B- _L area, and C- _L area), each _-R area (A- _R area, B- _R area, and C-R area) ), each +L region (A _+L region, B _+L region, and C _+L region segmented regions), and each +R region (A _+R region, B _+R region, and C _+R region segmented regions). The maximum peak intensity ratio obtained by dividing the maximum peak intensity of X-ray multiple diffraction peaks in at least one of the regions) by the background intensity is 500 or more. The gallium nitride crystal substrate of the present embodiment has high symmetry and a large maximum peak intensity ratio compared to conventional gallium nitride crystal substrates, and thus has high crystal quality. From the viewpoint of high crystal quality, the maximum peak intensity ratio is preferably 1000 or more, more preferably 2000 or more.

Here, "symmetry" in the rotational symmetry and reflection symmetry of the X-ray multiple diffraction pattern means that the positions of the multiple peak regions are symmetrical, and each X-ray in the multiple peak region It does not imply that the positions and intensities of the multiple diffraction peaks are symmetrical.

The GaN crystal substrate of the present embodiment has a {0001} plane as a principal surface, and a _single -color CuKα1 X-ray beam point-focused by a 0.3 mm × 0.3 mm cross slit using an X-ray lens on the incident side. 0001 symmetrical forbidden Bragg in X-ray diffraction measurements using a 0.27° collimator with a 0.27° slit on the receiving side for reflectance measurements and a graphite plate monochromator using a divergent incident X-ray parallel beam. φ scan pattern measurements around the [0001] axis at the incident angle ω and diffraction angle 2θ of reflection (i.e., incident angle ω: 8.5428° and diffraction angle 2θ: 17.0856°) provide detailed and accurate X From the viewpoint of obtaining a line multiplex diffraction pattern, the measurement conditions are such that the measurement φ angle step is 0.02° and the measurement time at each step ranges from 1 second to 50 seconds depending on the measurement φ angle range.

The rotation axis is the [0001] axis at the incident angle ω and the diffraction angle 2θ of the 0001 symmetrical forbidden Bragg reflection of the GaN crystal substrate of the present embodiment (that is, the incident angle ω: 8.5428° and the diffraction angle 2θ: 17.0856°). 12 to 21 show typical examples of φ scan pattern measurement. 12 and 13 both show examples of X-ray multiplex diffraction patterns in the region where φ is from −180° to 180°. Axes are logarithmic. From FIG. 12, the positions and intensity values of the X-ray multiple diffraction peaks can be easily read, but the background intensity is difficult to read. Here, the intensity of the X-ray multiple diffraction peak whose unit is the number of counts per second (cps) is almost unchanged in the range of 1 to 50 seconds for each step with φ of 0.02°. , the multiple diffraction peak measurement is measured at, for example, 3 seconds for the purpose of shortening the peak time. As for the background intensity measurement, the approximate value of the background intensity can be read from FIG. 13, but the variation is large. Therefore, in the measurement of the background intensity, the measurement time for each step is increased to 10 seconds or longer (preferably in the range of 15 to 50 seconds), and the unit is the number of counts per second (cps). Try to get more than one digit of precision. By increasing the number of measurements and performing measurements related to the background within a narrow range of φ, the variation can be reduced, the value can be read with an accuracy of two significant figures, and the measurement time can be shortened. .

The maximum peak intensity ratio of the X-ray multiple diffraction peaks is obtained by dividing the maximum peak intensity of the X-ray multiple diffraction peaks by the background intensity. Here, the maximum peak intensity reads the intensity value of the apex of the peak. The background intensity is the value of the intensity of the flat portion of the background on the low φ side of the peak with accuracy of 2 digits or more and the intensity of the flat portion of the background on the high φ side of the peak with accuracy of 2 digits or more. Read the values and take their average.

From FIG. 12 and the like, the maximum peak intensity of the X-ray multiple diffraction peak is about 6900 cps and the background intensity is about 3 cps, so the maximum peak intensity ratio is about 2300, which is 500 or more.

FIG. 14 shows an X-ray multiplex diffraction pattern of region A, which is a 120° rotationally symmetrical peak region from −60° to 60° in φ scanning of a GaN crystal substrate, which is a basic pattern. In the A region where φ is from −60° to 60°, there are the A ₋ region consisting of the A _−L region and the A _−R region and the A + region consisting of the A _+L region and the A _{+ R} region in this order. do. Here, the A ₋ region and the A ₊ region have mirror symmetry, the A _−L region and the A _−R region have rotational symmetry, and the A _+L region and the A _+R region and have rotational symmetry with each other. That is, if the A _{+ R} region is the basic segmental region, the A _{+ L} region is the rotational basic segmental region that has a rotationally symmetrical relationship with the A + _R region, and the A − _L region is a mirror image of the A _{+ R} region. It is a reflection basic segmented area having a symmetrical relationship, and the A- _R area is a rotational reflection basic segmental area having a relationship of reflection symmetry with the A _+L area. Considering such symmetry, as described above, if the two multiple peak groups present in the A _{+ R} region are denoted by mp1 and mp2, the two multiple peak groups present in the A _{+ L} region are It can be represented as mp1′ and mp2′ using the “′” symbol to indicate antisymmetry. The two multiple peak groups present in the A- _L region can be represented as (mp1) _m and (mp2) _m using the "() _m " symbol indicating reflection symmetry. The two multiple peak groups present in the A- _R region are labeled (mp1′) _m and (mp2′) _m using the “′” symbol for rotational symmetry and the “() _m ” symbol for mirror symmetry. It can be expressed as.

The pattern of the A ₊ region (the region where φ is from 0° to 60°) and the A ₋ region (the region where φ is from −60° to 0°) of the basic pattern shown in FIG. FIG. 16 shows a comparison with the pattern obtained. As shown in FIG. 16, (mp2′) _m , (mp1′) _m , (mp1) _m , and (mp2) _m , which are X-ray multiplex diffraction peaks in the A ₋ region, and X-ray multiple diffraction peaks in the A ₊ region. Since the positions of the X-ray multiple diffraction peaks of the diffraction peak groups mp2′, mp1′, mp1, and mp2 are well matched, (mp2′) _m and (mp1′) in the A ₋ region _m , (mp1) _m , and (mp2) _m , and mp2′, mp1′, mp1, and mp2 of the A ₊ region may each have mirror symmetry about the φ=0° axis. I can confirm.

FIG. 17 shows a comparison between the pattern of the A ₊ region (the region where φ is −60° to 0°) of the basic pattern shown in FIG. 14 and the pattern in which the A ₊ region is inverted around φ=30°. . As shown in FIG. 17, mp1′ and mp2′, which are X-ray multiple diffraction peak groups in the A _{+ L} region, and mp1 and mp2, which are X-ray multiple diffraction peak groups in the A _{+ R} region, each X-ray The positions of multiple diffraction peaks are in good agreement, indicating that mp1′ and mp2′ in the A _+L region and mp1 and mp2 in the A _+R region are rotationally opposite to each other about the φ=30° axis. It can be confirmed that it has symmetry.

In the above, the position of each X-ray multiple diffraction peak group is well matched means that the difference in the number of peaks of each X-ray multiple diffraction peak group having a symmetrical relationship is 2 or less, and that the symmetrical relationship It means that there are two or less X-ray multiple diffraction peak groups whose peak positions differ by more than 2°. Here, the fact that the positions of the respective X-ray multiple diffraction peak groups match well means that the GaN crystal substrate has high symmetry.

As described above, the GaN crystal substrate of this embodiment has high symmetry and a high maximum peak intensity ratio, and thus has high crystal quality.

FIG. 18 shows mp1 and mp2, which are X-ray multiple diffraction peak groups in the A _+R region, FIG. 19 shows mp1′ and mp2′, which are X-ray multiple diffraction peak groups in the A _+L region, and FIG. (mp1') _m and (mp2') _m , which are X-ray multiple diffraction peaks in the _-R region, and (mp1) _m and (mp2), which are X-ray multiple diffraction peaks in the A- _L region _m is indicated and peak numbers 1-14 are assigned to each peak. Here, the A _{+ R} region is the basic segmental region described above, the A _{+ L} region is the rotational basic segmental region that has a rotationally symmetrical relationship with the A _{+ R} region, and the A + _R region is the A _{+ R} region. is a rotationally symmetrical basic _segmental area that has a relationship of reflection symmetry with the A _{+L area} , which is a rotationally symmetrical basic segmental area with a rotationally symmetrical relationship with It is a reflection basic segmented area having the relationship of In FIGS. 18 to 21, the dashed ellipse indicates the number and/or shape of peaks applied between mp1 and mp1′, (mp1′) _m and (mp1) _m in a symmetrical relationship therewith. shows the part where the changes, and the part attached with the solid ellipse is the number and / or shape of the peaks applied between mp2 and mp2', (mp2') _m and (mp2) _m in a symmetrical relationship with this indicates the part where changes.

Referring to FIG. 18, in the A _+R region, which is the basic segmented region, mp1 has six X-ray multiple diffraction peaks, and mp2 has eight X-ray multiple diffraction peaks. In FIG. 18, two peaks of peak number 2 and peak number 3 overlap. Referring to FIG. 19, in the A _+L region, which is the basic segmented region of rotation, mp1′ has 6 X-ray multiple diffraction peaks, and mp2′ has 7 X-ray multiple diffraction peaks. be done. In FIG. 19, two peaks of peak number 2 and peak number 3 of mp1' overlap and appear to be one peak, but it is confirmed from peak analysis that they are two peaks. It can be seen that the peak of peak number 11 does not appear in mp2'. Referring to FIG. 20, in the A- _R region, which is the basic segmented region of retroreflection, six X-ray multiple diffraction peaks are observed at (mp1′) _m and seven at (mp2′) _m . of X-ray multiple diffraction peaks can be seen. In FIG. 20, it can be seen that the peak of peak number 11 does not appear in (mp2′) _m . Further, referring to FIG. 21, in the A- _L region, which is the reflection basic segmental region, six X-ray multiple diffraction peaks are observed at (mp1) _m , and eight X-ray diffraction peaks are observed at (mp2) _m . Line multiple diffraction peaks are seen.

The values normalized by dividing the intensity of the peaks numbered 1 to 14 by the intensity of the peak numbered 8, which has the maximum intensity, and the shape of each peak are the structures of the GaN crystal in each sectioned region of the GaN crystal substrate. depends on It is believed that the difference in the number of peaks and the shape of the peaks between peak numbers 10 and 12 is caused by the circulatory relationship. Here, in the X-ray multiple diffraction measurement of an actual crystal, it is conceivable that the number of X-ray multiple diffraction peaks may be reduced due to overlapping or lack of X-ray multiple diffraction peaks. Therefore, in a perfect crystal (a crystal without defects theoretically derived from the crystal structure), mp1 and mp1′, (mp1)m and (mp1′) _m which are in a symmetrical relationship with this have six X It is believed that mp2 and mp2′, (mp2) _m and (mp2′) _m having a symmetrical relationship therewith have eight X-ray multiple diffraction peaks.

The positions, relative intensities, indexing and multiple diffraction types (at least two diffraction planes involved in multiple diffraction) of the mp1 and mp2 X-ray multiple diffraction peaks in the A _+R region are shown in Table 2. For the indexing of X-ray multiple diffraction peaks and identification of multiple diffraction types, refer to J. Blasing and A. Krost, “X-ray multiple diffraction (Umweganregung) in wurtzite-type GaN and ZnO epitaxial layers”, phys. sol., (a) 201, No. 4, 2004, pp. R17-R20 (Non-Patent Document 2). P1-P3 and P5-P10 in the indexed peak column in Table 2 indicate the indexed peaks shown in the above literature. In the above document, in the GaN crystal substrate of the present embodiment, the X-ray of peak number 8 derived from (1-100) and (-1101), which are the maximum peak intensities in the range of 360° scanning in many cases Multiple diffraction peaks are indexed as P6. Also described is an indexed peak P4 that has not been found at present in the GaN crystal substrate of this embodiment. Such indexed peak P4 has approximately the same φ as indexed P5 (0.002° smaller φ compared to P5) and is of multiple diffraction type (02-22)/(0-22-1). ing. Here, in the range where φ is 30°, only 10 X-ray multiple diffraction peaks are obtained in the above document, but 14 X-ray multiple diffraction peaks are found in this embodiment. . In the above literature, the occurrence of multiple reflection was calculated by Bragg reflection from two crystal planes, but the actual multiple reflection may occur from two or more crystal planes, and the new peak is generated from two or more crystal planes. This is considered to be the case caused by the surface.

18 to 21 and Table 2, one example of the A _+R region, A _+L region, A _-R region, and A _-L region in the GaN crystal substrate of the present embodiment is the maximum The X-ray multiple diffraction peak of peak number 8, which indicates the peak intensity ratio, is indexed peak P6, and is an X-ray multiple diffraction peak derived from at least (1-100) and (-1101). Since (1-100) and (-1101) are crystal planes with high diffraction intensity, it is considered that the closer the crystal is to a perfect crystal, the higher the intensity of such X-ray multiple diffraction peaks. Therefore, each -L region (each segmented region of A- _L region, B- _L region, and C- _L region), each -R region (each of A- _R region, B- _R region, and C- _R region) each +L region (A _+L region, B _+L region, and C _+L region), and each +R region (A _+R region, B _+R region, and C _+R region). In each segmented region), the peak ratio varies depending on the crystal structure of the GaN crystal substrate and its crystal quality. X-ray multiple diffraction peaks from at least (1-100) and (-1101) of number 8 (indexed peak P6) are preferred. Furthermore, each -L region (A _-L region, B _-L region, and C _-L region), each -R region (A _-R region, B _-R region, and C _-R region) each +L region (A _+L region, B _+L region, and C _+L region), and each +R region (A _+R region, B _+R region, and C _+R region). Each of the X-ray multiple diffraction peaks showing the maximum peak intensity ratio in each segmented region) is an X-ray multiple diffraction peak derived from at least (1-100) and (-1101) of peak number 8 (indexed peak P6). It is more preferable to have

Also, as shown in Table 2, the indexed peaks are assigned multiple diffraction types, each indicating at least two diffraction planes involved in multiple diffractions. The closer the crystal is to a perfect crystal, the peak intensity of the reference peak derived from a plurality of arbitrarily specified crystal planes among the indexed peaks clearly attributed to multiple diffraction type divided by the background intensity. It is the relative ratio of the contrast peak intensity ratio obtained by dividing the peak intensity of the contrast peak derived from another predetermined crystal plane arbitrarily specified other than the reference peak to the reference peak intensity ratio obtained by dividing the background intensity by the background intensity. It is considered that the variation in the contrast peak relative intensity ratio is reduced. Therefore, the crystal quality of the gallium nitride crystal substrate can be evaluated from the small variation of the contrast peak relative intensity ratio with respect to the reference peak intensity ratio.

The gallium nitride crystal substrate of this embodiment is preferably a self-supporting substrate. Such gallium nitride crystal substrates are self-supporting substrates with high crystal quality, so they are suitably used for manufacturing semiconductor devices. Here, the gallium nitride crystal substrate preferably has a thickness of 250 μm or more in order to be a self-supporting substrate.

The thickness of the gallium nitride crystal substrate of the present embodiment is preferably 250 μm or more, more preferably 300 μm or more, and even more preferably 400 μm or more. Such a gallium nitride crystal substrate has a high crystal quality and a thickness of 250 μm or more, and is therefore suitably used for manufacturing semiconductor devices. The diameter of the gallium nitride crystal substrate of the present embodiment is preferably 50 mm or more, more preferably 75 mm or more, and even more preferably 100 mm or more. Such a gallium nitride crystal substrate has a high crystal quality and a diameter of 50 mm or more, and is therefore suitably used for manufacturing semiconductor devices.

<Manufacturing Method of GaN Crystal Substrate>
The method for manufacturing a GaN crystal substrate according to the present embodiment includes the steps of growing a GaN crystal on a base substrate, and heat-treating the grown GaN crystal under high temperature and high pressure conditions of 900° C. or higher and 1800 atm or higher in a nitrogen gas atmosphere. and processing the heat-treated GaN crystal into a GaN crystal substrate. In the method of manufacturing a GaN crystal substrate according to the present embodiment, a GaN crystal substrate with high crystal quality is obtained by processing a GaN crystal that has been heat-treated under high-temperature and high-pressure conditions of 900° C. or higher and 1800 atm or higher.

The heat treatment of the grown GaN crystal is performed in a nitrogen atmosphere at a treatment temperature of 900° C. or higher, preferably 1000° C., from the viewpoint of improving the crystal quality of the GaN crystal substrate obtained by processing the GaN crystal after the heat treatment. ° C. or higher, and the treatment pressure is 1800 atm or higher, preferably 2000 atm or higher. Further, from the specifications of the manufacturing apparatus, the treatment temperature is preferably 1200° C. or less, and the treatment pressure is preferably 2000 atmospheres or less.

(Example 1)
1. Preparation of GaN Crystal Substrate A template substrate having a C-plane as a main surface was prepared by growing GaN on a sapphire substrate with a diameter of 76 mm by MOCVD (metal organic chemical vapor deposition). It was placed on a 20 mm thick SiC-coated carbon substrate holder and placed in the reactor of a HVPE (Hydride Vapor Phase Epitaxy) apparatus. After heating the inside of the reactor to 1020° C., GaCl gas generated by reacting HCl gas with Ga and NH ₃ gas were supplied into the reactor. In the step of growing the GaN layer on such a base substrate, the reactor temperature was maintained at 1020° C. for 29 hours, the growth pressure was 1.01×10 ⁵ Pa, and the partial pressure of GaCl gas was 6.52×10. ² Pa, the partial pressure of NH ₃ gas was 7.54×10 ³ Pa, and the partial pressure of HCl gas was 3.55×10 ¹ Pa. After completion of the GaN layer growth step, the temperature inside the reactor was lowered to room temperature to obtain a GaN crystal. The obtained GaN crystal was crystal-grown with the C-plane as the growth plane, the surface state of the growth plane was a mirror surface, and the thickness measured with a stylus-type film thickness meter was 3.5 mm.

The GaN crystal obtained by processing the outer circumference and the front and back surfaces of the GaN crystal was heat-treated. A hot isostatic press (HIP) apparatus manufactured by Kobe Steel was used for the heat treatment. The heat treatment conditions were 1200° C. for 20 hours in a nitrogen atmosphere of 2000 atmospheres. After the heat treatment, the GaN crystal was recovered from the HIP device and processed into a GaN crystal substrate having a diameter of 2 inches (50.8 mm) and a thickness of 400 μm.

2. X-Ray Diffraction of GaN Crystal Substrate The GaN crystal substrate was placed on a sample stage of an X-ray diffraction apparatus (PANalytical X'Pert MRD manufactured by Spectris Co., Ltd.) so that the <1-100> direction of the GaN crystal substrate coincided with the direction of φ=0°. Using a small divergence angle incident X-ray parallel beam (X-ray lens parallel beam, beam diameter: 0.3 mm × 0.3 mm, using a flat plate monochromator with a collimator (using a 0.27 ° slit)), 0002 Offset calibration of ω and z was calibrated. At the calibrated and determined ω-2θ and z, the χ axis was aligned by adjusting the χ angle so that χ = 0° and the φ axis and the [0001] axis were aligned (that is, the crystal The axis of rotation and the crystallographic axis were aligned). Here, the setting of the χ axis by adjusting the angle of χ means that the angle of χ is changed in the range of 2° to 3° in steps of 0.3°, and ω scanning is repeatedly performed at the position of the 0002 symmetrical Bragg reflection, This means executing an optimization program that finds the value of χ that maximizes the peak value in each scan, and setting the offset value of the value of χ so that χ=0°. Next, the position of the 0001 symmetrical forbidden Bragg reflection (incidence angle φ scan was measured at ω: 8.5428°, diffraction angle 2θ: 17.0856°. The results are summarized in Tables 2 and 3.

In the 12 segmented regions where φ is every 30°, the difference in the number of peaks in each X-ray multiplex diffraction peak group having a symmetrical relationship is 2 or less, and each X-ray multiplexed diffraction peak having a symmetrical relationship Since there were no more than two diffraction peak groups with a peak position difference of more than 2°, the positions of the respective X-ray multiple diffraction peak groups were well matched, and the GaN crystal substrate had high symmetry. In addition, in all of the 12 segmented regions with φ every 30 °, the X-ray multiplex diffraction peak showing the maximum peak intensity ratio corresponds to peak number 8 (indexed peak P6) (1-100) and It was an X-ray multiple diffraction peak derived from (-1101).

3. Formation of Epitaxial Layer on GaN Crystal Substrate and Photoluminescence Measurement A 3 μm thick GaN layer was grown as an epitaxial layer on the GaN crystal substrate by MOCVD. The emission intensity of photoluminescence of the grown epitaxial layer was measured, and the results are summarized in Table 4.

(Example 2)
A GaN crystal substrate was prepared in the same manner as in Example 1, except that the thickness of the GaN crystal substrate was 250 μm. °, diffraction angle 2θ: 17.0856°), and the results are summarized in Tables 2 and 3. In the 12 segmented regions where φ is every 30°, the difference in the number of peaks in each X-ray multiplex diffraction peak group having a symmetrical relationship is 2 or less, and each X-ray multiplexed diffraction peak having a symmetrical relationship Since there were no more than two diffraction peak groups with a peak position difference of more than 2°, the positions of the respective X-ray multiple diffraction peak groups were well matched, and the GaN crystal substrate had high symmetry. In addition, in all of the 12 segmented regions with φ every 30 °, the X-ray multiplex diffraction peak showing the maximum peak intensity ratio corresponds to peak number 8 (indexed peak P6) (1-100) and It was an X-ray multiple diffraction peak derived from (-1101). Also, in the same manner as in Example 1, an epitaxial layer was grown on a GaN crystal substrate, and the photoluminescence emission intensity of the epitaxial layer was measured.

(Example 3)
A GaN crystal substrate was prepared in the same manner as in Example 1, except that the heat treatment was performed at 1000° C. for 20 hours in a nitrogen atmosphere of 2000 atm. φ scan was measured at ω: 8.5428°, diffraction angle 2θ: 17.0856°), and the results are summarized in Tables 2 and 3. In the 12 segmented regions where φ is every 30°, the difference in the number of peaks in each X-ray multiplex diffraction peak group having a symmetrical relationship is 2 or less, and each X-ray multiplexed diffraction peak having a symmetrical relationship Since there were no more than two diffraction peak groups with a peak position difference of more than 2°, the positions of the respective X-ray multiple diffraction peak groups were well matched, and the GaN crystal substrate had high symmetry. In addition, in all of the 12 segmented regions with φ every 30 °, the X-ray multiplex diffraction peak showing the maximum peak intensity ratio corresponds to peak number 8 (indexed peak P6) (1-100) and It was an X-ray multiple diffraction peak derived from (-1101). Also, in the same manner as in Example 1, an epitaxial layer was grown on a GaN crystal substrate, and the photoluminescence emission intensity of the epitaxial layer was measured.

(Example 4)
A GaN crystal substrate was prepared in the same manner as in Example 1, except that the heat treatment was performed at 900° C. for 20 hours in a nitrogen atmosphere of 2000 atm. φ scan was measured at ω: 8.5428°, diffraction angle 2θ: 17.0856°), and the results are shown in Tables 2 and 3. In the 12 segmented regions where φ is every 30°, the difference in the number of peaks in each X-ray multiplex diffraction peak group having a symmetrical relationship is 2 or less, and each X-ray multiplexed diffraction peak having a symmetrical relationship Since there were no more than two diffraction peak groups with a peak position difference of more than 2°, the positions of the respective X-ray multiple diffraction peak groups were well matched, and the GaN crystal substrate had high symmetry. In addition, in all of the 12 segmented regions with φ every 30 °, the X-ray multiplex diffraction peak showing the maximum peak intensity ratio corresponds to peak number 8 (indexed peak P6) (1-100) and It was an X-ray multiple diffraction peak derived from (-1101). Also, in the same manner as in Example 1, an epitaxial layer was grown on a GaN crystal substrate, and the photoluminescence emission intensity of the epitaxial layer was measured.

(Reference example 1)
A GaN crystal substrate was prepared in the same manner as in Example 1, except that the heat treatment was performed at 1200° C. for 20 hours in a nitrogen atmosphere of 1 atm. φ scan was measured at ω: 8.5428°, diffraction angle 2θ: 17.0856°), and the results are shown in Table 3. Also, in the same manner as in Example 1, an epitaxial layer was grown on a GaN crystal substrate, and the photoluminescence emission intensity of the epitaxial layer was measured.

(Reference example 2)
A GaN crystal substrate was produced in the same manner as in Example 1, except that no heat treatment was performed. : 17.0856°), and the results are shown in Table 3. Also, in the same manner as in Example 1, an epitaxial layer was grown on a GaN crystal substrate, and the photoluminescence emission intensity of the epitaxial layer was measured.

As described above, the GaN crystal substrate wafers obtained in Examples 1 to 4 had high symmetry and high maximum peak intensity ratio. Further, with reference to Table 4, it was found that the epitaxial layers grown on the GaN crystal substrate wafers obtained in Examples 1 to 4 had high photoluminescence emission intensity and high crystal quality. rice field.

The embodiments and examples disclosed this time are illustrative in all respects and should not be considered restrictive. The scope of the present invention is indicated by the scope of the claims rather than the above-described embodiments and examples, and is intended to include meanings equivalent to the scope of the claims and all modifications within the scope.

Claims (3)

A method for evaluating the crystal quality of a gallium nitride crystal substrate having a {0001} principal plane by X-ray diffraction measurement using a small divergence angle incident X-ray parallel beam, comprising:
a step of placing the gallium nitride crystal substrate on a sample stage such that its [1-100] direction coincides with the direction of φ=0°;
0002 calibrating the position of the gallium nitride crystal substrate by ω-scan and z-scan at the diffraction angle 2θ of the symmetrical Bragg reflection and by aligning the χ-axis;
performing a φ scan pattern measurement with the [0001] axis as the rotation axis at the incident angle ω and the diffraction angle 2θ of the 0001 symmetrical forbidden Bragg reflection;
From the high symmetry of the X-ray multiple diffraction peaks obtained from the φ scan pattern measurement and the maximum peak intensity ratio obtained by dividing the maximum peak intensity of the X-ray multiple diffraction peaks by the background intensity, nitriding A method for evaluating the crystal quality of a gallium nitride crystal substrate, which evaluates the crystal quality of a gallium nitride crystal substrate. In the φ scan pattern measurement, when φ is measured from -180° to 180°, φ is from -180° to -150°, from -150° to -120°, from -120° to -90° , -90° to -60°, -60° to -30°, -30° to 0°, 0° to 30°, 30° to 60°, 60° to 90°, 90° to 120°, 120° to 150°, and 150° to 180° every 30° into 12 subregions,
A reference peak intensity obtained by dividing the peak intensity of a reference peak derived from a plurality of arbitrarily specified predetermined crystal planes by the background intensity in the X-ray multiple diffraction peaks of at least two regions of each of the segmented regions. Contrasting peak relative, which is the relative ratio of the contrasting peak intensity ratio obtained by dividing the peak intensity of the contrasting peak derived from a plurality of other predetermined crystal planes arbitrarily specified other than the reference peak with respect to the ratio, by the background intensity 2. The method for evaluating the crystal quality of a gallium nitride crystal substrate according to claim 1 , wherein the crystal quality of the gallium nitride crystal substrate is evaluated from the smallness of variation in intensity ratio. In the φ scan pattern measurement, when φ is measured from -180° to 180°, φ is from -180° to -150°, from -150° to -120°, from -120° to -90° , -90° to -60°, -60° to -30°, -30° to 0°, 0° to 30°, 30° to 60°, 60° to 90°, 90° to 120°, from 120° to 150°, and from 150° to 180°. 3. The gallium nitride according to claim 1 or 2 , wherein the angle at which the φ axis corresponding to each segmented region is set is the central φ angle of the φ range of the segmented region. A method for evaluating the crystal quality of a crystal substrate.

Images