JP2006324434A - Semiconductor light-emitting device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a semiconductor light-emitting device capable of improving external quantum efficiency by efficiently taking out light from the lateral side of each light-emitting unit in the semiconductor light-emitting device in which a plurality of light-emitting units are formed into monolithic and the light-emitting units are connected in series and parallel, thereby making the units have high luminance. <P>SOLUTION: Semiconductor layers are laminated so as to form a light-emitting layer on a substrate 1 to form a semiconductor laminate 17, the semiconductor laminate layer 17 is electrically separated into a plurality of pieces, and a pair of electric connection portions 19, 20 to a conductive type layer are provided on each of the sections 17, thereby forming a plurality of light-emitting units 1 (1a, 1b), and the units are connected to one another by a wiring film 3. An eliminating section 17f is formed by eliminating the semiconductor laminate 17 sandwiched between the adjacent light-emitting units 1 to a portion near a substrate 11 other than the lower side of the wiring film 3. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  In the present invention, a plurality of light emitting portions are formed on a substrate and connected in series and parallel, so that, for example, a commercial AC power supply of 100V can be used in place of an illumination lamp or a fluorescent tube, or a sign or an automobile can be illuminated. The present invention relates to a semiconductor light emitting device that can be used for, for example. More specifically, the present invention relates to a semiconductor light emitting device having a structure that can improve the light extraction efficiency from each light emitting unit and increase the luminance for the same input.

  In recent years, with the advent of blue light emitting diodes (LEDs), LEDs have been used as light sources for displays and signal devices, and LEDs have been used in place of electric lamps and fluorescent tubes. When an LED is used instead of the lamp or the fluorescent tube, it is preferable to operate as it is with an AC drive of 100 V. For example, as shown in FIG. 6, the LEDs are connected in series and parallel and connected to an AC power supply 71. Things are known. S represents a switch (see, for example, Patent Document 1).

On the other hand, it is also considered to monolithically integrate light emitting devices in which such LED units are connected in series and parallel (see, for example, Patent Document 2). For example, as shown in FIG. 7, this structure has an i-GaN layer 61, an n-GaN contact layer 62, an n-AlGaN cladding layer 63, an active layer 64 composed of an InGaN multiple quantum well, p− The AlGaN cladding layer 65 and the p-GaN contact layer 66 are sequentially stacked, and a part of the semiconductor stacked portion is etched so that the n-GaN contact layer 62 is exposed, and further, the boundary portion of the adjacent LED is the i-GaN layer. The trench 70 is formed by etching until reaching 61, the SiO ₂ film 67 is formed in the trench 70, the transparent electrode 68 is formed on the p-GaN contact layer 66, the n-GaN contact layer 62 and the transparent electrode The metal electrode 69 is provided so as to be connected to 68. Then, every other electrode is connected to the first power supply wiring and the second power supply wiring to connect to the AC power supply 71, and each electrode is connected in parallel in the reverse direction.
Japanese Patent Laid-Open No. 10-083701 (FIG. 3) JP 2000-101136 A (FIG. 6)

  As described above, in order to monolithically form a light-emitting device in which a plurality of LEDs are connected in series and parallel, after a semiconductor layer is stacked on one substrate, a separation groove is formed to electrically separate each light-emitting unit. Although many light emitting units are formed, the light emitted from the LED not only travels to the upper surface side but travels in all directions laterally and downward. On the other hand, if a light-emitting device with high luminance is intended, for example, one side of the chip is as large as about 1500 μm square (the planar shape of each unit is about 80 μm × 180 μm, for example). For this reason, the light emitted from the unit inside the chip and traveling in the lateral direction travels through the semiconductor layer, and many of the light disappears before leaving the outside.

  The present invention has been made to solve such a problem. In a semiconductor light emitting device in which a plurality of light emitting units are monolithically connected, and the plurality of light emitting units are connected in series and parallel to increase the brightness, An object of the present invention is to provide a semiconductor light emitting device capable of improving the external quantum efficiency by efficiently extracting light from the side.

  A semiconductor light emitting device according to the present invention includes a substrate and a semiconductor layer stacked to form a light emitting layer on the substrate to form a semiconductor stacked portion, and the semiconductor stacked portion is electrically separated into a plurality of portions. A plurality of light emitting units each provided with an electrical connection to a pair of conductivity type layers, and the plurality of light emitting units connected to the electrical connection for connecting in series and / or in parallel, respectively. Each of the plurality of light emitting units is formed with at least a portion having an interval of 10 μm or more between the adjacent light emitting units, and the adjacent light emitting units including the interval of 10 μm or more. The semiconductor stacked portion sandwiched between the two is removed to the vicinity of the substrate except for the lower side of the wiring film. Here, up to the vicinity of the substrate means that it may reach the substrate or a part of the lower layer side of the semiconductor stacked portion may remain.

  A groove reaching the vicinity of the substrate in the semiconductor stacked portion sandwiched between the adjacent light emitting units instead of removing the semiconductor stacked portion sandwiched between the adjacent light emitting units including the interval of 10 μm or more. It is also possible to adopt a structure in which a plurality of columns are formed in parallel. At least a part of the grooves formed in parallel in the plurality of rows is formed so that the light emitting units are electrically insulated except for the electrical connection by wiring, thereby improving the insulation characteristics between the units. Can be made.

  It is preferable that the grooves formed in parallel in a plurality of rows are formed in a zigzag shape instead of a straight line in a planar shape, so that the traveling direction of light can be changed randomly and easily go out.

  It is preferable that the groove is filled with an insulator having a refractive index smaller than that of the semiconductor material constituting the semiconductor stacked portion, so that light is easily emitted from the light emitting unit to the gap portion side.

  The removed portion of the semiconductor stacked portion or the grooves formed in the plurality of rows are formed outside a dummy region of the semiconductor stacked portion provided through an isolation groove formed around each light emitting unit. Therefore, it is preferable that the light emitting unit is not directly damaged, and the outer periphery of the dummy region is formed in a zigzag shape in a planar shape, so that the incident angle of light to the interface changes, and thus light from the semiconductor stacked portion is emitted. It becomes easy to come out and is preferable.

  More specifically, the semiconductor stacked portion includes an n-type layer, an active layer, and a p-type layer made of a nitride semiconductor, and is formed so as to form a light emitting layer, whereby a blue or ultraviolet light emitting device. Since it is easy to obtain white light by applying a light emitting color conversion material or mixing it with the light of other green and red light emitting elements, it is convenient for a lighting device or the like.

  Here, the nitride semiconductor means a compound in which a group III element Ga and a group V element N or a part or all of a group III element Ga is substituted with other group III elements such as Al and In, and / or Alternatively, it refers to a semiconductor made of a compound (nitride) in which a part of N of the group V element is substituted with another group V element such as P or As.

  According to the present invention, at least a gap of 10 μm or more is formed between adjacent light emitting units of the light emitting unit, and the semiconductor laminated portion of the gap is removed, or a narrow groove is formed between the adjacent light emitting units. Since multiple rows are formed, the light emitted from the light emitting unit and traveling in the lateral direction exits into the voids where the semiconductor stack is removed and is taken out upward, or exits into the grooves and is refracted. When the direction is changed and taken out upward, the amount of light coming out on the upper surface side is greatly increased. In addition, since the light that has traveled to the substrate side is also reflected on the bottom surface side and returned easily from the side surface, the light extraction efficiency is greatly improved. By forming a dummy region between the groove and the light emitting unit and the semiconductor laminated portion and making the planar shape on the removal side of the semiconductor laminated portion into a zigzag shape, the incident angle of light to the interface changes, and further Light can be taken out from the semiconductor stack.

  Next, the semiconductor light emitting device of the present invention will be described with reference to the drawings. The semiconductor light-emitting device according to the present invention has a light-emitting layer on the substrate 11 as shown in FIG. 1 as a cross-sectional explanatory view (cross-section AA and BB in FIG. 1C) and a plane explanatory view of one embodiment. The semiconductor layers are stacked to form the semiconductor stacked portion 17, and the semiconductor stacked portion 17 is electrically separated into a plurality of parts, and each of the electrical connection portions 19 to the pair of conductivity type layers, 20 is provided to form a plurality of light emitting units 1 (1a, 1b), and the plurality of light emitting units 1 are wired to the electrical connection portions 19 and 20 so as to be connected in series and / or in parallel, respectively. The membrane 3 is connected.

  In the present invention, each of the plurality of light emitting units 1 has at least a portion having a distance d1 of 10 μm or more between the adjacent light emitting units 1, and the adjacent light emitting units 1 including the space d1 of 10 μm or more. 3 is removed by removing the semiconductor stacked portion 17 sandwiched between the first and second portions to the vicinity of the substrate 11 except for the lower side of the wiring film 3, or adjacently as shown in FIG. The semiconductor laminated portion 17 sandwiched between the light emitting units 1 is characterized in that a groove portion 17g (see FIG. 3) is formed in which a plurality of rows of grooves 17h are formed with the remaining portion 17i interposed therebetween. In the example shown in FIGS. 1 and 3, the semiconductor stacked portion 17 is removed so that a part of the undoped high-temperature buffer layer 13 remains, but may be removed until the substrate 11 is exposed. In addition, the space | interval d1 of 10 micrometers or more means that it is an area | region clearly different from what is called the isolation | separation groove | channel 17a (0.6-5 micrometers).

  That is, as shown in FIG. 1 (c), which is a plan view showing an example of each light emitting unit 1, a first separation groove 17a is formed around the light emitting unit 1, leaving a dummy region 17c around it. In addition, a wiring dummy region (second dummy region) 17d is left below the wiring film 3 via the second isolation groove 17b, and the surrounding semiconductor stacked portion 17 is removed to the vicinity of the substrate 1. Has been. In FIG. 1C, the hatched portion indicates the portion where the semiconductor stacked portion 17 is removed. As shown in FIG. 1C, the periphery of each light emitting unit 1 is removed by etching, leaving the dummy regions 17c and 17d, and the surrounding semiconductor stacked portion 17 removed. The semiconductor laminated portion 17 around the light emitting unit 1 may be directly removed by etching without leaving the dummy region 17c. However, if there are many etched portions, there are many scattered matters at that time, so there is contamination on the side surface of the light emitting unit 1. It is better to leave the dummy area 17c because the nation tends to adhere. Further, the distance d1 + 2c between the normal light emitting units 1 is about 8 μm or more as the dummy regions 17c and 17d because of the relationship between the width of the groove and the sum of the flat portions, but at least somewhere around the light emitting units, the adjacent light emitting units 1 It is preferable from the viewpoint of improving the external quantum efficiency that each light emitting unit 1 is formed so that the distance d1 is 10 μm or more, and the semiconductor laminated portion 17 at the distance is removed by etching as described above. That is, the removal portion 17f of the semiconductor laminated portion is formed without worrying about disconnection of the wiring film 3 because the wiring film is not formed thereon, unlike the separation groove of about 2 μm, preferably 5 μm or more. Is a sufficiently wide part of 10 μm or more.

  In the example shown in FIG. 1, a blue light emitting unit 1 (hereinafter also simply referred to as an LED) is formed by stacking nitride semiconductors, and a YAG (yttrium aluminum garnet) phosphor (not shown) is formed on the surface thereof. By providing a light emitting color conversion member made of Sr—Zn—La phosphor or the like, a light emitting device that emits white light is formed. For this reason, the semiconductor stacked portion 17 is formed by stacking nitride semiconductor layers, as shown in FIG. However, it is possible to form a light emitting unit of three primary colors of red, green, and blue so that it becomes white light, and it is not always necessary to make white light, and it can be formed in a light emitting portion of a desired light emitting color. it can.

As the substrate 11, sapphire (Al ₂ O ₃ single crystal) or SiC is used to stack a nitride semiconductor, but in the example shown in FIG. 1, sapphire (Al ₂ O ₃ single crystal) is used. ing. However, the substrate is selected from the viewpoint of the lattice constant and the thermal expansion coefficient according to the semiconductor layer to be laminated.

For example, as shown in FIG. 1, the semiconductor laminated portion 17 laminated on the sapphire substrate 11 has a low-temperature buffer layer 12 made of GaN of about 0.005 to 0.1 μm, and then a high-temperature buffer layer 13 made of undoped GaN. N-type layer 14 formed by a contact layer made of n-type GaN doped with Si and a barrier layer made of an n-type AlGaN-based compound semiconductor layer (layer having a large band gap energy), etc. A material having a band gap energy smaller than that of the barrier layer, for example, a well layer made of 1 to 3 nm In _0.13 Ga _0.87 N and a barrier layer made of 10 to 20 nm GaN; The active layer 15 having a multiple quantum well (MQW) structure in which 3 to 8 pairs are stacked is about 0.05 to 0.3 μm, and is p-type. The p-type barrier layer (layer having a large band gap energy) made of an lGaN-based compound semiconductor layer and the p-type layer 16 composed of a contact layer made of p-type GaN are combined in order of about 0.2 to 1 μm, for example, 0.6 μm. It is formed by being laminated. In the figure, although the height from the exposed surface of the n-type layer 14 to the surface of the translucent conductive layer 18 appears to be large, the height is actually about 1.5 to 2 μm, and the semiconductor laminated portion 17 The total thickness is much smaller than about 6.5 μm.

  In the example shown in FIG. 1, a high-temperature buffer layer 13 made of undoped and semi-insulating GaN is formed. When the substrate is made of an insulating substrate such as sapphire, there is no problem if a separation groove described later is formed up to the substrate even if it is not semi-insulating. In addition, since a semi-insulating semiconductor layer is provided, when the light emitting portions are electrically separated, they are electrically separated without completely etching up to the substrate surface. This is preferable. When the substrate 11 is made of a semiconductor substrate such as SiC, an undoped, semi-insulating high-temperature buffer layer 13 is formed so that the adjacent light emitting portions are electrically separated from each other, thereby making each light emitting portion independent. It is necessary for.

  In addition, the n-type layer 14 and the p-type layer 16 are two types of barrier layers and contact layers. However, a layer containing Al is provided on the active layer 6 side from the viewpoint of the carrier confinement effect. However, only a GaN layer or an AlGaN-based compound layer may be used. Moreover, these can also be formed with another nitride semiconductor layer, and another semiconductor layer may further intervene. Furthermore, in this example, the active layer 15 is sandwiched between the n-type layer 14 and the p-type layer 16, but a pn junction structure in which the n-type layer and the p-type layer are directly joined is also possible. Good. In addition, a p-type AlGaN compound layer is grown directly on the active layer 15, but a pit generation layer is formed below the active layer 15 by growing an undoped AlGaN compound layer of about several nm. Leakage due to contact between the p-type layer and the n-type layer can also be prevented while embedding the pits formed in 15.

  On the semiconductor laminated portion 17, a light-transmitting conductive layer 18 made of, for example, ZnO and capable of making ohmic contact with the p-type semiconductor layer 16 is provided in a thickness of about 0.01 to 0.5 μm. The translucent conductive layer 18 is not limited to ZnO, and even a thin alloy layer of about 2 to 100 nm of ITO or Ni and Au can diffuse current throughout the chip while transmitting light. it can. A part of the semiconductor laminated portion 17 is removed by etching to expose the n-type layer 14, and the n-type layer 14 and the high-temperature buffer layer 13 on the end side of the exposed portion of the n-type layer 14 are disposed in the vicinity of the substrate 1. The first separation groove 17a is formed by etching with a width of 0.6 to 5 μm (w), for example, about 2 to 3 μm. In the example shown in FIG. 1, the semiconductor laminated portion 17 is etched to the vicinity of the substrate 1 by leaving the dummy region 17 c of the interval c from the first separation groove 17 a and the lower side of the wiring film 3. Has been removed. The interval c between the dummy regions 17c is set within a range of about 1 to 30 μm. The separation groove 17a is formed by dry etching or the like, and is formed around the light emitting unit 1 with a width w as narrow as possible within a range that can be electrically separated, and is about 0.6 to 5 μm, for example, about 1 to 3 μm. (The depth is about 5 μm). If the width w is too large, it is likely to lead to disconnection of the wiring provided thereon, and therefore, the width w is formed as narrow as possible in an electrically separable range.

Then, an insulating film 21 such as SiO ₂ is formed on the entire surface of the translucent conductive layer 18 and the exposed n-type layer 14, and the p-side connection portion 19 and the n-side connection portion 20 are exposed by patterning. Thereafter, Ti that is a material of the wiring film 3 is formed by a lift-off method, and an Al film is provided on the n-side connection portion 20 that is an electrical connection portion with the n-type layer 14 to be alloyed. By forming the Au film on the other wiring portion, the wiring film 3 for connecting the light emitting units in series or in parallel is formed. The wiring film 3 is formed to a thickness of about 0.3 to 1 μm by vacuum deposition or sputtering of a metal film such as Au or Al. That is, the electrode and the wiring film are formed of the same material at the same time. It is also possible to form the wiring film by separately forming the n-side electrode and the p-side electrode by, for example, a Ti—Al alloy, Ti / Au laminated structure.

  For example, as shown in FIG. 1B, the n-side connection portion 20 of one light-emitting portion 1a separated by the separation groove 17a and the p-side connection portion 19 of the adjacent light-emitting portion 1b should be sequentially connected. For example, it can be connected in series and connected until the total operating voltage of 3.5 to 5 V per unit is close to 100 V (strictly, it can be adjusted by connecting resistors and capacitors in series) By connecting the sets in parallel and in parallel so that the connection direction of pn is reversed, a bright light source that operates with 100 V AC drive can be obtained. Further, as shown in FIG. 1 (c) as an example of the arrangement example of the light emitting unit 1, a set of two light emitting units each having a pn junction connected in parallel in the opposite direction are connected in series to obtain a total operating voltage. May be connected in series until the voltage approaches 100V. An equivalent circuit diagram of such an arrangement is as shown in FIG. If the brightness is not sufficient by this connection, these sets can be further formed and connected in parallel. As shown in FIG. 1 (c), when two light emitting units are connected in reverse parallel to form one set and further connected in series, not between the vertical direction but between the adjacent light emitting units 1 Therefore, it is necessary to connect the n-side connecting portion 20 and the p-side connecting portion 19 with the wiring film 3, and a place for forming the wiring film 3 is required between the light emitting units 1. As this space, the aforementioned dummy region 17c and the removed portion of the semiconductor stacked portion 17 can be formed.

Next, a method for manufacturing the semiconductor light emitting device having the structure shown in FIG. 1 will be described. Reactive gases such as trimethyl gallium (TMG), ammonia (NH ₃ ), trimethylaluminum (TMA), trimethylindium (TMIn), and n-type, together with carrier gas H ₂ , by metalorganic chemical vapor deposition (MOCVD) SiH ₄ as a dopant gas in the case of forming a p-type gas, and a necessary gas such as cyclopentadienyl magnesium (Cp ₂ Mg) or dimethyl zinc (DMZn) as a dopant gas in the case of a p-type is supplied to sequentially grow.

  First, on the substrate 11 made of sapphire, for example, a low-temperature buffer layer 12 made of a GaN layer is formed at a low temperature of about 400 to 600 ° C., for example, about 0.005 to 0.1 μm, and then the temperature is about 600 to 1200 ° C. The semi-insulating high-temperature buffer layer 13 made of undoped GaN is formed to about 1 to 3 μm, and the n-type layer 14 made of Si-doped n-type GaN and AlGaN compound semiconductor is formed to about 1 to 5 μm. To do.

Next, the growth temperature is lowered to a low temperature of 400 to 600 ° C., and, for example, 3 to 8 pairs of well layers made of In _0.13 Ga _0.87 N of 1 to 3 nm and barrier layers made of GaN of 10 to 20 nm are stacked. An active layer 6 having a quantum well (MQW) structure is formed to a thickness of about 0.05 to 0.3 μm.

  Next, the temperature in the growth apparatus is raised to about 600 to 1200 ° C., and the p-type AlGaN compound semiconductor layer and the p-type layer 16 made of GaN are combined and laminated to about 0.2 to 1 μm.

After that, an insulating film such as Si ₃ N ₄ is provided on the surface, and annealing is performed at about 400 to 800 ° C. for about 10 to 60 minutes to activate the p-type dopant. For example, a ZnO layer is subjected to MBE, sputtering, or vacuum deposition. The translucent conductive layer 18 is formed by forming a film with a thickness of about 0.1 to 0.5 μm by a method such as PLD or ion plating. Next, in order to form the n-side electrode 20, a part of the stacked semiconductor stacked portion 17 is etched by reactive ion etching using chlorine gas or the like so that the n-type layer 14 is exposed. At the time of mask formation at this time, the outside of the dummy region 17c around the light emitting unit 1 is opened and etched. Further, in order to electrically isolate the light emitting units 1 in the vicinity where the n-type layer 14 is exposed, the semiconductor stacked portion 17 is similarly formed with a width w of about 1 μm next to the exposed portion of the n-type layer 14. Etching is performed by dry etching to reach the high-temperature buffer layer 13 of the semiconductor stacked portion 17 to be electrically separated. At this time, the region outside the dummy region 17c can also be etched to the vicinity of the substrate 1 at the same time. When the separation groove 17c is formed away from the exposed portion of the n-type layer 14, etching outside the separation groove 17a and the dummy region 17c is performed separately from the exposure etching of the n-type layer 14. Also good.

Next, an insulating film 21 such as SiO ₂ is formed on the entire surface and patterned to expose a part of the translucent conductive layer 18 and a part of the n-type layer 14. The n-side connection portion 20 is formed. Then, Ti and Al are successively deposited by sputtering or vacuum deposition by a lift-off method or the like by about 0.1 μm and 0.3 μm, respectively, and heat-treated at about 600 ° C. for 5 seconds by RTA heating. The wiring film 3 is formed so as to be alloyed and to be connected to the n-side connection portion 20 and the p-side connection portion 19. As described above, the wiring film 3 is formed so that the light emitting unit 1 has a desired connection. If the ohmic contact of the p-side connection portion 19 is not sufficient, Ti and Au may be vacuum deposited on the p-side connection portion 18 by about 0.1 μm and 0.3 μm, respectively, and connected to the wiring film 3. Good. Alternatively, an Au film may be formed on the entire surface of the wiring film 3. These can be easily formed by a lift-off method. Then, the chip | tip of a semiconductor light-emitting device is obtained by chip-forming from the wafer for every light-emitting unit group which consists of a plurality of light-emitting units 1. When forming the wiring film 3, as shown in FIG. 1C, the electrode pad 4 for connection to the outside is formed simultaneously with the same material as the wiring film 3.

  In the above-described example, the dummy region 17c is left between the separation groove 17a and the removal portion 17f of the semiconductor stacked portion. However, since the dummy region 17c exists, the light emitting unit is etched when the removal portion 17f is etched. Although the side wall of 1 is not exposed to the etching gas atmosphere and the adhesion of impurities hardly occurs, the dummy region 17c may be omitted.

  In the above-described example, the semiconductor laminated portion 17 outside the dummy region 17c is removed by etching to form the removed portion 17f. However, the groove portion 17g can be formed without being completely removed. An example thereof is shown in FIG. 3 in the same diagram as FIG. 1 (a), and a partial plan view thereof is shown in FIG. 4 (a).

  That is, in FIG. 3, the above-mentioned removal portion is not completely removed, and a groove portion 17g is formed by alternately forming the grooves 17h and the remaining portions 17i. Since other structures such as the semiconductor stacked portion 17 are the same as those in the example shown in FIG. 1, the same portions are denoted by the same reference numerals and the description thereof is omitted. The width a of the groove 17h is about 2 to 5 μm, for example, about 3 μm, and the width b of the remaining portion is about 2 to 20 μm, for example, about 15 μm. It is desirable that at least a part of the groove 17h is formed so that the light emitting units are electrically insulated except for connection by wiring. By forming such a groove portion 17g, the light emitted from the semiconductor multilayer portion 17 and traveling in the lateral direction is caused by the difference between the refractive index of the groove 17h and the refractive index of the remaining portion (semiconductor multilayer portion) 17i. Whether the traveling direction changes, whether it travels or reflects, the traveling direction of the light changes significantly, which makes it easier to extract light to the outside. From the viewpoint of increasing the number of times the light traveling direction is changed as much as possible and facilitating extraction of light to the outside, it is preferable that the groove portions 17g are formed with the width and interval as described above.

  For example, as shown in FIG. 4 (b), the groove portion 17g has a planar zigzag groove 17h so that the light traveling direction is more easily changed. The light extraction efficiency is preferably improved. Similarly to the formation of the removal portion 17f, the formation of the groove portion 17g is similar to the formation of the removal portion 17f described above. When the semiconductor laminated portion 17 is etched by dry etching or the like, a linear groove or a zigzag groove opening is formed on the surface. By simply forming a mask of a resist film or the like on which a pattern is formed, it can be easily formed by the same etching.

In the example shown in FIG. 3, the insulating film 21 formed on the surface is also embedded in the groove 17h of the groove portion 17g. The insulating film 21 has a refractive index of about 1.5 to 1.7, which is smaller than the refractive index 2.5 to 2.7 of a nitride semiconductor layer such as SiO ₂ , for example. The light that travels in the lateral direction from the light easily exits into the groove 17g, and the light changes its direction due to reflection and refraction between the groove 17h and the remaining part 17i, and easily exits outside.

  FIG. 5 is an explanatory plan view of one light emitting unit portion showing a further modification of the semiconductor light emitting device of the present invention. That is, this example is characterized in that the outer periphery of the dummy portion 17c provided around the light emitting unit 1 via the separation groove 17a is etched so as to be zigzag in a planar shape. With such a zigzag shape, the traveling direction of the light is easily changed and the light can be easily taken out as in the case shown in FIG. 4B. This outer peripheral portion serves as the removal portion 17f of the semiconductor laminated portion shown in FIG. 1 described above, or the groove portion 17g shown in FIG. 3 and FIG. By forming the dummy portion 17c in a zigzag shape, there is no damage to the light emitting unit 1, and it is easy to prevent contamination from adhering, so the zigzag shape is formed directly on the outer periphery of the light emitting unit 1. More preferable than the case.

It is explanatory drawing of the cross section and plane of one Embodiment of the semiconductor light-emitting device by this invention. It is a figure which shows the equivalent circuit schematic which shows the connection of the light emission unit of FIG. It is sectional explanatory drawing which shows other embodiment of the semiconductor light-emitting device by this invention. FIG. 4 is an explanatory plan view of a groove portion shown in FIG. 3 and a diagram showing a modification thereof. It is a partial plane explanatory view showing a modification of a semiconductor light emitting device according to the present invention. It is a figure which shows the example of the conventional circuit which forms an illuminating device using LED. It is a figure which shows the example of the conventional structure which forms an LED unit monolithically and forms an illuminating device.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Light emitting unit 3 Wiring film 4 Electrode pad 11 Substrate 13 High temperature buffer layer 14 N-type layer 15 Active layer 16 p-type layer 17 Semiconductor laminated part 17a First separation groove 17b Second separation groove 17c Dummy area 17d Dummy area for wiring 17f Removal portion of semiconductor stacked portion 17g Groove portion 18 Translucent conductive layer 19 P-side connection portion 20 N-side connection portion 21 Insulating film

Claims (7)

  A semiconductor laminated portion is formed by laminating a semiconductor layer so as to form a light emitting layer on the substrate, and the semiconductor laminated portion is electrically separated into a plurality of layers, and a pair of conductivity type layers are respectively provided A plurality of light emitting units provided with an electrical connection to the wiring, and a wiring film connected to the electrical connection for connecting the plurality of light emitting units in series and / or in parallel, respectively. Each of the plurality of light emitting units includes at least a portion having an interval of 10 μm or more between adjacent light emitting units, and the semiconductor sandwiched between adjacent light emitting units having an interval of 10 μm or more A semiconductor light emitting device in which a stacked portion is removed to the vicinity of the substrate except under the wiring film.   A semiconductor laminated portion is formed by laminating a semiconductor layer so as to form a light emitting layer on the substrate, and the semiconductor laminated portion is electrically separated into a plurality of layers, and a pair of conductivity type layers are respectively provided A plurality of light emitting units provided with an electrical connection to the wiring, and a wiring film connected to the electrical connection for connecting the plurality of light emitting units in series and / or in parallel, respectively. Each of the plurality of light emitting units is a semiconductor light emitting device in which a plurality of rows of grooves reaching the vicinity of the substrate are formed in parallel in the semiconductor stacked portion sandwiched between adjacent light emitting units.   3. The semiconductor light emitting device according to claim 2, wherein the grooves formed in parallel in a plurality of rows are formed in a zigzag shape instead of a straight shape in a planar shape.   4. The semiconductor light emitting device according to claim 2, wherein the groove is filled with an insulator having a refractive index smaller than that of a semiconductor material constituting the semiconductor stacked portion.   A portion to be removed of the semiconductor stacked portion or a groove formed in the plurality of rows is formed outside a dummy region of the semiconductor stacked portion provided through an isolation groove formed around each light emitting unit. The semiconductor light-emitting device according to claim 1.   6. The semiconductor light emitting device according to claim 5, wherein the outer periphery of the dummy region is formed in a zigzag shape with a planar shape.   The semiconductor light emitting device according to claim 1, wherein the semiconductor stacked portion includes an n-type layer, an active layer, and a p-type layer made of a nitride semiconductor, and is formed to form a light emitting layer. .

Images