JP2006128727A - Semiconductor light-emitting element - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a light-emitting element of a GaN-based compound semiconductor as a light-emitting device by packaging by flip-chip junction that can effectively utilize light generated inside the element. <P>SOLUTION: In the semiconductor light-emitting element, an n-type layer, a light-emitting layer, and a p-type layer are laminated successively on one surface of a light-transmitting support substrate, and the GaN-based compound semiconductor layer, which has a recess for forming an n-type electrode on the n-type layer, is included. The recess has an inclined surface with p-type and n-type layer sides as a wide opening and a narrow bottom surface, respectively, and a reflection layer, which reflects light radiated from the light-emitting layer to the side of the support substrate, is provided at least on the inclined surface. The reflection layer is formed on the inclined surface from the n-type layer to the p-type layer, and is formed on a transparent conductive layer made of ITO, IZO, ZnO In<SB>2</SB>O<SB>3</SB>, SnO<SB>2</SB>, or CdO. The recess is formed in a comb shape in a plan view, thus composing comb-type p-type and n-type electrodes in a bitten manner. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a light emitting device using a GaN-based compound semiconductor.

  As a conventional GaN (gallium nitride) compound semiconductor light emitting device, for example, one having a structure disclosed in Patent Document 1 is known. Further, for example, Patent Documents 2 to 4 disclose semiconductor light-emitting elements that have the structure disclosed in Patent Document 1 and have an improved electrode configuration.

  A basic structure of a conventional semiconductor light emitting device as disclosed in these patent documents is shown in FIG. An n layer 102 made of an n-type semiconductor, a light emitting layer 103, and a p layer 104 made of a p-type semiconductor are sequentially stacked on a support substrate 101 made of sapphire or the like, and a p-type electrode 105 is formed on the p layer 104. Is provided. An exposed surface is provided from the p layer 104 side so that the n layer 102 is exposed, and an n-type electrode 106 is provided on the exposed surface.

Then, by applying a forward bias between the p-type electrode 105 and the n-type electrode 106 (that is, applying a positive voltage to the p-type electrode), electrons and holes are combined in the light emitting layer 103, and blue Or ultraviolet light is generated.
Japanese Patent Laid-Open No. 9-36422 (Claims, FIG. 1, etc.) Japanese Patent No. 3187284 (claims, etc.) Japanese Patent No. 3244010 (claims, etc.) Japanese Patent No. 3269070 (Claims etc.)

  By the way, in the conventional semiconductor light emitting device as shown in FIG. 10, among the light generated in the light emitting layer, the light traveling toward the p-type electrode is blocked by the p-type electrode, and the light traveling outward from the device side surface is not transmitted. Or absorbed by the n-type electrode (B2 in FIG. 10), the amount of light that can be taken out of the device and used effectively is reduced.

  Further, when the semiconductor light emitting element is, for example, flip-chip bonded to form a semiconductor light emitting device, light is extracted from the support substrate side, but most of the light traveling from the light emitting layer to the p type electrode is the p type electrode. Since the light is absorbed by the surface, the ratio of light (B1 in FIG. 10) that can be reflected to the support substrate side and taken out from the element is reduced.

  Due to such a phenomenon, the semiconductor light emitting device having the structure shown in FIG. 10 has a problem that the amount of light that can be taken out of the device and effectively used is smaller than the amount of light generated in the light emitting layer.

  The present invention has been made in view of the above circumstances, and an object thereof is a light-emitting element using a GaN-based compound semiconductor for mounting as a semiconductor light-emitting device by flip-chip bonding, and is generated inside the element. An object of the present invention is to provide a semiconductor light emitting device capable of more effectively utilizing the light to be emitted.

In the semiconductor light emitting device of the present invention, an n layer, a light emitting layer, and a p layer are sequentially laminated on one surface of a translucent support substrate, and a recess for forming an n-type electrode on the n layer is formed. A semiconductor light emitting device including a GaN-based compound semiconductor layer, wherein the recess has an inclined surface having a wide opening on the p layer side and a narrow bottom surface on the n layer side, and at least the light emitting layer on the inclined surface A reflection layer for reflecting the light emitted from the support substrate side, the reflection layer being formed on the inclined surface from the n layer to the p layer, ITO, IZO, ZnO , In ₂ O ₃ , SnO _2, or CdO is formed on a transparent conductive layer, and the shape of the concave portion is formed in a comb shape in a plan view, whereby a comb-shaped p-type electrode and a comb-shaped electrode are formed. n-type electrodes configured to bite each other It is characterized in a certain place.

The semiconductor light-emitting device of the present invention is configured as described above, and the light emitted from the light-emitting layer, in particular, leaks from the side of the light-emitting layer due to the action of the reflective layer provided on the inclined surface. The light absorbed by each of the constituent members can be reflected and taken out to the support substrate side. As a result, it is possible to increase the proportion of the light generated inside the element that can be extracted outside and effectively used (hereinafter sometimes referred to as “light extraction efficiency”). The reflective layer is formed on an inclined surface from the n layer to the p layer, and is formed on a transparent conductive layer made of ITO, IZO, ZnO, In ₂ O ₃ , SnO ₂ or CdO. Since the transparent conductive layer of these compounds can be electrically connected to the p layer but cannot be electrically connected to the n layer, the presence of the transparent conductive layer prevents current leakage by the reflective layer. Thus, the light emission efficiency of the semiconductor light emitting device can be increased. Furthermore, since the shape of the recess is formed in a comb shape in plan view, the p-type electrode and the n-type electrode of the comb shape are configured to be engaged with each other. The current distribution in the region can be made uniform, whereby the light emission itself of the semiconductor light emitting element can be improved.

<First Embodiment>
FIG. 1 is a schematic cross-sectional view showing a first embodiment of the semiconductor light emitting device of the present invention. In the semiconductor light emitting device 10 a according to the first embodiment, a semiconductor layer in which an n layer 102, a light emitting layer 103, and a p layer 104 are sequentially stacked is provided on a support substrate 101, and further on the p layer 104. A p-type electrode 105 is formed. In this semiconductor layer, a recess having an inclined surface with a wide opening on the p layer 104 side and a narrow bottom surface on the n layer 102 side is formed, and an n-type is formed on the bottom surface (exposed surface of the n layer 102) of the recess. An electrode 106 is provided.

  By applying a forward bias between the p-type electrode 105 and the n-type electrode, electrons and holes are combined in the light-emitting layer 103 to generate blue or ultraviolet light (wavelength: 500 to 250 nm). At this time, the light toward the concave portion of the light emitting layer 103 is reflected to the support substrate 101 side by the reflective layer (first reflective layer) 107 (A2 and A3 in FIG. 1). Thereby, the ratio of the light which can be extracted outside and used effectively among the light generated in the semiconductor light emitting device can be increased. The angle formed by the inclined surface of the recess and the extension line of the bottom surface of the recess (hereinafter referred to as “taper angle”) is preferably 20 to 80 °.

  As described above, the semiconductor light-emitting element of the present invention is mounted by flip-chip bonding to constitute a semiconductor light-emitting device, so that light is extracted from the support substrate 101 side. Therefore, the support substrate is composed of a material that transmits light generated inside the element, and sapphire is generally used as the material. The thickness of the support substrate is usually 30 to 500 μm.

The n layer 102 is a layer made of an n-type GaN compound semiconductor, and is specifically represented by Al _a In _b Ga _1-ab N (0 ≦ a, 0 ≦ b, a + b ≦ 1). Consists of compounds. For example, Al _a Ga _1-a N where a is 0.5 or less and In _b Ga _1-b N where b is 0.5 or less are preferable.

Although the n layer 102 is shown as a single layer in FIG. 1, the n layer has a laminated structure in which an n-type contact layer and an n-type cladding layer are laminated in this order from the support substrate side. Also good. In this case, the recess is formed so that the n-type contact layer is exposed on the bottom surface. As the n-type contact layer, GaN is common among the above-mentioned compounds constituting the n layer. As the n-type cladding layer, Al _0.3 Ga _0.7 N, AlN, GaN, etc. are common among the above-mentioned compounds constituting the n layer.

  In addition, it is desirable to dope n-type impurities such as Si, Ge, and S into the n-layer having the single-layer structure, the n-type contact layer, and the n-type cladding layer because the carrier concentration can be increased.

  The thickness of the n layer is generally 0.5 to 10 μm when the n layer has a single layer structure. Further, when the n layer has a laminated structure of an n-type contact layer and an n-type cladding layer, the thickness of the n-type contact layer is generally 0.2 to 10 μm, and the thickness of the n-type cladding layer is particularly limited. Usually, it is 0.01 to 3 μm.

The light emitting layer 103 is a layer made of a GaN-based compound semiconductor. Specifically, the light emitting layer 103 is made of a compound represented by Al _c In _d Ga _1-cd N (0 ≦ c, 0 ≦ d, c + d ≦ 1). Composed. In this light emitting layer, for example, by adjusting the composition ratio of In and Al in the above compound, the light emitting layer is doped with n-type impurities such as Si, Ge and S and p-type impurities such as Mg and Zn. Thus, the wavelength of the generated light can be adjusted in the range from blue to ultraviolet.

  In FIG. 1, the light emitting layer 103 is shown as a single layer. For example, the light emitting layer has a stacked structure including a plurality of layers, and a so-called multiple quantum well structure in which the composition of the compounds constituting each layer is changed. Is also preferable. The total thickness of the light emitting layer is generally 0.001 to 0.5 μm.

  The p layer 104 is a layer made of a p-type GaN compound semiconductor. In FIG. 1, the p layer 104 is shown as a single layer, but generally has a laminated structure in which a p-type cladding layer and a p-type contact layer are laminated in this order from the light emitting layer side.

The p-type cladding layer is made of, for example, a compound represented by Al _e Ga _1-e N (0 <e ≦ 1), and is further doped with a p-type impurity. e is preferably 0.05 or more. The p-type contact layer is made of, for example, p-type GaN doped with p-type impurities. Examples of the p-type impurity doped in the p-type cladding layer and the p-type contact layer include Mg and Zn. The thickness of the p-type cladding layer is usually 0.001 to 1.5 μm, and the thickness of the p-type contact layer is generally 0.01 to 2 μm.

  The material constituting the p-type electrode 105 is required to be in ohmic contact with the p-layer, and examples thereof include Pd, Ni, V, and alloys thereof. The thickness of the p-type electrode is usually 0.05 to 5 μm.

  The material constituting the n-type electrode 106 is required to have an ohmic contact with the n layer, and examples thereof include Al and its alloys. The thickness of the n-type electrode is usually 0.05 to 5 μm.

  The first reflecting layer 107 is for reflecting light leaking from the side surface of the light emitting layer 103 (light traveling toward the inclined surface side) out of the light generated from the light emitting layer 103 to the support substrate 101 side. For example, the reflective layer preferably has a reflectance of at least 85% of the light emitted from the light emitting layer 103, and a reflective layer having such a reflectance is selected from materials that can constitute the reflective layer described later. It is recommended to select what can be formed. In this case, the amount of light that can be reflected to the support substrate 101 side can be further increased. Typical materials for the reflective layer that can achieve the above reflectivity include Al, Rh, Ag, and the like.

  However, as shown in FIG. 1, the first reflective layer 107 is generally formed from the n layer 102 to the p layer 104 on the inclined surface of the concave portion. In this case, it is preferable that the material constituting the first reflective layer 107 cannot be in ohmic contact with either the n layer or the p layer.

  Further, for example, when the first reflective layer 107 is made of a material that can be in ohmic contact with only the p layer 104 and the p-type electrode 105 is made of the same material as the reflective layer 107, the light emitting layer 103 Of the generated light, the light traveling toward the p-type electrode 105 can be reflected toward the support substrate 101 (A1 and A4 in FIG. 1). Therefore, the extraction efficiency of the generated light can be further increased. Examples of such a material include Rh and its alloys.

  In addition, when the first reflective layer 107 is made of a material that can be in ohmic contact with only the n layer 102 and the n-type electrode 106 is made of the same material as the first reflective layer 107, the light emitting layer Of the light generated from 103, the light traveling toward the n-type electrode 106 can be reflected toward the support substrate 101. Therefore, the extraction efficiency of the generated light can be further increased in such a case. Examples of such a material include Al and alloys thereof. Further, in this case, a reflective layer (second reflective layer, not shown) is provided between the p-layer 104 and the p-type electrode 105, or a p-type material that can ensure the same reflectivity as the reflective layer. It is also preferable to form the electrode 105, whereby light traveling toward the p-type electrode 105 can also be reflected toward the support substrate 101, so that the light extraction efficiency can be further increased. The second reflective layer between the p layer 104 and the p-type electrode 105 is formed on the p layer via a transparent conductive layer (not shown) described in detail in the second to fifth embodiments. It is also preferable.

  The thickness of the reflective layer is 0.001 to 5 μm regardless of whether it is provided on the inclined surface of the recess (first reflective layer) or provided between the p layer and the p-type electrode (second reflective layer). Preferably, the thickness is 0.01 to 1 μm.

  In forming the n layer 102, the light emitting layer 103, and the p layer 104, a known film forming method such as a known vapor phase growth method [metal organic vapor phase growth method (MOCVD method)] can be employed. The doping of the n-type impurity into the n layer and the doping of the p-type impurity into the p layer are performed simultaneously with the formation of these layers.

  Although not shown in FIG. 1, when the support substrate is a sapphire substrate, it is difficult to directly form the n layer. Therefore, it is preferable to provide a buffer layer between the n layer and the support substrate. Such a buffer layer has a function of relaxing the lattice mismatch between the support substrate and the n layer, and the presence of this buffer layer improves the formation of the n layer (growth of the n-type GaN compound semiconductor crystal). It is possible to proceed. Examples of the buffer layer include GaN, AlN, GaAlN, and ZnO. By the way, when forming an n layer using GaN by vapor phase growth, a two-step growth method is adopted in which a buffer layer is first formed by forming a film under low temperature conditions, and then the n layer is formed by raising the temperature. By doing so, the buffer layer and the n layer can be formed more easily. The thickness of the buffer layer is usually 0.001 to 1 μm.

  The formation method of the said recessed part is mentioned later. Further, the p-type electrode, the n-type electrode, and the reflective layer may be formed by a known vapor deposition method or the like.

Second Embodiment
FIG. 2 is a schematic cross-sectional view showing a second embodiment of the semiconductor light emitting device of the present invention. Parts having the same functions as those of the first embodiment are denoted by the same reference numerals to avoid redundant description (the same applies to the embodiments described later). The second embodiment is characterized by the following points.

  In the semiconductor light emitting device 10b according to the second embodiment, the reflective layer 107 is formed on the inclined surface of the concave portion via the insulating film 109. When a reflective layer is provided on the inclined surface, current leakage is caused by the reflective layer, which may reduce the light emission efficiency of the semiconductor light emitting device. However, the presence of the insulating film can prevent the occurrence of such a leakage current, so that the light emission efficiency of the semiconductor light emitting element can be increased.

  The insulating film preferably has a transmittance of light emitted from the light emitting layer of at least 80%. More preferably, it is 90% or more. Thus, when the light transmittance in the insulating film is large, the ratio of light absorbed by the insulating film is small. Extraction efficiency is improved.

As the material for the insulating film, silicon oxide (SiO ₂ ), silicon nitride (Si ₃ N ₄ ), or the like is preferable. The insulating film can be formed by a known vapor deposition method using these materials. The thickness is preferably 30 to 10000 mm, and more preferably 50 to 1000 mm (for example, 100 mm). Such an insulating film can be formed by a known vapor deposition method (for example, the MOCVD method).

  FIG. 2 shows an example of an n-type electrode 106 in which a Ti layer 112, an Al layer 110, and an Au layer 111 are sequentially laminated from the bottom surface of the recess. For example, for the n-type electrode, it is preferable to use the same material as the first reflective layer 107, the second reflective layer 107a, and the p-type electrode 105. In this case, the reflective layer and the n-type electrode, or p Since the mold electrode and the n-type electrode can be formed at the same time, the manufacturing process of the semiconductor light emitting device can be simplified. However, for example, when Al having excellent reflectivity is used as the material of the first reflective layer 107 and the second reflective layer 107a, there is a problem that the adhesion between the n layer and Al is relatively low. . Therefore, in such a case, as shown in FIG. 2, a layer made of a material capable of improving the adhesion such as the Ti layer 112 is provided between the n layer 102 and the Al layer 110. Is recommended. When the n-type electrode has the three-layer structure shown in FIG. 2, the thickness of each layer is Ti layer: 10 to 2000 mm, Al layer: 10 to 3000 mm, Au layer: 30 to 3000 mm, and the total thickness of the n-type electrode Is recommended to be 50-8000cm. The Ti layer, the Al layer, and the Au layer may be layers made of materials other than Ti, Al, and Au, as long as they can perform the same function. The total thickness of the thickness and the n-type electrode may be within the above preferred range.

  In addition, FIG. 2 shows an example in which the p-type layer 104 and the p-type electrode 105 are also provided with the second reflective layer 107a, and the light reflected toward the p-type electrode 105 side is transmitted by the second reflective layer 107a. The light extraction efficiency can be improved by reflecting the light toward the support substrate 101 (A1 and A4 in FIG. 2). On the other hand, even if the second reflective layer 107a is not provided, in the p-type electrode, for example, when the reflectance of light emitted from the light emitting layer is at least 85%, the second reflective layer 107a. The effect similar to the aspect which provided can be ensured.

Further, in FIG. 2, the second reflective layer 107 a is a p-type electrode through a transparent conductive layer 108 a (second transparent conductive layer. The first transparent conductive layer will be described in a third embodiment described later). 105 is formed. The second transparent conductive layer 108a is made of ITO (indium-tin composite oxide), IZO (indium-zinc composite oxide), ZnO, In ₂ O ₃ , SnO ₂ , CdO, or the like.

  In addition, the second transparent conductive layer 108a is composed of a metal containing Ni, Pd, Pt, Cr, Mn, Ta, Cu, or Fe (for example, these metals alone or an alloy containing these elements). But you can. When the transparent conductive layer has such a configuration, the resistance value between the p layer 104 and the reflective layer 107a can be greatly reduced. However, in the case of a transparent conductive layer composed of a metal containing each of the above elements, the light emitted from the light-emitting layer 103 is reflected by the second reflective layer and the light traveling toward the transparent conductive layer. It is necessary to ensure transparency to the extent that light traveling toward the support substrate 101 can be satisfactorily transmitted. For this reason, in the case of a transparent conductive layer composed of a metal containing each of the above elements, it is desirable to reduce the thickness, for example, 20 mm or more, more preferably 50 mm or more, and 200 mm or less, more preferably It is recommended to keep it below 100mm.

Further, in FIG. 2, the second transparent conductive layer 108a is shown as a single layer, but it has a laminated structure composed of a plurality of layers (for example, two layers, three layers, four layers, etc.). It is also preferable. For example, in the case of a two-layer structure, the layer (A layer) on the p layer 104 side is a metal containing Ni, Pd, Pt, Cr, Mn, Ta, Cu, or Fe (for example, these simple metals or these in layer composed of an alloy) containing an element, the layer of the second reflective layer 107a side (B layer), ITO, IZO, In2 O3, ZnO, be a layer composed of SnO ₂ or CdO is recommended The When the second transparent conductive layer has such a structure, the resistance value between the p layer 104 and the second reflective layer 107a is greatly reduced, and the total thickness of the second transparent conductive layer is also compared. In addition to being able to be a highly reliable layer, the second transparent conductive layer can be configured to further increase the light extraction efficiency (described in a fifth embodiment described later).

  On the other hand, even if the second reflective layer 107a is not provided, the p-type electrode may be formed on the p-layer via the transparent conductive layer. By interposing such a transparent conductive layer, a material that cannot make ohmic contact with the p layer can be selected as a material of the p-type electrode.

  Examples of the material of the p-type electrode when the second reflective layer 107a is not provided and the p-type electrode is also provided with reflectivity include Al, Ag, and Rh.

  Details of the case where the second reflective layer 107a and the second transparent conductive layer 108a are formed between the p layer 104 and the p-type electrode 105 will be described later in third and fourth embodiments.

<Third Embodiment>
FIG. 3 is a schematic cross-sectional view showing a third embodiment of the semiconductor light emitting device of the present invention. The third embodiment is characterized by the following points.

  In the semiconductor light emitting device 10 c according to the third embodiment, the first reflective layer 107 is formed on the inclined surface of the recess through the first transparent conductive layer 108.

Examples of the constituent material of the first transparent conductive layer include ITO, IZO, ZnO, In ₂ O ₃ , SnO _{2, and} CdO. These compounds that can constitute the first transparent conductive layer can be electrically connected to the p layer, but cannot be electrically connected to the n layer. Therefore, current leakage due to the reflective layer can be prevented, and the light emission efficiency of the semiconductor light emitting element can be increased. As a constituent material of the first transparent conductive layer, ITO or IZO is particularly preferable.

  FIG. 3 shows an example in which a second reflective layer 107a is provided between the p-layer 104 and the p-type electrode 105 via the second transparent conductive layer 108a on the p-type electrode 105 side. The reflection layer reflects light traveling toward the p-type electrode 105 toward the support substrate 101 (A1 and A4 in FIG. 3), thereby increasing the light extraction efficiency. However, in this case, as shown in FIG. 3, it is necessary to provide a gap in the first transparent conductive layer 108 to prevent a short circuit between the p-type electrode 105 and the p-layer 104 and the n-type electrode 106. . Such a gap can be formed by applying a mask at the time of forming the transparent conductive layer or by performing known etching using plasma or an etching solution containing iron chloride on the transparent conductive layer.

  Both the first and second transparent conductive layers can be formed by a known vapor deposition method or the like, but it is desirable to form the transparent conductive layer by a sputtering method. When formed by sputtering, the adhesiveness between the transparent conductive layer and the p layer is good.

  Further, when the transparent conductive layer is formed by sputtering, it is recommended to use a magnetron sputtering apparatus. FIG. 4 is a conceptual diagram when a transparent conductive layer is formed using a magnetron sputtering apparatus. In FIG. 4, 113 is a sample for forming a transparent conductive layer, 114 is a target for forming a transparent conductive layer (that is, ITO target, IZO target, etc.), 115 is a magnet, and 116 is a sample holder. Sputtering particles are emitted from the surface of the target 114 by applying the plasma 118 to the target 114. In this magnetron sputtering apparatus, the plasma 118 can be confined between the opposing targets 114 and 114 with high density by the magnet 115. . The generated sputtered particles collide with the plasma 118 and fly away in a random direction. Among these sputtered particles, those facing the sample 113 (117 in the figure) are deposited on the surface of the sample 113, and a transparent conductive layer is formed.

  When the magnetron sputtering apparatus is used in this way, the sample (semiconductor light emitting element intermediate) does not have to be directly exposed to plasma, and thus damage to the sample (semiconductor layer) due to the plasma can be avoided.

  The thickness of the transparent conductive layer thus formed is preferably 0.001 to 2 μm, and more preferably 0.005 to 0.5 μm in both the first and second layers. In addition, when the 2nd transparent conductive layer is said laminated structure, the total thickness should just be in this range. However, when the second transparent conductive layer is made of a metal containing Ni, Pd, Pt, Cr, Mn, Ta, Cu, or Fe, as described above, the thickness is preferably 20 mm or more, more preferably It is recommended that it is 50 cm or more and 200 cm or less, more preferably 100 cm or less.

  Furthermore, it is desirable that the first and / or second transparent conductive layer is doped with a p-type dopant (p-type impurity), thereby reducing the contact resistance between the p layer and the transparent conductive layer. Further, from this point of view, the p-type dopant concentration in the transparent conductive layer is the reflection layer (the first reflection layer in the first transparent conductive layer and the second reflection layer in the second transparent conductive layer). It is preferable that the vicinity of the other side (near the p layer) is larger than the vicinity of the contact surface. More preferably, the p-type dopant concentration is gradually or gradually increasing from the vicinity of the contact surface with the reflective layer toward the vicinity of the other surface.

  Examples of the p-type dopant doped into the transparent conductive layer include Mg, Zn, Cd, Ca, Be, and C. In addition, the method for doping the transparent conductive layer with the p-type dopant is not particularly limited, and a known method can be employed. For example, it is desirable to employ an ion implantation method. In this ion implantation method, by adjusting the acceleration voltage, the dopant to be implanted can be collected at a specific thickness position. Therefore, the p-type dopant concentration is set closer to the other surface side than the contact surface with the reflective layer. It is easy to make it larger. In addition, when the p-type dopant concentration is increased stepwise or gradually from the vicinity of the contact surface with the reflective layer to the vicinity of the other surface by ion implantation, the acceleration voltage is changed stepwise during doping. It is sufficient to adopt a method to do this.

  Further, the first and / or second transparent conductive layer is heat-treated at a temperature of 200 to 800 ° C. in at least one gas atmosphere selected from the group consisting of a rare gas, nitrogen and oxygen. Is also preferable. By performing such heat treatment, the carrier density in the transparent conductive layer can be increased, so that the resistivity can be lowered.

The heat treatment is generally applied to the semiconductor light emitting device intermediate before forming the reflective layer after the transparent conductive layer is formed on the inclined surface (and the p layer surface), but the surface of the transparent conductive layer is oxidized by this heat treatment. Thus, an oxide layer may be formed. In this case, it is preferable to remove the oxide layer, whereby the contact resistance with the reflective layer formed on the transparent conductive layer can be lowered and the adhesion can be improved. The oxide layer can be removed by etching using, for example, a mixed solution of iron chloride (FeCl ₃ ) and hydrochloric acid.

  As described above, the first and / or second reflective layer preferably has a light reflectance of 85% or more. However, in the third embodiment, the concave portion is interposed via the transparent conductive layer. Therefore, a material that cannot make ohmic contact with the p layer can be selected as a constituent material of the reflective layer. Therefore, Al or Ag is suitable as a constituent material of the reflective layer, but Al is particularly suitable because Ag has a problem of being easily corroded.

  In addition, the material of the p-type electrode 105 and the n-type electrode 106 in the third embodiment is not particularly limited, but Au is suitable. As described above, since the semiconductor light emitting device of the present invention is flip-chip bonded, Au of these electrodes can be used as a gold bump. However, regardless of the embodiment, in the semiconductor light emitting device of the present invention, the bonding means in the case of mounting is not limited to bonding by gold bumps, and can be bonded by, for example, solder bumps.

<Fourth embodiment>
FIG. 5 is a schematic cross-sectional view showing a fourth embodiment of the semiconductor light emitting device of the present invention. The fourth embodiment is characterized by the following points.

  In the semiconductor light emitting element 10d according to the fourth embodiment, the second transparent conductive layer 108a is interposed only in a part between the second reflective layer 107a and the p layer 104, and the second reflective layer 107a. And a portion 119 where the p layer 104 is in direct contact.

  In the second and third embodiments described above, the second transparent conductive layer 108a is interposed on the entire surface between the second reflective layer 107a and the p layer 104 (FIGS. 2 and 3). In this case, of the light generated from the light emitting layer 103, all the light directed to the p-type electrode 105 side passes through the second transparent conductive layer 108a and reaches the second reflective layer 107a. The light reaching the interface between the second reflective layer 107a and the second transparent conductive layer 108a becomes reflected light and travels toward the support substrate 101. All the reflected light also passes through the second transparent conductive layer 108a. Light from the light emitting layer 103 and the reflected light are slightly absorbed when passing through the second transparent conductive layer 108a. Therefore, among the light generated from the light emitting layer 103, the light that can be extracted from the support substrate 101 side is also reduced.

  Therefore, in the semiconductor light emitting device 10d according to the fourth embodiment, the degree of interposition of the second transparent conductive layer 108a for achieving electrical connection between the second reflective layer 107a and the p layer 104 is reduced. The amount of light absorbed by the second transparent conductive layer 108a is reduced by increasing the area in which the second reflective layer 107a and the p-layer 104 are in direct contact with each other, by suppressing the current to a level that can secure a sufficient amount of current. ing.

  The form of the second transparent conductive layer according to the fourth embodiment is not particularly limited as long as a sufficient amount of current for light emission can be secured. For example, in a plan view, a continuous shape such as a lattice shape or a linear shape (for example, a shape in which a plurality of lines are arranged in parallel), a form having a plurality of locations where the p layer can be exposed, a protrusion (dot), etc. May be discontinuously scattered in a plan view (so-called discontinuous layer). The degree of the intervention of the second transparent conductive layer is, for example, 40 area% or more, more preferably 60 area% or more with respect to the total area in a plan view between the second reflective layer and the p layer. Thus, it is desirable that the area be 100 area% or less (or 100 area% if light absorption by the second transparent conductive layer is not a problem), more preferably 80 area% or less.

  In order to form the second transparent conductive layer according to the fourth embodiment, a mask is applied to a portion where the p-layer should be exposed in plan view when the formation method described in the third embodiment is performed. Alternatively, after forming the transparent conductive layer, a method of forming a portion where the p layer is exposed in a plan view can be adopted by a known etching method using an etching solution containing plasma or iron chloride.

  FIG. 5 shows a mode in which the first reflective layer 107 is formed on the inclined surface with the insulating film 109 interposed therebetween, but the first transparent conductive layer is replaced with the first transparent conductive layer instead of the insulating film 109. The reflective layer 107 and the inclined surface may be interposed, and the n-type electrode 106 may have the structure shown in FIG.

<Fifth Embodiment>
FIG. 6 is a schematic view showing a fifth embodiment of the semiconductor light emitting device of the present invention, and FIG. 6A is a cross-sectional view showing a laminated structure of each layer. The fifth embodiment is characterized by the following points.

  In the semiconductor light emitting device 10e according to the fifth embodiment, the second transparent conductive layer 108a has a laminated structure (represented as a single layer in FIG. 6) composed of the A layer and the B layer, and B The layer side surface (the second reflective layer 107a side) has an uneven structure. The presence of the uneven structure can suppress light scattering at the interface between the second reflective layer 107a and the second transparent conductive layer 108a. Therefore, light emitted from the light emitting layer 103 toward the p-type electrode 105 side can be reflected to the support substrate 101 side with high efficiency, and the amount of light that can be extracted from the light emitting element to the outside is increased. Can do.

  In order to form the concavo-convex structure on the surface of the second transparent conductive layer 108a, for example, a known method using plasma or an etching solution containing iron chloride after the second transparent conductive layer 108a (B layer) is formed. Etching may be performed. Then, by forming the second reflective layer 107a, the interface between the second transparent conductive layer 108a and the second reflective layer 107a becomes uneven.

  In the semiconductor light emitting device 10e according to the fifth embodiment, the surface of the n layer 102 on the support substrate 101 side also has a concavo-convex structure. FIG. 6B shows an example in which the surface of the n layer 102 on the support substrate side has an uneven structure. Due to the presence of the concavo-convex structure, light traveling toward the support substrate 101 can be prevented from being scattered at the interface between the n layer 102 and the support substrate 101. For this reason, the light quantity which can be taken out to the exterior of a light emitting element through the support substrate 101 can be increased.

  For example, a method in which a part of the n layer 102 is decomposed by irradiating a laser beam from the support substrate 101 side after the n layer 102 is formed on the support substrate 101 can be employed for the uneven structure on the surface of the n layer 102. The decomposition product of the n layer 102 is diffused to the outside from the side surface of the semiconductor light emitting element. Therefore, when such a concavo-convex structure forming method is adopted, the concave portion becomes a void.

  Examples of the concavo-convex structure on the surface of the second transparent conductive layer and the concavo-convex structure on the surface of the n layer include a form in which convex portions having a shape such as a triangular pyramid shape or a cylindrical shape are periodically formed. It is done. The pitch of the convex portions is not particularly limited. For example, when the shape of the convex portion is a triangular pyramid or a cylinder, a distance (for example, the same length) is approximately the same as the diameter length of the bottom surface of the triangular pyramid or the cylinder. It can be illustrated. When the convex portion is a triangular pyramid, it is desirable that the angle formed by the line segment passing through the apex of the triangular pyramid and the center point of the bottom surface and the side surface of the triangular pyramid is, for example, 30 to 60 °.

  Note that although FIG. 6 shows a mode in which both the surface of the second transparent conductive layer 108a and the surface of the n layer 102 have a concavo-convex structure, only one of them may have a concavo-convex structure. . However, from the viewpoint of further increasing the light extraction efficiency, it is desirable that the surface of the second transparent conductive layer 108a and the surface of the n layer 102 have an uneven structure.

  FIG. 6 shows a mode in which the first reflective layer 107 is formed on the inclined surface with the insulating film 109 interposed therebetween, but the first transparent conductive layer is replaced with the first transparent conductive layer instead of the insulating film 109. The reflective layer 107 and the inclined surface may be interposed, and the n-type electrode 106 may have the structure shown in FIG.

<Sixth embodiment>
FIG. 7 is a perspective view showing a sixth embodiment of the semiconductor light emitting device of the present invention. In the semiconductor light emitting device 10f according to the sixth embodiment, the shape of the recess is formed in a comb shape in plan view, whereby the comb p-type electrode 105 and the comb n-type electrode 106 are seen in plan view. And are configured to bite each other. In FIG. 7, in order to facilitate understanding of the configuration of each layer of the semiconductor light emitting element, the reflective layer, the transparent conductive layer, and the insulating film are not shown, but two opposing comb teeth of the p-type electrode 105 The cross section including the recess sandwiched between the semiconductor light emitting devices has the cross-sectional structure (first embodiment to fifth embodiment) of the semiconductor light emitting device of the present invention as shown in FIGS. 1 to 3, 5, or 6. ing.

  In the case of the sixth embodiment, the semiconductor light emitting device of the present invention prevents the light emission from concentrating directly under the electrodes by equalizing the resistance of each current path and widening the current distribution in the light emitting layer 103. In addition, the emission area can be increased. Therefore, it is possible to increase the amount of light generated.

<The formation method of the said recessed part>
Next, the formation method of the said recessed part is demonstrated. FIG. 8 shows a cross-sectional structure of a semiconductor light-emitting element intermediate that has undergone each step of forming a recess. First, as shown in FIG. 8A, a GaN-based compound semiconductor layer in which an n layer 102, a light emitting layer 103, and a p layer 104 are sequentially stacked is formed on a support substrate 101.

Next, an etching mask layer 200 is formed on the p layer 104 [FIG. 8B]. The etching mask layer 200 is formed by using, for example, SiO ₂ , SiN, Ni, resist resin, or the like by a known vapor deposition method (SiO ₂ , SiN, Ni, etc.) or curing by application / exposure (resist resin). Can be formed. The thickness is preferably 0.1 to 5 μm.

  Next, a resist resin (preferably a positive photoresist) is applied on the etching mask layer 200, exposed and developed, and a pattern (opening) for etching the etching mask layer 200 is formed. FIG. 8C shows a semiconductor light emitting element intermediate having a resist film 201 in which an opening for etching the etching mask layer 200 is formed.

Next, the etching mask layer is etched to form a pattern (opening) for providing the recess in the GaN-based compound semiconductor layer. For the etching when the etching mask layer is made of SiO ₂ , for example, buffer hydrofluoric acid (BHF, a mixture of NH ₄ F and HF) can be used as an etching solution. At this time, the etching length in the side surface direction of the etching mask layer becomes larger in the vicinity of the resist film side of the etching mask layer where the exposure time to the etching solution becomes longer, and on the other hand, in the vicinity of the p layer side of the etching mask layer. The etching length in the side surface direction is reduced. Therefore, by this etching, an opening having a wide opening on the resist film side (on the surface side after removing the resist film) and a narrow bottom surface on the p layer side is formed in the etching mask layer. By adjusting the shape of the opening (the opening area and the inclination angle of the inclined surface of the opening), the shape of the recess formed in the GaN-based compound semiconductor layer (the opening area and the inclination angle of the inclined surface) can be controlled. .

The shape of the opening of the etching mask layer can be controlled by the concentration of BHF and the etching time. For example, BHF has a concentration of NH ₄ F: HF = 10 to 0.1: 1 (volume ratio, the same applies hereinafter) and diluted 1 to 10 times (volume times, the same applies hereinafter) with pure water. In addition, although the etching time depends on the thickness of the etching mask layer, for example, when the thickness is 0.5 μm, 2 to 4 minutes is preferable. What is necessary is just to select suitably according to a shape.

  After the opening is formed in the etching mask layer, the resist film is removed. As a method, the solvent film is generally dissolved and removed using a solvent capable of dissolving the resist film. Although the type of solvent depends on the type of resist resin, ketones such as acetone, fuming nitric acid and the like are usually used. FIG. 8D shows a semiconductor light emitting device intermediate in which the resist film is removed after the opening is formed in the etching mask layer 200. For example, as shown in FIG. 8C, even if the opening provided in the resist film 201 is relatively small, the etching mask layer is formed as shown in FIG. The opening area of 200 can be increased.

Next, as shown in FIG. 8E, etching is performed with plasma 201 containing chlorine to form the recess. For this etching, a known plasma processing apparatus [reactive ion etching apparatus (RIE) or the like] may be used. In addition to the shape of the opening of the etching mask layer 200 described above, the shape of the recess (opening area and inclination angle of the inclined surface) adjusts the thickness of the etching mask layer 200 and the generation power of plasma during etching. Can also be controlled. In this etching, a part of the etching mask layer 200 is also etched. However, in the etching mask layer made of, for example, SiO _{2, the} etching rate of the GaN-based compound semiconductor layer is about 2 to 4 times faster. . Therefore, if the thickness of the etching mask layer 200 and the inclination angle of the opening are appropriate, for example, the etching mask layer 200 having the opening shape as shown in FIG. A recess having a shape as shown in e) can be formed.

After the etching with chlorine-containing plasma is completed, the remaining etching mask layer is removed to obtain a semiconductor light emitting device intermediate [FIG. 8 (f)] in which the recess is formed in the GaN compound semiconductor layer. . The method for removing the etching mask layer is not particularly limited, and dissolution removal using a solvent capable of dissolving the material constituting the etching mask layer can be employed. Etching mask layer is, for example, when configured for SiO ₂ may be used such as BHF acid or hydrofluoric acid.

<Step after forming the recess>
Next, of the semiconductor light emitting device of the present invention, the process after forming the recess will be described by taking the second embodiment as an example. In the case of the second embodiment, first, a transparent conductive layer (the second transparent conductive layer) and an insulating film are formed. The formation order of these is not particularly limited. FIG. 9 shows an example in which the transparent conductive layer 108a is formed [FIG. 9A], and then the insulating film 109 is formed [FIG. 9B]. After the formation of the transparent conductive layer 108a, it is preferable to perform heat treatment under the above conditions or doping with a p-type dopant. In addition, when the heat treatment is performed, it is recommended to remove the oxide layer formed on the surface of the transparent conductive layer.

  Thereafter, the Ti layer 112, the reflective layers 107 and 107a, and the Al layer 110 constituting the n-type electrode are formed [FIG. 9C]. When the reflective layers 107 and 107a are made of Al, they can be formed simultaneously with the Al layer 110. Subsequently, the p-type electrode 105 is formed to obtain the semiconductor light emitting device of the second embodiment shown in FIG.

  About the layer formation method in each said process, heat processing, and doping, what is necessary is just to follow the method and conditions demonstrated by the term of the above-mentioned 1st Embodiment to 5th Embodiment.

  As described above, in the semiconductor light emitting device of the present invention, in addition to being able to increase the light extraction efficiency by the reflective layer (first reflective layer) provided on the inclined surface, the reflective layer is made of an insulating film. Accordingly, the generation of leakage current due to the reflective layer can be suppressed, and the light emission efficiency of the semiconductor light emitting element can be increased.

  On the other hand, the reflective layer (first reflective layer) having the inclined surface is electrically connected to the p layer via the first transparent conductive layer, and the second reflection is also formed between the p-type electrode and the p layer. The second reflective layer also has a good ohmic property between the second reflective layer and the p layer, particularly when the second reflective layer is electrically connected to the p layer via the transparent conductive layer. Thus, loss at the p-type electrode portion can be suppressed. In addition, when the semiconductor light emitting element is mounted on a circuit board or the like by flip chip bonding for use in a semiconductor light emitting device or the like, the absorption of light at the p-type electrode is suppressed by providing this second reflective layer. The light extraction efficiency from the support substrate (sapphire substrate) can be increased.

  In addition, by interposing the second transparent conductive layer only in a part between the second reflective layer and the p layer, the electrical connection between the second reflective layer and the p layer is maintained at a level sufficient for light emission. However, the area where the second reflective layer and the p layer are in direct contact can be increased. Therefore, since the amount of light absorbed by the second transparent conductive layer among the light generated from the light emitting layer can be reduced, the light extraction efficiency can be further increased.

  Furthermore, by making the second transparent conductive layer a thin film (for example, having a thickness of 20 to 200 mm) made of a metal containing the above-mentioned specific element, the light generated from the light emitting layer and the reflected light thereof While suppressing absorption by the transparent conductive layer, the resistance between the p-layer and the second reflective layer can be lowered, and a semiconductor light-emitting element with a low operating voltage can be obtained.

  In addition, since the second transparent conductive layer has the above-described specific laminated structure, the resistance between the p layer and the second reflective layer is reduced, and the operating voltage of the semiconductor light emitting element is reduced. The total thickness of the transparent conductive layer 2 can be made relatively large. Therefore, the surface of the second transparent conductive layer (second reflective layer side surface) can have an uneven structure. In this case, light emitted from the light emitting layer and traveling toward the second reflective layer can be efficiently reflected to the support substrate side while suppressing scattering at the interface between the second reflective layer and the second transparent conductive layer. The light extraction efficiency can be further increased.

  In addition, by providing a concavo-convex structure on the n-layer support substrate side surface, it is possible to suppress scattering of light emitted from the light emitting layer toward the support substrate side at the n layer-support substrate interface. For this reason, the light extraction efficiency can be further increased.

  Further, in the semiconductor light emitting device that employs the above-described layer configuration and in addition the configuration in which the comb-type n-type electrode and the comb-type p-type electrode are arranged to be engaged with each other, in the p-type electrode region, The current distribution can be made uniform, thereby improving the light emission itself of the device, and as described above, the light directed to the side surface of the light emitting layer and the p-type electrode by the action of the first reflective layer and the second reflective layer. Can be taken out from the supporting substrate, so that the brightness of the element can be further increased.

  Hereinafter, the present invention will be described in detail based on examples. However, the following examples are not intended to limit the present invention, and all modifications made without departing from the spirit of the preceding and following descriptions are included in the technical scope of the present invention.

<Evaluation experiment of light extraction efficiency>
Experiment 1
Example 1
A semiconductor light emitting element intermediate A having a GaN-based compound semiconductor layer in which an n layer, a light emitting layer, and a p layer are sequentially laminated on a support substrate was prepared. Table 1 shows the compounds constituting the support substrate, the n layer, the light emitting layer, and the p layer, and the thickness of each layer.

An etching mask layer (thickness: 1 μm) made of SiO ₂ was formed on the surface of the p layer of the semiconductor light emitting device intermediate A. The etching mask layer was formed by a MOCVD method at a temperature of 450 ° C. using a mixed gas of silane gas (SiH ₄ ) and oxygen.

  Further, a positive photoresist (“iP1800” manufactured by Tokyo Ohka Kogyo Co., Ltd.) is applied on the etching mask layer by a spin coating method (500 rotations for 10 seconds, then 2000 rotations for 30 seconds), and a hot surface having a surface temperature of 110 ° C. After being placed on the plate and baked for 4 minutes, it was exposed for 10.5 seconds using an i-line stepper device ("Manual Aligner MA-10" manufactured by Mikasa). Thereafter, an alkali developer ("NMD3" manufactured by Tokyo Ohka Kogyo Co., Ltd.) The development time was 90 seconds, followed by washing with water for 5 minutes, followed by post-baking at 110 ° C. for 2 minutes to form a resist film having openings as shown in FIG.

Next, the opening of the resist film was etched using BHF at room temperature for 6 minutes to form an opening having a wide opening on the resist film side surface and a narrow bottom surface on the p layer side. BHF was prepared by adding HF to a product of NH ₄ F: HF = 6: 1 (volume ratio) manufactured by Kanto Chemical Co., Ltd. to make NH ₄ F: HF = 3: 1, and further diluted 10 times with pure water. Things were used. Thereafter, the resist film was dissolved and removed using acetone.

  Next, plasma etching was performed using RIE to form the recesses in the GaN-based compound semiconductor layer, whereby a semiconductor light emitting device intermediate B was obtained. Etching was performed by adding Ar to chlorine gas and supplying 200 W as RF power. The taper angle of this concave portion (the angle formed by the inclined surface and the extension line of the concave bottom surface) is 30 °.

Thereafter, an insulating film (SiO ₂ ) film (thickness: 0.1 μm) was formed on the inclined surface. The insulating film is formed on the surface of the GaN-based compound semiconductor layer of the semiconductor light-emitting element intermediate B by a MOCVD method at a temperature of 450 ° C. using a mixed gas of silane gas (SiH ₄ ) and oxygen, and then etched. Etching was performed in the same manner as the mask layer was etched, so that the insulating film remained only on the inclined surface as shown in FIG. 9B.

  Next, a GaN-based compound semiconductor layer of the semiconductor light emitting device intermediate B having an insulating film formed on the inclined surface and a transparent conductive layer (100Å) made of ITO were formed on the insulating film. The transparent conductive layer was formed under the condition of RF power: 100 W in an environment in which 2% by volume of oxygen was added to Ar and the degree of vacuum was 0.07 Pa using a magnetron sputtering apparatus. Thereafter, a resist film having an opening for etching the transparent conductive layer is provided on the surface of the transparent conductive layer by the same method as that provided on the etching mask layer. Ar flow rate: 50 sccm, RF power: Etching was performed under the condition of 100 W, so that the transparent conductive layer remained in the regions as shown in FIGS. 7A and 7B to obtain a semiconductor light emitting device intermediate C.

Next, ion implantation (doping) was performed on the transparent conductive layer of the semiconductor light emitting device intermediate C under the conditions of a source source: MgCl ₂ , a beam current: 200 μA, and a degree of vacuum: 1 × 10 ⁻⁶ Torr. The dose was 1 × 10 ¹³ / cm ² .

Next, the semiconductor light-emitting element intermediate C after doping is heat-treated at 500 ° C. for 1 minute in a mixed gas atmosphere of nitrogen and oxygen (O ₂ concentration: 2 vol%, flow rate: 50 sccm) using a lamp annealing apparatus. gave. Thereafter, the oxide layer formed on the surface of the transparent conductive layer was removed with a mixed solution of iron chloride (FeCl ₃ ) and hydrochloric acid.

Subsequently, a Ti layer (thickness: 500 mm) is formed on the surface of the semiconductor light emitting device intermediate C after the heat treatment and oxide layer removal, and an etching process similar to that for the transparent conductive layer is performed, as shown in FIG. Only the concave surface portion was left. Thereafter, an Al layer (thickness: 8000 mm) is further formed, and only the Ti layer, the insulating film, and the transparent conductive layer are left as shown in FIG. 9C by the same etching process as that of the transparent conductive layer. I made it. Both the Ti layer and the Al layer were formed using an electron beam evaporation apparatus under the conditions of an emission current of 30 mA and a vacuum degree of 1 × 10 ⁻⁶ Torr.

  After the formation of the Al layer, heat treatment was performed at 440 ° C. for 15 minutes. This heat treatment improves the adhesion between the Ti layer and the Al layer, improves the adhesion between the p layer and the second transparent conductive layer, and reduces the contact resistance between the p layer and the second transparent conductive layer. It is carried out for the purpose. In the semiconductor light emitting device of the present invention, when an electrode in which Ti and Al are laminated is adopted, such heat treatment is preferably performed from the viewpoint of improving interlayer adhesion and reducing contact resistance between Ti and Al.

Thereafter, using a sputtering apparatus, an Au layer (thickness: 1.5 μm) is formed under the conditions of Ar flow rate: 50 sccm, RF power: 200 W, vacuum degree: 1 × 10 ⁻⁶ Torr, and etching treatment similar to that for the transparent conductive layer Thus, as shown in FIG. 2, only the p-type electrode (105) and the n-type electrode on the Al layer (111) were left to obtain a semiconductor light emitting device.

  The obtained semiconductor light-emitting device was placed on a photodetector (photodiode), a current of 2 mA was supplied to the semiconductor light-emitting device, and light generated outside the device from the support substrate side was measured with a photodetector. In this photodetector, the detected light is output as a current value. The results are shown in Table 2.

Examples 2, 3 and Comparative Example 1
In etching of an etching mask layer for obtaining a semiconductor light emitting device intermediate B, a product of NH ₄ F: HF = 6: 1 from Kanto Chemical Co., Inc. diluted 10 times with pure water as BHF (Example 2) In addition, NH ₄ F was added to the product of Kanto Chemical Co. to prepare NH ₄ F: HF = 10: 1 and diluted 10 times with pure water (Example 3), or the product of Kanto Chemical Co., Ltd. A semiconductor light-emitting device was obtained in the same manner as in Example 1 except that NH ₄ F was added to the sample to change to NH ₄ F: HF = 10: 1 (Comparative Example 1). The taper angle is 45 ° in Example 2, 60 ° in Example 3, and 90 ° in the comparative example. That is, the comparative example corresponds to a conventional semiconductor light emitting element. For these semiconductor light emitting devices, the light intensity (current value) was measured in the same manner as in Example 1. The results are shown in Table 2.

  In Table 2, as the light intensity ratio, the photocurrent values of the semiconductor light emitting devices of Examples 1 to 3 are shown as relative values when the photocurrent value obtained by the semiconductor light emitting device of the comparative example is 1. These results are also shown in the graph of FIG. In the graph of FIG. 11, the vertical axis represents the light intensity ratio, and the horizontal axis represents the taper angle of the concave portion of the semiconductor light emitting element. In the semiconductor light emitting devices of Examples 1 to 3, the light directed toward the support substrate by the reflective layer (Al layer) provided on the inclined surface of the recess or the p-type electrode made of Al, that is, the light extracted from the support substrate In addition, the current leakage is prevented by the insulating film and the supplied current is efficiently used for light emission. The results shown in Table 2 and FIG. 11 are obtained by these effects.

  Further, as Example 4, a semiconductor light emitting device was fabricated in the same manner as in Example 2 except that the ion implantation (doping) described in Example 1 was not performed on the transparent conductive layer. The operation voltage was compared. As a result, the semiconductor light emitting device of Example 2 had an operating voltage of 0.2 V lower than that of Example 4, and the resistivity reduction effect by doping the transparent conductive layer was confirmed.

Experiment 2
Example 5, Comparative Example 2
The semiconductor light-emitting device (Example 5) having the structure shown in FIG. 3 with the configuration shown in Table 3 according to the same method and conditions as in Example 2, and Table 3 according to the same method and conditions as in Comparative Example 1 were used. A semiconductor light emitting device (Comparative Example 2) having the structure shown in FIG. The p-type electrode and the n-type electrode of the semiconductor light emitting device of Comparative Example 2 both have a laminated structure, and Au is the outermost layer.

  The emission intensity (photocurrent value) of these semiconductor light emitting devices was measured in the same manner as in Example 1. The results are shown in Table 4.

  As can be seen from Table 4, in the semiconductor light emitting device of Example 5, the photocurrent with respect to the input current is larger than that of the semiconductor light emitting device of the comparative example, and the light emission intensity is high. In particular, under the condition of a supply current of 2 mA, the semiconductor light emitting device of Example 5 has a light emission intensity of about 1.7 times that of Comparative Example 2.

  In the semiconductor light emitting device of Example 5, the ratio of light directed to the support substrate side, that is, the light extracted from the support substrate by the reflective layer (Al layer) provided on the inclined surface of the recess or the p-type electrode made of Al In addition, the transparent conductive layer made of ITO provided on the inclined surface prevents current leakage and efficiently uses the supplied current for light emission. The results shown are obtained.

<Effect confirmation experiment by heat treatment of transparent conductive layer>
A semiconductor light emitting device was produced in the same manner as in Example 2 except that the heat treatment conditions of the semiconductor light emitting device intermediate C after forming the transparent conductive layer and doping were changed every 200 ° C. at 200 to 900 ° C., The relationship between the resistance value of the transparent conductive layer and the heat treatment temperature was investigated. The resistance value of the transparent conductive layer was measured by a four-terminal method. The transparent conductive layer on the p layer is made square in plan view, and probes are applied one by one to the four corners of the transparent conductive layer. Two adjacent probes are paired, a constant current is passed through one pair of probes, and the voltage is measured with the other pair of probes. The resistance value of the transparent conductive layer is calculated from the relationship between current (I) and voltage (V). As a measuring device, “Semiconductor Parameter Analyzer 4145B” manufactured by Agilent Technologies was used. The results are shown in FIG.

  In the graph of FIG. 12, the vertical axis represents the resistance value of the transparent conductive layer, and the horizontal axis represents the heat treatment temperature. As can be seen from FIG. 12, when the heat treatment temperature of the transparent conductive layer is 200 to 800 ° C., the resistance value is very small, indicating that the current supplied to the semiconductor light emitting device can be used for light emission more efficiently. ing.

It is a cross-sectional schematic diagram which shows 1st Embodiment of the semiconductor light-emitting device of this invention. It is a cross-sectional schematic diagram which shows 2nd Embodiment of the semiconductor light-emitting device of this invention. It is a cross-sectional schematic diagram which shows 3rd Embodiment of the semiconductor light-emitting device of this invention. It is a conceptual diagram in the case of forming a transparent conductive layer with a magnetron sputtering apparatus. It is a cross-sectional schematic diagram which shows 4th Embodiment of the semiconductor light-emitting device of this invention. It is a schematic diagram which shows 5th Embodiment of the semiconductor light-emitting device of this invention, (a) It is sectional drawing which shows the laminated structure of each layer, (b) It is sectional drawing which shows the structure of the support substrate side surface of n layer. It is a perspective view which shows 6th Embodiment of the semiconductor light-emitting device of this invention. It is a figure for demonstrating the process of forming the recessed part of the semiconductor light-emitting device of this invention. It is a figure for demonstrating the process of manufacturing the semiconductor light-emitting device (2nd Embodiment) of this invention. It is a cross-sectional schematic diagram which shows the structure of the conventional semiconductor light-emitting device. It is a graph which shows the result of Experiment 1 of an example. It is a graph which shows the heat processing effect of a transparent conductive layer.

Explanation of symbols

10a 10b 10c 10d 10e 10f 20 Semiconductor light emitting element 101 Support substrate 102 n layer 103 light emitting layer 104 p layer 105 p type electrode 106 n type electrode 107 first reflective layer 107a second reflective layer 108 first transparent conductive layer 108a Second transparent conductive layer 109 Insulating film 110 Al layer 111 Au layer 112 Ti layer 113 Sample 114 forming transparent conductive layer Target 115 for forming transparent conductive layer Magnet 116 Sample holder 117 Sputtered particle path 118 Plasma 119 p layer and first layer Contact surface 120 of the second reflective layer 120 Uneven structure 121 on the second reflective layer side surface of the second transparent conductive layer 200 Uneven structure 200 on the surface of the n-side support substrate Etching mask layer 201 Resist film 202 Plasma A1 containing chlorine A2 A3 A4 Path B1 of light reflected by the reflective layer toward the support substrate Light path B2 reflected from the p-type electrode surface Light path from the light emitting layer to the device side surface

Claims (9)

A GaN-based compound semiconductor layer having a recess for forming an n-type electrode on the n-layer is sequentially formed on one surface of a light-transmitting support substrate. A semiconductor light emitting device,
The concave portion has an inclined surface having a wide opening on the p-layer side and a narrow bottom surface on the n-layer side, and reflects at least the light emitted from the light emitting layer to the supporting substrate side on the inclined surface. Having a reflective layer,
The reflective layer is formed on the inclined surface from the n layer to the p layer, and is formed on a transparent conductive layer made of ITO, IZO, ZnO, In ₂ O ₃ , SnO ₂ or CdO,
The semiconductor is characterized in that the shape of the concave portion is formed in a comb shape in a plan view so that the comb-shaped p-type electrode and the comb-shaped n-type electrode are configured to be engaged with each other. Light emitting element.   2. The semiconductor light emitting device according to claim 1, further comprising a second reflective layer for reflecting light emitted from the light emitting layer toward the support substrate between the p layer and the p-type electrode. element.   The semiconductor light emitting element according to claim 2, wherein the second reflective layer is electrically connected to the p layer via a second transparent conductive layer.   4. The semiconductor light emitting element according to claim 3, wherein the second transparent conductive layer is interposed in a part between the second reflective layer and the p layer. 5. The semiconductor light emitting element according to claim 1, wherein the second transparent conductive layer is made of ITO, IZO, ZnO, In ₂ O ₃ , SnO _2, or CdO. The second transparent conductive layer is made of a metal containing Ni, Pd, Pt, Cr, Mn, Ta, Cu or Fe, and has a thickness of 20 to 200 mm,
6. A B layer composed of ITO, IZO, ZnO, In ₂ O ₃ , SnO ₂ or CdO, and the B layer is in contact with the second reflective layer. A semiconductor light-emitting device according to 1.   The semiconductor light emitting element according to claim 6, wherein the B layer related to the second transparent conductive layer has a concavo-convex structure on the surface of the second reflective layer.   8. The semiconductor light emitting device according to claim 1, wherein the first and / or second transparent conductive layer is doped with a p-type dopant.   The p-type dopant concentration in the first and / or second transparent conductive layer is larger in the vicinity of the other surface side than in the vicinity of the contact surface with the reflective layer. Semiconductor light emitting device.

Images