JP2002026443A - Nitride-based semiconductor element and its manufacturing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a nitride-based semiconductor element in which a substrate and a nitride-based semiconductor layer formed on the substrate can be divided into pieces without deteriorating the characteristics of the element and a method of manufacturing the element. SOLUTION: In this nitride-based semiconductor element, an n-electrode 1 is formed in an area other than a stripe-like area formed along a resonator end face A side and another stripe-like area formed along another resonator end face B side on the rear surface of an n-GaN transparent substrate 100 and a p-electrode 2 is formed on the whole upper surface of a p-GaN contact layer 28.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

TECHNICAL FIELD The present invention relates to a gallium nitride (G)
The present invention relates to a nitride-based semiconductor device in which a nitride-based semiconductor layer is formed on a conductive substrate such as aN) and silicon carbide (SiC), and a method for manufacturing the same.

[0002] Here, the nitride-based semiconductor is BN
III-V such as (boron nitride), GaN (gallium nitride), AlN (aluminum nitride), InN (indium nitride), TIN (thallium nitride), or a mixed crystal thereof
It is a group nitride semiconductor.

[0003]

2. Description of the Related Art Conventionally, conductive and transparent GaN,
In a nitride-based semiconductor laser device in which a nitride-based semiconductor layer is formed on a substrate made of SiC or the like, see, for example, Applied Physics, Vol. 68, No. 7 (1999), p. 797-8
As described in No. 00, an electrode patterned in a predetermined shape is formed on a nitride-based semiconductor layer. On the other hand, the electrodes formed on the back surface of the substrate are not subjected to patterning, and the electrodes are formed on the entire back surface of the substrate.

FIG. 9 is a schematic perspective view showing an example of a conventional nitride semiconductor laser device in which a nitride semiconductor layer is formed on a GaN substrate.

[0005] In the semiconductor laser device of FIG. 9, n-GaN is provided on a transparent substrate 51 made of n-GaN having conductivity.
N-cladding layer 52 made of AlGaN, n-light guide layer 53 made of n-GaN, MQW active layer 54 having a multiple quantum well structure, p-cap layer 55 made of p-AlGaN, p-GaN made of p-GaN The light guide layer 56 and the p-clad layer 57 made of p-AlGaN are sequentially laminated. The p-cladding layer 57 has a stripe-shaped ridge portion 50 formed therein. An n-current block layer 58 made of n-AlGaN is formed on a flat portion of the p-cladding layer 57. Furthermore, this n-
A p-contact layer 59 made of p-GaN is formed on the current block layer 58 and the p-cladding layer 57 of the ridge 50.
Are formed.

An n-electrode 60 is formed on a surface of the transparent substrate 51 opposite to the crystal growth layer side (hereinafter referred to as a back surface) so as to cover the entire surface.

On the other hand, on the p-contact layer 59, both end faces in the resonator length direction (hereinafter referred to as resonator end faces).
The p-electrode 61 is formed in a region excluding a stripe-shaped region along a side edge and a stripe-shaped region along an end surface (hereinafter, referred to as a side surface) side in a direction perpendicular to the cavity length direction. As described above, in the above-described semiconductor laser device, the p-electrode 61 is patterned into a predetermined shape, and the p-contact layer is formed in the stripe regions on both cavity end faces and the stripe regions on both side surfaces. 59 are exposed.

At the time of manufacturing the above-described semiconductor laser device, first, the layers 52 to 59 are grown on the transparent substrate 51. Then, in the semiconductor wafer having the layers 52 to 59 formed on the transparent substrate 51, the n-electrode 60 is formed on the entire back surface of the transparent substrate 51, and is formed in a predetermined shape on the p-contact layer 59 of the semiconductor wafer. Patterned p
An electrode 61 is formed. In this case, each p electrode 6
A plurality of p-electrodes 61 patterned in a square shape are formed on the p-contact layer 59 of the semiconductor wafer such that the p-contact layer 59 is exposed in a grid pattern between the p-electrodes 61.

After the n-electrode 60 and the p-electrode 61 are formed as described above, the region of the p-contact layer 59 exposed between the p-electrodes 61 is shifted from the p-contact layer 59 side in a direction perpendicular to the resonator length direction. Scribe along to form the wound. Then, the semiconductor wafer is cleaved along the scratches and divided into a plurality of stripe-shaped semiconductor wafers. In this manner, the resonator end face is exposed by cleavage, and a resonator is manufactured. Then, a protective film is formed on the end face of the resonator.

Further, in the divided striped semiconductor wafer, the region of the p-contact layer 59 exposed between the p-electrodes 61 is scribed from the p-contact layer 59 side in a direction parallel to the resonator length direction. Form a wound.
Then, the stripe-shaped semiconductor wafer is further cleaved along the scribe flaws to perform element isolation, thereby forming the semiconductor laser element shown in FIG.

As described above, at the time of manufacturing the above-described semiconductor laser device, the patterned p-electrode 61 and the p-contact layer 59 exposed between the p-electrodes 61 are used as marks for a portion to be scribed.

[0012]

As described above, when fabricating a semiconductor laser device, scribing is performed from the p-contact layer 59 side to form a flaw in the cavity fabrication process and device isolation process.

Here, since the nitride semiconductor is a very hard material, it is necessary to apply a large pressure from the p-contact layer 59 side when scribing. Further, in order to easily cleave in the resonator fabrication process and the device isolation process, the p-contact layer 59 is formed.
It is necessary to form a deep wound from the side. For this reason, if the scratch is formed by scribing from the p-contact layer 59 side as described above, there is a possibility that a part of the semiconductor laser device is damaged or damaged.

By the way, the depth from the upper surface of the p-contact layer 59 to the MQW active layer 54 is usually 1 μm.
Or less than 1 μm. For this reason, as described above, if a deep flaw is formed by applying a large pressure from the p-contact layer 59 side and performing scribe,
-The MQW active layer 54 formed relatively close to the contact layer 59 may be significantly damaged. When the MQW active layer 54 is damaged as described above, device characteristics of the semiconductor laser device deteriorate.

An object of the present invention is to provide a nitride semiconductor device capable of dividing a substrate and a nitride semiconductor layer formed on the substrate without deteriorating device characteristics, and a method of manufacturing the same. is there.

[0016]

Means for Solving the Problems and Effects of the Invention A nitride-based semiconductor device according to the present invention comprises a first conductive type first nitride-based semiconductor layer on one surface of a conductive substrate. An element region and a nitride semiconductor element having a second conductivity type second nitride semiconductor layer formed thereon, wherein the substrate is exposed such that a predetermined region along at least one side of the other surface of the substrate is exposed. A first ohmic electrode is formed on the other surface, and a second ohmic electrode is formed on the second nitride-based semiconductor layer.

The active element region of the nitride-based semiconductor device in this case is, for example, an active layer or a light-emitting layer of a light-emitting diode device or a semiconductor laser device, a core layer of a waveguide device,
It corresponds to an I (intrinsic; intrinsic) layer of an N photodiode, a pn junction of a photodiode or an HBT (heterojunction bipolar transistor), a channel of an FET (field effect transistor), and the like.

In the nitride semiconductor device according to the present invention, a first ohmic electrode is formed on the other surface of the substrate so that a predetermined region along at least one side of the other surface of the substrate is exposed. . When manufacturing such a nitride-based semiconductor device, the substrate is divided along with the first and second nitride-based semiconductor layers and the active device region in a predetermined region of the substrate exposed between the first ohmic electrodes. As described above, a predetermined region of the substrate exposed between the first ohmic electrodes can be used as a division mark when the nitride-based semiconductor device is manufactured.

As described above, at the time of manufacturing the nitride semiconductor device according to the present invention, the substrate is divided from the other surface side of the substrate together with the first and second nitride semiconductor layers and the active device region. . Here, in the nitride-based semiconductor device, the thickness of the substrate is thicker than the first nitride-based semiconductor layer, and the active element region is formed at a position relatively far from the other surface of the substrate. Therefore, when the division is performed from the other surface side of the substrate, even if a part of the element is damaged or damaged by the division, the active element region is not adversely affected. Therefore, good device characteristics can be realized in the nitride-based semiconductor device.

[0020] The second ohmic electrode may be formed on the entire upper surface of the second nitride-based semiconductor layer. In this case, the contact area between the second ohmic electrode and the second nitride-based semiconductor layer is increased, so that the contact resistance between the second ohmic electrode and the second nitride-based semiconductor layer is reduced. Can be. Therefore, in such a nitride-based semiconductor device, the operating voltage can be reduced, and the amount of heat generated during operation can be reduced. Thereby,
The life of the element is prolonged.

[0021] The second ohmic electrode may be formed on the second nitride-based semiconductor layer such that a predetermined region along at least one side of the second nitride-based semiconductor layer is exposed.
Further, the first and second ohmic electrodes may have the same shape, and the second ohmic electrode may be formed in a region of the second nitride-based semiconductor layer above the first ohmic electrode.

[0022] The substrate may be transparent. In this case, the second ohmic electrode formed on the second nitride-based semiconductor layer can be transmitted through the substrate and observed from the other surface of the substrate. Therefore, when such an element is mounted junction-down with the substrate side up, it is possible to position the element while observing the shape of the second ohmic electrode. As described above, when assembling a device by mounting the above-described nitride-based semiconductor element, assembly accuracy and assembly yield are improved.

In the case where the second ohmic electrode is formed in the same shape as the first ohmic electrode, the second ohmic electrode is formed in accordance with the shape of the first ohmic electrode which can be observed through the substrate. The electrodes can be patterned.

The substrate may be made of gallium nitride, and the substrate may be made of silicon carbide. First
The second nitride-based semiconductor layer may include at least one of gallium, aluminum, indium, boron, and thallium.

According to the method of manufacturing a nitride semiconductor device of the present invention, a first nitride semiconductor layer of a first conductivity type, an active device region, and a second conductivity type are formed on one surface of a substrate having conductivity. Forming a second nitride-based semiconductor layer in order, and forming, on the other surface of the substrate, a first ohmic electrode patterned so as to expose at least divided regions arranged substantially in parallel at predetermined intervals. Forming, forming a second ohmic electrode on the second nitride-based semiconductor layer, and dividing the substrate along the divided region together with the first and second nitride-based semiconductor layers and the active element region And a step of performing

In the method for manufacturing a nitride-based semiconductor device according to the present invention, the first substrate is provided on one surface of a conductive substrate.
A conductive type first nitride-based semiconductor layer, an active element region, and a second conductive type second nitride-based semiconductor layer are sequentially formed;
On the other surface of the substrate, a first ohmic electrode patterned so as to expose at least divided regions arranged substantially in parallel at a predetermined interval is formed, and a first ohmic electrode is formed on the second nitride-based semiconductor layer. Two ohmic electrodes are formed.

As described above, the substrate is divided from the other surface side of the substrate together with the first and second nitride-based semiconductor layers and the active element region by using the divided region exposed between the first ohmic electrodes as a mark. it can. Here, the thickness of the substrate is larger than that of the first nitride-based semiconductor layer, and the active element region is formed at a position relatively far from the other surface of the substrate. In this case, even if a part of the element is damaged or damaged due to the division, the active element region is not adversely affected. Therefore, good device characteristics can be realized in the nitride-based semiconductor device.

The step of forming the second ohmic electrode includes:
Forming a second ohmic electrode patterned on the second nitride-based semiconductor layer such that at least a divided region arranged substantially parallel to the second nitride-based semiconductor layer at a predetermined interval is exposed. May be.

The substrate may be transparent. In this case, the second ohmic electrode formed on the second nitride-based semiconductor layer can be transmitted through the substrate and observed from the other surface of the substrate. Therefore, when such an element is mounted junction-down with the substrate side up, it is possible to position the element while observing the shape of the second ohmic electrode. As described above, when assembling a device by mounting the above-described nitride-based semiconductor element, assembly accuracy and assembly yield are improved.

The step of dividing the substrate includes a step of scribing the divided area and a step of cleaving the substrate along with the first and second nitride-based semiconductor layers and the active element region along the scratch formed by the scribing. May be included.

In this case, the other surface of the substrate is scratched by scribing, and the substrate can be cleaved from the other surface side of the substrate together with the first and second nitride-based semiconductor layers and the active element region. Elements can be easily divided without adversely affecting the active element region.

[0032]

DESCRIPTION OF THE PREFERRED EMBODIMENTS A nitride semiconductor laser device will be described below as an example of a nitride semiconductor device according to the present invention.

FIG. 1 is a schematic perspective view showing a first example of a nitride-based semiconductor laser device according to the present invention.

As shown in FIG.
0 has a thickness of 1 on a transparent substrate 100 made of n-GaN.
n-cladding layer 2 of n-Al _a Ga _1-a N of μm
1. n-light guide layer 22 made of n-GaN having a thickness of 0.1 μm, MQW active layer 23 having a multiple quantum well (MQW) structure, and p-Al _b Ga _1-b N having a thickness of 0.02 μm P-cap layer 24, 0.1 μm thick p-Ga
_{P-Al c Ga 1-c} N consisting p- cladding layer 26 made of N p- light guide layer 25 and the thickness of 0.5μm is formed in this order. A stripe-shaped ridge portion 29 extending in the resonator length direction is formed in the p-clad layer 26, and a 0.35 μm-thick n-Al _d Ga is formed on a flat portion of the p-clad layer 26. _An n-current blocking layer 27 made of _1-dN is formed. The p-cladding layer 26 on the n-current blocking layer 27 and the ridge portion 29
A p-contact layer 28 made of p-GaN having a thickness of 0.1 μm is formed thereon.

In this embodiment, Si is used as an n-type dopant, and Mg is used as a p-type dopant.

The above-described semiconductor laser device 500 has a stripe structure. In the semiconductor laser device 500, the cavity end face A serves as a laser light emission face. A protective film is formed on the resonator end faces A and B.

In the above, the n-cladding layer 21 and the p-
The composition of AlGaN constituting each of the cap layer 24, the p-cladding layer 26, and the n-current blocking layer 27, that is, the values of a to d are 0 ≦ a <d, 0 ≦ c <d, and 0 ≤c <b. For example, in this case, a = 0.07, b = 0.25, and c =
0.07 and d = 0.12.

The MQW active layer 23 has a thickness of 8
A multiple quantum well (M) in which three quantum well layers made of In _0.13 Ga _0.87 N of 0 ° and four quantum barrier layers made of In _0.05 Ga _0.95 N having a thickness of 160 ° are alternately stacked.
QW) structure.

In the above-described semiconductor laser device 500, the n-electrode 1 is formed on the surface of the transparent substrate 100 opposite to the crystal growth layer side (hereinafter referred to as the back surface). The p-electrode 2 is formed on the p-contact layer 28. Hereinafter, details of the n-electrode 1 and the p-electrode 2 will be described.

FIG. 2 is a schematic plan view when the semiconductor laser device 500 of FIG. 1 is observed from above and below. Here, FIG. 2A shows a case where the semiconductor laser device 500 is observed from the p-contact layer 28 side. FIG. 2B shows a semiconductor laser device 500.
Is observed from the back side of the transparent substrate 100. In the n-electrode 1 and the p-electrode 2, 2
The stripe region 29a surrounded by the dashed line indicates a region below and above the stripe-shaped ridge portion 29.

As shown in FIG. 2A, in the semiconductor laser device 500, the p-electrode 2 is formed on the entire upper surface of the p-contact layer 28. As described above, in the semiconductor laser device 500 of this example, the p-electrode 2 is not patterned.

On the other hand, as shown in FIG. 2B, on the back surface of the transparent substrate 100 of the semiconductor laser device 500, a stripe-shaped region along the side on the resonator end face A side and the side on the resonator end face B side The n-electrode 1 is formed in a region excluding a stripe-shaped region along.

As described above, the semiconductor laser device 50 of the present embodiment
At 0, the n-electrode 1 formed on the rear surface of the transparent substrate 100 is patterned. In this case, the transparent substrate 10 is formed in the stripe-shaped region on the side of the resonator end faces A and B.
0 is exposed.

The semiconductor laser device 500 of this embodiment is manufactured by the following method. When the semiconductor laser device 500 is manufactured, first, the layers 21 to 28 are formed on one surface of the transparent substrate 100. Thus, the transparent substrate 1
A semiconductor wafer 600 (see FIG. 3 described later) in which the respective layers 21 to 28 are formed on the substrate 00 is manufactured. Thereafter, an n-electrode 1 patterned into a predetermined shape is formed on the back surface of the transparent substrate 100 of the semiconductor wafer 600, and a p-type electrode is formed.
The p electrode 2 is formed on the entire upper surface of the contact layer 28.

FIG. 3 is a schematic plan view when the semiconductor wafer 600 on which the n-electrode 1 is formed in a predetermined shape is observed from the back side of the transparent substrate 100.

As shown in FIG. 3, a plurality of stripe-shaped n
The electrodes 1 are arranged substantially in parallel at predetermined intervals, and each n
In a striped region between the electrodes 1, the transparent substrate 10
0 is exposed.

The transparent substrate 10 exposed between the n-electrodes 1
A broken line 30 shown in the center of the region 0 is a dividing line of the semiconductor wafer 600 in a later-described resonator manufacturing process. In this case, the transparent substrate 10 exposed between the n-electrodes 1
The region of 0 corresponds to the divided region.

In this embodiment, the n-electrode 1 is patterned by, for example, an etching method or a lift-off method.
Here, for example, the n-electrode 1 in this case is formed by stacking a Ti film, an Al film and a Ti film in this order, by stacking a Ti film and an Al film in this order, or by forming a Ni film and a Ti film. And an Au film are laminated in this order, or a Ti film, a P film
The t film and the Au film are laminated in this order.

On the other hand, in the step of forming the p-electrode 2, the p-electrode 2 is formed on the entire upper surface of the p-contact layer 28 without performing patterning. For example, the p electrode 2 in this case is
A Ni film, a Ti film, a Pt film, and an Au film laminated in this order; a Pd film, a Pt film, and an Au film laminated in this order; a Ni film, an Au film, and a Ti film And the Au film are stacked in this order, the Ni film and the Au film are stacked in this order, or the Ni film, the Pt film, and the Au film are stacked in this order.

The material constituting the n-electrode 1 and p
The combination with the material constituting the electrode 2 is arbitrary. The order of forming the n-electrode 1 and the p-electrode 2 is arbitrary.

As described above, the n-electrode 1 and the p-electrode 2
Is formed, each region of the transparent substrate 100 exposed between the striped n-electrodes 1 is separated from the transparent substrate 100 side by a broken line 3.
Scribing along 0 creates a flaw. Then, the semiconductor wafer 600 is cleaved along the scratch. Thus, the semiconductor wafer 600a is divided into a plurality of stripe-shaped semiconductor wafers 600a as shown in FIG. 3B, and the resonator end faces A and B are exposed to manufacture a resonator. Note that FIG.
A broken line 31 in (b) is an element separation line in an element separation step described later.

As described above, in the present embodiment, the patterned n-electrode 1 and the area of the transparent substrate 100 exposed between the n-electrodes 1 are used as marks for a portion to be scribed.

As described above, in the resonator manufacturing process of this embodiment, scribing is performed from the back surface side of the transparent substrate 100 to form a flaw. Here, in this case, the distance from the back surface of the transparent substrate 100 to the MQW active layer 23 is about 50 to 250 μm, and the MQW active layer 23 is formed at a position relatively far from the back surface of the transparent substrate 100. I have. For this reason, even if scribe is performed by applying a large pressure from the back side of the transparent substrate 100, the transparent substrate 10
0, the MQW active layer 23
There is no danger of large defects or damages occurring. Also,
The defect or damage of the element caused by the scribe
There is no possibility that the W active layer 23 will be adversely affected.

After manufacturing the resonator as described above, the semiconductor wafer 600a in the form of a stripe is further divided by cleavage along the broken line 31 in a direction parallel to the resonator length direction, as shown in FIG. . Thus, the semiconductor laser elements 500 are separated.

Here, in the above-described element isolation step,
The semiconductor wafer 600 is scribed from the n-electrode 1 side.
a may be cleaved or the semiconductor wafer 600a may be cleaved by scribing from the p-electrode 2 side. p
When performing scribe from the electrode 2 side, the MQW
Although there is a possibility that a defect or damage may occur on the side surface of the active layer 23 and a region in the vicinity thereof, the defect or damage generated on the side surface of the MQW active layer 23 and a region in the vicinity thereof hardly affects the element characteristics.

As described above, the above semiconductor laser device 5
In the manufacturing method of No. 00, a scratch is formed by scribing from the back surface side of the transparent substrate 100 which is located relatively far from the MQW active layer 23. For this reason, the MQW active layer 23 does not suffer from a defect or damage due to the scribe, and the MQW active layer 23 does not have an adverse effect even if a part of the element is lost or damaged due to the scribe. Therefore, in the semiconductor laser device 500 manufactured by such a method, good device characteristics can be obtained.

In the semiconductor laser device 500, the p electrode 2 is formed on the entire upper surface of the p-contact layer 28. Therefore, the p-electrode 2 and the p-contact layer 2
8 has a large contact area. Therefore, in the semiconductor laser device 500, the p-electrode 2 and the p-contact layer 28
And the contact resistance between them is reduced, and a good ohmic contact can be obtained. Thus, the operating voltage can be reduced, and the amount of heat generated during operation can be reduced. As a result, in the semiconductor laser element 500, deterioration of the element is prevented, and a longer life can be realized.

FIG. 4 is a schematic plan view showing a second example of the semiconductor laser device according to the present invention. FIG. 4 (a)
The semiconductor laser element 501 of this example is replaced with the p-contact layer 28.
FIG. 4B is a diagram when the semiconductor laser element 501 of the present example is observed from the back surface side of the transparent substrate 100.

The semiconductor laser device 501 of this embodiment has the same structure as the semiconductor laser device 500 of FIG. 1, except for the following points.

As shown in FIG. 4A, in the semiconductor laser device 501, the vicinity of both ends of the stripe region 29a of the p-contact layer 28 and the stripe region 29
The p-electrode 4 is formed on a predetermined rectangular area including a. In this case, the rectangular predetermined area is
This is a region excluding a stripe-shaped region along the sides on the resonator end faces A and B side and a stripe-shaped region along the sides on the side surface of the element.

In the semiconductor laser device 501 of this example, since the p-electrode 4 is formed so as to cover the entire stripe region 29a, the stripe region 2
The current can be uniformly injected into the entire portion 9a.

The semiconductor laser device 501 of this example is manufactured by the same method as the manufacturing method of the semiconductor laser device 500 except for the following points.

At the time of manufacturing the semiconductor laser device 501, after the layers 21 to 28 are grown on the transparent substrate 100, the n-electrode 3 is patterned by the same method as when forming the n-electrode 1 of the semiconductor laser device 500. I do. And
The p-electrode 4 is further patterned in accordance with the position of the pattern of the n-electrode 3.

Here, the pattern of the n-electrode 3 formed on the back side of the transparent substrate 100 is
1 to 28 can be seen through. Therefore, by patterning the p-electrode 4 while observing the pattern of the n-electrode 3 that can be seen through, it is possible to easily perform alignment between the n-electrode 3 and the p-electrode 4.

After patterning the p-electrode 4 as described above, the patterned n-electrode 3 is patterned in the same manner as in the case of the semiconductor laser device 500.
A scratch is formed by scribing along a stripe-shaped n-electrode 3 in a predetermined region of the transparent substrate 100 exposed between them.
Further, the semiconductor wafer is divided along the scratches to expose the cavity end faces A and B, thereby producing the cavity.

Here, in the semiconductor laser element 501 of this example, the p-electrode 4 is not formed in the region near the cavity end surfaces A and B except for the end of the stripe region 29a.
In this region, p-contact layer 28 is exposed. Therefore, in this case, the cleavage can be performed at the portion of the p-contact layer 28 exposed between the p electrodes 4 except for the end of the stripe region 29a. Therefore, it is not necessary to divide the p-electrode 4 over a long distance. For this reason, in this case, it is possible to manufacture the resonator by easily performing cleavage. Further, in this case, there is no possibility that the p-electrode 4 covers the resonator end faces A and B.

After the resonator is manufactured as described above, the stripe-shaped semiconductor wafer is further cleaved along the direction parallel to the resonator length direction and divided, and separated into individual semiconductor laser elements 501.

In the element isolation step, scribing may be performed from the back side of the transparent substrate 100, or scribing may be performed from the p electrode 4 side. p electrode 4
When scribing is performed from the side, both ends (that is, both sides of the element) of the MQW active layer 23 in a direction perpendicular to the resonator length direction and regions near the element may be damaged or damaged. The defects and damages occurring on the side surfaces have less influence on the device characteristics as compared with the defects and damages occurring on the resonator end face of the device.

Here, in this case, the p-electrode 4 is not formed in the striped area near the resonator end faces A and B, and the p-contact layer 28 is exposed in this area. Therefore, the exposed p-contact layer 2
8 can be cleaved. Therefore, since it is not necessary to divide the p-electrode 6, it is possible to easily cleave and perform element isolation.

As described above, the semiconductor laser device 5 of this embodiment
01, n formed on the back surface of the transparent substrate 100
The electrode 3 is patterned, and using the pattern of the n-electrode 3 as a mark, the transparent substrate 100 exposed between the n-electrodes 3 from the back surface side of the transparent substrate 100 and the layer above the transparent substrate 100 are scribed to form scratches. Then, the semiconductor wafer is cleaved along the scratches to produce a resonator. For this reason, even if a part of the element is lost or damaged when the scratch is formed by the scribe, the MQW active layer 23 at a position relatively far from the rear surface of the scratched transparent substrate 100 is not adversely affected. , MQW active layer 23 have good cavity end faces A and B. Therefore, in the semiconductor laser device 501, good device characteristics can be obtained.

When the semiconductor laser element 501 is mounted junction-down with the transparent substrate 100 facing upward, since the transparent substrate 100 is transparent, the end of the stripe region 29a on the p-contact layer 28 side is used. It is possible to assemble the semiconductor laser device by accurately positioning the sub-mount or the like using the part as a mark. Therefore, in the semiconductor laser device using such a semiconductor laser element 501, the position of the light emitting point formed on the MQW active layer 23 can be controlled with high accuracy. Therefore, the assembling accuracy of the semiconductor laser device is improved, and the assembling yield is improved.

The semiconductor laser device 501 including the transparent substrate 100 made of GaN is usually assembled in a junction-down state because the substrate itself has low heat radiation. Therefore, in this case, the above effect is effective.

FIG. 5 is a schematic plan view showing a third example of the semiconductor laser device according to the present invention. FIG. 5 (a)
The semiconductor laser element 502 of this example is replaced with the p-contact layer 28.
FIG. 5B is a diagram when the semiconductor laser device 502 of the present example is observed from the back surface side of the transparent substrate 100.

The semiconductor laser device 502 of this embodiment has the same structure as the semiconductor laser device 500 of FIG. 1 except for the following points.

As shown in FIG. 5B, in the semiconductor laser device 502, a stripe-shaped region along the side of the cavity end faces A and B on the back surface of the transparent substrate 100 and both side surfaces of the device The n-electrode 5 is formed in a region excluding a stripe-shaped region along the side of.

The semiconductor laser device 502 is manufactured by the same method as the method of manufacturing the semiconductor laser device 500 except for the following points.

At the time of manufacturing the semiconductor laser element 502, the n-electrode 5 is patterned into a stripe by the same patterning method as the n-electrode 1 of the semiconductor laser element 500, and a stripe-shaped region exposed between the n-electrodes 5 is further formed. Transparent substrate 1 in a stripe shape in a direction orthogonal to
The n-electrode 5 is patterned so that 00 is exposed. Thereby, a plurality of rectangular n-electrodes 5 are formed on the back surface of the transparent substrate 100 of the semiconductor wafer, and the transparent substrate 100 is exposed in a lattice pattern between the n-electrodes 5.

In the fabrication of the semiconductor laser element 502 of this example, the n-electrode 5 patterned as described above and the n-electrode 5 are exposed during the formation of a scribed flaw in the resonator fabrication process and the device isolation process. The transparent substrate 100 thus obtained is used as a mark for the formation position of the scratch.

That is, after patterning the n-electrode 5 as described above, first, the transparent substrate 100 exposed in a grid pattern between the individual patterned n-electrodes 5 is formed in the same manner as in the case of the semiconductor laser device 500. A scribe is made along one of the stripe-shaped regions in two directions orthogonal to each other to form a flaw. Further, the semiconductor wafer is divided into stripes along the scratches to expose the resonator end faces A and B, thereby manufacturing a resonator. Further, a flaw is formed in a region of the transparent substrate 100 exposed between the side surfaces of the adjacent n-electrode 5 along a direction parallel to the cavity length direction. Then, the semiconductor wafer is divided along the scratches and separated into individual semiconductor laser elements 502.

Here, in the above-described element isolation step,
Cleavage can be performed at the exposed region of the transparent substrate 100. Therefore, in this case, it is not necessary to divide the n-electrode 5, and it is possible to easily cleave and perform element isolation.

As described above, the semiconductor laser device 5 of this embodiment
02, n formed on the back surface of the transparent substrate 100
The electrode 5 is patterned, and using the pattern of the n-electrode 5 as a mark, the transparent substrate 100 exposed between the n-electrodes 6 from the rear surface side of the transparent substrate 100 and the layer above the transparent substrate 100 are scribed to form scratches. Then, the semiconductor wafer is cleaved along the scratches to manufacture a resonator and perform element isolation.

In such a semiconductor laser device 502, since the MQW active layer 23 is located at a position relatively far from the back surface of the scratched transparent substrate 100, part of the device is lost when the scratch is formed by scribing. Alternatively, even if the MQW active layer 23 is damaged, the MQW active layer 23 is not adversely affected.
Resonator end faces A and B of the active layer 23 are good end faces. Therefore, in such a semiconductor laser device 502, good device characteristics can be obtained.

In the semiconductor laser device 502, the p-electrode 6 is formed on the entire upper surface of the p-contact layer 28. Therefore, the p-electrode 6 and the p-contact layer 2
8 has a large contact area. Therefore, in the semiconductor laser device 502, the p-electrode 6 and the p-contact layer 28
Is reduced, and a good ohmic contact between p electrode 6 and p-contact layer 28 can be obtained. Thus, the operating voltage can be reduced, and the amount of heat generated during operation can be reduced. As a result, in the semiconductor laser element 502, deterioration of the element is prevented, and a longer life can be realized.

By the way, as described above, the transparent substrate 100
Since GaN or SiC is a very hard material, it is necessary to apply a large pressure to the transparent substrate 100 to form a flaw when scribing. For this reason, such scribing tends to cause a defect, particularly on the side surface of the element (the end surface in the direction perpendicular to the resonator length direction), and the side surface tends to be uneven.

In particular, when cleavage is performed on the (1-100) plane, which is the main cleavage plane of the GaN substrate and the GaN nitride-based semiconductor layer, and this plane is used as the resonator end face, the side face of the formed element is used. Are (11-20) planes, so that irregularities are easily formed on the side surfaces. In this case, if the n-electrode is formed on the entire back surface of the substrate, both sides of the n-electrode also have an uneven shape.

Normally, when a semiconductor laser device is mounted junction-down, the side surface of the semiconductor laser device is recognized by an image recognition technique, so that the semiconductor laser device is positioned on a submount or the like with reference to the side surface. However, when the side surface of such a semiconductor laser device has an uneven shape, it is difficult to accurately position the semiconductor laser device based on the side surface. Therefore, the assembly yield of the semiconductor laser device using the semiconductor laser element is reduced.

On the other hand, the semiconductor laser device 5
In No. 02, since the n-electrode 5 is formed excluding the stripe-shaped regions on both sides of the transparent substrate 100, n
The side of the electrode 5 becomes linear. Therefore, when the semiconductor laser element 502 is mounted junction-down, not the side surface of the element but the side of the n-electrode 5 is recognized by the image recognition technology, and the semiconductor laser element 502 is mounted on the n-electrode 5.
Can be accurately positioned on the sub-mount or the like with reference to the side of. As a result, the assembly accuracy and the assembly yield of the semiconductor laser device using the semiconductor laser element are improved.

FIG. 6 is a schematic plan view showing a fourth example of the semiconductor laser device according to the present invention. FIG. 6 (a)
FIG. 6B is a diagram when the semiconductor laser element 503 of this example is observed from the p-GaN contact layer 28 side, and FIG.
FIG. 4 is a diagram when the semiconductor laser element 503 of the present example is observed from the back side of the transparent substrate 100.

The semiconductor laser device 503 of this embodiment is
Except for the following points, it has the same structure as the semiconductor laser device 502 of FIG.

As shown in FIG. 6A, in the semiconductor laser device 503, the p-contact layer 28 has a p-electrode on a predetermined region having a rectangular shape including the region near both ends of the stripe region 29a and the stripe region 29a. 8 are formed. In this case, the rectangular predetermined region is a region excluding a stripe region along the sides on the resonator end faces A and B side and a stripe region along the sides on both side surfaces of the element. .

In the semiconductor laser device 503 of this example, since the p-electrode 8 is formed so as to cover the entire stripe region 29a, the stripe region 2
A current can be uniformly applied to the entire surface 9a.

The semiconductor laser device 503 of this example is manufactured by the same method as the manufacturing method of the semiconductor laser device 502 except for the following points.

In manufacturing the semiconductor laser element 503, the p-electrode 8 is formed by the same method as the p-electrode 4 of the semiconductor laser element 501 of FIG. 4, and the same as the n-electrode 5 of the semiconductor laser element 502 of FIG. The n-electrode 7 is formed by the method. Here, since the transparent substrate 100 and each of the layers 21 to 28 are transparent, the n formed on the rear surface side of the transparent substrate 100
The pattern of the electrode 7 is changed to the transparent substrate 100 and the respective layers 21 to 2.
8 can be seen through. Therefore, by patterning the p-electrode 8 while looking at the pattern of the n-electrode 7 that can be seen through, it is possible to easily perform the alignment between the n-electrode 7 and the p-electrode 8.

After the p-electrode 8 is patterned as described above, a predetermined area of the transparent substrate 100 exposed between the patterned n-electrodes 7 is scribed in the same manner as in the case of the semiconductor laser element 502. And a semiconductor wafer is divided along the scratch.

Here, in the semiconductor laser device 503 of this example, except for the end of the stripe region 29a, the p-electrode 8 is not formed in the stripe region on the side of the resonator end surfaces A and B. The p-contact layer 28
Is exposed. Therefore, except for the end of the stripe region 29a, the cleavage can be performed in the region of the p-contact layer 28 exposed between the p electrodes 8. Therefore, since it is not necessary to divide the p-electrode 8 over a long distance, it is possible to easily cleave and manufacture a resonator.
Further, in this case, there is no possibility that the p-electrode 8 covers the resonator end faces A and B.

After the resonator is manufactured as described above, the adjacent n-electrode 7 is formed as in the case of the semiconductor laser element 502.
The area of the transparent substrate 100 exposed between the side surfaces is scribed along a direction parallel to the cavity length direction to form a flaw. Then, the semiconductor wafer is divided along the scratches and separated into individual semiconductor laser elements 503.

Here, in the above-described element isolation step,
Cleavage can be performed at the exposed region of the transparent substrate 100. Therefore, since there is no need to divide the n-electrode 7, it is possible to easily cleave and perform element isolation.

As described above, the semiconductor laser device 5
03, n formed on the back surface of the transparent substrate 100
The electrode 7 is patterned, and using the pattern of the n-electrode 7 as a mark, the transparent substrate 100 exposed between the n-electrodes 7 from the back side of the transparent substrate 100 and a layer thereabove are scribed to form scratches. Then, the semiconductor wafer is cleaved along the scratches to manufacture a resonator and perform element isolation.
Here, in the semiconductor laser element 503, since the MQW active layer 23 is located at a position relatively far from the back surface of the transparent substrate 100 in which the flaw is formed, a part of the element is damaged or damaged when the flaw is formed by the scribe. Even MQW active layer 2
3 is not adversely affected, and the cavity end faces A and B of the MQW active layer 23 are excellent end faces. Therefore, in such a semiconductor laser device 503, good device characteristics can be obtained.

When the semiconductor laser element 503 is mounted junction-down, since the transparent substrate 100 is transparent, the end of the stripe region 29a on the p-contact layer 28 side is used as a mark. It is possible to accurately assemble a semiconductor laser device that positions the element 503 on a submount or the like. Therefore, in a semiconductor laser device using such a semiconductor laser element 503, the position of a light emitting point formed on the MQW active layer 23 can be controlled with high accuracy.
Therefore, the assembling accuracy of the semiconductor laser device is improved, and the assembling yield is improved.

The semiconductor laser device 503 provided with the transparent substrate 100 made of GaN is usually assembled in a junction-down state because the substrate itself has low heat radiation. Therefore, in this case, the above effects can be obtained effectively.

Further, in the above-described semiconductor laser device 503, as in the case of the semiconductor laser device 502 in FIG.
The side of the electrode 7 becomes linear. Therefore, when such a semiconductor laser element 503 is mounted junction-down, not the side face of the element but the side of the n-electrode 7 is recognized by the image recognition technology, and the semiconductor laser element 503 is recognized.
Can be accurately positioned on a submount or the like with reference to the side of the n-electrode 7. As a result, the assembly accuracy and the assembly yield of the semiconductor laser device using the semiconductor laser element 503 are improved.

FIG. 7 is a schematic plan view showing a fifth example of the semiconductor laser device according to the present invention. FIG. 7 (a)
FIG. 7B is a diagram when the semiconductor laser element 504 of the present example is observed from the p-GaN contact layer 28 side.
FIG. 4 is a diagram when the semiconductor laser element 504 of the present example is observed from the back side of the transparent substrate 100.

The semiconductor laser device 504 of this embodiment is similar to that of FIG.
As shown in (b), the stripe region 2 of the transparent substrate 100
The semiconductor laser device 502 has the same structure as the semiconductor laser device 502 of FIG. 5 except that the n-electrode 9 is also formed on the region near both ends of the semiconductor laser device 9a.

As shown in FIG. 7B, in the semiconductor laser device 504 of this example, the transparent substrate 100 and the p-type
An n-electrode 9 and a p-electrode 10 are formed so as to cover the entire stripe region 29a of the contact layer 28.
Therefore, in such a semiconductor laser element 504, a current can be uniformly injected into the entire stripe region 29a.

The semiconductor laser device 504 of this example is manufactured by the same method as the manufacturing method of the semiconductor laser device 502 except that the method of patterning the n-electrode 9 is different.

In the above-described semiconductor laser device 504, the n-electrode 9 formed on the back surface of the transparent substrate 100 is patterned, and the pattern of the n-electrode 9 is used as a mark between the n-electrode 9 from the back surface of the transparent substrate 100. The transparent substrate 100 and the layer above the transparent substrate 100 are scribed to form scratches. Then, the semiconductor wafer is cleaved along the scratches to form a resonator and perform element isolation.

Here, such a semiconductor laser device 50
In No. 4, since the MQW active layer 23 is located relatively far from the side of the transparent substrate 100 where the flaw is formed, the MQW active layer 23 has an adverse effect even if a part of the element is lost or damaged when the flaw is formed by the scribe. Not receive, MQ
Resonator end faces A and B of the W active layer 23 are good end faces.
Therefore, in such a semiconductor laser device 504, good device characteristics can be obtained.

In the semiconductor laser device 504, the p-electrode 10 is formed on the entire upper surface of the p-contact layer 28. Therefore, the contact area between p electrode 10 and p-contact layer 28 is large. Therefore, in the semiconductor laser element 504, the contact resistance between the p-electrode 10 and the p-contact layer 28 is reduced, and a good ohmic contact can be obtained. Thus, the operating voltage can be reduced, and the amount of heat generated during operation can be reduced. As a result, in the semiconductor laser element 504, deterioration of the element is prevented, and a longer life can be realized.

Further, in the above-described semiconductor laser element 504, the sides of the n-electrode 9 are linear as in the case of the semiconductor laser element 502 of FIG. 5 and the semiconductor laser element 503 of FIG. Therefore, such a semiconductor laser element 5
When mounting the semiconductor device 04 by junction down, the side of the n-electrode 9 is recognized by the image recognition technology instead of the side surface of the device, and the semiconductor laser device 504 is accurately mounted on a submount or the like with the side of the n-electrode 9 as a reference. Can be positioned. As a result, the assembling accuracy and the assembling yield of the semiconductor laser device using the semiconductor laser element 504 are improved.

FIG. 8 is a schematic plan view showing a sixth example of the semiconductor laser device according to the present invention. FIG. 8 (a)
The semiconductor laser element 505 is connected to the p-GaN contact layer 28.
FIG. 8B is a diagram when viewed from the side, and FIG. 8B is a diagram when the semiconductor laser element 505 is viewed from the back surface side of the transparent substrate 100.

The semiconductor laser device 505 of this example has the same structure as the semiconductor laser device 504 of FIG. 7, except for the following points.

As shown in FIG. 8A, in the semiconductor laser device 505, a p-electrode 12 having the same shape as the n-electrode 11 is formed. That is, the p-electrode 12 is formed in the vicinity of both ends of the stripe region 29a on the p-contact layer 28 and on a predetermined rectangular region including the stripe region 29a. In this case, the rectangular predetermined region is a region excluding a stripe region along the sides on the resonator end faces A and B side and a stripe region along the sides on both side surfaces of the element. is there.

As described above, the semiconductor laser device 50 of this embodiment is
In 5, the n-electrode 11 and the p-electrode 12 are formed so as to cover the transparent substrate 100 and the entire stripe region 29a of the p-contact layer 28. Therefore, in such a semiconductor laser element 505, a current can be uniformly injected into the entire stripe region 29a.

The semiconductor laser device 505 of this example is manufactured by the same method as the manufacturing method of the semiconductor laser device 504 except that the method of patterning the p-electrode 12 is different.

That is, at the time of manufacturing the semiconductor laser device 505, after patterning the n-electrode 11,
The p-electrode 12 is further patterned according to the pattern position of the n-electrode 11. Here, the transparent substrate 100
And since the layers 21 to 28 are transparent, the transparent substrate 10
The pattern of the n-electrode 11 formed on the back surface side of 0 can be seen through the transparent substrate 100 and the layers 21 to 28. Therefore, by patterning the p-electrode 12 while observing the pattern of the n-electrode 11 that can be seen through, it is possible to easily perform the alignment between the n-electrode 11 and the p-electrode 12.

In the semiconductor laser device 505 of this example, the p-electrode 12 is not formed in the stripe-shaped region on the side of the cavity end faces A and B except for the end of the stripe region 29a. The contact layer 28 is exposed; For this reason, in the resonator manufacturing process of the semiconductor laser element 505, the semiconductor laser device 505 can be cleaved at the portion of the p-contact layer 28 exposed between the p-electrodes 12. Therefore, in this case, since it is not necessary to divide the p-electrode 12 over a long distance, it is possible to easily cleave and manufacture a resonator. Further, in this case, there is no possibility that the p-electrode 12 will cover the resonator end faces A and B.

On the other hand, in the element isolation step of the semiconductor laser element 505, since the cleavage can be performed at the exposed portions of the transparent substrate 100 and the p-contact layer 28, n
It is not necessary to divide electrode 11 and p-electrode 12. For this reason, it is possible to easily perform cleavage and perform element isolation.

As described above, in the semiconductor laser device 505, the n-electrode 11 formed on the back surface of the transparent substrate 100
Is patterned, and using the pattern of the n-electrode 11 as a mark, the transparent substrate 100 exposed between the n-electrodes 11 from the rear surface side of the transparent substrate 100 and a layer thereabove are scribed to form scratches. Then, the semiconductor wafer is cleaved along the flaw to produce resonance and perform element isolation.

Here, the MQW of the semiconductor laser device 505
Since the active layer 23 is located relatively far away from the back surface of the scratched transparent substrate 100, the MQW active layer 23 is adversely affected even if a part of the element is lost or damaged when the scratch is formed by scribe. However, the cavity end faces A and B of the MQW active layer 23 are excellent end faces. Therefore, in such a semiconductor laser device 505, good device characteristics can be obtained.

In the case where the semiconductor laser element 505 is mounted in a junction-down state, since the transparent substrate 100 and each of the layers 21 to 28 are transparent, the p-type
Using the end of the stripe region 29a on the contact layer 28 side as a mark, the semiconductor laser device 505 can be accurately positioned on a submount or the like to assemble a semiconductor laser device.

Therefore, such a semiconductor laser device 5
In the semiconductor laser device using No. 01, the position of the light emitting point formed on the MQW active layer 23 can be controlled with high accuracy. Therefore, the assembling accuracy of the semiconductor laser device is improved, and the assembling yield is improved.

In the semiconductor laser device 505 provided with the transparent substrate 100 made of GaN, since the substrate itself has low heat dissipation, it is usually assembled in a junction-down manner. Therefore, in this case, the above effects can be obtained effectively.

Further, in the above-described semiconductor laser device 505, the semiconductor laser device 502 of FIG. 5, the semiconductor laser device 503 of FIG. 6, and the semiconductor laser device 50 of FIG.
Similarly to 4, the sides of the n-electrode 11 are linear. Therefore, when such a semiconductor laser element 505 is mounted junction-down, the side of the n-electrode 11 is recognized by the image recognition technology instead of the side of the element, and the side of the n-electrode 11 is recognized by the image recognition technology. It can be accurately positioned on a submount or the like as a reference. As a result, the assembly accuracy and the assembly yield of the semiconductor laser device using the semiconductor laser element 505 are improved.

In the above-described semiconductor laser devices 500 to 505, the composition of each of the layers 21 to 28 is not limited to the above. Each of the layers 21 to 28 may be made of a nitride-based semiconductor containing at least one of Ga, Al, In, B and Tl.

In the above description, an n-type layer, an active layer, and a p-type layer are formed on the transparent substrate 100, and a patterned n-electrode is formed on the back surface of the transparent substrate 100. This n-electrode pattern is marked. The case where scribing is performed from the transparent substrate side by using as described above, a p-type layer, an active layer and an n-type layer are sequentially formed on the transparent substrate, and a patterned p-electrode is formed on the back surface of the transparent substrate. The scribe may be performed from the transparent substrate side using the pattern of the p-electrode as a mark.

Although the semiconductor laser device having the buried ridge stripe structure has been described above, the present invention is applied to a ridge stripe structure and a self-aligned structure employing an insulating film such as SiO ₂ as a current blocking layer. The present invention may be applied to a semiconductor laser device having another structure.

In the above, the case where the present invention is applied to a semiconductor laser device has been described. However, the present invention may be applied to a light emitting device other than a semiconductor laser device such as a light emitting diode. Further, the present invention can be applied to semiconductor devices other than light emitting devices.

For example, when the present invention is applied to a pn junction type light emitting diode, since the n-type layer of the light-emitting diode has a property of emitting light brighter than the p-type layer,
A junction-down assembly is performed by forming, for example, an n-electrode patterned in a lattice pattern on the back surface of the transparent substrate of the mold. As a result, light can be sufficiently extracted from the transparent substrate side, and a light emitting diode with high luminance is realized.

[Brief description of the drawings]

FIG. 1 shows a first example of a nitride-based semiconductor laser device according to the present invention.
It is a typical see-through perspective view which shows the example of.

FIG. 2 is a schematic plan view when the semiconductor laser device of FIG. 1 is observed from above and below.

FIG. 3 is a schematic plan view when a semiconductor wafer having an n-electrode formed in a predetermined shape is observed from the back surface side of a transparent substrate.

FIG. 4 is a schematic plan view showing a second example of the semiconductor laser device according to the present invention.

FIG. 5 is a schematic plan view showing a third example of the semiconductor laser device according to the present invention.

FIG. 6 is a schematic plan view showing a fourth example of the semiconductor laser device according to the present invention.

FIG. 7 is a schematic plan view showing a fifth example of the semiconductor laser device according to the present invention.

FIG. 8 is a schematic plan view showing a sixth example of the semiconductor laser device according to the present invention.

FIG. 9 is a schematic perspective view showing an example of a conventional nitride-based semiconductor laser device.

[Explanation of symbols]

1,3,5,7,9,11 n electrode 2,4,6,8,10,12 p electrode 21 n-cladding layer 22 n-light guide layer 23 MQW active layer 24 p-cap layer 25 p-light Guide layer 26 p-cladding layer 27 n-current blocking layer 28 p-contact layer 29 a stripe region 100 transparent substrate 500, 501, 502, 503, 504, 505 semiconductor laser element 600, 600a semiconductor wafer A, B cavity end face

Continued on the front page F term (reference) 5F041 AA37 AA41 CA33 CA34 CA40 CA76 CA77 CA93 DA04 5F073 AA09 AA13 AA45 AA61 AA74 CA07 CB22 DA30 DA32 DA34

Claims (12)

[Claims] A first substrate is provided on one surface of a conductive substrate.
A nitride-based semiconductor device comprising a conductive type first nitride-based semiconductor layer, an active element region, and a second conductive-type second nitride-based semiconductor layer formed on the other surface of the substrate. A first ohmic electrode is formed on the other surface of the substrate such that a predetermined region along at least one side is exposed, and a second ohmic electrode is formed on the second nitride-based semiconductor layer A nitride semiconductor device characterized by the above-mentioned. 2. The second ohmic electrode includes a second ohmic electrode,
2. The nitride semiconductor device according to claim 1, wherein the nitride semiconductor device is formed on the entire upper surface of the nitride semiconductor layer. 3. The second ohmic electrode includes a second ohmic electrode.
2. The nitride semiconductor device according to claim 1, wherein said nitride semiconductor device is formed on said second nitride semiconductor layer such that a predetermined region along at least one side of said nitride semiconductor layer is exposed. 4. The first and second ohmic electrodes have the same shape, and the second ohmic electrode is formed on a region above the first ohmic electrode in the second nitride-based semiconductor layer. 4. The method as claimed in claim 3, wherein
The nitride-based semiconductor device according to the above. 5. The nitride-based semiconductor device according to claim 1, wherein said substrate is transparent. 6. The nitride semiconductor device according to claim 1, wherein said substrate is made of gallium nitride. 7. The nitride-based semiconductor device according to claim 1, wherein said substrate is made of silicon carbide. 8. The semiconductor device according to claim 1, wherein the first and second nitride-based semiconductor layers include at least one of gallium, aluminum, indium, boron, and thallium. Nitride based semiconductor device. 9. A first substrate on one surface of a substrate having conductivity.
Forming a conductive type first nitride-based semiconductor layer, an active element region, and a second conductive type second nitride-based semiconductor layer in this order; and substantially parallel to the other surface of the substrate at a predetermined interval. Forming a first ohmic electrode patterned so as to expose at least a stripe-shaped divided region disposed on the substrate; and forming a second ohmic electrode on the second nitride-based semiconductor layer. Dividing the substrate along with the first and second nitride-based semiconductor layers and the active element region along the divided region. 10. The step of forming the second ohmic electrode includes patterning the second nitride-based semiconductor layer such that at least a stripe-shaped divided region arranged substantially in parallel at a predetermined interval is exposed. 10. The method according to claim 9, further comprising the step of forming the second ohmic electrode. 11. The method for manufacturing a nitride-based semiconductor device according to claim 9, wherein said substrate is transparent. 12. A step of dividing the substrate, wherein the step of scribing the divided area includes the step of scribing the substrate along the first and second nitride-based semiconductor layers and the active layer along the scratch formed by the scribing. 12. The method of manufacturing a nitride-based semiconductor device according to claim 9, further comprising a step of cleaving together with the device region.