JP2004087930A - Nitride semiconductor light-emitting device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a nitride semiconductor light-emitting device, in which the extraction efficiency of light is improved and external quantum efficiency is markedly improved. <P>SOLUTION: In a nitride semiconductor light-emitting device, an n-type gallium nitride based compound semiconductor layer 21. an active layer 20 and a p-type gallium nitride based compound semiconductor layer 22 are laminated in this order on a substrate 1. In the same plane, at least electrodes 31, 4 of p-n single pair are formed, and an n electrode 4 is formed on an n electrode 4 forming surface, where a part of the semiconductor layer 21 is exposed. The p electrode 31 consists of a p main electrode and a p auxiliary electrode 32 for diffusing a current over the whole surface of the semiconductor layer 22 from the p main electrode. In this nitride compound semiconductor light-emitting device, a current blocking trench 51 is formed in the vicinity of the p electrode, and is set at least deeper than the n electrode 4 forming surface. <P>COPYRIGHT: (C)2004,JPO

Description

[0001] The present invention relates to a nitride semiconductor light emitting device, and more particularly to improving light extraction efficiency.
[0002]
2. Description of the Related Art Blue light-emitting diodes (LEDs), ultraviolet LEDs, and blue-violet semiconductor lasers (LDs) using GaN-based compound semiconductors have been developed. It is expected as a new light source to replace a vacuum tube illumination light source such as a fluorescent lamp. However, even now, in order to use these light-emitting elements for lighting applications, it is necessary to further increase the output of the elements, and various studies have been made for that purpose.
[0003] Incidentally, the GaN-based compound semiconductor described above has a characteristic that it is basically difficult to grow a thick film. Therefore, in a general GaN-based compound semiconductor light-emitting device, the distance from the main electrode for wire bonding to the light-emitting layer is inevitably extremely short, and is performed with another material-based semiconductor light-emitting device. Such means for uniformizing light emission using the current diffusion layer (uniformity in the sense that light emission occurs uniformly over the entire surface of the light emitting layer) cannot be usually employed. For this reason, the ohmic electrode is a so-called transparent electrode (also referred to as a second p auxiliary electrode in the present specification) having a thin film that transmits light, and the transparent electrode is formed on almost the entire surface (of the p-type layer) of the device. In order to obtain uniform light emission by making the current spread over the entire surface of the light-emitting layer, various measures have been taken.
[0004] By employing the above-mentioned transparent electrode, the entire surface of the light emitting layer is effectively utilized, and the amount of light emitted inside the device is greatly increased. In addition, by suppressing dislocation defects and the like, injected carriers can be converted into photons at a high rate, and as a result, the internal quantum efficiency can be greatly improved.
[0005]
However, from the viewpoint of the light extraction efficiency, there are various disadvantages from the viewpoint of the element structure. First, the size of an LED die cut out from a wafer is 200 to 400 μm square as a general-purpose product, though it differs depending on the application. With this size, the positions of the pn main electrodes are formed at a pair of corners located diagonally to the epi plane. Here, the n-electrode is naturally formed on the n-type semiconductor layer by etching the epi-surface. Here, the portion having the highest current density naturally emits the best light, but the portion having the highest current density is the portion from the p main electrode to the n electrode. Even if a full-surface electrode (which is a transparent electrode and corresponds to the second p-auxiliary electrode in the present specification) is employed, current is not completely and uniformly distributed over the entire surface. This is because the current tends to flow over the shortest distance if the resistance value is the same, and as a result, the current concentrates on the portion from the p main electrode to the n electrode. Therefore, despite the presence of the entire surface electrode, the periphery of the epi is less luminous than the center.
Another problem is the absorption of light by the p main electrode. Although it depends on the mounting method and the like, it is difficult to extract light emitted immediately below the p main electrode in face-up mounting. There is also a problem that light is emitted in the vicinity of the p main electrode, propagates directly below the p main electrode, and is absorbed as it is. Further, there is a problem of light absorption by the transparent electrode as the second p auxiliary electrode. That is, the light transmittance of the transparent electrode is only about 50%, and since the transparent electrode is formed on almost the entire surface, a factor that deteriorates the efficiency of extracting light to be emitted from the vertical (surface) direction of the device. It has become. However, since this problem has a front-to-back relationship with the improvement of the carrier injectability by the transparent electrode, for example, non-use of the transparent electrode does not solve the fundamental problem.
SUMMARY OF THE INVENTION Accordingly, an object of the present invention is to provide a nitride semiconductor light emitting device in which light extraction efficiency is enhanced and external quantum efficiency is significantly improved.
[0008]
According to a nitride semiconductor light emitting device of the present invention, at least an n-type gallium nitride compound semiconductor layer, an active layer, and a p-type gallium nitride compound semiconductor layer are sequentially laminated on a substrate. At least a pair of pn1 electrodes are formed in the same plane, the n-electrode is formed on the n-electrode formation surface partially exposing the n-type gallium nitride-based compound semiconductor layer, and the p-electrode is formed of a p-main electrode and a p-electrode. In a nitride semiconductor light emitting device comprising a p auxiliary electrode for diffusing current from the main electrode to the entire surface of the p-type gallium nitride based compound semiconductor layer, a current blocking groove is formed close to the p main electrode, and the groove is at least n It is characterized by being deeper than the electrode formation surface. With this configuration, a portion of the shortest distance as a current from the center of the p main electrode toward the n electrode is electrically disconnected, so that concentration of the current is suppressed and diffusion to the peripheral portion is promoted. it can. In addition, light emission immediately below the p main electrode can be suppressed. As a result, the light emission in the peripheral portion increases, and the light extraction efficiency increases. In addition, the element breakdown occurs where the highest voltage is applied, and as a result, the current flows the most, and the part that emits the most light, that is, the part from the p main electrode to the n electrode is electrostatically damaged. Since the current can be diffused, the electrostatic withstand voltage of the entire die is improved, and the life of the element is extended. Further, as a further effect of adopting this configuration, since the diffusion of electricity is promoted to the p auxiliary electrode side, light emission immediately below the p main electrode can be suppressed. As a result, it is possible to suppress the light emission that could not be efficiently extracted even when the light was emitted, and to improve the light emission capability of the entire device by emitting light from the portion that can efficiently emit light. In addition, by adopting such a configuration, when light emitted immediately below the p main electrode travels toward the n electrode, it is almost absorbed by the transparent electrode or the like before being emitted from the end face of the epi. However, by surrounding the p main electrode with the side surface of the die and the current blocking groove, light can be reliably extracted even if light is emitted immediately below the p main electrode. Conversely, light propagating in the nitride semiconductor layer in the direction of the p main electrode can be extracted without being absorbed by the p main electrode.
Further, in the nitride semiconductor light emitting device of the present invention, in addition to the current blocking groove, a light extraction groove is formed on a line connecting the p main electrode and the n electrode. By adopting such a configuration, for example, when light emitted near the diagonal epi end face where the pn both main electrodes are not formed proceeds to the other epi end face, almost no light is emitted before being emitted from the epi end face. Although it was absorbed by the transparent electrode and changed into heat and could not be taken out, it could be taken out of the epi from the groove before that.
Further, in the nitride semiconductor light emitting device of the present invention, a light extraction groove for further dividing the light emitting surface is added substantially perpendicular to the device end face. If the groove is increased so as to divide the light emitting surface evenly, the light emitted from the active layer can be extracted from the groove portion to the outside of the epi at a short distance no matter which direction the light travels. Further, by forming the groove perpendicular to the first p auxiliary electrode, it is possible to emit light over the entire surface without obstructing the flow of current.
Further, in the nitride semiconductor light emitting device of the present invention, the total area of the two groove portions is 5 to 50% with respect to the light emitting surface. If it is less than this, the effect is small, and if it is more than this, the light emitting region is conversely reduced, resulting in an increase in current density and a decrease in luminous efficiency and an increase in drive voltage.
Further, in the nitride semiconductor light emitting device of the present invention, the depth of the two grooves reaches the substrate. In the present invention, the substrate material is not particularly limited, and sapphire, SiC, GaN, AlN, or the like can be used as the substrate, and the refractive index differs depending on the material of the substrate. . Here, the refractive index of each material is about 1.8 for sapphire, about 2.8 for SiC, about 2.5 for GaN, and about 2.2 for AlN. When sapphire is used as a substrate, light has a lower refractive index than GaN to be laminated, so that light is guided preferentially on the epi side having a higher refractive index. Therefore, in order to extract the guided light to the outside of the epi from the groove, it is effective to form a groove up to the substrate.
Further, in the nitride semiconductor light emitting device according to the present invention, the two grooves penetrate the substrate. Furthermore, in the nitride semiconductor light emitting device of the present invention, the groove penetrates the substrate. If a substrate having a high refractive index, such as SiC, is used as the substrate, light is guided preferentially through the substrate. Therefore, if a groove is formed so as to penetrate the substrate, it is more effective to improve the light extraction efficiency. Even when a substrate having a low refractive index such as sapphire is used, not all light is guided in the epi, but also guided in the substrate, and the thickness of the substrate is smaller than the epi thickness. Is so thick that the total amount of light guided in the substrate cannot be ignored. Therefore, the light extraction efficiency is improved regardless of the refractive index of the substrate.
Further, in the nitride semiconductor light emitting device of the present invention, the width (w) of each of the two grooves is equal to or larger than the thickness of the laminated film on the substrate. Light guided in the epi is guided between the substrate interface and the epi surface, that is, in the epi thickness. Therefore, by setting the groove width to be equal to or larger than the epi thickness, the probability that light emitted from the groove enters the epi again is reduced.
Further, in the nitride semiconductor light emitting device of the present invention, the relationship between the width (w) of the two grooves and the depth (d) of the grooves is w / d ≧ 1. With such a configuration, the probability that the light emitted from the groove enters the epi again is reduced.
Further, in the nitride semiconductor light emitting device according to the present invention, both the grooves have a taper angle. By having a taper angle, the light extraction efficiency can be further improved. In this case, the width (w) of the groove is defined by the outermost surface, not the bottom of the groove.
Further, in the nitride semiconductor light emitting device of the present invention, a protective film is formed on both side surfaces of the groove. Since the protective film prevents the active layer from being exposed to the short circuit, the life can be improved.
Furthermore, in the nitride semiconductor light emitting device of the present invention, the substrate has irregularities on the growth surface side. The use of a specially processed substrate as shown in FIG. 6 further increases the light extraction efficiency. In FIG. 6, a hatched portion is a concave portion. It has irregularities of regular triangle, rhombus, or regular hexagon, and has the effect of scattering, diffracting, and refracting light on the irregular surface. It is important that the step of the unevenness has a size of not less than 50 nm and not more than the thickness of the semiconductor layer grown on the substrate. The reason for this is that when at least the emission wavelength (for example, 206 nm to 632 nm in the case of an AlGaInN-based emission layer) is λ, if there is no depth or a step of λ / 4 or more, light can be sufficiently scattered or diffracted. On the other hand, when the depth of the concave portion or the step of the convex portion exceeds the thickness of the semiconductor layer grown on the substrate, the current does not easily flow in the lateral direction in the laminated structure, and the luminous efficiency decreases. It is. Therefore, the surface of the semiconductor layer may be concave and / or convex. In order to sufficiently scatter or diffract light, it is preferable that the depth or the step is λ / 4 or more, but if the depth or the step is λ / 4n (n is the refractive index of the semiconductor layer) or more. Scattering or diffraction effects can be obtained.
In addition, the size of the concave portion and / or the convex portion (that is, the length of one side which is a constituent side of the concave portion and / or the convex portion) and the interval between each other are at least λ when the emission wavelength in the semiconductor is λ. It is important that the size is / 4 or more. Conversely, if the size of the unevenness or the mutual interval is too large, the scattering surface is reduced, which is not appropriate. Therefore, 0.2 to 20 μm is preferable.
Further, if a taper angle of 40 to 50 ° is formed as compared with the case where the unevenness is formed perpendicular to the epi-growth surface, the extraction efficiency is further improved.
Further, in the nitride semiconductor light emitting device of the present invention, the two grooves are formed by etching. Etching is the easiest to form, with little variation.
[0020]
The present inventors have found that in a nitride semiconductor light emitting device, light emitted immediately below the p main electrode cannot be effectively extracted, and conversely, the inside of the nitride semiconductor layer extends in the direction of the p main electrode. The propagating light is absorbed by the p main electrode, the light output is concentrated at the shortest distance between the pn main electrodes, and the light is not uniformly emitted over the entire surface, and the ratio of the light emitting component emitted from the end face Was relatively large, and the present invention was completed. That is, according to the first aspect of the present invention, the end face of the light emitting region is etched inside the p main electrode, in addition to the first end face light emitting section in the peripheral area of the element, which is the end face light emitting section in the ordinary light emitting element. By forming the second end face light emitting portion (light extraction groove and current constriction groove) exposed by the processing, the second end face light emission portion functions as a light extraction window for light emitted immediately below the p main electrode, and the auxiliary electrode The light emitted below propagates directly below the p main electrode and acts so as not to be absorbed by the p main electrode, and also prevents an electrical short circuit between the pn electrodes, thereby preventing electrons from flowing into the lower part of the p main electrode. Suppression suppresses light emission immediately below the p main electrode, and as a result, current can be diffused over the entire surface of the epi and the entire surface of the epi can be illuminated. Therefore, the light extraction efficiency is improved, and the total amount of the actual light emission is also improved. As a result, the external quantum efficiency can be significantly improved.
[0021]
Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 shows an embodiment of the nitride semiconductor light emitting device of the present invention. FIG. 1 (a) is a plan view of the nitride semiconductor light emitting device viewed from the lamination surface side, and FIG. FIG. 4 shows a cross-sectional view taken along the line A ′. In the figure, 1 is a substrate, 21 is an n-type gallium nitride compound semiconductor layer, 22 is a p-type gallium nitride compound semiconductor layer, and 20 is an active layer. Hereinafter, the production method of the present invention will be described in detail.
[0022]
As a semiconductor wafer, a nitride semiconductor layer having a structure to be an LED (light emitting diode) was formed on a spinel substrate. Specifically, a GaN buffer layer, an n-type GaN contact layer, an n-type AlGaN cladding layer, an InGaN active layer having a multiple quantum well structure, a p-type GaN cap layer, a p-type AlGaN And a p-type GaN contact layer. A current blocking groove and a light extraction groove are formed from the epi-face side of the semiconductor wafer. Etching (including both dry etching and wet etching) is suitable as a method for forming the groove, and an optical method using laser irradiation or a mechanical method such as a dicer or a scriber is also possible.
[0023]
(Nitride semiconductor wafer 100, 200, 300, 400, 500, 600)
The nitride semiconductor wafers 100, 200, 300, 400, 500, and 600 have a structure in which a nitride semiconductor 2 is formed on a substrate 1. Examples of the substrate 1 of the nitride semiconductor 2 include various materials such as sapphire, spinel, silicon carbide, zinc oxide, and gallium nitride single crystal. In order to form a nitride semiconductor layer with good productivity and good crystallinity, sapphire is used. Substrates, spinel substrates and the like are preferably used.
[0024]
Nitride semiconductor _{(In X Ga Y Al 1-} X-Y N, 0 ≦ X, 0 ≦ Y, X + Y ≦ 1) can be variously formed by a MOCVD method or HVPE method. A nitride semiconductor can be used as a semiconductor element by forming a PN junction, a PIN junction, and a MIS junction. The structure of the semiconductor can be variously selected such as a homo junction, a hetero junction, and a double hetero junction. In addition, a single quantum well structure or a multiple quantum well structure in which the semiconductor layer is thin enough to produce a quantum effect can be used.
[0025]
[0026]
The order of forming the grooves may be immediately before or after forming the electrodes. If the groove is formed immediately before forming the electrode, it is necessary to mask the electrode material during the formation of the second p auxiliary electrode so that the electrode material does not enter the groove. In addition, if the groove is formed immediately after the formation of the electrode, it is necessary to form the groove together with the second p auxiliary electrode, and the method of forming the groove is limited.
[0027]
The current confinement groove and the light extraction groove may be formed separately, but it is not particularly necessary to form them separately, and the steps can be simplified by forming them simultaneously. As a prior art, there is an example in which a light extraction groove is formed at the time of exposing an n-contact layer for forming an n-electrode (JP-A-2002-164574, JP-A-2002-26386, etc.). Since the current confinement groove needs to be deeper than the exposed surface of the n-contact layer in order to exhibit the effect, it is difficult to form the groove in the same process. Therefore, it is preferable to perform the step in a step different from the step of exposing the n-contact layer, although it depends on the method of forming the groove.
[0028]
【Example】
(Example 1)
Using a 425 μm-thick and washed sapphire substrate as a substrate, a nitride semiconductor was laminated by MOCVD to form a nitride semiconductor wafer. The nitride semiconductor was formed as a multilayer film so as to be a light emitting element. First, a buffer layer having a thickness of about 200 Å was formed at 510 ° C. by flowing NH ₃ (ammonia) gas, TMG (trimethylgallium) gas, and hydrogen gas as a carrier gas as source gases.
[0029]
Next, after stopping the inflow of the TMG gas, the temperature of the reactor was raised to 1050 ° C., and again a TMG gas and a SiH ₄ (silane) gas as a dopant gas were allowed to flow, so that GaN having a thickness of about 6 μm serving as an n-type contact layer was formed. A layer was formed.
[0030]
The active layer is once made only of carrier gas and NH _3, and the temperature of the reactor is kept at 800 ° C., then NH ₃ (ammonia) gas, TEG gas, TMI (trimethylindium) as a source gas and nitrogen gas as a carrier gas. , A barrier layer made of undoped GaN is grown to a thickness of 200 angstroms, then the temperature is set to 800 ° C., and a well layer made of undoped In _0.4 Ga _0.6 N using TMG, TMI, and ammonia. Is grown to a thickness of 30 angstroms. An active layer 20 having a multiple quantum well structure having a total thickness of 1120 Å is formed by alternately laminating five barrier layers and four well layers in the order of barrier + well + barrier + well... + Barrier. Grow.
[0031]
After the flow of TMG and TMI was stopped to form a cladding layer on the active layer and the temperature of the reactor was maintained at 1050 ° C., NH ₃ (ammonia) gas, TMA (trimethylaluminum) gas, TEG gas, A Cp ₂ Mg (cyclopentadiermagnesium) gas as a dopant gas and a nitrogen gas as a carrier gas were passed to form a GaAlN layer having a thickness of about 0.1 μm as a p-type cladding layer.
[0032]
Finally, the temperature of the reactor is maintained at 1050 ° C., and NH ₃ (ammonia) gas, TMG gas, Cp ₂ Mg gas as a dopant gas, and hydrogen gas as a carrier gas are flowed as a source gas, and a thickness of about 0 is formed as a p-type contact layer. A 0.5 μm GaN layer was formed. (Note that the p-type nitride semiconductor layer has been annealed at 400 ° C. or higher.)
The nitride semiconductor 2 thus formed is exposed to the n-contact layer to expose the n-type contact layer, and then the second p auxiliary electrode 33 is formed, and the first p auxiliary electrode 32 is formed thereon. The p main electrode is formed at the same time. Thereafter, an n-electrode is formed, and a protective film is formed. Up to this point, it is the same as the conventional case, and a chip cut out as a 350 μm square chip is referred to as Reference A.
[0033]
Based on the reference A, a current confinement groove was formed vertically to the substrate without making an angle (FIG. 1A). As a result, the extraction efficiency of 5% light was increased as compared with Reference A.
[0034]
(Example 2)
In addition to Example 1, a light extraction groove was formed vertically on the line connecting the p main electrode and the n electrode to the substrate as shown in FIG. As a result, the light extraction efficiency increased by 12% as compared with Reference A.
[0035]
(Example 3)
In addition to Example 2, a light extraction groove was formed almost perpendicularly to the element end face to the substrate without making an angle as shown in FIG. As a result, 16% light extraction efficiency was increased as compared with Reference A.
[0036]
(Example 4)
In Example 3, the current constriction groove and the light extraction groove were formed perpendicular to the epitaxial growth surface side of the substrate, but were tapered. A taper angle of 50 ° was formed with respect to the vertical plane. As a result, 24% light extraction efficiency was increased as compared with Reference A.
[0037]
(Example 5)
The same processing as in Example 4 was performed and formed on a specially processed substrate. Using the pattern shown in FIG. 6A, the length of one side of an equilateral triangle is 10 μm, the interval between adjacent triangles is 4 μm, the depth of the unevenness is 0.8 μm, and the taper angle of the unevenness is 40 ° from the vertical direction. Was. As a result, the light extraction efficiency increased by 41% compared to Reference A.
[0038]
(Example 6)
While Examples 1 to 5 were cut out as chips of 350 μm square, hereafter, examples using 1 mm square chips will be described. The semiconductor layers are stacked in the same manner as in the first embodiment, and both pn electrodes are formed in an electrode arrangement as shown in FIG. This state is referred to as reference B. As shown in FIG. 4, a current constriction groove and a light extraction groove are formed in the reference B in a mesh shape. At this time, the groove width was 29 μm, the groove depth was up to the substrate, and the groove was formed perpendicular to the substrate without forming a taper. As a result, 40% light extraction efficiency was increased as compared with Reference B.
[0039]
(Example 7)
In the same manner as in Example 6, a groove was formed except that a taper angle of 50 ° was formed in a direction perpendicular to the substrate. As a result, the light extraction efficiency was increased by 62% as compared with Reference B.
[0040]
(Example 8)
The same processing as in Example 7 was performed and formed on a specially processed substrate. Using the pattern shown in FIG. 6A, the length of one side of an equilateral triangle is 25 μm, the interval between adjacent triangles is 10 μm, the depth of the unevenness is 0.8 μm, and the taper angle of the unevenness is 40 ° from the vertical direction. Was. As a result, the light extraction efficiency increased by 66% as compared with Reference A.
[0041]
【The invention's effect】
The nitride light-emitting device of the present invention, according to the first aspect of the present invention, further includes an inner surface of a p-side main electrode in addition to a first end-surface light-emitting portion in a peripheral portion of the device, which is an end-surface light-emitting portion in a normal light-emitting device. Forming the second end face light emitting portion (light extraction groove and current constriction groove) in which the end face of the light emitting region is exposed by etching, the second end face light emitting portion suppresses the inflow of electrons directly below the p main electrode. Further, light emission immediately below the p main electrode can be suppressed, and light emitted below the second p auxiliary electrode can be prevented from entering directly below the p main electrode. Even if light is emitted immediately below the p main electrode, light is emitted. It acts as an extraction window and prevents an electrical short circuit between the pn electrodes. As a result, current can be diffused over the entire surface of the epi and the entire surface of the epi can be illuminated. Therefore, the light extraction efficiency is improved, and the total amount of the actual light emission is also improved. As a result, the external quantum efficiency can be significantly improved.
[Brief description of the drawings]
FIG. 1 is a view showing an example of a nitride semiconductor light emitting device of the present invention, wherein FIG. 1 (a) is a schematic plan view thereof, and FIG. 1 (b) is a schematic view taken along line AA ′ of FIG. 1 (a). It is sectional drawing.
FIG. 2 is a schematic plan view showing another example of the nitride semiconductor light emitting device of the present invention.
FIG. 3 is a schematic plan view showing another example of the nitride semiconductor light emitting device of the present invention.
FIG. 4 is a schematic plan view showing another example of the nitride semiconductor light emitting device of the present invention.
FIG. 5 is a schematic plan view showing an example of a conventional nitride semiconductor light emitting device.
FIG. 6 is a schematic pattern of a specially processed substrate that can be used for the nitride semiconductor light emitting device of the present invention.
FIG. 7 is a schematic plan view showing an example of a conventional nitride semiconductor light emitting device.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Substrate 2 ... Nitride semiconductor 20 ... Active layer 21 ... n-type gallium nitride compound semiconductor 22 ... p-type gallium nitride compound semiconductor 31 ... p main electrode 32 ... A first p auxiliary electrode 33 a second p auxiliary electrode 4 an n electrode 41 an n main electrode 42 an n auxiliary electrode 51 a current constriction groove 52 light extraction Grooves 100, 200, 300, 400, 500, 600 ... nitride semiconductor wafer

Claims (12)

At least an n-type gallium nitride-based compound semiconductor layer, an active layer, and a p-type gallium nitride-based compound semiconductor layer are sequentially stacked on a substrate, and at least one pair of pn electrodes are formed in the same plane, and the n-electrode is an n-type. A p-electrode is formed on the n-electrode formation surface partially exposing the gallium nitride-based compound semiconductor layer, and a p-electrode for diffusing current from the p-main electrode to the entire surface of the p-type gallium nitride-based compound semiconductor layer. In a nitride semiconductor light emitting device including a p auxiliary electrode,
A nitride semiconductor light emitting device, wherein a current blocking groove is formed close to a p main electrode, and the groove is deeper than at least an n electrode forming surface. 2. The nitride semiconductor light emitting device according to claim 1, wherein in the nitride semiconductor light emitting device, a light extraction groove is formed on a line connecting the p main electrode and the n electrode in addition to the current blocking groove. 3. The nitride semiconductor light emitting device according to claim 1, wherein a light extraction groove for dividing a light emitting surface is further added substantially perpendicularly to an end surface of the device. 4. The nitride semiconductor light emitting device according to claim 1, wherein in the nitride semiconductor light emitting device, a total area of the both groove portions is 5 to 50% with respect to a light emitting surface. 5. 5. The nitride semiconductor light emitting device according to claim 1, wherein in the nitride semiconductor light emitting device, the depths of both the grooves reach the substrate. 5. The nitride semiconductor light emitting device according to claim 1, wherein in the nitride semiconductor light emitting device, both the groove portions penetrate the substrate. 6. 7. The nitride semiconductor light emitting device according to claim 1, wherein in the nitride semiconductor light emitting device, a width (w) of each of the grooves is equal to or larger than a layer thickness on a substrate. 8. 7. The nitride semiconductor light emitting device according to claim 1, wherein a relationship between a width (w) of each of the grooves and a depth (d) of the grooves is w / d ≧ 1. 9. The nitride semiconductor light emitting device according to claim 1, wherein said grooves have a taper angle. 10. The nitride semiconductor light emitting device according to claim 1, wherein a protective film is formed on both side surfaces of the groove in the nitride semiconductor light emitting device. 11. The nitride semiconductor light emitting device according to claim 1, wherein the substrate has irregularities on a growth surface side. The nitride semiconductor light emitting device according to claim 1, wherein the two grooves are formed by etching in the nitride semiconductor light emitting device.

Images