JP3322179B2 - Gallium nitride based semiconductor light emitting device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a light emitting diode covering a wavelength range from blue to ultraviolet or a semiconductor laser diode and a gallium nitride based semiconductor light emitting device used for a light emitting device in the same wavelength range. The present invention relates to an excellent method for doping impurities in a gallium nitride-based semiconductor.

[0002]

2. Description of the Related Art A short-wavelength light-emitting element having a wavelength shorter than blue is expected as a light source for an optical disk capable of full-color display and high-density recording.
Research has been actively made using group III compound semiconductors and group III-V compound semiconductors such as SiC and GaN. In particular, blue light-emitting diodes and blue-violet laser diodes have recently been realized using GaN, GaInN, or the like, and light-emitting elements using gallium nitride-based semiconductors have attracted attention. As a method for depositing a gallium nitride based semiconductor crystal, a metal organic chemical vapor deposition method (M
The OVPE method and the molecular beam epitaxy method (MBE method) are generally used.

When manufacturing the above-mentioned semiconductor light emitting device, it is indispensable to manufacture p-type GaN and n-type GaN by controlling the conductivity type by doping impurities. Of these, p-type GaN is particularly difficult to produce, and low-resistance p-type GaN has conventionally been produced by performing heat treatment or electron beam irradiation after MOVPE crystal growth. Further, N is added to p-type GaN in order to inject a current into the semiconductor light emitting device.
An electrode is formed by depositing i or the like. At this time, a method of doping p-type GaN has been devised in order to reduce the contact resistance.

For example, Japanese Patent Application Laid-Open No. 8-97471 discloses
Is 500 ° of the outermost surface of p-type GaN in contact with the Ni electrode.
(Angstrom) region, high density (1 × 1
0e²⁰cm^-3From 1 × 10e^{twenty one}cm^-3) To Mg
Low contact resistance by providing a first contact layer
And increase the carrier concentration to increase the operating voltage of the device.
It is described that the pressure is reduced (FIG. 8A). I
However, here, the high doping concentration of Mg is increased.
If too high, the hole concentration will be rather high
Occurs on the active layer side of the first contact layer.
P-type GaN second core with lower Mg concentration than contact layer
A contact layer is formed. And this second contact
In order to increase the hole concentration, the Mg concentration of the
0e¹⁹cm ^-3From 5 × 10e²⁰cm^-3Dopi within range
Is considered to be desirable (FIG. 8B).
reference).

[0005]

However, when a semiconductor light emitting device was actually manufactured within the above-described conventional range, the following two problems occurred.

The first is a problem due to the diffusion of Mg.
If the p-type GaN second contact layer is actually doped with Mg by about 5 × 10e ²⁰ cm ^{−3 in} order to obtain the maximum hole concentration, the Mg atoms become the light emitting layer during the growth of the first and second contact layers. We found a phenomenon that diffused into the active layer. Mg diffused into the active layer formed a non-emission center, and as a result, the luminous efficiency of the device was reduced.

The second problem is the problem of the hole concentration. In order to avoid the diffusion of Mg into the active layer as described above, Mg
When the doping concentration of is less than 1 × 10e ²⁰ cm ⁻³ ,
This time the hole concentration has dropped.

In view of the above problems, the present invention reduces contact resistance without diffusing Mg into an active layer while using two contact layers, a first contact layer and a second contact layer. An object of the present invention is to provide a semiconductor element in which the hole concentration in a contact layer is maximized.

[0009]

In order to achieve the above object, a semiconductor light emitting device of the present invention has a thickness of 0.1 μm or more and 1.0 μm or less.
a first contact layer having a thickness equal to or less than m and in contact with the p-electrode; a second contact layer in contact with the first contact layer and formed on the active layer side of the first contact layer; The structure is such that the layer is doped with an acceptor impurity at a higher concentration than the second contact layer. According to this configuration, since the first contact layer has a sufficient thickness, diffusion of the acceptor impurity from the first contact layer to the second contact layer can be easily performed.

In the above configuration, it is preferable that the thickness of the first contact layer is 0.1 μm or more and 0.5 μm or less.

The first contact layer has an acceptor impurity concentration of 5 × 10 e ²⁰ cm ⁻³ or more, and the second contact layer has an acceptor impurity concentration of 1 × 10 e ²⁰ cm ⁻³ or more.
It is preferably less than × 10e ²⁰ cm ⁻³ . According to this configuration, the hole concentration of the contact layer can be increased while suppressing diffusion of the acceptor impurity into the active layer.

The semiconductor light emitting device of the present invention has a first contact layer in contact with the p-electrode, and a second contact layer in contact with the first contact layer and formed on the active layer side of the first contact layer. , And Si, O, S, Se, Te, C,
At least one element selected from Zn is simultaneously doped. According to this configuration, it is possible to prevent the hole concentration from decreasing even if the acceptor impurity is introduced at a high concentration.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present inventors have conducted intensive studies and as a result, have found that a GaN semiconductor light emitting device using two contact layers, a first contact layer and a second contact layer, has the above-mentioned structure.
By properly controlling the state of acceptor impurities such as Mg introduced into one contact layer, the contact resistance can be reduced without diffusing Mg into the active layer, and the hole concentration in the second contact layer can be maximized. Semiconductor device can be obtained. That is, the present invention considers the diffusion of the acceptor impurity from the first contact layer to the second contact layer and the diffusion of the acceptor impurity from the second contact layer to the active layer. By setting the concentration of the acceptor impurity, the hole concentration of the contact layer is increased while suppressing the diffusion of the acceptor impurity into the active layer. In the semiconductor light emitting device of the present invention, an acceptor impurity is introduced into the first contact layer at a higher concentration than the second contact layer. Therefore, hereinafter, a semiconductor light emitting device and a method for manufacturing the same according to an embodiment of the present invention will be described with reference to the drawings.

(Embodiment 1) FIG. 1 is a cross-sectional view of a semiconductor light emitting device according to Embodiment 1 of the present invention.
It is formed by crystal growth using a metal organic chemical vapor deposition (MOVPE) method. Therefore, the structure and the specific manufacturing method of the semiconductor light emitting device of the present invention will be sequentially described below with reference to FIG.

First, before the vapor phase growth, the sapphire C-plane substrate 1 is set on a susceptor in a reaction furnace, evacuated, and then heated at 1050 ° C. in a hydrogen atmosphere of 70 Torr at 1050 ° C.
The substrate surface is cleaned by heating for 5 minutes. 600
After cooling to ℃, trimethylgallium (TMG)
A polycrystalline GaN buffer layer 2 is deposited to a thickness of 50 nm by flowing 0 μmol / min, 2.5 L / min of ammonia, and 2 L / min of carrier hydrogen. Next, only the supply of TMG is stopped,
After the temperature was raised to 950 ° C., TMG was added at 20 μmol / mol.
And mono-silane at a rate of 10 cc / min.
An N single crystal layer 3 is formed. Thereafter, 5 μmol / min of trimethylaluminum (TMA) was further added, and n-Al
A GaN cladding layer 4 is deposited.

Next, the supply of TMA is stopped and nG
An aN guide layer 5 is deposited. Then, the substrate temperature is lowered to 750 ° C. in a mixed atmosphere of ammonia, hydrogen and nitrogen to reach a constant temperature, and then trimethylindium (TMI)
Was added at 10 μmol / min, and TMG at 1 μmol / min.
aN well layer, supply of TMI was stopped, and TMG was reduced to 10
GaN barrier layers are alternately produced by supplying μmol / min, and In
A GaN-based active layer 6 is deposited. The GaN barrier layer may contain In.

Next, supply of TMI and TMG is stopped, and the growth temperature is set to 1090 ° C. in a mixed atmosphere of ammonia and hydrogen.
After the temperature was raised to a constant temperature, TMG was added at 20 μmol / mol.
Min, biscyclopentadienyl magnesium (Cp2M
g) is added at 1 μmol / min, and the p-GaN guide layer 7 and further 5 μmol / min of TMA are added to grow the p-AlGaN cladding layer 8.

Thereafter, only the supply of TMA is stopped and p-
The GaN second contact layer 9 and the p-GaN first contact layer 10 are formed. This p-GaN second contact layer 9
And the supply amount of Cp2Mg during the growth of the p-GaN first contact layer 10 is about 1 μmol / min and about 1 μmol / min, respectively.
At 0 μmol / min, the film thickness was 0.5 μm and 0.1 μm, respectively.
μm. Finally, a Ni positive electrode 11 is selectively deposited on the p-GaN first contact layer 10 side of the wafer on which the blue-violet semiconductor laser structure is deposited by photolithography or the like. At this time, the positive electrode 11 may be an alloy containing Ni.

After depositing various films as described above, the process proceeds to an etching process for making the semiconductor laser function as a semiconductor laser.

First, the Ni positive electrode 11 shown in FIG.
Is used as an etching mask, a part of the p-GaN first and second contact layers 9 and 10 and a part of the p-AlGaN cladding layer 8 are removed by etching in a plasma atmosphere using a mixed gas of hydrogen and chlorine as a raw material. At this time, the atmospheric pressure is 1 Torr
r, the substrate temperature is room temperature, the mixing ratio of hydrogen and chlorine is preferably 1 to 1, and the typical etching rate is about 50 nm.
/ Min.

Next, a mask of SiO ₂ or the like is selectively deposited on the wafer, and as shown in FIG.
The selective etching is performed until the mask is removed, and then the Al negative electrode 12 is formed. The etching is performed in the same plasma atmosphere as described above. After that, etching is performed in the same plasma atmosphere as described above so that the cross section of the element in FIG. In this case, it is necessary to reflect light on this surface to cause resonance, and flatness of unevenness of about 1 nm or less is required. Finally, the wafer is placed in a nitrogen atmosphere at 700 ° C. for 30 minutes.
A heat treatment for activating Mg is performed. The pressure is one atmosphere. The atmosphere gas is preferably only nitrogen, but may be a mixed gas of hydrogen and nitrogen. The heat treatment temperature may be 500 ° C. or higher.

Next, the results of evaluating the characteristics of the semiconductor light emitting device (blue-violet semiconductor laser) formed as described above will be described with reference to the drawings.

FIG. 2 shows Mg doping characteristics of p-type GaN (having a thickness of 0.1 μm) by the MOVPE method. As is clear from FIG. 2, the Mg concentration taken into GaN monotonically increases with an increase in the supply amount of Cp2Mg, but the p carrier concentration (hole concentration) becomes highest when the Mg concentration is 5 × 10e ²⁰ cm ^−3. When the amount of Cp2Mg was further taken in, it decreased in contrast to the increase in the supply of Cp2Mg. This phenomenon is considered to be due to the fact that even if Mg is doped to a concentration of 5 × 10e ²⁰ cm ⁻³ or more, Mg does not enter a group III site serving as an acceptor impurity and forms a deep level at an interstitial position. However, when the Mg concentration is 5 × 10e ²⁰ c
When it is set to m ^-3 or more, it can be said that Mg atoms are in a state where they are easily moved by the subsequent heat treatment or the like. Therefore, the Mg of the second contact layer finally completed by the diffusion of Mg from the first contact layer to the second contact layer so that the diffusion of Mg from the second contact layer to the active layer does not occur.
In order to increase the concentration of Mg, the first contact layer is doped with Mg at a concentration of 5 × 10e ²⁰ cm ⁻³ or more, and the second contact layer is doped with 5
It can be said that it is desirable to dope Mg at a concentration of less than × 10e ²⁰ cm ⁻³ .

Next, p-Ga doped with Mg at a high concentration is used.
The diffusion of Mg from the N first contact layer to the p-GaN second contact layer and the p-AlGaN / InGaN-based active layer was determined by SIMS (Secondary Ion Mass).
Spectroscopy) is shown in FIG.
Shown in FIG. 3 shows the profile of the Mg concentration in the blue-violet semiconductor laser structure in the depth direction.

The thickness of the p-GaN first contact layer is set to 0.
When the thickness is 1 μm or more and 0.5 μm or less, Mg diffuses into the p-GaN second contact layer and becomes a high concentration of 1-5 × 10e ²⁰ cm ⁻³ , thereby increasing the hole concentration and increasing the light emission efficiency.
However, even if the p-type GaN first contact layer is doped with Mg at a high concentration with a thickness of 500 ° as in the prior art described in JP-A-8-97471, the diffusion of Mg hardly occurs and the hole concentration increases. Turned out not to be.

That is, if the first contact layer is doped with Mg only at a high concentration, M
Since the diffusion of g is insufficient, the hole concentration of the contact layer cannot be increased, and in order to sufficiently diffuse Mg into the second contact layer, the thickness of the first contact layer is made thicker than before. Turned out to be extremely important. Here, the thickness of the first contact layer is desirably 0.1 μm or more and 0.5 μm or less, as evident from the results of FIG. 3, but may be 1 μm or less. The reason will be described below.

P-type Ga having an Mg concentration of 1 × 10e ²¹ cm ⁻³
FIG. 4 shows the dependence of the hole concentration and the crack density on the crystal surface on the film thickness in N. Mg concentration of 5 × 10e ²⁰ cm ^-3
In the high-concentration doping region described above, the lattice distortion increases as the thickness of the Mg in the interstitial position increases and the film thickness increases, and defects and cracks are generated to deteriorate the crystallinity, so that the hole concentration decreases. When the crack density increases, leakage occurs when current is injected into the light emitting element. From this, p-GaN
It can be said that the thickness of the first contact layer is preferably 1 μm or less.

The semiconductor light emitting device according to the first embodiment of the present invention has been described above. However, in actually fabricating the semiconductor light emitting device, a slight error occurs in the thickness of the contact layer. It is about 20%. Therefore, it is necessary to consider this error in the film thickness within the range specified in the present invention.

As described above, the second contact layer
Mg doping concentration of 1 × 10e ²⁰cm^-3Less than
And the hole concentration decreases,
Doping of second contact layer of semiconductor light emitting device
Density is 1 × 10e²⁰cm^-3It is hoped that
No.

Embodiment 2 Next, a semiconductor light emitting device according to Embodiment 2 of the present invention will be described with reference to the drawings.

In FIG. 2, 5 × 10
e ²⁰ cm although ^-3 or more Mg rather hole concentration when doping showed a tendency to decrease, the present embodiment the carrier concentration by doping the 5 × 10e ²⁰ cm ^-3 or more Mg in the contact layer is reduced Not to provide a way.

FIG. 5 shows a p-type GaN produced by the MOVPE method.
Is the dependence of the hole concentration and the Mg concentration on the flow rate of Cp2Mg. The supply amounts of TMG and NH ₃ are the same as the above flow rates, but 50 ppm of SiH ₄ is added at 5 sccm Cp 2.
Supplied simultaneously with Mg, and about 1 × 10e ¹⁹ cm ^-3 of Si
g and doping at the same time. Mg concentration of 5 × 10e ²⁰ c
In the region where m ⁻³ or more, a decrease in hole concentration was not observed when Si was co-doped, and increased to a maximum of about 1 × 10e ²⁰ cm ⁻³ with the supply of Cp2Mg. This is presumably because the strain was alleviated by co-doping with heteroatoms such as Si, and the inactive Mg atoms conventionally located at the interstitial position were accommodated in the group III sites that can serve as acceptors. From experiments, it was found that the concentration of Si co-doped with Mg is preferably about 1/2 to 1/10 of the Mg concentration. When this was used for the p-GaN second contact layer, the hole concentration increased and the resistance was significantly reduced. The impurity used for the simultaneous doping is not limited to Si, but may be any element such as C, O, S, Se, Te, and Zn that can enter the group III site.

FIG. 6 shows current-voltage characteristics of the blue-violet semiconductor laser. By using the p-GaN first and second contact layers of the present invention, a laser diode capable of greatly reducing the resistance of the device and operating at a low voltage was manufactured.

In this embodiment, a blue-violet semiconductor laser using a sapphire substrate has been described. However, a blue-violet semiconductor laser using an n-type SiC substrate as shown in FIG. Has been confirmed to have the same effect.

In this embodiment, the active layer is made of InGa.
As described in the case of the N-system multiple quantum well, the general formula is Al _x G
a _y In _z N (x = 0, y = 0, z = 0 when x + y + z = 1)
It is needless to say that the present invention also provides the same effect for a light emitting element such as a laser diode or a light emitting diode in which a contact layer, a cladding layer, and an active layer are formed of a gallium nitride-based mixed crystal represented by

[0036]

As described above, according to the present invention,
The contact resistance is reduced by using the p-type GaN first and second contact layers, and the hole concentration in the second contact layer can be almost maximized without diffusing Mg into the active layer, A blue-violet semiconductor laser having low resistance, high efficiency, and excellent optical characteristics can be manufactured.

[Brief description of the drawings]

FIG. 1 is a sectional view of a semiconductor light emitting device according to an embodiment of the present invention.

FIG. 2 is a diagram showing the relationship between the doping amount of Mg and the hole concentration of p-type GaN.

FIG. 3 is a diagram showing a profile of a Mg concentration in a depth direction in a blue-violet semiconductor laser structure.

FIG. 4 is a diagram showing the dependence of the hole concentration and the crack density on the crystal surface on the film thickness in p-type GaN having a Mg concentration of 1 × 10e ²¹ cm ⁻³ .

FIG. 5 is a view showing the Cp2Mg flow rate dependence of the hole concentration and the Mg concentration in p-type GaN produced by the MOVPE method.

FIG. 6 is a diagram showing current-voltage characteristics of a blue-violet semiconductor laser of the present invention.

FIG. 7 is a sectional view of a semiconductor light emitting device according to an embodiment of the present invention.

FIG. 8 is a diagram showing a hole concentration range when a p-contact layer is formed by a conventional method.

[Explanation of symbols]

Reference Signs List 1 sapphire substrate 2 GaN buffer layer 3 n-GaN layer 4 n-AlGaN clad layer 5 n-GaN guide layer 6 InGaN-based MQW active layer 7 p-GaN guide layer 8 p-AlGaN clad layer 9 p-GaN second contact layer REFERENCE SIGNS LIST 10 p-GaN first contact layer 11 Ni positive electrode 12 Al negative electrode 21 SiC substrate 22 AlN buffer layer 23 n-GaN layer 24 n-AlGaN cladding layer 25 n-GaN guide layer 26 InGaN-based MQW active layer 27 p-GaN Guide layer 28 p-AlGaN cladding layer 29 p-GaN second contact layer 30 p-GaN first contact layer 31 Ni positive electrode 32 Al negative electrode 33 SiO ₂ insulating film

Continued on the front page (72) Inventor Yoshihiro Hara 1006 Kazuma Kadoma, Osaka Prefecture Matsushita Electric Industrial Co., Ltd. (72) Inventor Isao Kidoguchi 1006 Oji Kadoma Kadoma City, Osaka Matsushita Electric Industrial Co. (72 Inventor Ayu Tsujimura 1006 Kazuma Kadoma, Kadoma City, Osaka Prefecture (72) Inside Matsushita Electric Industrial Co., Ltd. 1006, Kazuma, Kamon, Fumonma-shi Matsushita Electric Industrial Co., Ltd. (56) References JP-A-9-153645 (JP, A) JP-A-7-15041 (JP, A) JP-A 8-97471 (JP, A) JP-A-8-330629 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) H01L 33/00 H01S 5/00-5/50

Claims (5)

(57) [Claims] 1. A gallium nitride based semiconductor light emitting device comprising: a first contact layer in contact with a p-electrode; and a second contact layer in contact with the first contact layer and formed on an active layer side of the first contact layer. The acceptor impurity concentration on the p-electrode side of the first contact layer is 5 × 10 ²⁰ cm ⁻³ or more, and the acceptor impurity concentration on the active layer side of the second contact layer is 1 × 10 ²⁰ cm ⁻³ or more. A gallium nitride-based material, wherein the concentration of the acceptor impurity monotonically increases from the active layer side of the second contact layer to the p-electrode side of the first contact layer, while being less than 5 × 10 ²⁰ cm ^−3. Semiconductor light emitting device. 2. The gallium nitride based semiconductor light emitting device according to claim 1, wherein the thickness of the first contact layer is 0.1 μm or more and 0.5 μm or less. 3. The method according to claim 1, wherein the second contact layer comprises S as an impurity.
2. The gallium nitride based semiconductor light emitting device according to claim 1, wherein the gallium nitride based semiconductor light emitting device contains at least one element selected from i, O, S, Se, Te, C and Zn. 4. The gallium nitride based semiconductor light emitting device according to claim 1, wherein the acceptor impurity is Mg. Nitrogen according to claim 1, wherein the first contact layer and a second contact layer, characterized in that a GaN
Gallium arsenide based semiconductor light emitting device.