Epitaxial growth substrate, method for manufacturing an epitaxial growth substrate, and semiconductor element

JP2026148251APending Publication Date: 2026-09-17DOWA ELECTRONICS MATERIALS CO LTD
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
JP2025036706
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2025-03-07
Publication Date
2026-09-17

AI Technical Summary

Benefits of technology

【0017】 本発明によれば、光半導体素子に優れた性能を付与可能なエピタキシャル成長用基板を提供できる。 また、本発明によれば、光半導体素子に優れた性能を付与可能なエピタキシャル成長用基板の製造方法を提供できる。 また、本発明によれば、性能に優れる光半導体素子を提供できる。

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2026148251000001_ABST
    Figure 2026148251000001_ABST
Patent Text Reader

Abstract

Provided is an epitaxial growth substrate capable of imparting excellent performance to an optical semiconductor element. 【Solution】The present invention provides an off-substrate having a lattice constant a g and an off-angle θ, and a plurality of compositionally graded layers laminated such that the difference in lattice constant from the off-substrate gradually increases from the off-substrate side, wherein the compositionally graded layer farthest from the off-substrate serves as a pseudo-substrate layer, using the lattice constant a g and the off-angle θ, the following formula (1): a g '=a g / cosθ···(1) a surface lattice constant a calculated from g ', a strained lattice constant b calculated from an X-ray diffraction measurement result of the pseudo-substrate layer p and the following formula (2): Xα p·g =(b p -a g ') / a g '···(2) a misfit degree Xα calculated from p·g is 0.5% or less, which is an epitaxial growth substrate.
Need to check novelty before this filing date? Find Prior Art

Description

[Technical Field]

[0001] The present invention relates to an epitaxial growth substrate, a method for manufacturing an epitaxial growth substrate, and a semiconductor device. [Background technology]

[0002] Conventionally, optical semiconductor devices are known that comprise a substrate and a semiconductor layer formed on the substrate by epitaxial growth, the semiconductor layer having a different lattice constant from the substrate.

[0003] In recent years, attempts have been made to improve the characteristics of optical semiconductor devices by improving the crystallinity of semiconductor layers such as the operating layer of the optical semiconductor device. For example, Patent Document 1 discloses an epitaxial wafer and a device manufactured using the same, in which the crystallinity of the operating layer of the epitaxial wafer has been improved to improve its characteristics. The epitaxial wafer comprises a substrate, a buffer layer formed on the substrate, an operating layer formed on the buffer layer and consisting of an epitaxial film having a lattice constant greater than the lattice constant of the material constituting the substrate, and a coating layer consisting of one or more layers formed on the operating layer and arranged to be in contact with the operating layer, wherein the lattice constant of the layer in contact with the operating layer is smaller than the larger of the lattice constant of the operating layer and the lattice constant of the buffer layer, and the thickness of the coating layer is 0.2 μm or more and 2.0 μm or less. Patent Document 1 also discloses a step layer consisting of multiple layers configured such that the lattice constant gradually increases from the substrate side, placed between the substrate and the buffer layer. [Prior art documents] [Patent Documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2006-80448 [Overview of the Initiative] [Problems that the invention aims to solve]

[0005] However, the epitaxial wafer described in Patent Document 1 and the optical semiconductor devices such as those manufactured using it are insufficient in terms of improving the crystallinity of semiconductor layers such as the working layer and further enhancing the performance of the optical semiconductor devices, and there is room for further improvement.

[0006] Accordingly, the present invention has been made in view of the above-mentioned problems, and aims to provide an epitaxial growth substrate capable of imparting excellent performance to optical semiconductor devices. Furthermore, the present invention aims to provide a method for manufacturing an epitaxial growth substrate that can impart excellent performance to an optical semiconductor device. Furthermore, the present invention aims to provide an optical semiconductor device with superior performance. [Means for solving the problem]

[0007] The inventors have diligently researched to solve the above-mentioned problems. Conventionally, even when a substrate 11 having an off-angle θ is used, epitaxial growth occurs so as to fill the step S in the lateral direction (direction of arrow A) as shown in Figure 1, and the lattice constant a of the off-angle substrate 11 g Traditionally, the difference in lattice constants between the off-substrate and the layer above it was considered based on the off-substrate, i.e., the off-angle θ was not taken into account. However, step layers (corresponding to "composition gradient layers" in the present invention described later) and buffer layers (corresponding to "pseudo-substrate layers" in the present invention described later), which increase the lattice constant in a stepwise manner, have compressive strain due to compression, and the actual lattice constant becomes smaller than the theoretical lattice constant calculated from the composition. When improving the crystallinity of the pseudo-substrate layer on the off-substrate 11 by reducing the degree of mismatch with the lattice constant of the pseudo-substrate layer above it which has compressive strain via the composition gradient layer, it was found that there is a difference between the case where the off-angle of the off-substrate is small and the case where the off-angle is large, and the conventional lattice constant a g We found that an examination based solely on that criterion was insufficient. According to the inventors, the lattice constant a gWhen a substrate for epitaxial growth is formed using an off-axis substrate having an off-cut angle θ, the lattice constant a of a pseudo surface that takes into account the off-cut angle θ of the off-axis substrate g ′ and the lattice constant b of the pseudo substrate layer after being compressed (that is, having compressive strain) p When the mismatch degree between the above is designed to fall within a specific range, the crystallinity of the pseudo substrate layer is improved, and the adverse effect caused by excessive strain on the semiconductor layer formed thereon is reduced, thereby improving the performance of an optical semiconductor element (for example, improvement in luminous efficiency for a light-emitting element, reduction in dark current for a light-receiving element, further improvement in mobility for an electronic device, etc.). The present inventors have found that this is possible, and completed the present invention described below.

[0008] That is, the gist configuration of the present invention is as follows.

[0009] [1] An off-axis substrate having a lattice constant a g and having an off-cut angle θ, and a plurality of composition gradient layers stacked such that a difference in lattice constant with the off-axis substrate gradually increases from the off-axis substrate side, wherein the composition gradient layer farthest from the off-axis substrate is used as a pseudo substrate layer, using the lattice constant a g and the off-cut angle θ, the following formula (1): a g ′=a g / cosθ ···(1) a surface lattice constant a calculated from the above g ′, and a strained lattice constant b calculated from an X-ray diffraction measurement result of the pseudo substrate layer p and, using the above, the following formula (2): Xα p·g =(b p -a g ′) / a g ′ ···(2) a mismatch degree Xα calculated from the above p·g is 0.5% or less, which is the substrate for epitaxial growth.

[0010] [2] The following formula (3): Xβ p·g =(b p -a g ) / a g ...(3) The degree of mismatch Xβ calculated from p·g However, it exceeds 0.5%, The aforementioned mismatch degree Xα p·g The epitaxial growth substrate described in [1], wherein the concentration is -3.0% or more and 0.5% or less.

[0011] [3] The epitaxial growth substrate according to claim [1] or [2], wherein the off-angle θ is 4° or more and 44° or less.

[0012] [4] The plurality of composition gradient layers are stacked such that the degree of mismatch Y of the lattice constant of the composition gradient layer with respect to the lattice constant of the layer adjacent to the off-substrate side is 0.05% or more and 0.3% or less, as an epitaxial growth substrate according to any one of [1] to [3].

[0013] [5] The epitaxial growth substrate according to any one of [1] to [4], wherein the off-substrate is an InP substrate and the composition gradient layer is an InAsP layer or an InGaAs layer.

[0014] [6] The epitaxial growth substrate according to any one of [1] to [5], wherein the pseudo-substrate layer is the thickest of the composition gradient layers, and the thickness of the pseudo-substrate layer is 300 nm or more.

[0015] [7] Lattice constant a g The process includes stacking a plurality of compositionally graded layers on an off-substrate having an off-angle θ such that the difference in lattice constants with respect to the off-substrate gradually increases from the off-substrate side, The composition gradient layer furthest from the off-substrate is designated as the pseudo-substrate layer. The pseudo-substrate layer is defined by the lattice constant a g And using the aforementioned off-angle θ, the following equation (1): a g '=a g / cosθ ···(1) The surface lattice constant a' calculated from, The strain lattice constant b calculated from the X-ray diffraction measurement results of the aforementioned pseudo-substrate layer. p and, Using the following equation (2): Xα p·g =(b p -a g ') / a g ' ···(2) The degree of inconsistency Xα calculated from p·g A method for manufacturing an epitaxial growth substrate, wherein the material content is 0.5% or less.

[0016] Obtained using an epitaxial growth substrate described in any of [8][1]~[6], An optoelectronic semiconductor device wherein a semiconductor layer lattice-matched with the pseudo-substrate layer is provided on the surface of the pseudo-substrate layer of the epitaxial growth substrate. [Effects of the Invention]

[0017] According to the present invention, it is possible to provide an epitaxial growth substrate that can impart excellent performance to optical semiconductor devices. Furthermore, according to the present invention, it is possible to provide a method for manufacturing an epitaxial growth substrate that can impart excellent performance to an optical semiconductor device. Furthermore, the present invention provides an optical semiconductor device with superior performance. [Brief explanation of the drawing]

[0018] [Figure 1] This is a schematic diagram illustrating epitaxial growth and off-angle. [Figure 2] This is a schematic diagram showing an example of an epitaxial growth substrate according to the present invention. [Figure 3] This is a schematic diagram showing an example of a semiconductor photodetector, which is one embodiment of the optical semiconductor device of the present invention. [Modes for carrying out the invention]

[0019] (Terminology and measurement methods) First, before describing the embodiments, the following points will be explained in advance.

[0020] In this specification, an X-ray diffraction (XRD) apparatus can be used to confirm the off-angle θ of an off-substrate. For example, if the off-substrate is an InP substrate, the off-angle is determined by measuring the rocking curve relative to the (400) plane of the InP substrate using an XRD apparatus. Specifically, the ω scan plane orientation is measured while rotating the φ axis in the XRD measurement relative to the substrate (half-scan + ω scan at four orientations: φ=0°, 90°, 180°, and 270°), and the displacement angle of the (400) plane relative to the substrate surface is determined from the angle of ω, and this displacement angle is defined as the off-angle θ. The upper limit of the off-angle θ for the off-board is typically 45°.

[0021] In this specification, the symbol 'a' represents the lattice constant calculated by Vegard's law based on the composition, and the layer to which the lattice constant belongs is indicated by a subscript attached to 'a'. For example, the lattice constant of the off-substrate (initial growth substrate) is 'a'. g The lattice constant of the pseudo substrate layer is a p、 The lattice constant of the composition-inclined layer is a c It should be written as follows. Furthermore, in this specification, the mismatch degrees Xα, Xβ, and Y are expressed as percentages (%) of the value calculated by "Δa / a" when the difference in lattice constants a between the two layers is Δa. For example, the mismatch degree Y of the L2 layer relative to the L1 layer is Y L2·L1 It can also be written as, "(a L2 -a L1 ) / a L1 It can be calculated using the formula: In this specification, grid matching means that the degree of mismatch is within ±0.1%.

[0022] In this specification, the surface lattice constant a of the off-substrate considering the off-angle θ. g ' is the lattice constant a of the off-substrate. g This is the value obtained by dividing by cosθ (where θ is the off-angle), and is specifically calculated by the following equation (1). a g'=a g / cosθ ···(1)

[0023] In this specification, the composition ratios of each layer can be those obtained by SIMS analysis. For example, for the composition ratios of multiple composition gradient layers in this specification, in the case of an epitaxial growth substrate, the values ​​obtained by performing SIMS analysis (quadrupole type) in the thickness direction from the pseudo-substrate layer can be used. On the other hand, in the case of an optoelectronic semiconductor device, the values ​​obtained by performing SIMS analysis (quadrupole type) in the thickness direction from the window layer can be used. Furthermore, for the SIMS analysis results, the average elemental concentration values ​​of the half-thickness range of each layer at the center of the thickness direction of each layer can be used. During manufacturing, the composition ratio is calculated using the lattice constant obtained by XRD measurement and the emission center wavelength obtained by photoluminescence (PL) measurement, converted to Eg (i.e., band gap) for a single-layer grown film (see, for example, procedure (A) in the calculation of the degree of relaxation R described later). This determines the growth conditions that result in the desired composition ratio, and layers with the desired composition ratio can be stacked using these growth conditions.

[0024] In this specification, unless otherwise specified, the lattice constant a shall be the value calculated by Vegard's law based on the composition ratio. First, we calculate the simple lattice constant of the mixed crystal according to Vegard's law. InGaAsP system (i.e., general formula: (In a Ga b )(P x As y To explain using )) as an example, the physical constant A abxy (The lattice constant according to Vegard's law) is the physical property constant B of the four binary mixed crystals that form the basis of the pseudoquaternary mixed crystal, given that the composition ratios of each component are known. ax ,B bx ,B ay ,B by It is calculated using the following formula (4) based on the lattice constants (reference values ​​in Table 1 below). A abxy = a × x × B ax +b × x × B bx +a × y × B ay +b×y×Bby ...(4)

[0025] [Table 1]

[0026] In the case of a pseudo-ternary mixed crystal, the general formula is: (In a Ga b Al c Taking (As) as an example, the band gap Eg and the lattice constant a according to Vegard's law can be calculated from the following equations (5) and (6). Eg abcy =(a×b×E aby +b×c×E bcy +c×a×E acy ) / (a×b+b×c+c×a) ···(5) A abcy = a × B ay +b×B by +c × B cy ...(6)

[0027] Furthermore, even when the III-V compound semiconductor is a ternary, pentary, or hexary system, the composition wavelength and lattice constant can be determined by rearranging the equations according to the same reasoning as described above. For binary systems, the values ​​described in the above-mentioned literature can be used.

[0028] In this specification, when "InGaAsP" is simply written without specifying the composition ratio, it means any compound in which the chemical composition ratio of Group III elements (total of In and Ga) to Group V elements (As and P) is 1:1, and the ratios of Group III elements In and Ga and Group V elements As and P are undefined. In this case, it includes cases where neither In nor Ga is included in the Group III elements, and also cases where neither As nor P is included in the Group V elements. That is, "InGaAsP" means "In y Ga 1-y As x P 1-x This can be written as , where x and y independently represent values ​​between 0 and 1 (inclusive). Furthermore, when written as "InGaP", it means that the above "InGaAsP" does not contain As except for unavoidable contamination during manufacturing, and when written as "InGaAs", it means that the above "InGaAsP" does not contain P except for unavoidable contamination during manufacturing. In other words, "InGaP" is "In y Ga 1-y It can be written as "P", and "InGaAs" is written as "In y Ga 1-y It can be written as "As", where y represents a value between 0 and 1 (inclusive). Similarly, when written as "InAsP", it means that Ga is not included in the above "InGaAsP" except for unavoidable contamination during manufacturing, and when written as "GaAsP", it means that In is not included in the above "InGaAsP" except for unavoidable contamination during manufacturing. In other words, "InAsP" is "InAs x P 1-x It can be written as "GaAsP" and "GaAs x P 1-x It can be written as " and in each case, x represents a value between 0 and 1 (inclusive). Furthermore, when written as "InP," it means that the above "InGaAsP" does not contain Ga and As, except for unavoidable contamination during manufacturing. The composition ratios of each component, such as InGaAsP and InGaAs, can be measured by photoluminescence and XRD measurements. Furthermore, "unavoidable contamination during manufacturing" as used here refers not only to unavoidable contamination in the manufacturing equipment that uses the raw material gas, but also to phenomena such as atomic diffusion at the interface of each layer during crystal growth and subsequent heat treatment.

[0029] In this specification, a layer that functions electrically as p-type is referred to as a p-type layer, and a layer that functions electrically as n-type is referred to as an n-type layer. On the other hand, the term "undoped" refers to the intentional absence of specific impurities such as Si, Zn, S, Sn, and Mg. Undoped InGaAsP layers may contain impurities that are unavoidable during the manufacturing process, specifically impurities with a low carrier density (e.g., 4 × 10⁻⁶). 16 / cm 3less), it is treated as "undoped" in the present specification. The values of impurity concentrations of impurities such as Si, Zn, S, Sn, and Mg are those obtained by SIMS analysis.

[0030] In the present specification, let b be the strained (with strain) lattice constant of strained lattice calculated from the XRD measurement results of each layer. For example, the strained lattice constant of the pseudo-substrate layer is defined as b p in this specification. Then, for the epitaxial growth substrate grown up to the pseudo-substrate layer, XRD measurement is performed with an X-ray diffractometer D8 manufactured by Bruker Corporation, Δθ of the pseudo-substrate layer is measured by reciprocal lattice space mapping (RSM) of the (004) plane or the like, further, the PL wavelength of the pseudo-substrate layer is measured with a photoluminescence (PL) measuring instrument RPMBlue manufactured by Ontoinnovation, then the relaxation degree R is calculated using the obtained Δθ and the PL wavelength, and the strained lattice constant b of the pseudo-substrate layer is calculated by analysis software (Jordan Valley Rads) using the obtained relaxation degree R p . For example, when the pseudo-substrate layer is an InAsP layer, the relaxation degree R is calculated according to the following procedures (A) to (E) using the obtained Δθ and PL wavelength, and the strained lattice constant b of the pseudo-substrate layer is calculated by analysis software (Jordan Valley Rads) using the obtained relaxation degree R p .

[0031] (A) The relational expression between energy E and wavelength λ is the following formula (7): E=hc / λ ···(7) (In formula (7), h represents Planck's constant, and c represents the speed of light.) , and the energy E for InAs x P 1-x is represented by the following formula (8): E=1.351-1.315x+0.32x 2 ···(8) as above (see Kenichi Iga, edited by the Japan Society of Applied Physics, "Semiconductor Laser", Ohmsha, Ltd., first edition 1994), therefore if the PL wavelength is obtained, the composition of InAs x P 1-x can be determined.

[0032] (B) The lattice constant a of a ternary mixed crystal can be approximately expressed using Vegard's law. InAs x P 1-x , the lattice constant a of InAs InAs and the lattice constant a of InP InP can be used to obtain the following formula (9): a InAsP = a InAs x + a InP (1-x) ···(9) as shown above.

[0033] (C) When the epitaxial film has fully compressive strain, that is, when the lattice constant of the epitaxial film matches that of the off-cut substrate, the c-axis lattice constant a strain:c is given by the following formula (10): a epi strain:c = a epi relax - 2C 12 / C 11 (a sub - a epi relax ) ···(10) as shown above. For InAs x P 1-x , a in formula (10) epi relax is given by the following formula (11): a epi relax = a InAsP ···(11) as shown above. The elastic constant C in formula (10) 11 represents the difficulty of strain in the x-axis direction when stress in the x-axis direction is applied to a plane perpendicular to the x-axis with the c-axis taken as the z-axis, and the elastic constant C 12 represents the difficulty of strain in the y-axis direction when stress in the x-axis direction is applied to a plane perpendicular to the x-axis with the c-axis taken as the z-axis. Elastic constant C 11 and elastic constant C 12 use values obtained by linearly combining the elastic constant values of InAs and InP, as shown in formulas (12) and (13): C 11 = C InAs 11x+C InP 11 (1-x) ···(12) C 12 =C InAs 12 x+C InP 12 (1-x) ···(13) As shown above, from equations (10) to (13), a epi strain:c It is possible to find this.

[0034] (D) The lattice constant of the epitaxial film is a epi In this case, the angle of X-ray diffraction is given by Bragg's equation (14): a epi sinθ epi =2λ ···(14) As stated above, when converted, the following formula (15): θ epi =arcsin(2λ / a epi ) ···(15) As such, the diffraction angle of the off-substrate is θ sub In this case, the theoretical value of Δθ is given by the following equation (16): Δθ = θ epi -θ sub ...(16) That is correct.

[0035] (E) And the theoretical values ​​of complete relaxation and complete compression strain (Δθ relax ,Δθ strain Depending on how close the measured value Δθ is to the perfect relaxation value, the degree of relaxation R is given by the following formula (17): R=(Δθ strain -Δθ) / (Δθ strain -Δθ relax ) ···(17) It is defined as follows. Using the relaxation degree R thus obtained, the lattice constant b of the strain of the pseudo-substrate layer is analyzed using analysis software (Jordan Valley Rads). p This can be calculated. Furthermore, if a semiconductor layer such as a light absorption layer is formed on the pseudo-substrate layer, the strain lattice constant b in that semiconductor layer can also be determined by performing XRD measurements on that semiconductor layer in the same manner as described above.

[0036] In this specification, the degree of mismatch between the pseudo-substrate layer and the off-substrate is Xα. p·g b is the lattice constant of the pseudo-substrate layer. p and the surface lattice constant a considering the off-angle of the off-substrate. g Using ', the following equation (2): Xα p·g =(b p -a g ') / a g ' ···(2) It is calculated using [a specific formula / method], and the value is expressed as a percentage (%).

[0037] In this specification, the degree of mismatch between the pseudo-substrate layer and the off-substrate is Xβ. p·g b is the lattice constant of the pseudo-substrate layer. p and lattice constant a without considering the off-angle of the off-substrate g Using the following formula (3): Xβ p·g =(b p -a g ) / a g ...(3) It is calculated using [a specific formula / method], and the value is expressed as a percentage (%).

[0038] In this specification, the total thickness of each layer formed can be measured using an optical interferometry film thickness analyzer. Furthermore, the thickness of each individual layer can be calculated from cross-sectional observation of the layer using an optical interferometry film thickness analyzer and a transmission electron microscope. In addition, when the thickness of each layer is small, such as in a superlattice structure, the thickness can be measured using TEM-EDS. In the cross-sectional view, if a given layer has an inclined surface, the thickness of that layer shall be the maximum height from the flat surface of the layer directly below it.

[0039] (Substrate for epitaxial growth) The epitaxial growth substrate of the present invention has a lattice constant a gThe present invention comprises an off-substrate having an off-angle θ and a plurality of composition gradient layers stacked such that the difference in lattice constants between the off-substrate and the off-substrate gradually increases from the off-substrate side. The epitaxial growth substrate of the present invention uses the composition gradient layer furthest from the off-substrate as a pseudo-substrate layer, and the lattice constant a of the off-substrate g And using the off-angle θ, the following equation (1): a g '=a g / cosθ ···(1) Surface lattice constant a calculated from g 'and the strain lattice constant b calculated from the X-ray diffraction measurement results of the pseudo-substrate layer. p Using and , the following equation (2): Xα p·g =(b p -a g ') / a g ' ···(2) The degree of inconsistency Xα calculated from p·g However, it is less than 0.5%. It is presumed that, because the crystallinity of the pseudo-substrate layer is improved in the epitaxial growth substrate described above, the adverse effects of excess strain on the semiconductor layer formed on it are reduced, and as a result, superior performance can be imparted to the optoelectronic semiconductor device. Furthermore, the epitaxial growth substrate of the present invention can be obtained by the method for manufacturing the epitaxial growth substrate of the present invention, as described later.

[0040] The epitaxial growth substrate of the present invention may optionally include other layers between the off-substrate and the composition gradient layer closest to the off-substrate (hereinafter sometimes referred to as the "proximity layer"). Other layers include, for example, a buffer layer.

[0041] <Properties of the substrate for epitaxial growth> In the epitaxial growth substrate of the present invention, the degree of mismatch Xα p·g The amount is 0.5% or less, preferably 0.15% or less, and more preferably -1.5% or less. Inconsistency degree Xα p·gIf the value is below the above upper limit, the crystallinity of the pseudo-substrate layer can be effectively improved. Also, the degree of inconsistency Xα p·g It is preferable that it be -3.0% or higher.

[0042] In the epitaxial growth substrate of the present invention, the mismatch degree Xβ p·g It is preferable that it exceeds 0.5%, and more preferably that it is 0.6% or more. The lattice constant a of the substrate without considering the off-angle in this way. g The degree of inconsistency with Xβ p·g When forming a pseudo-substrate layer with a large off-angle, the mismatch degree Xα takes into account the above off-angle. p·g Setting it below the above upper limit has a significant effect. Note: Inconsistency degree Xβ p·g For example, this could be 3.0% or less, but 2.0% or less is also acceptable.

[0043] <Structure of a substrate for epitaxial growth> The structure of the epitaxial growth substrate will be described below with reference to Figure 2, but the epitaxial growth substrate of the present invention is not limited to this. In Figure 2, for the sake of explanation, the aspect ratio of the substrate and each layer is exaggerated regardless of the actual thickness, but the horizontal width of each layer relatively represents the magnitude of the lattice constant.

[0044] Figure 2 is a schematic diagram showing an example of an epitaxial growth substrate according to the present invention. This schematic diagram illustrates the change in lattice constant, and as described above, the horizontal width of each layer relatively represents the magnitude of the lattice constant. The epitaxial growth substrate 1 shown in Figure 2 comprises an off-substrate 11 and a plurality of composition gradient layers 12 (where n is a natural number greater than or equal to 2). The composition gradient layer 12 furthest from the off-substrate 11 (the Nth composition gradient layer 12) is a pseudo-substrate layer 121, and the composition gradient layer 12 closest to the off-substrate 11 (the 1st composition gradient layer 12) is a neighboring layer 122. Here, the number of composition gradient layers 12 corresponds to the value of N. Although not shown in the diagram, other layers may be provided between the off-plate 11 and the adjacent layer 122.

[0045] <Off-board> Off-board has a mismatch degree of Xα p·g The above are not particularly limited as long as they satisfy the specified requirements, and compound semiconductor substrates such as InP substrates, GaAs substrates, GaSb substrates, InAs substrates, and InSb substrates can be appropriately selected depending on the composition of the semiconductor layer of the optical semiconductor device manufactured using the epitaxial growth substrate, with InP substrates being preferred. The off-board can be either an n-type or a p-type board.

[0046] The off-plate substrate may be doped with sulfur (S), undoped, or semi-insulating, but if conductivity is required, it is preferable to use one doped with sulfur. If the off-substrate needs to be conductive and is doped, the carrier density is, for example, 1.0 × 10⁻⁶. 16 / cm 3 That is all, 1.0 × 10 17 / cm 3 The above is also acceptable. For example, 1.0 × 10 20 / cm 3 The following is true: 1.0 × 10 19 / cm 3 The following is also acceptable.

[0047] The substrate diameter can typically be selected between 2 and 6 inches. The thickness of the off-plate substrate can be 100 μm or more, depending on the substrate diameter, and preferably 300 μm or more. It is preferable to use an off-plate substrate that conforms to SEMI standards.

[0048] The off-angle θ of the off-board is equal to the mismatch degree Xα. p·g While the above are not particularly limited as long as they meet the specified requirements, a value of 4° or more is preferred, and 5° or more is more preferred. Furthermore, a value of 44° or less is preferred, and 25° or less is more preferred.

[0049] <Multiple compositional gradient layers> Multiple composition gradient layers are stacked on the off-substrate such that the lattice constant difference with the off-substrate gradually increases from the off-substrate side. The number of composition gradient layers is N, and this number is appropriately set according to the lattice constant difference between the off-substrate and the pseudo-substrate layer. Here, the "lattice constant difference with the off-substrate" refers to the lattice constant a of each composition gradient layer. c And the lattice constant a of the off-substrate g The difference between (Δa=a c -a g This means that the composition gradient layers are arranged in a stepped manner. In addition, in the case of multiple composition gradient layers, the composition gradient layer furthest from the off-substrate (the Nth) is designated as the pseudo-substrate layer, and the composition gradient layer closest to the off-substrate (the 1st) is designated as the adjacent layer. The multiple composition gradient layers as a whole may have a continuously gradient composition, a stepped gradient, or a combination thereof, but a stepped gradient is preferred. In the case of a stepped gradient, the composition in each individual composition gradient layer may be constant, but other configurations are also permitted, such as forming thin layers with different compositions between the composition gradient layers, or having a composition gradient in a part of the layer. In the following embodiment, we will explain the case where the composition in the composition gradient layers is constant and the composition gradients in a stepped manner from the adjacent layer to the pseudo-substrate layer. By providing multiple such composition gradient layers, the crystallinity of the pseudo-substrate layer can be increased, and the performance of the semiconductor layer can be effectively improved. Note that the lattice constant a of the pseudo-substrate layer c (a p ) is the largest among the multiple composition gradient layers, while the lattice constant a of the adjacent layer is c This is the smallest among the multiple compositional gradient layers.

[0050] Here, the total of N compositional gradient layers have a lattice constant (a) of the layer adjacent to the off-substrate side (hereinafter sometimes referred to as the "lower layer"). c ) lattice constant (a c or a p The degree of mismatch Y of the nth composition gradient layer, in other words, the lattice constant a cn Lattice constant a of the (n+1)th composition gradient layer cn+1 Inconsistency Y cn+1·cnHowever, it is preferable that the layers are laminated so that the concentration is preferably 0.05% or more, preferably 0.3% or less, and more preferably 0.15% or less. Inconsistency degree Y cn+1·cn If multiple composition gradient layers are stacked so that the result is above the lower limit mentioned above, the mismatch with the substrate Xβ can be reduced without making the N number too large. p·g This makes it possible to form a large pseudo-substrate layer. On the other hand, the degree of inconsistency Y cn+1·cn If multiple compositionally graded layers are stacked so that the ratio is below the above upper limit, the occurrence of through-dislocations in the compositionally graded layers can be suppressed.

[0051] A composition gradient layer containing a pseudo-substrate layer has a mismatch degree of Xα. p·g While there are no particular limitations as long as the above meets the specified requirements, if the off-base substrate is an InP substrate, it is preferable that it be an InAsP layer or an InAlAs layer. The compositional gradient layer can be either an n-type or a p-type layer.

[0052] The composition gradient layer, which includes the pseudo-substrate layer, is preferably doped with Si (silicon) or the like. The carrier density of the above composition gradient layer is, for example, 1.0 × 10⁻⁶. 16 / cm 3 That is all, 1.0 × 10 17 / cm 3 The above is also acceptable. For example, 1.0 × 10 19 / cm 3 The following is true: 1.0 × 10 18 / cm 3 The following is also acceptable.

[0053] The number of composition gradient layers including the pseudo-substrate layer (corresponding to N in Figure 1) is preferably a natural number of 3 or more, and more preferably a natural number of 5 or more. Using a natural number of 3 or more allows for greater crystallinity of the pseudo-substrate layer and more effectively improves the performance of the semiconductor layer. On the other hand, the upper limit of N depends on the relationship between the lattice constant of the off-substrate and the lattice constant of the pseudo-substrate layer, and the relationship between the lattice constant of the pseudo-substrate layer and the lattice constant of the semiconductor layer described later, but for example it is 15 or less, and may also be 9 or less.

[0054] Multiple composition gradient layers may have misfit dislocations on the off-substrate side. Here, all of the multiple composition gradient layers may have misfit dislocations, or some of the composition gradient layers may have misfit dislocations, but it is preferable that the adjacent layers have misfit dislocations on the off-substrate side. Misfit dislocations in compositionally graded layers can be identified, for example, by observing the cross-section of the compositionally graded layer using a transmission electron microscope (TEM).

[0055] The pseudo-substrate layer is the composition gradient layer that is furthest from the off-substrate among the multiple composition gradient layers. Here, the pseudo-substrate layer is a layer that can serve as a substrate for epitaxial growth of a semiconductor layer in the manufacturing of an optoelectronic semiconductor device.

[0056] From the viewpoint of improving crystallinity, the pseudo-substrate layer is the thickest of the composition gradient layers, and is preferably 300 nm or thicker, preferably 500 nm or thicker, more preferably 1 μm or thicker, and even more preferably 2 μm or thicker. Furthermore, to avoid excessively long growth times, the thickness of the pseudo-substrate layer is, for example, 10 μm or less.

[0057] The thickness of each compositionally graded layer other than the pseudo-substrate layer is preferably such that it exceeds the critical film thickness with respect to the lattice constant difference between it and the adjacent layer, thereby relieving internal stress while generating misfit dislocations. Specifically, the thickness of each composition gradient layer is preferably 300 nm or more, more preferably 600 nm or more. Furthermore, it is preferably 1500 nm or less, and more preferably 1200 nm or less.

[0058] <buffer layer> A buffer layer is a layer that may be placed between the off-substrate and the adjacent layer.

[0059] The buffer layer is typically a lattice-matched layer with the off-substrate, and it is preferable that it has the same composition as the off-substrate. For example, if the off-substrate is an InP substrate, it is preferable that the buffer layer be an InP layer. The buffer layer can be either an n-type or a p-type layer.

[0060] The buffer layer is preferably doped with Si or the like. The carrier density of the buffer layer is, for example, 1.0 × 10⁻⁶. 16 / cm 3 That is all, 1.0 × 10 17 / cm 3 The above is also acceptable. For example, 1.0 × 10 19 / cm 3 The following is true: 1.0 × 10 18 / cm 3 The following is also acceptable.

[0061] The thickness of the buffer layer may be, for example, 300 nm or more, or 600 nm or more. Alternatively, it may be, for example, 2000 nm or less, or 1500 nm or less.

[0062] (Method for manufacturing epitaxial growth substrates) The present invention relates to a method for manufacturing an epitaxial growth substrate (hereinafter sometimes simply referred to as the "manufacturing method"), and the lattice constant a g The manufacturing method of the present invention includes a step of stacking multiple composition gradient layers on an off-substrate having an off-angle θ such that the difference in lattice constants with respect to the off-substrate gradually increases from the off-substrate side (composition gradient layer stacking step). Furthermore, the manufacturing method of the present invention uses the composition gradient layer furthest from the off-substrate as a pseudo-substrate layer, and the pseudo-substrate layer has a lattice constant a g And using the off-angle θ, the following equation (1): a g '=a g / cosθ ···(1) The surface lattice constant a' was calculated from the above, and the strain lattice constant b was calculated from the X-ray diffraction measurement results of the pseudo-substrate layer. p Using and , the following equation (2): Xα p·g =(b p -a g ') / ag ' ···(2) The degree of inconsistency Xα calculated from p·g Form it so that the percentage is 0.5% or less. With the manufacturing method described above, an epitaxial growth substrate capable of imparting excellent performance to optical semiconductor devices can be obtained. Furthermore, with the manufacturing method described above, the epitaxial growth substrate of the present invention can be obtained.

[0063] The manufacturing method of the present invention may optionally include a step of forming a buffer layer on the surface of the off-substrate on which the composition gradient layers are to be laminated (buffer layer formation step) before laminating the composition gradient layers.

[0064] The preferred configurations of the epitaxial growth substrate, off-substrate, composition gradient layer, and optional buffer layer are as described above. The following describes the composition gradient layer lamination process and the optional buffer layer formation process.

[0065] <Composition gradient layer lamination process> In the composition gradient layer lamination process, multiple composition gradient layers are laminated on a predetermined off-substrate such that the lattice constant difference with the off-substrate gradually increases from the off-substrate side. The composition gradient layer furthest from the off-substrate is designated as the pseudo-substrate layer, and the pseudo-substrate layer is given a mismatch degree of Xα. p·g The material is formed to have a content of 0.5% or less, preferably 0.15% or less, more preferably -0.15% or less, and preferably -3.0% or more.

[0066] The method for forming a compositionally graded layer including a pseudo-substrate layer is not particularly limited as long as it can be epitaxially grown on an off-substrate or on any buffer layer. Examples include metal-organic chemical vapor deposition (MOCVD), molecular beam epitaxy (MBE), and sputtering.

[0067] For example, if the composition gradient layer containing the pseudo-substrate layer is an InAsP layer, it can be formed by supplying trimethylindium (TMIn) as the In source, arsine (AsH3) as the As source, and phosphine (PH3) as the P source into the growth furnace in a predetermined mixing ratio using a carrier gas, and growing it in the vapor phase on an off-substrate or any buffer layer. If an n-type or p-type layer is desired, additional dopant source gases may be used as appropriate.

[0068] Here, the degree of inconsistency Xα mentioned above. p·g Composition gradient layers and pseudo-substrate layers that meet the above requirements can be formed by controlling the composition ratio by changing the mixing ratio of the raw material gases. Furthermore, the thickness of each layer can be adjusted by the ratio of Group III raw material gas to Group V raw material gas, the growth temperature, and the growth time.

[0069] <Buffer layer formation process> In the buffer layer formation process, a buffer layer is formed on the surface of the off-substrate to which the composition gradient layers will be stacked, before stacking the composition gradient layers.

[0070] The method for forming the buffer layer is not particularly limited as long as it can be epitaxially grown on an off-substrate; for example, the method for forming the composition gradient layer described above can be used.

[0071] For example, if the buffer layer is an InP layer, it can be formed by supplying trimethylindium (TMIn) as the In source and phosphine (PH3) as the P source into a growth furnace using a carrier gas in a predetermined mixing ratio, and growing it in the vapor phase on an off-substrate. If an n-type or p-type layer is to be formed, an additional dopant source gas may be used as appropriate.

[0072] The thickness of the buffer layer can be adjusted by the ratio of Group III raw material gas to Group V raw material gas, the growth temperature, and the growth time.

[0073] (Optical semiconductor device) The optical semiconductor device of the present invention is obtained using the epitaxial growth substrate of the present invention described above. In the optical semiconductor device of the present invention, a semiconductor layer is provided on the surface of the pseudo-substrate layer of the epitaxial growth substrate of the present invention described above, which is lattice-matched with the pseudo-substrate layer. Here, the optical semiconductor device of the present invention may have an off-substrate removed by etching or the like. That is, the optical semiconductor device of the present invention may comprise an off-substrate, a plurality of composition gradient layers, and a light absorption layer in this order, or it may comprise a plurality of composition gradient layers and a light absorption layer. The optical semiconductor device of the present invention may optionally include components other than the epitaxial growth substrate and semiconductor layer (hereinafter sometimes referred to as "other components"). Examples of other components include a window layer, an anti-reflective layer, electrodes, and the like.

[0074] Here, the optical semiconductor device of the present invention may be either a semiconductor photodetector or a semiconductor light-emitting element. When the optical semiconductor device of the present invention is a semiconductor photodetector, the semiconductor layer is usually a light-absorbing layer. On the other hand, when the optical semiconductor device of the present invention is a semiconductor light-emitting element, the semiconductor layer is usually a light-emitting layer.

[0075] In the following description, a semiconductor photodetector will be used as an example to explain the optical semiconductor device of the present invention, but the optical semiconductor device of the present invention is not limited to this.

[0076] <Substrate for epitaxial growth> As the substrate for epitaxial growth, the epitaxial growth substrate of the present invention described above is used.

[0077] <Light-absorbing layer> The light-absorbing layer is provided on the surface of the pseudo-substrate layer so as to be lattice-matched with the pseudo-substrate layer. It is more preferable that the lattice constant a of the light-absorbing layer is slightly smaller than that of the pseudo-substrate layer. Furthermore, the lattice constant b of the light-absorbing layer, as determined by XRD measurement, is equal to the lattice constant b of the pseudo-substrate layer. p It is more preferable that the lattice constant difference is close to zero while maintaining lattice matching. Here, the composition of the light-absorbing layer can be appropriately selected according to the absorption center wavelength. For example, the absorption center wavelength can be 2000 to 3000 nm. In this case, the composition of the light-absorbing layer can be formed from a III-V compound semiconductor consisting of a total of 2 to 6 elements selected from Group III elements (Al, Ga, In) and Group V elements (P, As, Sb), such as InGaAs. InAs, InGaAlAs, or InGaAsP may also be used. Furthermore, it is preferable that the light-absorbing layer be undoped. Furthermore, the light-absorbing layer can be either an n-type or a p-type layer.

[0078] The thickness of the light-absorbing layer is not particularly limited; for example, it may be 500 nm or more, or 1000 nm or more. Alternatively, it may be 5000 nm or less, or 3000 nm or less.

[0079] The method for forming the light-absorbing layer is not particularly limited and can be formed on the surface of the pseudo-substrate layer by conventionally known methods.

[0080] <Window layer> An arbitrary window layer can be provided on the surface of the light-absorbing layer (on the opposite side from the pseudo-substrate layer). The composition of the window layer can be selected as appropriate, but if the pseudo-substrate layer is an InAsP layer, it is preferable that the window layer be an InAsP layer, and it is preferable that the lattice constant a of the window layer be lattice-matched with that of the pseudo-substrate layer, and more preferably that it have the same composition as the pseudo-substrate layer. The light-absorbing layer can be either an n-type or p-type layer.

[0081] The window layer is preferably doped with Si or the like. The carrier density of the window layer is, for example, 1.0 × 10⁻⁶. 16 / cm 3 That is all, 1.0 × 10 17 / cm 3 The above is also acceptable. For example, 1.0 × 10 19 / cm 3 The following is true: 1.0 × 10 18 / cm 3 The following is also acceptable.

[0082] The thickness of the window layer is not particularly limited; for example, it may be 300 nm or more, or 600 nm or more. Alternatively, it may be 2000 nm or less, or 1500 nm or less.

[0083] The method for forming the window layer is not particularly limited and can be formed on the surface of the light-absorbing layer (opposite the pseudo-substrate layer) by conventionally known methods.

[0084] Here, optionally, after forming the window layer, Zn may be diffused from the window layer to the light-absorbing layer using a method such as MOCVD. For example, diethylzinc can be used as the Zn source.

[0085] <Anti-reflection layer> Any anti-reflective layer can be applied to the surface of the window layer (opposite the light-absorbing layer). The anti-reflective layer can be selected as appropriate, but examples include SiN films.

[0086] The thickness of the anti-reflective layer is not particularly limited; for example, it may be 50 nm or more, or 100 nm or more. Alternatively, it may be 600 nm or less, or 300 nm or less.

[0087] The method for forming the anti-reflective layer is not particularly limited and can be formed on the surface of the window layer (opposite the light-absorbing layer) by conventionally known methods such as CVD or coating. The anti-reflective layer is typically formed by creating a pattern on the surface of the window layer using photolithography or similar methods with a resist.

[0088] <Electrode> Any electrode is provided to be electrically connected to the epitaxial growth substrate. For example, electrodes can be provided on the surface of the window layer, the surface of the off-substrate of the epitaxial growth substrate, etc. The electrode may be either a p-type electrode or an n-type electrode.

[0089] The method for forming the electrodes is not particularly limited, and the electrodes can be formed by conventionally known methods such as sputtering, electron beam deposition, and resistance heating.

[0090] <An example of an optoelectronic semiconductor device> The following description will refer to Figure 3 to explain a semiconductor photodetector, which is one embodiment of the optical semiconductor device of the present invention; however, the optical semiconductor device of the present invention is not limited to this embodiment. In Figure 3, for the sake of explanation, the aspect ratios of the substrate and each layer are exaggerated and do not reflect their actual thickness.

[0091] Figure 3 is a schematic diagram showing an example of a semiconductor photodetector, which is one embodiment of the optical semiconductor device of the present invention. The semiconductor photodetector 2 shown in Figure 3 comprises an epitaxial growth substrate 1, a light-absorbing layer 21, and a window layer 22 in that order. The epitaxial growth substrate 1 comprises an off-substrate 11 and a plurality of composition gradient layers 12, the composition gradient layer 12 furthest from the off-substrate 11 being a pseudo-substrate layer 121, and the composition gradient layer 12 closest to the off-substrate 11 being a proximity layer 122. The light-absorbing layer 21 is provided on the surface of the pseudo-substrate layer 121. In other words, the semiconductor photodetector 2 comprises an off-substrate 11, a plurality of composition gradient layers 12, a light absorption layer 21, and a window layer 22 in this order, with the light absorption layer 21 provided on the surface of the pseudo-substrate layer 121. [Examples]

[0092] The present invention will be described in more detail using the following examples, but the present invention is not limited to the following examples. Note that lattice constants a and b p This was measured or calculated using the method described above.

[0093] (Example 1) <Manufacturing of substrates for epitaxial growth> First, trimethylindium (TMIn) was prepared as the In source, trimethylgallium (TMG) as the Ga source, arsine (AsH3) as the As source, and phosphine (PH3) as the P source as the raw material gases.

[0094] An n-type InP buffer layer was deposited on the (100) plane of an n-type InP substrate (3 inches, substrate thickness: 625 μm, lattice constant: 0.5869 nm, off-angle in the

[0110] direction from the (100) plane: 6°) using the MOCVD method. Next, on the surface of the n-type InP buffer layer, trimethylindium arsine and phosphine are used as raw material gases, and while varying the amount of P relative to the total amount of As and P contained in the raw material gas (P gas phase ratio), n-type InAs is grown by vapor deposition (MOCVD). x P 1-x Composition-graded layers 1 to 6 were deposited to obtain a substrate for epitaxial growth. Here, composition-graded layer 6, which is furthest from the n-type InP substrate, was used as the pseudo-substrate layer. Furthermore, the n-type InP substrate, n-type InP buffer layer, and n-type InAs x P 1-x Table 2 shows the various formation conditions and lattice constant values ​​for epitaxial growth substrates with composition gradient layers 1 to 6 (up to the pseudo-substrate layer).

[0095] <Manufacturing of semiconductor photodetectors> In addition to the epitaxial growth substrate described above, an undoped InGaAs light-absorbing layer (semiconductor layer) and an n-type InAsP window layer were sequentially deposited on the surface of the pseudo-substrate layer of the epitaxial growth substrate obtained in the same manner as described above. The various formation conditions and lattice constant values ​​for the InGaAs light-absorbing layer and InAsP window layer are shown in Table 2. Next, a mask made of Si3N4 is formed on the InAsP window layer and etched to form a mask opening. Then, a Zn source is flowed using the MOCVD method, and an average Zn concentration of 5.0 × 10⁻¹⁶ is obtained from the n-type InAsP window layer at the mask opening to a part of the n-type InGaAs light absorption layer (to a depth of 0.3 μm from the interface between the window layer and the light absorption layer toward the light absorption layer). 18 / cm 3 The material was converted to the p-type by diffusing Zn in this manner. DEZn (diethylzinc) was used as the Zn source. Next, a 0.2 μm Si3N4 film (anti-reflective layer) was deposited over the entire surface by CVD, and a pattern was formed by photolithography using a resist to create an exposed area that exposed a portion of the outer periphery of the p-type region. Then, a p-type electrode was formed by electron beam deposition to connect to the window layer which had been p-typed in the exposed area, and a back electrode was formed on the back surface of the n-type InP substrate to obtain a semiconductor photodetector with an electrode.

[0096] (Example 2) An epitaxial growth substrate and an optoelectronic semiconductor device were obtained in the same manner as in Example 1, except that an n-type InP substrate with an off-angle of 10° was used as the off-substrate, and the various formation conditions and properties of each layer were changed as shown in Table 6.

[0097] (Example 3) An epitaxial growth substrate and an optoelectronic semiconductor device were obtained in the same manner as in Example 1, except that an n-type InP substrate with an off-angle of 15° was used as the off-substrate, and the various formation conditions and properties of each layer were changed as shown in Table 7.

[0098] (Comparative Example 1) An epitaxial growth substrate and an optoelectronic semiconductor device were obtained in the same manner as in Example 1, except that an n-type InP substrate with an off-angle of 2° was used as the off-substrate, and the various formation conditions and properties of each layer were changed as shown in Table 3.

[0099] (Comparative Example 2) An epitaxial growth substrate and an optoelectronic semiconductor device were obtained in the same manner as in Example 1, except that an n-type InP substrate with an off-angle of 0° was used instead of an off-angle substrate, and the various formation conditions and properties of each layer were changed as shown in Table 5.

[0100] (Full width at half maximum of the rocking curve of the pseudo-substrate layer) Using the epitaxial growth substrates obtained in the examples and comparative examples, the full width at half maximum (FWHM) of the rocking curve of the pseudo-substrate layer was calculated by X-ray crystallography (XRC) using the following method. Specifically, reciprocal lattice space mapping (RSM) of the (004) plane was performed using a Bruker D8 X-ray diffractometer, and the full width at half maximum (FWHM) of the rocking curve was determined from the resulting ω-direction profile using the analysis software LEPTOS. The results are shown in Table 3. Furthermore, a smaller half-width of the rocking curve indicates that a semiconductor layer with superior crystallinity can be obtained.

[0101] (Measurement of dark current) The dark current of the semiconductor photodetectors obtained in Example 1 and Comparative Example 1 was measured. Specifically, the measurement was performed using the two-terminal method with a Keithley "model-2635B" source measure unit, applying bias voltages of Vr = -0.01V and -0.1V. The results are shown in Table 4. Furthermore, the smaller the dark current of a semiconductor photodetector, the better the performance of that semiconductor photodetector.

[0102] In Tables 2 and 3, "Lattice constant a" is the lattice constant a of the off-substrate. g and the lattice constants of each layer according to Begard's law (lattice constant a of the compositionally graded layers) c , lattice constant a of the pseudo-substrate layer p (meaning, etc.) "Surface lattice constant a g '」 is the lattice constant a g The value obtained by dividing by cosθ (where θ is the off-angle), that is, "a g It means "cosθ". "Lattice constant b p " is the lattice constant b of the strained pseudo-substrate layer calculated from the XRD measurement results. p Meaning, "Y" represents the degree of mismatch between each layer and the adjacent layer (lower layer) on the off-board side. "Xα p·g " is the degree of mismatch between the pseudo-substrate layer and the off-substrate, and "(b p -a g ') / a g This means ' and takes the off-angle into consideration. "Xβ" p·g " is the degree of mismatch between the pseudo-substrate layer and the off-substrate, and "(b p -a g ) / a g This means "[...] and does not take the off-angle into consideration."

[0103] [Table 2]

[0104] [Table 3]

[0105] [Table 4]

[0106] As is clear from Table 3, the epitaxial growth substrates of Examples 1 to 3 have a smaller half-width of the rocking curve of the pseudo-substrate layer than the epitaxial growth substrates of Comparative Examples 1 and 2, resulting in improved crystallinity of the pseudo-substrate layer and, consequently, superior performance can be imparted to the optoelectronic device. This can also be understood from the fact that the semiconductor photodetector of Example 1, which has a smaller half-width of the rocking curve than the semiconductor photodetector of Comparative Example 1, exhibits a smaller dark current, as shown in Table 4. [Industrial applicability]

[0107] According to the present invention, it is possible to provide an epitaxial growth substrate that can impart excellent performance to optical semiconductor devices. Furthermore, according to the present invention, it is possible to provide a method for manufacturing an epitaxial growth substrate that can impart excellent performance to an optical semiconductor device. Furthermore, the present invention provides an optical semiconductor device with superior performance. [Explanation of Symbols]

[0108] 1. Substrate for epitaxial growth 11 Offset board 12 Compositionally graded layer 121 Pseudo substrate layer 122 Proximity Layer 2. Semiconductor photodetector 21 Light-absorbing layer 22 window layers

Claims

1. Lattice constant a g An off-substrate having and having an off-angle θ, The system comprises a plurality of compositionally graded layers stacked such that the lattice constant difference with the off-substrate gradually increases from the off-substrate side, The composition gradient layer furthest from the off-substrate is designated as the pseudo-substrate layer. The lattice constant a g And using the off-angle θ, the following equation (1): a g ’=a g / cosθ ・・・(1) Surface lattice constant a calculated from g 'and, The strain lattice constant b calculated from the X-ray diffraction measurement results of the aforementioned pseudo-substrate layer. p and, Using the following formula (2): Xα p・g =(b p -a g ’) / a g ’ ・・・(2) The degree of mismatch Xα calculated from p・g However, the content is 0.5% or less, for epitaxial growth substrates.

2. The following formula (3): Xβ p・g =(b) p --] g ) / ] g ・・・(3) The degree of mismatch Xβ calculated from p・g However, it exceeds 0.5%, The aforementioned mismatch degree Xα p・g The epitaxial growth substrate according to claim 1, wherein the amount is -3.0% or more and 0.5% or less.

3. The epitaxial growth substrate according to claim 1, wherein the off-angle θ is 4° or more and 44° or less.

4. The epitaxial growth substrate according to claim 1, wherein the plurality of composition gradient layers are stacked such that the degree of mismatch Y of the lattice constant of the composition gradient layer with respect to the lattice constant of the layer adjacent to the off-substrate side is 0.05% or more and 0.3% or less.

5. The epitaxial growth substrate according to claim 1, wherein the off-substrate is an InP substrate and the composition gradient layer is an InAsP layer or an InGaAs layer.

6. The epitaxial growth substrate according to claim 1, wherein the pseudo-substrate layer is the thickest of the composition gradient layers, and the thickness of the pseudo-substrate layer is 300 nm or more.

7. Lattice constant a g The process includes stacking a plurality of compositionally graded layers on an off-substrate having an off-angle θ such that the difference in lattice constants with respect to the off-substrate gradually increases from the off-substrate side, The composition gradient layer furthest from the off-substrate is designated as the pseudo-substrate layer. The pseudo-substrate layer is defined by the lattice constant a g And using the off-angle θ, the following equation (1): a g ’=a g / cosθ ・・・(1) The surface lattice constant a' calculated from, The strain lattice constant b calculated from the X-ray diffraction measurement results of the aforementioned pseudo-substrate layer. p and, Using the following formula (2): Xα p・g =(b p -a g ’) / a g ’ ・・・(2) The degree of mismatch Xα calculated from p・g A method for manufacturing an epitaxial growth substrate, wherein the content is formed to be 0.5% or less.

8. Obtained using an epitaxial growth substrate according to any one of claims 1 to 6, An optoelectronic semiconductor device wherein a semiconductor layer lattice-matched with the pseudo-substrate layer is provided on the surface of the pseudo-substrate layer of the epitaxial growth substrate.

Citation Information

Patent Citations

  • Epitaxial wafer and element

    JP2006080448A