WO2015159844A1 - Method for producing substrate for epitaxial growth, substrate for epitaxial growth obtained by same, and light emitting element using said substrate - Google Patents

Abstract

This method for producing a substrate for epitaxial growth comprises: a patterning step wherein an inorganic material film (60) of a predetermined pattern is formed on a base (40) by screen printing; and a curing step wherein the inorganic material film (60) is cured. Provided are: a method for efficiently producing a substrate for epitaxial growth; a substrate for epitaxial growth, which is obtained by the method; and a light emitting element which uses this substrate.

Description

Epitaxial growth substrate manufacturing method, epitaxial growth substrate obtained therefrom, and light-emitting device using the substrate

The present invention relates to a substrate manufacturing method for epitaxially growing a semiconductor layer and the like, a substrate manufactured by the manufacturing method, and a light emitting element in which a semiconductor layer is formed on the substrate.

Semiconductor light emitting devices generally include light emitting diodes (LEDs) and laser diodes (LDs), and are widely used in various light sources used for backlights, lighting, traffic lights, large displays, and the like.

In a light emitting device having a semiconductor layer such as a nitride semiconductor, normally, a buffer layer, an n-type semiconductor layer, an active layer, and a p-type semiconductor layer are epitaxially grown on a light-transmitting substrate in this order, and each of the n-type and p-type semiconductors. It is configured by forming an n-side electrode and a p-side electrode that are electrically connected to the layer. In this light emitting element, light generated in the active layer is emitted to the outside of the semiconductor layer from the externally exposed surface (upper surface, side surface) of the semiconductor layer, the exposed surface (back surface, side surface) of the substrate, and the like. In such a light emitting device, when light generated in the active layer is incident on the interface between the semiconductor layer and the electrode or the interface between the semiconductor layer and the substrate at an angle greater than a predetermined critical angle, the semiconductor layer repeats total reflection. Propagating in the horizontal direction, a part of the light is absorbed during that time, and the light extraction efficiency decreases.

Therefore, Patent Documents 1 and 2 disclose that the semiconductor layer growth surface of the substrate is etched to form a concavo-convex pattern, thereby improving the light extraction efficiency of the light-emitting element. Further, Patent Document 2 discloses that by providing such a concavo-convex pattern on the growth surface of the semiconductor layer of the substrate, the dislocation density of the semiconductor layer is reduced and deterioration of the characteristics of the light emitting element can be suppressed.

JP 2010-206230 A Japanese Patent Laid-Open No. 2001-210598

The semiconductor light emitting device as described above is required to be manufactured with higher production efficiency. Accordingly, an object of the present invention is to provide a manufacturing method for efficiently manufacturing an epitaxial growth substrate used in a light emitting device such as a semiconductor light emitting device, an epitaxial growth substrate manufactured by the manufacturing method, and light emission using the epitaxial growth substrate. It is to provide an element.

According to the first aspect of the present invention, there is provided a method for manufacturing an epitaxial growth substrate, comprising: a patterning step of forming an inorganic material film having a predetermined pattern on a substrate by screen printing; and a curing step of curing the inorganic material film. Provided.

In the method for manufacturing an epitaxial growth substrate, the inorganic material film may be made of a sol-gel material.

The method for manufacturing the epitaxial growth substrate may further include a step of etching a region where the surface of the base material is exposed.

In the method for manufacturing a substrate for epitaxial growth, a buffer layer may be formed on the base material having the inorganic material film.

In the method for manufacturing a substrate for epitaxial growth, a buffer layer may be formed on the base material before the patterning step.

In the manufacturing method of the substrate for epitaxial growth, a convex portion and a concave portion generated by forming the inorganic material film of the predetermined pattern,
i) each having an elongated shape extending in a wavy manner in plan view; and ii) the extending direction, the bending direction, and the length may be non-uniform.

In the method for manufacturing a substrate for epitaxial growth, the base material may be a sapphire substrate.

According to the second aspect of the present invention, there is provided an epitaxial growth substrate having a concavo-convex pattern obtained by the epitaxial growth substrate manufacturing method of the first aspect.

The substrate for epitaxial growth is
i) The projections or depressions of the concavo-convex pattern surface of the epitaxial growth substrate each have an elongated shape extending in a wavy manner in plan view, and
ii) The convex part or the concave part of the concavo-convex pattern surface of the epitaxial growth substrate may be uneven in extension direction, bending direction and length.

In the epitaxial growth substrate, the extending direction of the protrusions is irregularly distributed in plan view,
The contour line in plan view of the convex portion included in the region per unit area of the uneven pattern may include more straight sections than curved sections.

In the epitaxial growth substrate, the width of the convex portion in a direction substantially orthogonal to the extending direction of the convex portion in a plan view may be constant.

In the epitaxial growth substrate, the curved section forms a plurality of sections by dividing a contour line in plan view of the convex portion by a length that is π (circumferential ratio) times an average value of the width of the convex portion. The ratio of the linear distance between the two end points to the length of the contour line between the two end points of the section is 0.75 or less,
The straight section may be a section that is not the curved section among the plurality of sections.

In the epitaxial growth substrate, the curved section forms a plurality of sections by dividing a contour line in plan view of the convex portion by a length that is π (circumferential ratio) times an average value of the width of the convex portion. In this case, one of the two angles formed by the line segment connecting one end of the section and the midpoint of the section and the line segment connecting the other end of the section and the midpoint of the section is 180 ° or less. It is a section where the angle is 120 ° or less,
The straight section is a section that is not the curved section among the plurality of sections,
The ratio of the curve section among the plurality of sections may be 70% or more.

In the epitaxial growth substrate, the extending direction of the protrusions is irregularly distributed in plan view,
The width of the convex portion in a direction substantially orthogonal to the extending direction of the convex portion in a plan view may be constant.

In the epitaxial growth substrate, a Fourier transform image obtained by subjecting the unevenness analysis image obtained by analyzing the unevenness pattern with a scanning probe microscope to a two-dimensional fast Fourier transform process has an absolute value of the wave number of 0 μm ⁻¹ . A circular or annular pattern having a substantially origin at a certain origin is shown, and the circular or annular pattern exists in a region where the absolute value of the wave number is in the range of 10 μm ⁻¹ or less. It's okay.

According to the third aspect of the present invention, there is provided a light emitting device comprising a semiconductor layer including at least a first conductivity type layer, an active layer, and a second conductivity type layer on the epitaxial growth substrate of the second aspect. .

In the method for manufacturing an epitaxial growth substrate according to the present invention, since the concave / convex pattern is formed on the base material by screen printing, the epitaxial growth substrate can be easily produced. In addition, since the concavo-convex pattern is formed by screen printing without using photolithography or a nanoimprint method that requires production of a mold, which requires expensive optical precision equipment and generates a large amount of waste liquid, the production of the epitaxial growth substrate of the present invention The method has a low manufacturing cost and a low environmental burden. In addition, since the epitaxial growth substrate of the present invention has a function as a diffraction grating substrate for improving light extraction efficiency, a light emitting device manufactured using this substrate has high light emission efficiency. Therefore, the epitaxial growth substrate of the present invention is extremely effective for the production of a light emitting device having excellent luminous efficiency.

It is a flowchart of the manufacturing method of the board | substrate for epitaxial growth of embodiment. 2A to 2C are diagrams conceptually showing a method for producing a screen printing plate according to the embodiment. 3A to 3C are diagrams conceptually showing each process of the method for manufacturing an epitaxial growth substrate according to the embodiment. 4A to 4C are schematic sectional views of an epitaxial growth substrate on which a buffer layer is formed. FIG. 5A is an example of an AFM image of the surface of the substrate obtained by the method for manufacturing an epitaxial growth substrate of the embodiment, and FIG. 5B is for epitaxial growth on a cutting line in the AFM image of FIG. The cross-sectional profile of a board | substrate is shown. It is a schematic sectional drawing of the optical element of embodiment. FIG. 7 is an example of a plane view analysis image (black and white image) of the substrate such as epitaxial growth according to the embodiment. FIGS. 8A and 8B are diagrams for explaining an example of a method for determining a branch of a convex portion in a planar view analysis image. FIG. 9A is a diagram used for explaining the first definition method of the curve section, and FIG. 9B is a diagram used for explaining the second definition method of the curve section.

Hereinafter, embodiments of a method for manufacturing an epitaxial growth substrate of the present invention, an epitaxial growth substrate obtained thereby, and a light-emitting element using the epitaxial growth substrate will be described with reference to the drawings.

[Method for producing substrate for epitaxial growth]
A method for manufacturing the epitaxial growth substrate will be described. As shown in FIG. 1, the manufacturing method of the substrate for epitaxial growth according to the embodiment mainly includes a solution preparation step P1 for preparing a sol-gel material, a film of the sol-gel material is formed on the substrate by screen printing, and the substrate is uneven. A pattern forming step (patterning step) P2 for forming a pattern and a curing step P3 for curing the film of the sol-gel material are included. In the following, first, a screen printing plate for forming an uneven pattern and a manufacturing method thereof will be described with reference to FIGS. 2A to 2C, and the above steps will be described with reference to FIGS. 3A to 3C. Will be described in order.

<Screen printing plate>
For example, as described in JP-A-2009-48163, a screen printing plate is obtained by applying a photosensitive resin composition on a screen or pasting a photosensitive resin composition on a screen. A resin layer can be formed by attaching, a pattern can be formed on the resin layer, and then the resin layer can be crosslinked and cured. Specifically, it can be produced by the following resin layer forming step, exposure step, development step, and post-exposure step. These are described in detail below.

First, as shown in FIG. 2A, the photosensitive resin composition is applied to the screen 12 and dried, and laminated with a predetermined thickness to form the resin layer 14 (resin layer forming step). The screen 12 can be used without particular limitation as long as it is a known screen printing screen. For example, a stainless steel or polyester screen is stretched on a flat screen mold made of a metal rod-like material such as aluminum in a square or rectangular shape.

As shown in FIG. 2B, the pattern is exposed by irradiating light onto the resin layer 14 on the screen 12 through the photomask 16 (exposure process). As a photomask 16, if it is well-known things, such as a film mask and a chrome mask, it can be used without limitation. As the light source, a combination of a known light source such as a mercury lamp or a halogen lamp that emits ultraviolet rays or visible light and an optical filter can be used. The exposure time can be freely set according to the used and photosensitive resin composition.

Resin layer modified by exposure as shown in FIG. 2C by immersing the screen 12 having the exposed resin layer 14 in a developer, rubbing with a waste cloth, or spraying the developer with a spray. (Or unexposed resin layer) 14a is developed and removed (development step). As a developing solution, if it is a well-known water or alkaline aqueous solution, it can be used without limitation. The development method and development time can be freely set according to the developer used and the photosensitive resin composition. The portion from which the resin layer 14a has been removed becomes an opening 18 where the screen 12 is exposed. In this way, the screen printing plate 10 is obtained. Further, the resin layer 14 may be further irradiated with light to crosslink and cure the unexposed portion of the resin layer 14 (post-exposure step).

The planar shape (planar pattern) of the resin layer (mask portion) 14 and the opening 18 of the screen printing plate 10 obtained as described above is not particularly limited, and is a regularly oriented pattern such as a stripe, a wavy stripe, or a zigzag. Or a regularly oriented pattern such as a dot pattern. Alternatively, the shape may be an elongated shape extending while undulating, and the extending direction, the direction of undulation, and the extending length may be irregular. The mask part 14 and the opening part 18 extending while undulating may be branched in the middle.

<Sol-gel material solution preparation process>
First, a solution of sol-gel material (inorganic material) is prepared. As the sol-gel material, silica, Ti-based material, ITO (indium-tin-oxide) -based material, sol-gel material such as ZnO, ZrO ₂ , Al ₂ O ₃ can be used. For example, when an uneven structure made of silica is formed on a substrate by a sol-gel method, a metal alkoxide (silica precursor) is prepared as a sol-gel material. As precursors of silica, tetramethoxysilane (TMOS), tetraethoxysilane (TEOS), tetra-i-propoxysilane, tetra-n-propoxysilane, tetra-i-butoxysilane, tetra-n-butoxysilane, tetra- Tetraalkoxide monomers represented by tetraalkoxysilane such as sec-butoxysilane, tetra-t-butoxysilane, methyltrimethoxysilane, ethyltrimethoxysilane, propyltrimethoxysilane, isopropyltrimethoxysilane, phenyltrimethoxysilane, Methyltriethoxysilane (MTES), ethyltriethoxysilane, propyltriethoxysilane, isopropyltriethoxysilane, phenyltriethoxysilane, methyltripropoxysilane, ethyltripro Xysilane, propyltripropoxysilane, isopropyltripropoxysilane, phenyltripropoxysilane, methyltriisopropoxysilane, ethyltriisopropoxysilane, propyltriisopropoxysilane, isopropyltriisopropoxysilane, phenyltriisopropoxysilane, tolyltriethoxy Trialkoxide monomers typified by trialkoxysilane such as silane, dimethyldimethoxysilane, dimethyldiethoxysilane, dimethyldipropoxysilane, dimethyldiisopropoxysilane, dimethyldi-n-butoxysilane, dimethyldi-i-butoxysilane, dimethyldi- sec-Butoxysilane, Dimethyldi-t-butoxysilane, Diethyldimethoxysilane, Diethyldiethoxysilane, Diethyl Dipropoxysilane, Diethyldiisopropoxysilane, Diethyldi-n-butoxysilane, Diethyldi-i-butoxysilane, Diethyldi-sec-butoxysilane, Diethyldi-t-butoxysilane, Dipropyldimethoxysilane, Dipropyldiethoxysilane, Di Propyldipropoxysilane, dipropyldiisopropoxysilane, dipropyldi-n-butoxysilane, dipropyldi-i-butoxysilane, dipropyldi-sec-butoxysilane, dipropyldi-t-butoxysilane, diisopropyldimethoxysilane, diisopropyldiethoxysilane, diisopropyl Dipropoxysilane, diisopropyldiisopropoxysilane, diisopropyldi-n-butoxysilane, diisopropyldi-i-butoxysilane, diisopro Pildi-sec-butoxysilane, diisopropyldi-t-butoxysilane, diphenyldimethoxysilane, diphenyldiethoxysilane, diphenyldipropoxysilane, diphenyldiisopropoxysilane, diphenyldi-n-butoxysilane, diphenyldi-i-butoxysilane Dialkoxide monomers typified by dialkoxysilanes such as diphenyldi-sec-butoxysilane and diphenyldi-t-butoxysilane can be used. Furthermore, alkyltrialkoxysilanes or dialkyldialkoxysilanes in which the alkyl group has C4-C18 carbon atoms can also be used. Monomers having a vinyl group such as vinyltrimethoxysilane, vinyltriethoxysilane, 2- (3,4-epoxycyclohexyl) ethyltrimethoxysilane, 3-glycidoxypropylmethyldimethoxysilane, 3-glycidoxypropyltrimethoxy Monomers having an epoxy group such as silane, 3-glycidoxypropylmethyldiethoxysilane, 3-glycidoxypropyltriethoxysilane, monomers having a styryl group such as p-styryltrimethoxysilane, 3-methacryloxypropylmethyl Monomers having a methacrylic group such as dimethoxysilane, 3-methacryloxypropyltrimethoxysilane, 3-methacryloxypropylmethyldiethoxysilane, 3-methacryloxypropyltriethoxysilane, 3-acryloxypropylene Monomers having an acrylic group such as trimethoxysilane, N-2- (aminoethyl) -3-aminopropylmethyldimethoxysilane, N-2- (aminoethyl) -3-aminopropyltrimethoxysilane, 3-aminopropyltri Monomers having amino groups such as methoxysilane, 3-aminopropyltriethoxysilane, 3-triethoxysilyl-N- (1,3-dimethyl-butylidene) propylamine, N-phenyl-3-aminopropyltrimethoxysilane, Monomers having a ureido group such as 3-ureidopropyltriethoxysilane, monomers having a mercapto group such as 3-mercaptopropylmethyldimethoxysilane, and sulfoxides such as bis (triethoxysilylpropyl) tetrasulfide A monomer having an alkyl group, a monomer having an isocyanate group such as 3-isocyanatopropyltriethoxysilane, a polymer obtained by polymerizing a small amount of these monomers, a composite material characterized by introducing a functional group or polymer into a part of the material, etc. The metal alkoxides may be used. In addition, some or all of the alkyl group and phenyl group of these compounds may be substituted with fluorine. Furthermore, metal acetylacetonate, metal carboxylate, oxychloride, chloride, a mixture thereof and the like can be mentioned, but not limited thereto. Examples of the metal species include, but are not limited to, Ti, Sn, Al, Zn, Zr, In, and a mixture thereof in addition to Si. What mixed suitably the precursor of the said metal oxide can also be used. Further, a mesoporous convex portion may be formed by adding a surfactant to these materials. Furthermore, a silane coupling agent having a hydrolyzable group having affinity and reactivity with silica and an organic functional group having water repellency can be used as a precursor of silica. For example, silane monomers such as n-octyltriethoxysilane, methyltriethoxysilane, and methyltrimethoxysilane, vinylsilanes such as vinyltriethoxysilane, vinyltrimethoxysilane, vinyltris (2-methoxyethoxy) silane, vinylmethyldimethoxysilane, Methacrylic silane such as 3-methacryloxypropyltriethoxysilane, 3-methacryloxypropyltrimethoxysilane, 2- (3,4-epoxycyclohexyl) ethyltrimethoxysilane, 3-glycidoxypropyltrimethoxysilane, 3-glycyl Epoxy silanes such as Sidoxypropyltriethoxysilane, 3-Mercaptopropyltrimethoxysilane, Mercaptosilanes such as 3-Mercaptopropyltriethoxysilane, 3-Octanoylthio-1-pro Sulfur silane such as rutriethoxysilane, 3-aminopropyltriethoxysilane, 3-aminopropyltrimethoxysilane, N- (2-aminoethyl) -3-aminopropyltrimethoxysilane, N- (2-aminoethyl)- Examples thereof include aminosilanes such as 3-aminopropylmethyldimethoxysilane and 3- (N-phenyl) aminopropyltrimethoxysilane, and polymers obtained by polymerizing these monomers.

When a mixture of TEOS and MTES is used as the sol-gel material solution, the mixing ratio thereof can be set to 1: 1, for example, as a molar ratio. This sol-gel material produces amorphous silica by performing hydrolysis and polycondensation reactions. In order to adjust the pH of the solution as a synthesis condition, an acid such as hydrochloric acid or an alkali such as ammonia is added. The pH is preferably 4 or less or 10 or more. Moreover, you may add water in order to perform a hydrolysis. The amount of water to be added can be 1.5 times or more in molar ratio with respect to the metal alkoxide species.

Solvents for the sol-gel material solution include, for example, alcohols such as methanol, ethanol, isopropyl alcohol (IPA) and butanol, aliphatic hydrocarbons such as hexane, heptane, octane, decane and cyclohexane, benzene, toluene, xylene, mesitylene and the like Aromatic hydrocarbons, ethers such as diethyl ether, tetrahydrofuran and dioxane, ketones such as acetone, methyl ethyl ketone, isophorone and cyclohexanone, butoxyethyl ether, hexyloxyethyl alcohol, methoxy-2-propanol and benzyloxyethanol Ether alcohols, glycols such as ethylene glycol and propylene glycol, ethylene glycol dimethyl ether, diethylene glycol dimethyl ether, Glycol ethers such as pyrene glycol monomethyl ether acetate, esters such as ethyl acetate, ethyl lactate and γ-butyrolactone, phenols such as phenol and chlorophenol, N, N-dimethylformamide, N, N-dimethylacetamide, N- Amides such as methylpyrrolidone, halogen-based solvents such as chloroform, methylene chloride, tetrachloroethane, monochlorobenzene, and dichlorobenzene, hetero-containing compounds such as carbon disulfide, water, and mixed solvents thereof can be mentioned. In particular, ethanol and isopropyl alcohol are preferable, and those in which water is mixed are also preferable.

As an additive of the sol-gel material solution, polyethylene glycol, polyethylene oxide, hydroxypropyl cellulose, polyvinyl alcohol for viscosity adjustment, alkanolamine such as triethanolamine which is a solution stabilizer, β diketone such as acetylacetone, β ketoester, Formamide, dimethylformamide, dioxane and the like can be used. Further, as an additive of the sol-gel material solution, a material that generates acid or alkali by irradiating light such as energy rays typified by ultraviolet rays such as excimer UV light can be used. By adding such a material, the sol-gel material solution can be cured by irradiation with light.

<Patterning process>
As shown in FIG. 3A, a sol-gel material film (convex portion) 60 having a predetermined pattern is formed on the substrate 40 by screen printing using the solution of the sol-gel material (inorganic material) prepared as described above. To do. As the base material 40, substrates having various translucency can be used. For example, glass, sapphire single crystal (Al ₂ O ₃ ; A plane, C plane, M plane, R plane), spinel single crystal (MgAl ₂ O ₄ ), ZnO single crystal, LiAlO ₂ single crystal, LiGaO ₂ single crystal, A substrate made of a material such as oxide single crystal such as MgO single crystal, Si single crystal, SiC single crystal, SiN single crystal, GaAs single crystal, AlN single crystal, GaN single crystal and boride single crystal such as ZrB ₂ is used. be able to. Of these, sapphire single crystal substrates and SiC single crystal substrates are preferred. In addition, the surface orientation of a base material is not specifically limited. In addition, the base material may be a just substrate having an off angle of 0 degrees or a substrate having an off angle.

The sol-gel material film can be formed as follows. First, the screen printing plate 10 is placed on the substrate 40, and a solution of the sol-gel material is applied to the entire screen printing plate 10. Next, while pressing the squeegee 22 against the screen printing plate 10 with a constant pressure, the squeegee 22 is moved at a constant speed in a direction parallel to the substrate 40 (a direction indicated by an arrow in FIG. 3A). Then, the sol-gel material solution oozes out from the opening 18 of the screen printing plate 10 through the screen 12, and the sol-gel material film (convex) at a position corresponding to the opening 18 of the screen printing plate 10 on the substrate 40. Part) 60 is formed. The shape (pattern) in plan view of the opening 18 of the screen printing plate 10 and the sol-gel material film (convex portion) 60 formed therefrom is not particularly limited, but is regularly oriented such as stripes, wavy stripes, and zigzags. It may be a regularly oriented pattern such as a pattern or a dot-like pattern. Or the convex part 60 may extend irregularly | swelling, and the extension direction, the direction of a wave | undulation, and extension length may be irregular on planar view. That is, the convex part 60 may have the characteristics that i) each has an elongated shape extending while undulating, and ii) the extending direction, the bending direction, and the length are not uniform. The convex portion 60 extending while undulating may be branched in the middle. Between the convex parts 60, the area | region (concave part 70) where the base-material surface was exposed is divided. The concave portion 70 extends along the convex portion 60, and similarly to the convex portion 60, the concave direction may have an elongated shape in which the extending direction, the waviness direction, and the extending length are irregular in plan view. The height (film thickness) of the convex portion (sol-gel material film) 60 is preferably in the range of 20 nm to 10 μm. In addition, in order to improve adhesiveness on the base material 40, you may provide a surface treatment, an easily bonding layer, etc.

<Curing process>
After the patterning step, the convex portion 60 made of a sol-gel material is cured. The convex part 60 can be hardened by carrying out main baking. By the main firing, the hydroxyl group contained in the silica (amorphous silica) constituting the convex portion 60 is detached, and the sol-gel material film (convex portion) becomes stronger. The main baking is preferably performed at a temperature of 600 to 1200 ° C. for about 5 minutes to 6 hours. In this way, as shown in FIG. 3B, the convex portion 60 is cured, and the epitaxial growth substrate 100 in which the convex portion 60 and the concave portion 70 formed on the base material 40 form the concave / convex pattern 80 is formed. Can do. At this time, when the convex portion 60 is made of silica, it becomes amorphous or crystalline, or a mixed state of amorphous and crystalline depending on the firing temperature and firing time. In addition, when a material that generates an acid or an alkali by irradiating light such as ultraviolet rays to the sol-gel material solution, energy represented by ultraviolet rays such as excimer UV light is used instead of firing the convex portion 60. The projection 60 can be cured by irradiating the line.

Further, the surface of the convex portion 60 may be subjected to a hydrophobic treatment. A known method may be used for the hydrophobizing treatment. For example, if the surface is silica, it can be hydrophobized with dimethyldichlorosilane, trimethylalkoxysilane, or the like, or trimethylsilyl such as hexamethyldisilazane. A method of hydrophobizing with an agent and silicone oil may be used, or a surface treatment method of metal oxide powder using supercritical carbon dioxide may be used.

Further, after the curing step, as shown in FIG. 3C, the exposed base material surface may be etched to form a recess 70 a in the base material 40. Thereby, the substrate 100a for epitaxial growth in which the uneven | corrugated pattern 80a which consists of the convex part 60 and the recessed part 70a was formed can be formed. In this epitaxial growth substrate 100a, since the concave portion 70a is formed in the base material 40, the concave / convex depth of the concave / convex pattern can be increased as compared with the substrate 100 in which the base material 40 is not etched. When a sapphire substrate is used as the base material 40, the base material 40 can be etched by RIE using a gas containing BCl ₃ or the like, for example.

A buffer layer may be further formed on the surface of the substrate on which the uneven patterns 80 and 80a are formed as described above (the surface on which the uneven pattern is formed). Thereby, the epitaxial growth substrates 100b and 100c having the buffer layer 20 on the surfaces of the concave and convex patterns 80 and 80a as shown in FIGS. 4A and 4B are obtained. When the cross-sectional shape of the concavo-convex pattern is a relatively gentle inclined surface and has a corrugated structure, a uniform buffer layer with few defects can be formed.

</ RTI> Before the patterning step, a buffer layer may be formed on the substrate. As a result, as shown in FIG. 4C, the convex portion 60 is formed on the buffer layer 20, and the region where the surface of the buffer layer 20 is exposed (the concave portion 70 b) is defined between the convex portions 60. Is done. As a result, an epitaxial growth substrate 100d on which the concavo-convex pattern 80b is formed is obtained.

The buffer layer 20 can be formed using a known method such as a low temperature MOCVD method or a sputtering method. The layer thickness of the buffer layer 20 is preferably in the range of 1 nm to 100 nm. When the semiconductor layer is epitaxially grown on the surfaces of the epitaxial growth substrates 100b, 100c, and 100d having the buffer layer, the difference in lattice constant between the substrate and the semiconductor layer is relaxed by the buffer layer, so that a highly crystalline semiconductor layer can be formed. When epitaxially growing a GaN-based semiconductor layer on the epitaxial growth substrate of the embodiment, the buffer layer can be composed of Al _X Ga _1-X N (0 ≦ x ≦ 1), and is not limited to a single layer structure. Alternatively, a multilayer structure of two or more layers in which two or more kinds having different compositions are laminated may be used.

In the manufacturing method of the substrate for epitaxial growth which forms an uneven | corrugated pattern using the normal nanoimprint method like the prior art described in patent document 1, it forms in the area | region corresponding to the convex part of the mold of a base material after mold peeling. The remaining film needs to be removed by etching by RIE or the like. However, in the method for manufacturing the epitaxial growth substrate of this embodiment, the inorganic material film is formed only at a predetermined position (position where the convex portion is formed) on the base material by screen printing. Since the concavo-convex pattern is formed by forming the film, there is no need for etching for removing the remaining film. Therefore, the substrate manufacturing time can be shortened by the epitaxial growth substrate manufacturing method of the present embodiment. In addition, when the substrate surface is exposed to etching, the substrate surface exposed by etching may be rough (damaged), and chemical processing may be required after etching. Since the substrate manufacturing method does not require etching, such damage does not occur, and there is no need for chemical treatment. Therefore, the substrate manufacturing process can be simplified and the manufacturing time can be shortened by the epitaxial growth substrate manufacturing method of the present embodiment.

In addition, the method for manufacturing an epitaxial growth substrate according to the present embodiment forms a concavo-convex pattern by screen printing as described above without using photolithography or nanoimprinting, thereby reducing the production cost of the epitaxial growth substrate and The load can be reduced.

In the epitaxial growth substrate 100 formed by the manufacturing method as described above, since the convex portion 60 is formed of an inorganic material, the epitaxial growth substrate 100 has excellent heat resistance.

It is to be noted that the epitaxial growth substrates 100 and 100a to 100d shown in FIGS. 3B and 4C and FIGS. 4A to 4C formed by the manufacturing method of the present embodiment are formed on the base material 40. The plurality of convex portions 60 and concave portions 70, 70 a, 70 b formed are concave / convex patterns 80, 80 a, 80 b.

FIG. 5A shows an example of an AFM image of the epitaxial growth substrate manufactured by the manufacturing method of the present embodiment, and FIG. 5B shows an epitaxial growth substrate on the cutting line in the AFM image of FIG. 5A. The cross-sectional profile of is shown.

The cross-sectional shape of the concavo-convex pattern of the substrate for epitaxial growth is not particularly limited, but as shown in FIGS. 3B, 3C, 4A, 4B, 5C, and 5B, It may be formed of a relatively gentle inclined surface, and may have a waveform (referred to as “corrugated structure” as appropriate in this application) upward from the substrate 40. That is, the convex part 60 may have a cross-sectional shape that becomes narrower from the bottom part on the base material side toward the top part.

The planar shape of the concavo-convex pattern of the substrate for epitaxial growth is not particularly limited, and may be a regularly oriented pattern such as a stripe, a wavy stripe, a zigzag or a regularly oriented pattern such as FIG. As shown in a) an example of the AFM image of the concavo-convex pattern on the substrate surface, the convex portion (white portion) and the concave portion are wavyly extended, and the extending direction, the direction of waviness and the extending length are flat. It may be irregular in appearance. That is, i) the convex portion and the concave portion each have an elongated shape extending while undulating, and ii) the convex portion and the concave portion have a feature that the extending direction, the bending direction, and the length are uneven in the concavo-convex pattern. You may have. When the concavo-convex pattern of the substrate for epitaxial growth has the above characteristics, the concavo-convex cross section appears repeatedly even if the concavo-convex pattern 80 is cut in any direction orthogonal to the surface of the substrate 40. Moreover, a convex part and a recessed part may be branched in part or in the middle by planar view (refer Fig.5 (a)). In FIG. 5A, the pitch of the convex portions and the concave portions appears to be uniform as a whole. Further, the concave portion 70 of the concavo-convex pattern 80 is partitioned by the convex portion 60 and extends along the convex portion 60, and similarly to the convex portion 60, the extending direction, the direction of waviness and the extending length are in plan view. It may be an irregular elongated shape.

When the epitaxial growth substrate is used as a substrate of a light emitting device formed of, for example, a GaN-based semiconductor material, in order to improve the light extraction efficiency of the light emitting device, the frequency distribution is such that the pitch of the unevenness becomes an annular shape in the Fourier transform image. In particular, an irregular pattern that has no directivity in the direction of the projections and depressions is preferable. In order for the epitaxial growth substrate 100 to function as a diffraction grating that improves the light extraction efficiency of the light-emitting element, the average pitch of the irregularities is preferably in the range of 100 nm to 10 μm, and more preferably in the range of 100 to 1500 nm. . If the average pitch of the unevenness is less than the lower limit, the pitch is too small with respect to the emission wavelength of the light-emitting element, so there is a tendency that light diffraction due to the unevenness does not occur, while if the upper limit is exceeded, the diffraction angle decreases, The function as a diffraction grating tends to be lost. The average pitch of the irregularities is more preferably in the range of 200 to 1200 nm.

The average value of the uneven depth distribution is preferably in the range of 20 nm to 10 μm. The average value of the depth distribution of the irregularities is more preferably in the range of 50 nm to 5 μm, and it is necessary because the average value of the depth distribution of the irregularities is less than the lower limit, the depth is too small with respect to the emission wavelength. On the other hand, if the upper limit is exceeded, the thickness of the semiconductor layer required for planarization of the surface of the semiconductor layer becomes large when a semiconductor layer is laminated on the substrate to produce a light emitting device. Therefore, the time required for manufacturing the light emitting element becomes longer. The average value of the uneven depth distribution is more preferably in the range of 100 nm to 2 μm. The standard deviation of the unevenness depth is preferably in the range of 10 nm to 5 μm. If the standard deviation of the depth of the unevenness is less than the lower limit, the required diffraction tends not to occur because the depth is too small with respect to the wavelength of visible light. On the other hand, if the upper limit is exceeded, the diffracted light intensity is uneven. Tend to occur. The standard deviation of the unevenness depth is more preferably in the range of 25 nm to 2.5 μm.

When the epitaxial growth layer is formed on the epitaxial growth substrate 100 on which the concavo-convex pattern having elongated convex portions and concave portions extending in an irregular direction while undulating is formed, there are the following advantages. First, since the concavo-convex inclined surface is relatively gentle, the epitaxial growth layer is uniformly laminated on the concavo-convex pattern 80, and an epitaxial layer with few defects can be formed. Furthermore, since the concavo-convex pattern has an irregular shape with no directivity in the direction of the concavo-convex, even if a defect due to the pattern occurs, a homogeneous epitaxial growth layer having no anisotropy in the defect can be formed.

Further, when a light emitting device is manufactured by epitaxially growing a semiconductor layer on the epitaxial growth substrate 100 having such an uneven pattern, there are the following advantages. First, since an epitaxial growth substrate having such a concavo-convex pattern has high light extraction efficiency, a light-emitting device manufactured using this substrate has high light emission efficiency. Second, since the light diffracted by the epitaxial growth substrate having such a concavo-convex pattern has no directivity, the light extracted from the light emitting element manufactured using this substrate goes in all directions without directivity. Third, the manufacturing time of the light emitting device can be shortened for the following reason. When a light-emitting element is manufactured using a substrate having a concavo-convex pattern, it is necessary to stack the semiconductor layer until the concavo-convex shape is filled with the semiconductor layer and the surface becomes flat, as will be described later. An epitaxial growth substrate on which a concavo-convex pattern having elongated convex portions and concave portions extending in irregular directions in a wavy manner has sufficient light extraction efficiency with a concavo-convex depth of the order of several tens of nanometers. Compared with the conventional substrate having a concavo-convex pattern with a concavo-convex depth on the order of submicron to micrometer as described in (1), the layer thickness for stacking the semiconductor layers can be reduced. Therefore, the growth time of the semiconductor layer can be shortened, and the manufacturing time of the light emitting element can be shortened.

In the present application, the average pitch of the unevenness means the average value of the unevenness pitch when the unevenness pitch on the surface where the unevenness is formed (adjacent protrusions or adjacent recesses). Say. The average value of the pitch of such irregularities is as follows using a scanning probe microscope (for example, product name “E-sweep” manufactured by Hitachi High-Tech Science Co., Ltd.):
Measuring method: Cantilever intermittent contact method Cantilever material: Silicon Cantilever lever width: 40μm
Cantilever tip tip diameter: 10 nm
After measuring the unevenness of the surface and measuring the unevenness analysis image, measure 100 or more intervals between any adjacent protrusions or adjacent recesses in the unevenness analysis image, and obtain the arithmetic average Can be calculated.

In the present application, the average value of the uneven depth distribution and the standard deviation of the uneven depth can be calculated as follows. The shape of the surface unevenness is measured by using an inspection probe microscope (for example, product name “E-sweep” manufactured by Hitachi High-Tech Science Co., Ltd.) to measure the unevenness analysis image. At the time of the unevenness analysis, an arbitrary 3 μm square (3 μm long, 3 μm wide) or 10 μm square (10 μm long, 10 μm wide) measurement is performed under the above-described conditions to obtain the uneven analysis image. In that case, the data of the unevenness | corrugation height in the measurement point more than 16384 points (vertical 128 points x horizontal 128 points) in a measurement area | region are each calculated | required by a nanometer scale. The number of such measurement points varies depending on the type and setting of the measurement device used. For example, the product name “E-sweep” manufactured by Hitachi High-Tech Science Co., Ltd. is used as the measurement device. In this case, 65536 points (256 vertical points × 256 horizontal points) can be measured (measured at a resolution of 256 × 256 pixels) in a measurement area of 10 μm square. Here, in order to increase the measurement accuracy, the unevenness analysis image may be subjected to flat processing including primary inclination correction. Moreover, in order to ensure sufficient measurement accuracy in various analyzes related to the uneven shape described below, the measurement region has a length of 15 times or more of the average value of the widths of the convex portions included in the measurement region. It is preferable to use a square region with a length of. And about the uneven | corrugated height (unit: nm) measured in this way, first, the height from the bottom face (surface on the opposite side of the surface where the uneven | corrugated pattern was formed) of all the measurement points is the height. The highest measurement point P is obtained. Then, a surface including such a measurement point P and parallel to the bottom surface of the base material is defined as a reference surface (horizontal plane), and each depth value from the reference surface (a height value from the base material bottom surface at the measurement point P is measured). The difference obtained by subtracting the height from the bottom surface of the substrate at the measurement point) is obtained as the data of the unevenness depth. Such unevenness depth data can be obtained by automatically calculating with software or the like in the measuring device depending on the measuring device (for example, product name “E-sweep” manufactured by Hitachi High-Tech Science Co., Ltd.) A value obtained by such automatic calculation can be used as the data of the unevenness depth.

In this way, after obtaining the unevenness depth data at each measurement point, the values that can be calculated by obtaining the arithmetic mean and standard deviation thereof are the average value of the unevenness depth distribution and the standard deviation of the unevenness depth, respectively. adopt. In this specification, the average pitch of the unevenness and the average value of the depth distribution of the unevenness can be obtained through the measurement method as described above regardless of the material of the surface on which the unevenness is formed.

In addition, in the present application, an “irregular uneven pattern” means that a Fourier transform image obtained by performing a two-dimensional fast Fourier transform process on an unevenness analysis image obtained by analyzing the surface unevenness shape has an absolute value of wave number. It includes a quasi-periodic structure that shows a circular or annular pattern having an origin substantially at the center of 0 μm ⁻¹ , that is, has a concavo-convex pitch distribution although it has no directivity in the direction of the concavo-convex. The circular or annular pattern may have an absolute value of wave number of 10 μm ⁻¹ or less (may be in the range of 0.1 to 10 μm ⁻¹ , and may further be in the range of 0.667 to 10 μm ⁻¹ , preferably May be within a range of 0.833 to 5 μm ⁻¹ ). The light scattered and / or diffracted from such a concavo-convex pattern has a relatively broad wavelength band, not light of a single or narrow band wavelength, and the scattered light and / or diffracted light is directed. There is no sex and heads in all directions. Therefore, a substrate having such a quasi-periodic structure is suitable for a substrate used for a light emitting element such as an LED as long as the uneven pitch distribution diffracts visible light.

In the Fourier transform image obtained by subjecting the unevenness analysis image to the two-dimensional fast Fourier transform process, a pattern is observed by gathering bright spots. Therefore, here, “Fourier transform image shows a circular pattern” means that the pattern of bright spots in the Fourier transform image looks almost circular, and part of the outer shape is convex or Includes those that appear to be concave. In addition, “the Fourier transform image shows an annular pattern” means that the pattern in which the bright spots are gathered in the Fourier transform image looks almost an annular shape, and the shape of the outer circle or inner circle of the ring is This includes those that appear to have a substantially circular shape, and those that appear to have a convex or concave part of the outer circle of the annulus and the inner circle. Further, “a circular or annular pattern may have an absolute value of a wave number of 10 μm ⁻¹ or less (within a range of 0.1 to 10 μm ⁻¹ , and further within a range of 0.667 to 10 μm ^−1. , Preferably within a range of 0.833 to 5 μm ⁻¹ ) ”means that 30% or more of the bright spots constituting the Fourier transform image have a wave number of 30% or more. Absolute value of 10 μm ⁻¹ or less (may be in the range of 0.1 to 10 μm ⁻¹ , more preferably in the range of 0.667 to 10 μm ⁻¹ , preferably in the range of 0.833 to 5 μm ⁻¹ . It may be present in a region that falls within the range. By forming the concavo-convex pattern so as to satisfy the above conditions, when the epitaxial growth substrate of the embodiment is used as a substrate of a light emitting device, the wavelength dependency and directivity of light emitted from the light emitting device (light emission in a certain direction strongly) Property) can be made sufficiently small.

The following is known about the relationship between the concave-convex pattern and the Fourier transform image. If the concavo-convex pattern itself has no distribution or directivity in the pitch, the Fourier transform image also appears as a random pattern (no pattern), but the concavo-convex pattern is isotropic in the XY direction as a whole, but the distribution in the pitch is In some cases, a circular or annular Fourier transform image appears. Further, when the concavo-convex pattern has a single pitch, the ring appearing in the Fourier transform image tends to be sharp.

The two-dimensional fast Fourier transform processing of the unevenness analysis image can be easily performed by electronic image processing using a computer equipped with two-dimensional fast Fourier transform processing software.

It should be noted that the planar analysis image (black and white image) as shown in FIG. 7 is obtained by processing the concave / convex analysis image so that the convex portion is displayed in white and the concave portion is displayed in black. FIG. 7 is a diagram showing an example of a planar view analysis image of the measurement region in the epitaxial growth substrate 100 according to the present embodiment.

The width of the convex portion (white display portion) of the planar view analysis image is referred to as “the width of the convex portion”. For the average value of the widths of such convex portions, arbitrary 100 or more locations are selected from the convex portions of the planar view analysis image, and the respective directions are substantially perpendicular to the extending direction of the convex portions in plan view. It can be calculated by measuring the length from the boundary of the convex part to the boundary on the opposite side and obtaining the arithmetic average thereof.

In addition, when calculating the average value of the width of the convex portion, as described above, the value at the position randomly extracted from the convex portion of the planar analysis image is used, but the position where the convex portion is branched. The value of may not be used. Whether or not a certain region is a region related to branching in the convex portion may be determined, for example, based on whether or not the region extends more than a certain amount. More specifically, the determination may be made based on whether or not the ratio of the extension length of the region to the width of the region is a certain value (for example, 1.5) or more.

8 (a) and 8 (b), whether or not the region branches in the middle of the convex portion extending in a certain direction and protruding in a direction substantially perpendicular to the extending axis of the convex portion. An example of a method for determining whether will be described. Here, the extending axis of the convex portion is a virtual axis along the extending direction of the convex portion determined from the shape of the outer edge of the convex portion when the region to be determined whether to branch is excluded from the convex portion. It is. More specifically, the extending axis of the convex portion is a line drawn so as to pass through the approximate center point of the width of the convex portion orthogonal to the extending direction of the convex portion. FIG. 8A and FIG. 8B are schematic diagrams for explaining only a part of the convex portion in the planar view analysis image, and the region S indicates the convex portion. In FIG. 8A and FIG. 8B, it is assumed that the regions A1 and A2 protruding at the midway position of the convex portion are determined as the determination target regions for branching. In this case, when the regions A1 and A2 are excluded from the convex portion, the extending axes L1 and L2 are defined as lines passing through the approximate center point of the width of the convex portion orthogonal to the extending direction of the convex portion. Such an extended axis may be defined by image processing by a computer, may be defined by an operator who performs analysis work, or is defined by both image processing by a computer and manual operation by an operator. May be. In FIG. 8A, the region A1 protrudes in a direction perpendicular to the extending axis L1 at a midway position of the convex portion extending along the extending axis L1. In FIG. 8B, the region A2 protrudes in a direction perpendicular to the extending axis L2 at a midway position of the convex portion extending along the extending axis L2. It should be noted that the region that inclines and protrudes with respect to the direction orthogonal to the extending axes L1 and L2 may be determined by using the same idea as that for the regions A1 and A2 described below. .

According to the above determination method, since the ratio of the extension length d2 of the region A1 to the width d1 of the region A1 is approximately 0.5 (less than 1.5), the region A1 is not a branching region. Determined. In this case, the length d3 in the direction passing through the region A1 and orthogonal to the extending axis L1 is one of the measurement values for calculating the average value of the widths of the protrusions. On the other hand, since the ratio of the extension length d5 of the region A2 to the width d4 of the region A2 is approximately 2 (1.5 or more), the region A2 is determined to be a branching region. In this case, the length d6 in the direction passing through the region A2 and orthogonal to the extending axis L2 is not one of the measurement values for calculating the average value of the widths of the protrusions.

In the epitaxial growth substrate 100 of the present embodiment, the width of the protrusions in a direction substantially orthogonal to the extending direction of the protrusions of the uneven pattern 80 in a plan view may be constant. Whether or not the width of the convex portion is constant can be determined based on the width of the convex portion of 100 points or more obtained by the above measurement. Specifically, an average value of the widths of the protrusions and a standard deviation of the widths of the protrusions are calculated from the widths of the protrusions of 100 points or more. Then, the value calculated by dividing the standard deviation of the width of the convex portion by the average value of the width of the convex portion (standard deviation of the width of the convex portion / average value of the width of the convex portion) is the variation coefficient of the width of the convex portion. It is defined as The variation coefficient becomes smaller as the width of the convex portion is constant (the variation in the width is smaller). Therefore, whether or not the width of the convex portion is constant can be determined depending on whether or not the variation coefficient is equal to or less than a predetermined value. For example, it can be defined that the width of the convex portion is constant when the variation coefficient is 0.25 or less.

Further, as shown in FIG. 7, in the epitaxial growth substrate 100 according to the present embodiment, the extending directions of the convex portions (white portions) included in the concave / convex pattern are irregularly distributed in plan view. That is, the convex portion has a shape extending in an irregular direction, not a regular stripe shape or a regularly arranged dot shape. In addition, in the measurement region, that is, the predetermined region of the concavo-convex pattern, the contour line in the plan view of the convex portion included in the region per unit area includes more straight sections than curved sections.

In the present embodiment, “including more straight sections than curved sections” means that the concave / convex pattern does not occupy a lot of sections that are winding in all sections on the contour of the convex portion. Whether or not the outline of the convex portion in plan view includes more straight sections than curved sections can be determined, for example, by using one of the following two methods of defining a curved section. .

<First definition method of curve section>
In the first definition method of the curved section, the curved section is divided into a plurality of sections by dividing the outline of the convex portion in plan view by a length that is π (circumferential ratio) times the average value of the width of the convex portion. When formed, it is defined as a section in which the ratio of the straight line distance between the two end points to the length of the contour line between the two end points of the section is 0.75 or less. Further, the straight section is defined as a section other than the curved section among the plurality of sections, that is, a section where the ratio is greater than 0.75. Hereinafter, with reference to FIG. 9A, an example of a procedure for determining whether or not the contour line of the convex portion in plan view includes more straight sections than curved sections using the first definition method. explain. FIG. 9A is a diagram illustrating a part of the planar view analysis image of the concavo-convex pattern, and the concave portions are shown in white for convenience. Region S1 represents a convex portion, and region S2 represents a concave portion.

Procedure 1-1
One convex portion is selected from the plurality of convex portions in the measurement region. An arbitrary position on the contour X of the convex portion is determined as a start point. In FIG. 9A, as an example, the point A is set as the start point. Reference points are provided at predetermined intervals on the contour line X of the convex portion from the start point. Here, the predetermined interval is a length that is π (circumferential ratio) / 2 times the average value of the widths of the convex portions. In FIG. 9A, point B, point C, and point D are sequentially set as an example.

Procedure 1-2
When the points A to D, which are reference points, are set on the contour line X of the convex portion, a determination target section is set. Here, the start point and the end point are reference points, and a section including a reference point serving as an intermediate point is set as a determination target. In the example of FIG. 9A, when the point A is selected as the start point of the section, the point C set second from the point A is the end point of the section. Since the distance from the point A is set to a length that is π / 2 times the average value of the width of the convex portion here, the point C is π of the average value of the width of the convex portion along the contour line X. It is a point away from the point A by a double length. Similarly, when the point B is selected as the start point of the section, the point D set second from the point B is the end point of the section. Here, it is assumed that the target section is set in the set order, and point A is the point set first. That is, first, the section between section A and point C (section AC) is set as a section to be processed. Then, the length La of the contour X of the convex portion connecting the points A and C and the linear distance Lb between the points A and C shown in FIG. 9A are measured.

Procedure 1-3
A ratio (Lb / La) of the linear distance Lb to the length La is calculated using the length La and the linear distance Lb measured in the procedure 1-2. When the ratio is 0.75 or less, it is determined that the point B that is the midpoint of the section AC of the contour line X of the convex portion is a point existing in the curve section. On the other hand, when the ratio is larger than 0.75, it is determined that the point B is a point existing in the straight section. In the example shown in FIG. 9A, since the ratio (Lb / La) is 0.75 or less, the point B is determined to be a point existing in the curve section.

Procedure 1-4
When each point set in the procedure 1-1 is selected as the start point, the procedure 1-2 and the procedure 1-3 are executed.

Step 1-5
Steps 1-1 to 1-4 are executed for all the convex portions in the measurement region.

Step 1-6
The contour of the convex portion in plan view when the proportion of the points determined to be in the straight line segment among all the points set for all the convex portions in the measurement region is 50% or more of the whole. It is determined that the line includes more straight sections than curved sections. On the other hand, when the proportion of the points determined to be in the straight line segment among all the points set for all the convex portions in the measurement region is less than 50% of the whole, the plan view of the convex portions It is determined that the upper contour line includes more curved sections than straight sections.

The processes in steps 1-1 to 1-6 may be performed by a measurement function provided in the measurement apparatus, may be performed by executing analysis software or the like different from the measurement apparatus, or may be performed manually. You may go on.

Note that the process of setting points on the contour of the convex portion in step 1-1 ends when it is no longer possible to set points by going around the convex portion or protruding from the measurement area. do it. Further, since the ratio (Lb / La) cannot be calculated for the section outside the first set point and the last set point, it may be excluded from the above determination. Moreover, what is necessary is just to exclude the convex part in which the length of an outline is less than (pi) times the average value of the width | variety of a convex part.

<Second definition method of curve section>
In the second definition method of the curved section, the curved section is divided into a plurality of sections by dividing an outline of the convex portion in plan view by a length that is π (circumferential ratio) times the average value of the width of the convex portion. When formed, a line segment (line segment AB) connecting one end (point A) of the section and the midpoint (point B) of the section, the other end (point C) of the section, and the midpoint (point of the section) Of the two angles formed by the line segment (line segment CB) connecting B), the smaller angle (the one that is 180 ° or less) is defined as a section in which the angle is 120 ° or less. The straight section is defined as a section other than the curved section among the plurality of sections, that is, a section in which the angle is larger than 120 °. Hereinafter, with reference to FIG. 9B, an example of a procedure for determining whether or not the contour line of the convex portion in plan view includes more straight sections than curved sections using the second definition method. explain. FIG. 9B is a diagram showing a part of the planar analysis image of the same concavo-convex pattern as FIG.

Procedure 2-1
One convex portion is selected from the plurality of convex portions in the measurement region. An arbitrary position on the contour X of the convex portion is determined as a start point. In FIG. 9B, as an example, the point A is set as the start point. Reference points are provided at predetermined intervals on the contour line X of the convex portion from the start point. Here, the predetermined interval is a length that is π (circumferential ratio) / 2 times the average value of the widths of the convex portions. In FIG. 9B, point B, point C, and point D are sequentially set as an example.

Procedure 2-2
When the points A to D, which are reference points, are set on the contour line X of the convex portion, a determination target section is set. Here, the start point and the end point are reference points, and a section including a reference point serving as an intermediate point is set as a determination target. In the example of FIG. 9B, when the point A is selected as the start point of the section, the point C set second from the point A becomes the end point of the section. Since the distance from the point A is set to a length that is π / 2 times the average value of the width of the convex portion here, the point C is π of the average value of the width of the convex portion along the contour line X. It is a point away from the point A by a double length. Similarly, when the point B is selected as the start point of the section, the point D set second from the point B is the end point of the section. Here, it is assumed that the target section is set in the set order, and point A is the point set first. That is, first, the section of point A and point C is set as a process target section. Then, the smaller angle θ (the one that is 180 ° or less) of the two angles formed by the line segment AB and the line segment CB is measured.

Procedure 2-3
When the angle θ is 120 ° or less, it is determined that the point B is a point existing in the curve section. On the other hand, when the angle θ is larger than 120 °, it is determined that the point B is a point existing in the straight line section. In the example shown in FIG. 9B, since the angle θ is 120 ° or less, the point B is determined as a point existing in the curve section.

Step 2-4
When each point set in the procedure 2-1 is selected as the start point, the procedure 2-2 and the procedure 2-3 are executed.

Step 2-5
Steps 2-1 to 2-4 are executed for all convex portions in the measurement region.

Step 2-6
The contour of the convex portion in plan view when the proportion of the points determined to be in the straight line segment among all the points set for all the convex portions in the measurement region is 70% or more of the whole. It is determined that the line includes more straight sections than curved sections. On the other hand, when the ratio of the points determined to be in the straight section among all the points set for all the convex portions in the measurement region is less than 70% of the whole, the plan view of the convex portions It is determined that the upper contour line includes more curved sections than straight sections.

The processing of steps 2-1 to 2-6 may be performed by a measurement function provided in the measurement device, or may be performed by executing analysis software or the like different from the measurement device. It may be done manually.

Note that the process of setting points on the contour of the convex part in step 2-1 above ends when it is no longer possible to set points by going around the convex part or protruding from the measurement area. do it. Further, since the angle θ cannot be calculated for the section outside the first set point and the last set point, it may be excluded from the above determination. Moreover, what is necessary is just to exclude the convex part in which the length of an outline is less than (pi) times the average value of the width | variety of a convex part.

As described above, by using one of the first and second definition methods of the curve section, the contour line X in the plan view of the convex portion includes more straight sections than the curve section in the measurement region. It can be determined whether or not. In addition, for the uneven pattern 80 of a certain substrate 100 for epitaxial growth, the determination of “whether the contour line in the plan view of the convex portion included in the region per unit area includes more straight sections than curved sections” is epitaxial growth. The determination may be made based on one measurement region that is randomly extracted from the region of the concave / convex pattern 80 of the substrate 100 for measurement, or a plurality of different measurements in the concave / convex pattern 80 of the same substrate 100 for epitaxial growth. The determination may be performed by comprehensively determining the determination result for the region. In this case, for example, the determination result of the larger one among the determination results for a plurality of different measurement regions is expressed as “the contour line in the plan view of the convex portion included in the region per unit area has more straight sections than the curved sections. You may employ | adopt as a determination result of "whether it is included".

In the above-described embodiment, a solution of sol-gel material such as TiO ₂ , ZnO, ZnS, ZrO, BaTiO ₃ , SrTiO ₂ or a fine particle dispersion may be used as the inorganic material solution used in the patterning step. Among, TiO ₂ is preferred from the relationship of the film forming property and refractive index. Among, TiO ₂ is preferred from the relationship of the film forming property and refractive index. The inorganic material film may be formed by using a liquid phase deposition (LPD) method or the like.

Moreover, you may use a polysilazane solution as an inorganic material used in a patterning process. The convex part formed using the polysilazane solution may be converted into ceramics (silica modification) in the curing step to form the convex part made of silica. “Polysilazane” is a polymer having a silicon-nitrogen bond, such as SiO ₂ , Si ₃ N _{4 made of} Si—N, Si—H, N—H, etc., and ceramics such as both intermediate solid solutions SiO _X N _Y. It is a precursor inorganic polymer. More preferred is a compound which is converted to silica by being ceramicized at a relatively low temperature as represented by the following general formula (1) described in JP-A-8-112879.

General formula (1):
—Si (R1) (R2) —N (R3) —
In the formula, R1, R2, and R3 each represent a hydrogen atom, an alkyl group, an alkenyl group, a cycloalkyl group, an aryl group, an alkylsilyl group, an alkylamino group, or an alkoxy group.

Among the compounds represented by the general formula (1), perhydropolysilazane (also referred to as PHPS) in which all of R 1, R 2 and R 3 are hydrogen atoms, and the hydrogen part bonded to Si is partially an alkyl group or the like. Substituted organopolysilazanes are particularly preferred.

As another example of polysilazane to be ceramicized at a low temperature, silicon alkoxide-added polysilazane obtained by reacting polysilazane with silicon alkoxide (for example, JP-A No. 5-23827), glycidol-added polysilazane obtained by reacting glycidol ( For example, JP-A-6-122852), an alcohol-added polysilazane obtained by reacting an alcohol (for example, JP-A-6-240208), a metal carboxylate-added polysilazane obtained by reacting a metal carboxylate ( For example, JP-A-6-299118), an acetylacetonate complex-added polysilazane obtained by reacting a metal-containing acetylacetonate complex (for example, JP-A-6-306329), and metal fine particles are added. Metal fine particles Pressurized polysilazane (e.g., JP-A-7-196986) and the like can also be used.

As the solvent of the polysilazane solution, hydrocarbon solvents such as aliphatic hydrocarbons, alicyclic hydrocarbons and aromatic hydrocarbons, halogenated hydrocarbon solvents, ethers such as aliphatic ethers and alicyclic ethers can be used. In order to promote the modification to a silicon oxide compound, an amine or metal catalyst may be added.

[Light emitting element]
A light emitting element can be manufactured using the substrate for epitaxial growth obtained by the manufacturing method of the substrate for epitaxial growth of the said embodiment. As shown in FIG. 6, the light emitting device 200 according to the embodiment is formed by stacking a first conductivity type layer 222, an active layer 224, and a second conductivity type layer 226 in this order on the epitaxial growth substrate 100. The semiconductor layer 220 is provided. Furthermore, the light emitting device 200 of the embodiment includes a first electrode 240 that is electrically connected to the first conductivity type layer 222 and a second electrode 260 that is electrically connected to the second conductivity type layer 226.

As a material of the semiconductor layer 220, a known material used for a light-emitting element may be used. As a material used for a light emitting element, for example, a GaN-based semiconductor material represented by a general formula In _x Al _y Ga _1-xy N (0 ≦ x ≦ 1, 0 ≦ y ≦ 1, 0 ≦ x + y ≦ 1) In the light emitting device of the present embodiment, a GaN-based semiconductor represented by the general formula Al _X Ga _Y In _{ZN 1-} AM _A is used without any limitation in the light-emitting element of this embodiment. be able to. GaN-based semiconductors can contain other group III elements in addition to Al, Ga, and In, and contain elements such as Ge, Si, Mg, Ca, Zn, Be, P, As, and B as required. You can also Furthermore, it is not limited to elements that are intentionally added, but may contain impurities that are inevitably contained depending on the growth conditions of the semiconductor layer, and trace impurities contained in the raw materials and reaction tube materials. In addition to the nitride semiconductor, other semiconductor materials such as GaAs, GaP-based compound semiconductor, AlGaAs, InAlGaP-based compound semiconductor can also be used.

The n-type semiconductor layer 222 as the first conductivity type layer is stacked on the substrate 100. The n-type semiconductor layer 222 may be formed of materials and structures known in the art, and may be formed of, for example, n-GaN. The active layer 224 is stacked on the n-type semiconductor layer 222. The active layer 224 may be formed of materials and structures known in the art, and may have, for example, a multiple quantum well (MQW) structure in which GalnN and GaN are stacked a plurality of times. The active layer 224 emits light by injection of electrons and holes. A p-type semiconductor layer 226 as a second conductivity type layer is stacked on the active layer 224. The p-type semiconductor layer 226 may have a structure known in the art, and may be formed of, for example, p-AlGaN and p-GaN. The method for stacking the semiconductor layers (n-type semiconductor layer, active layer and p-type semiconductor layer) is not particularly limited, and MOCVD (metal organic chemical vapor deposition), HVPE (hydride vapor deposition), MBE (molecular beam epitaxy). A known method that can grow a GaN-based semiconductor can be applied. The MOCVD method is preferable from the viewpoint of layer thickness controllability and mass productivity.

Although a concavo-convex pattern 80 is formed on the surface of the substrate 100 for epitaxial growth, the surface is flattened by lateral growth of the semiconductor layer as described in JP-A-2001-210598 during the epitaxial growth of the n-type semiconductor layer. Progresses. Since the active layer needs to be formed on a flat surface, it is necessary to stack an n-type semiconductor layer until the surface becomes flat. The substrate for epitaxial growth according to the embodiment has a relatively gentle cross-sectional shape of the concavo-convex pattern, and has a corrugated structure, so that the surface flattening progresses quickly and the thickness of the n-type semiconductor layer is reduced. Can do. The growth time of the semiconductor layer can be shortened.

The n-electrode 240 as the first electrode is formed on the n-type semiconductor layer 222 exposed by etching a part of the p-type semiconductor layer 226 and the active layer 224. The n-electrode 222 may be formed of a material and structure known in the art, and is made of, for example, Ti / Al / Ti / Au or the like, and is formed by a vacuum deposition method, a sputtering method, a CVD method, or the like. A p-electrode 260 as the second electrode is formed on the p-type semiconductor layer 226. The p-electrode 226 may be formed of a material and structure known in the art, and may be formed of, for example, a translucent conductive film made of ITO or the like and an electrode pad made of a Ti / Au laminated body or the like. The p-electrode 260 may be formed from a highly reflective material such as Ag or Al. The n-electrode 240 and the p-electrode 260 can be formed by any film forming method such as a vacuum deposition method, a sputtering method, a CVD method, or the like.

In addition, when a voltage is applied to the first conductivity type layer and the second conductivity type layer, the active layer includes at least a first conductivity type layer, an active layer, and a second conductivity type layer. The layer structure of the semiconductor layer is arbitrary as long as it emits light.

The optical element 200 of the embodiment configured as described above may be a face-up optical element that extracts light from the p-type semiconductor 226 side. In that case, a light-transmitting conductive material is used for the p-electrode 260. It is preferable. The optical element 200 of the embodiment may be a flip-chip optical element that extracts light from the substrate 100 side. In that case, it is preferable to use a highly reflective material for the p-electrode 260. In any method, the light generated in the active layer 224 can be effectively extracted outside the device by the diffraction effect of the concave / convex pattern 80 of the substrate.

Further, in the optical element 200, since the concave / convex pattern 80 is formed on the substrate 100, the semiconductor layer 220 having a low dislocation density is formed, and deterioration of the characteristics of the light emitting element 200 is suppressed.

As mentioned above, although this invention was demonstrated by embodiment, the manufacturing method and optical element of the substrate for epitaxial growth of this invention are not limited to the said embodiment, It changes suitably within the range of the technical idea described in the claim. can do.

The epitaxial growth substrate can be produced continuously at a high speed by the epitaxial growth substrate manufacturing method of the present invention. Further, since photolithography is not used to form the concavo-convex pattern, the manufacturing cost is low and the burden on the environment is small. Furthermore, since the substrate for epitaxial growth obtained by the manufacturing method of the present invention has a function as a diffraction grating substrate that improves the light extraction efficiency, a light emitting device manufactured using this substrate has high light emission efficiency. Therefore, the epitaxial growth substrate obtained by the production method of the present invention is extremely effective for the production of a light emitting device having excellent light emission efficiency, and contributes to energy saving.

DESCRIPTION OF SYMBOLS 10 Screen printing plate, 20 Buffer layer, 40 Base material 60 Convex part, 70 Concave part, 80 Concave / convex pattern 100 Epitaxial growth substrate, 200 Light emitting element, 220 Semiconductor layer

Claims (15)

1. A patterning step of forming an inorganic material film of a predetermined pattern on a substrate by screen printing;
   A method for manufacturing an epitaxial growth substrate, comprising: a curing step for curing the inorganic material film.
2. The method for manufacturing a substrate for epitaxial growth according to claim 1, wherein the inorganic material film is made of a sol-gel material.
3. 3. The method for manufacturing an epitaxial growth substrate according to claim 1, further comprising a step of etching a region where a surface of the base material is exposed.
4. The method for producing a substrate for epitaxial growth according to any one of claims 1 to 3, wherein a buffer layer is formed on the base material having the inorganic material film.
5. 3. The method for manufacturing a substrate for epitaxial growth according to claim 1, wherein a buffer layer is formed on the base material before the patterning step.
6. Convex portions and concave portions generated by forming the inorganic material film of the predetermined pattern on the base material,
   6. The method according to claim 1, wherein i) each has an elongated shape extending in a wavy manner in plan view, and ii) the extending direction, the bending direction, and the length are not uniform. The manufacturing method of the board | substrate for epitaxial growth as described in a term.
7. The method for manufacturing a substrate for epitaxial growth according to any one of claims 1 to 6, wherein the base material is a sapphire substrate.
8. An epitaxial growth substrate having a concavo-convex pattern obtained by the method for manufacturing an epitaxial growth substrate according to any one of claims 1 to 7.
9. i) The projections or depressions of the concavo-convex pattern surface of the epitaxial growth substrate each have an elongated shape extending in a wavy manner in plan view, and
   The substrate for epitaxial growth according to claim 8, wherein the projecting portion or the recessed portion of the concavo-convex pattern surface of the substrate for epitaxial growth has a non-uniform extension direction, bending direction, and length.
10. The extending direction of the protrusions is irregularly distributed in plan view,
    The epitaxial growth substrate according to claim 9, wherein a contour line in plan view of the convex portion included in a region per unit area of the uneven pattern includes more straight sections than curved sections.
11. The substrate for epitaxial growth according to claim 10, wherein a width of the convex portion in a direction substantially orthogonal to the extending direction of the convex portion in a plan view is constant.
12. In the case where the curved section is formed by dividing a contour line in plan view of the convex portion by a length that is π (circumferential ratio) times the average value of the width of the convex portion, A section in which a ratio of a linear distance between the two end points to a length of the contour line between the two end points is 0.75 or less;
    The epitaxial growth substrate according to claim 10, wherein the straight section is a section that is not the curved section among the plurality of sections.
13. In the case where the curved section is formed by dividing a contour line in plan view of the convex portion by a length that is π (circumferential ratio) times the average value of the width of the convex portion, Of the two angles formed by the line segment connecting one end and the midpoint of the section and the line segment connecting the other end of the section and the midpoint of the section, the angle that is 180 ° or less is 120 ° or less. And
    The straight section is a section that is not the curved section among the plurality of sections,
    The ratio of the curve section among the plurality of sections is 70% or more.
    The substrate for epitaxial growth according to claim 10 or 11.
14. A Fourier transform image obtained by performing a two-dimensional fast Fourier transform process on a concavo-convex analysis image obtained by analyzing the concavo-convex pattern with a scanning probe microscope is substantially centered on the origin where the absolute value of the wave number is 0 μm ^−1. A circular or annular pattern is shown, and the circular or annular pattern exists in a region where the absolute value of the wave number is in the range of 10 μm ⁻¹ or less. The substrate for epitaxial growth according to any one of the above.
15. 15. A light emitting device comprising a semiconductor layer including at least a first conductivity type layer, an active layer, and a second conductivity type layer on the epitaxial growth substrate according to any one of claims 8 to 14.

Images