CN114122199B - Method for producing p-type group III nitride semiconductor - Google Patents

Abstract

The present invention relates to a method for producing a p-type group III nitride semiconductor including Al as an essential component. The carrier concentration of a p-type group III nitride semiconductor containing Al as an essential component is increased. An electron blocking layer (14) made of Mg-doped AlGaN is formed on the light emitting layer (13) by MOCVD. The electron blocking layer (14) is formed such that the ratio H/Mg of H concentration to Mg concentration before the p-type heat treatment is 50-100%. The H/Mg may be controlled by the growth temperature of the electron blocking layer (14), the V/III ratio, the Mg concentration, etc.

Description

Method for producing p-type group III nitride semiconductor

Technical Field

The present invention relates to a method for producing a p-type Group III nitride semiconductor containing Al as an essential constituent.

Background Art

In a light-emitting element that emits ultraviolet light and is composed of a Group III nitride semiconductor, an electron blocking layer composed of p-AlGaN is formed on a light-emitting layer to prevent electrons from diffusing to the p-side, thereby improving light emission efficiency.

Non-patent document 1 shows that the formation energy boundary of H ⁺ and H ^- in GaN and AlN (the maximum value of the formation energy relative to the Fermi level) is located near 2.5 eV from the conduction band of AlN in Figures 2 and 3, and the formation energy boundary position of GaN and H ⁺ and H ^- (0.5 eV from the conduction band) is very different. In addition, it is shown that when Mg is doped into GaN and AlN, the formation of Mg-H is lower than that of Mg alone, so it is easily doped into the III group site.

Non-patent document 2 shows that for GaN, when the Mg concentration is 3×10 ¹⁹ cm ^-3 or less, the Mg concentration is equal to the H concentration, and a Mg-H complex is formed. After the Mg-H complex is formed, H is separated from GaN by heat treatment to obtain a p-type. It is shown that the net acceptor concentration is equal to the H concentration separated by heat treatment. On the other hand, it is shown that when the Mg concentration is 3×10 ¹⁹ cm ^-3 or more, the Mg concentration is not equal to the H concentration, a part of it forms a Mg-H complex, and the rest may form a single Mg or Mg cluster. In addition, the separation of H caused by heat treatment is not observed. Therefore, it is shown that a high carrier concentration cannot be obtained by compensating carriers by Mg ₂ N ₃ and nitrogen vacancies generated by excessive doping of Mg.

Non-Patent Document 3 mentions the possibility of suppressing various defects by controlling the Fermi level of a Group III nitride semiconductor.

Prior art literature

Patent Literature

Non-patent document 1: PHYSICAL REVIEW LETTERS 108, 156403 (2012)

Non-patent document 2: APPLIED PHYSICS LETTERS 98, 213505 (2011)

Non-patent document 3: JOURNAL OF APPLIED PHYSICS 120, 185704 (2016)

Summary of the invention

In order to increase the output of ultraviolet light-emitting elements, it is necessary to increase the carrier concentration of p-AlGaN. In addition, if a deep ultraviolet light-emitting element can be realized without using a contact layer composed of p-GaN, the light absorption of the p-layer can be suppressed, and the light extraction efficiency when GaN is used for the p-layer can be theoretically increased from about 7% to 80%. However, in AlGaN, the activation energy of Mg is high, and the acceptor is compensated to the nitrogen vacancy, so it is difficult to increase the carrier concentration of p-AlGaN, and it is also difficult for p-AlGaN to contact with metal.

Therefore, an object of the present invention is to increase the carrier concentration of a Mg-doped p-type Group III nitride semiconductor containing Al as an essential constituent.

The inventors have repeatedly studied to increase the carrier concentration of p-AlGaN, and found that in AlGaN, unlike GaN, the Mg concentration and the H concentration are not equal. That is, the bond Mg-H between Mg and H is generated in GaN, and Mg easily enters the Ga site, thereby achieving an increase in the carrier concentration. However, it is known that the H concentration in AlGaN is lower than the Mg concentration, and the Mg-H decreases, so the carrier concentration does not increase. In addition, the inventors found that by suppressing the generation of nitrogen vacancies and making the ratio of H concentration to Mg concentration close to 100%, the carrier concentration can be increased and the resistance can be reduced. The present invention is completed based on the above findings of the inventors.

The present invention is a method for manufacturing a p-type Group III nitride semiconductor, characterized in that, in the method for manufacturing a p-type Group III nitride semiconductor doped with Mg and having an Al composition of 60% to 90%, before the heat treatment for p-type conversion, the Group III nitride semiconductor doped with Mg and having an Al composition of 60% to 90% is formed in a manner such that the ratio of the H concentration to the Mg concentration in the crystal becomes 50 to 100%.

The Group III nitride semiconductor containing Al as an essential component is AlN, AlGaN, AlInN or AlGaInN.

The growth temperature of the Group III nitride semiconductor is preferably 1100° C. or higher.

The V/III ratio of the Group III nitride semiconductor is preferably 2,000 to 10,000.

The Mg concentration of the Group III nitride semiconductor is preferably 1×10 ¹⁸ to 2×10 ²⁰ /cm ³ , and more preferably 7×10 ¹⁸ to 7×10 ¹⁹ /cm ³ .

The ratio of the peak intensity at 3 to 3.8 eV to the peak intensity at 4.37 to 4.6 eV in the CL (cathodoluminescence) spectrum is preferably 3 or less. The peak at 4.37 to 4.6 eV indicates transitions due to Mg in the group III site, and the peak at 3 to 3.8 eV indicates transitions from defect levels due to group V vacancies.

The growth pressure of the Group III nitride semiconductor is preferably 10 to 60 kPa.

The C concentration of the Group III nitride semiconductor is preferably 1×10 ¹⁷ /cm ³ or less.

According to the present invention, the carrier concentration of a Mg-doped p-type Group III nitride semiconductor containing Al as an essential component can be increased, and the resistance can be reduced.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a diagram showing the structure of a light emitting element according to Example 1. In FIG.

FIG. 2 is a graph showing the relationship between H/Mg and growth temperature.

FIG. 3 is a graph showing the relationship between H/Mg and V/III ratio.

FIG. 4 is a graph showing the relationship between H/Mg and the peak intensity ratio.

FIG. 5 is a graph showing the relationship between the output P0 of the light emitting element and the V/III ratio.

FIG. 6 is a graph showing the relationship between the forward voltage Vf of a light emitting element and the V/III ratio.

FIG. 7 is a graph showing the relationship between H/Mg and Mg concentration.

FIG. 8 is a graph showing the relationship between the H concentration and the Mg concentration.

FIG. 9 is a graph showing the relationship between C concentration and growth temperature.

FIG. 10 is a graph showing the relationship between C concentration and growth pressure.

FIG. 11 is a graph showing the relationship between the C concentration and the V/III ratio.

Explanation of symbols

10: Substrate

11: Buffer layer

12: n contact layer

13: Luminous layer

14: Electron blocking layer

15: p-contact layer

16: Transparent electrode

17: p electrode

18: n electrode

DETAILED DESCRIPTION

Hereinafter, specific examples of the present invention will be described with reference to the drawings, but the present invention is not limited to the examples.

Example 1

1 is a diagram showing the structure of a light emitting element of Example 1. It is a flip-chip ultraviolet light emitting element having a substrate 10 , a buffer layer 11 , an n-contact layer 12 , a light emitting layer 13 , an electron blocking layer 14 , a p-contact layer 15 , a transparent electrode 16 , a p-electrode 17 , and an n-electrode 18 .

(Composition of each layer)

First, the structure of each layer of the light-emitting element of Example 1 is described.

The substrate 10 is a growth substrate made of sapphire. The thickness of the substrate 10 is, for example, 900 μm. In addition to sapphire, AlN, Si, SiC, ZnO, etc. may also be used.

The buffer layer 11 is located on the substrate 10. The buffer layer 11 has a structure in which three layers, namely a core layer, a low-temperature buffer layer, and a high-temperature buffer layer, are stacked in sequence. The core layer is composed of undoped AlN grown at a low temperature, and is a layer that serves as a crystal nucleus for crystal growth. The thickness of the core layer is, for example, 10 nm. The low-temperature buffer layer is a layer composed of undoped AlN grown at a higher temperature than the core layer. The thickness of the low-temperature buffer layer is, for example, 0.3 μm. The high-temperature buffer layer is a layer composed of undoped AlN grown at a higher temperature than the low-temperature buffer layer. The thickness of the high-temperature buffer layer is, for example, 2.7 μm. By providing such a buffer layer 11, the density of through-dislocations in AlN can be reduced.

The n-contact layer 12 is located on the buffer layer 11. The n-contact layer 12 is composed of n-Al _x Ga _y N (where x+y=1). The Si concentration is preferably 1×10 ¹⁹ to 5×10 ¹⁹ /cm ³ . This is because both the improvement of the luminous efficiency and the reduction of the resistance of the light-emitting element of Example 1 can be achieved. As an example of the composition of the n-contact layer 12, the Al composition 100x mol% (hereinafter, the composition means mol% (AlN mol%). x is the composition ratio) is 62 mol%, the thickness is 1.3 μm, and the Si concentration is 2×10 ¹⁹ /cm ³ .

The light emitting layer 13 is located on the n-contact layer 12. The light emitting layer 13 is an MQW structure having two well layers. That is, it is a structure in which the first barrier layer, the first well layer, the second barrier layer, the second well layer, and the third barrier layer are stacked in sequence. The first well layer and the second well layer are composed of n-Al _x Ga _y N (where x+y=1). The Al composition is set according to the desired emission wavelength. For example, the wavelength is 210 to 360 nm, particularly 250 to 300 nm. The first barrier layer, the second barrier layer, and the third barrier layer are composed of n-Al _x1 Ga _y1 N (where x1+y1=1, x≤x1) having a higher Al composition than the first well layer and the second well layer. As an example of the composition of the first well layer and the second well layer, the Al composition is 40 mol%, the thickness is 2.4 nm, and the Si concentration is 9×10 ¹⁸ /cm ³ . An example of the first and second barrier layers has an Al composition of 55 mol%, a thickness of 11 nm, and a Si concentration of 9×10 ¹⁸ /cm ³ . An example of the third barrier layer has an Al composition of 55 mol%, a thickness of 4 nm, and a Si concentration of 5×10 ¹⁸ /cm ³ .

The electron blocking layer 14 is located on the light emitting layer 13. The electron blocking layer 14 is composed of Mg-doped p-AlGaN having a higher Al composition than the third barrier layer. The electron diffusion to the p-contact layer 15 side is suppressed by the electron blocking layer 14. The electron blocking layer 14 does not need to be a single layer and can be composed of multiple layers. For example, it can be a structure in which two layers with different Al compositions are alternately and repeatedly stacked. In addition, it can also be a structure in which the Al composition is inclined from the third barrier layer to the p-contact layer 15.

The electron blocking layer 14 preferably has the following structure: Al composition of 60-90 mol%, thickness of 10-50 nm, Mg concentration of 7×10 ¹⁸ -1×10 ²⁰ /cm ³ , and carrier concentration of 1×10 ¹⁶ -1×10 ¹⁷ /cm ³ . As an example, Al composition of 80 mol%, thickness of 25 nm, and Mg concentration of 5×10 ¹⁹ /cm ³ .

The p-contact layer 15 is located on the electron blocking layer 14. The p-contact layer 15 has a structure in which a first p-contact layer 15A and a second p-contact layer 15B are stacked in sequence. The first p-contact layer 15A is composed of p-AlGaN doped with Mg, and the second p-contact layer 15B is composed of p-GaN doped with Mg. The Al composition of the first p-contact layer 15A is smaller than the Al composition of the electron blocking layer 14 and is set in a manner not to absorb light of the emission wavelength. By making the p-contact layer 15 into such a two-layer structure, light absorption caused by the p-contact layer 15 is suppressed and light output is improved. In the past, it was difficult to make the p-contact layer 15 into such a structure because the carrier concentration of p-AlGaN could not be increased. However, according to the method described later, p-AlGaN with a high carrier concentration can be formed, so such a structure of the p-contact layer 15 can be adopted.

The first p-contact layer 15A preferably has the following configurations: Al composition of 60-90 mol%, thickness of 10-50 nm, Mg concentration of 7×10 ¹⁸ -1×10 ²⁰ /cm ³ , and carrier concentration of 1×10 ¹⁶ -1×10 ¹⁷ /cm ³ . As an example, Al composition of 50 mol%, thickness of 50 nm, and Mg concentration of 5×10 ¹⁹ /cm ³ .

The second p-contact layer 15B preferably has the following structure: The thickness is 10 to 30 nm, and the Mg concentration is 1×10 ²⁰ to 6×10 ²⁰ /cm ³ . As an example, the thickness is 18 nm, and the Mg concentration is 4×10 ²⁰ /cm ³ .

A groove is provided in a part of the surface of the p-contact layer 15. The groove has a depth that penetrates the p-contact layer 15 and the light-emitting layer 13 and reaches the n-contact layer. The n-electrode 18 is provided in the groove.

The transparent electrode 16 is located on the p-contact layer 15. The material of the transparent electrode 16 is, for example, a transparent conductive oxide such as IZO, ITO, ICO, ZnO, or MgZnO. Transparency here means high transmittance in the visible light wavelength region. As an example, a thin film ITO with a film thickness of 2 nm is used.

The p-electrode 17 is located on the transparent electrode 16. The p-electrode 17 is Rh, Ni/Al, Pd/Al, Pd/Al/Au, Ni/Au, Al, etc. As an example, Al is used.

The n-electrode 18 is located on the n-contact layer 12 exposed at the bottom surface of the groove. The n-electrode 18 is made of Ti/Al/Ni, V/Al/Ni, V/Al/Ru, or the like.

As described above, in the light emitting element of Example 1, the electron blocking layer 14 and the first p-contact layer 15A composed of p-AlGaN have high carrier concentrations, thereby reducing resistance. In addition, the second p-contact layer 15B composed of GaN is thin, so that the absorption of light radiated from the light emitting layer 13 is small. Therefore, according to the light emitting element of Example 1, light output can be increased and forward voltage can be reduced.

(Regarding the manufacturing process of the light emitting element)

Next, the manufacturing process of the light-emitting element of Example 1 is described. It should be noted that the MOCVD method is used for the crystal growth of the Group III nitride semiconductor, ammonia is used as the nitrogen source, trimethylgallium or triethylgallium is used as the Ga source, and trimethylaluminum is used as the Al source. In addition, silane is used as the n-type dopant gas, and bis(cyclopentadienyl)magnesium is used as the p-type dopant gas. In addition, hydrogen and nitrogen are used as carrier gases.

First, a substrate 10 made of sapphire is prepared. Then, a buffer layer 11 is formed on the substrate 10. The buffer layer 11 is formed by first forming a core layer made of AlN by MOCVD. The growth temperature is, for example, 880°C. Next, a low-temperature buffer layer and a high-temperature buffer layer made of AlN are sequentially formed on the core layer by MOCVD.

Next, an n-contact layer 12 made of n-AlGaN is formed on the buffer layer 11 by the MOCVD method.

Next, the light emitting layer 13 is formed on the n-contact layer 12 by MOCVD. The light emitting layer 13 is formed by sequentially stacking a first barrier layer, a first well layer, a second barrier layer, a second well layer, and a third barrier layer.

Next, an electron blocking layer 14 made of Mg-doped AlGaN is formed on the light emitting layer 13 by MOCVD. Here, the electron blocking layer 14 is formed so that the ratio H/Mg of the H concentration (number/cm ³ ) to the Mg concentration (number/cm ³ ) before the heat treatment for p-type conversion is 50 to 100% (H/Mg×100%, hereinafter H/Mg is also expressed in %). A more preferred range of H/Mg is 70 to 100%, and a further preferred range is 80 to 100%. The reason for setting H/Mg to such a range is as follows.

In GaN, the boundary of the formation energy of H ⁺ and H ^- is close to the conduction band, and H ⁺ is easily formed. Therefore, it is easy to form a bond Mg-H between Mg and H, and Mg easily enters the Ga site, thereby increasing the carrier concentration. On the other hand, the inventors found that in AlGaN, the H concentration is lower than the Mg concentration before the heat treatment for p-type conversion. If the H concentration is lower than the Mg concentration, the carrier concentration of p-AlGaN becomes lower. The reasons are considered to be as follows. The boundary of the formation energy of H ⁺ and H ^- in AlGaN (the maximum value of the formation energy relative to the Fermi level) is close to the valence band. Therefore, the transition energy of electrons relative to the conduction band of H becomes larger, so it is not easy to form H ⁺ . Since it is not easy to form H ⁺ , the influence of donors caused by nitrogen defects in AlGaN becomes larger, and the Fermi level is located closer to the conduction band than the boundary of the formation energy of H ⁺ and H ^- . Since the Fermi level is close to the conduction band, the formation of H ⁺ is further hindered. From these reasons, it can be considered that the H concentration in the crystal grown in AlGaN is lower than the Mg concentration. As a result, Mg-H is not easily formed, Mg is not easy to enter the Ga site, and the carrier concentration does not increase.

Therefore, in Example 1, the H/Mg ratio in the crystal is set to 50-100% before the heat treatment for p-type conversion, so that the H concentration is close to the Mg concentration. If so, the Fermi level of AlGaN can be brought close to the valence band, and the nitrogen defects in AlGaN can be reduced. As a result, H ⁺ is easily formed in AlGaN, the bond between Mg and H, i.e., Mg-H, can be increased, and Mg is easily doped into the acceptor site. As a result, the carrier concentration of AlGaN can be increased.

H/Mg can be controlled by the growth temperature, V/III ratio, Mg concentration, etc. of the electron blocking layer 14. For example, H/Mg can be controlled to 50 to 100% by setting the growth temperature, V/III ratio, and Mg concentration to the following ranges.

The growth temperature is 1100° C. or higher. The upper limit of the growth temperature may be within a range in which AlGaN does not evaporate, and is, for example, 1400° C. or lower. It is more preferably 1250° C. or higher, and even more preferably 1300° C. or higher.

The V/III ratio is 2000 to 10000, more preferably 3000 to 9000, and even more preferably 4000 to 8000.

The Mg concentration is 1×10 ¹⁸ to 2×10 ²⁰ /cm ³ . When more than 2×10 ²⁰ /cm ³ of Mg is doped, the arrangement of group III (Al and Ga) and group V (N) is reversed, and polarity inversion occurs. When polarity inversion occurs, impurities such as C and O are easily doped, which deteriorates electrical properties. More preferably, it is 7×10 ¹⁸ to 7×10 ¹⁹ /cm ³ .

In addition, H/Mg can also be controlled by the peak intensity ratio of the CL spectrum. Specifically, if the ratio of the peak intensity of 3 to 3.8 eV to the peak intensity of 4.37 to 4.6 eV is 3 or less, H/Mg can be 50 to 100%. It should be noted that the peak of 4.37 to 4.6 eV represents the transition caused by Mg in the group III site, and the peak of 3 to 3.8 eV represents the transition from the defect level caused by the group V vacancy. The CL spectrum is preferably a sufficiently low temperature, for example, a value below 10K.

The growth pressure is preferably 10 to 60 kPa. This is because in the case of AlGaN, if the growth pressure is high, Al and NH ₃ will undergo a parasitic reaction and will not grow smoothly. In addition, in order to control the C concentration within the range described below, it is also preferable to set the growth pressure to the above range.

The C concentration is preferably 1×10 ¹⁷ /cm ³ or less. C replaces N at the group V site, acts like a donor, and compensates the acceptor. Therefore, it is believed that the Fermi level moves to the conduction band side, becoming an important factor that hinders the formation of H ⁺ . That is, the C concentration is considered to be an important factor in carrier compensation. If the C concentration is high, H/Mg decreases. The lower the C concentration, the better. The lower limit is not specifically specified, but about 1×10 ¹⁶ /cm ³ is the background level. The higher the growth temperature, the higher the V/III ratio, and the higher the growth pressure, the lower the C concentration can be. For example, by setting the growth temperature to 1050-1300°C, the V/III ratio to 1000-10000, and the growth pressure to 10-60kPa, the C concentration can be controlled to be below 1×10 ¹⁷ /cm ^3. It should be noted that the intensity of carrier compensation caused by the C concentration is believed to be weaker than the carrier compensation caused by the group V vacancies.

Next, the p-contact layer 15 is formed on the electron blocking layer 14 by the MOCVD method. The p-contact layer 15 is formed by sequentially stacking a first p-contact layer 15A composed of Mg-doped AlGaN and a second p-contact layer 15B composed of Mg-doped GaN. Here, the first p-contact layer 15A is formed in a manner such that H/Mg is 50 to 100% in the same manner as the electron blocking layer 14. This can increase the carrier concentration of the first p-contact layer 15A.

Then, heat treatment is performed in a nitrogen atmosphere to convert the electron blocking layer 14 and the p-contact layer 15 to p-type. Since the H/Mg ratio before the heat treatment for converting to p-type is 50 to 100%, the carrier concentration of the electron blocking layer 14 and the first p-contact layer 15A after the heat treatment for converting to p-type can be higher than that of the conventional manufacturing method.

Next, a predetermined region on the surface of the p-contact layer 15 is dry-etched to form a groove having a depth reaching the n-contact layer 12 .

Next, a transparent electrode 16 is formed on the p-contact layer 15. Next, a p-electrode 17 is formed on the transparent electrode 16, and an n-electrode is formed on the n-contact layer 12 exposed at the bottom surface of the groove. The transparent electrode 16, the p-electrode 17, and the n-electrode 18 are formed by sputtering, evaporation, etc. The light-emitting element of Example 1 is manufactured in the above manner.

Next, various experimental examples related to Example 1 are described.

(Experiment 1)

A sample was prepared by sequentially stacking a buffer layer consisting of AlN, undoped AlGaN, and p-AlGaN doped with Mg on a substrate consisting of sapphire. The H concentration, Mg concentration, C concentration, and H/Mg of p-AlGaN were measured by secondary ion mass spectrometry (SIMS), and the transitions caused by group V vacancies and group III site Mg were measured by CL spectroscopy.

FIG2 is a graph showing the relationship between H/Mg and the growth temperature of p-AlGaN. The Al composition 100x of p- _{AlxGa1 -xN} is 35 mol%, 60 mol%, 80 mol%, and 100 mol% (i.e., AlN). It should be noted that the growth temperature is a value obtained by measuring the substrate using a thermocouple, and the substrate surface temperature is 80 to 100°C lower than the measured growth temperature. As shown in FIG2, it can be seen that the higher the growth temperature, the higher the H/Mg, and the closer it is to 100%.

FIG3 is a graph showing the relationship between H/Mg and the V/III ratio when forming p-AlGaN. The Al composition of p-AlGaN is 35 mol%, 50 mol%, and 85 mol%. As shown in FIG3, it can be seen that the higher the V/III ratio, the higher the H/Mg, and the closer it is to 100%. In addition, it can be inferred from FIG3 that if the V/III ratio is 2000 to 10000, H/Mg can be made 50% or more. In addition, it can be inferred from FIG2 and FIG3 that the H/Mg can be further increased when the Al composition is in the range of 20 mol% to 40 mol% or 70 mol% to 90 mol%. It is believed that it is not easy to increase H/Mg when the Al composition is near 50 mol%, which is related to the defect or dislocation density in AlGaN. When the Al composition of the AlGaN layer is high, the lattice matching with the underlying AlN layer is not large, and misfit dislocations rarely occur. On the other hand, when the Al composition of the AlGaN layer is low, misfit dislocations often occur at the beginning of AlGaN growth, but as the growth film thickness increases, the misfit dislocations overlap and decrease. Therefore, it is considered that H/Mg is not easy to increase at around 50 mol% of Al composition which is an intermediate composition.

FIG4 is a graph showing the relationship between H/Mg and the peak intensity ratio. The peak intensity ratio indicates the amount of group V vacancies in AlGaN, and is the ratio of the peak intensity of 3 to 3.8 eV to the peak intensity of 4.37 to 4.6 eV in the CL spectrum of p-AlGaN with an Al composition of 50 mol%. The peak of 4.37 to 4.6 eV indicates the transition caused by Mg (MgIII) at the group III site, and the peak of 3 to 3.8 eV indicates the transition from the defect level caused by the group V vacancy (VN). As shown in FIG4 , it can be seen that as the peak intensity ratio decreases, H/Mg increases, and if the peak intensity ratio is 3 or less, H/Mg can be made 50% or more.

FIG7 is a graph showing the relationship between H/Mg and Mg concentration, and FIG8 is a graph showing the relationship between H concentration and Mg concentration. As shown in FIG7 , it can be seen that the higher the Mg concentration, the higher the H/Mg. In addition, as shown in FIG8 , it can be seen that the H concentration increases approximately linearly with the increase in Mg concentration until the Mg concentration reaches about 2×10 ¹⁹ /cm ³ , but the H concentration is saturated when it exceeds 2×10 ¹⁹ /cm ^3. Considering this result and the practical Mg concentration in actual equipment, it is considered that the optimal Mg concentration is in the range of 7×10 ¹⁸ to 7×10 ¹⁹ /cm ³ .

FIG9 is a graph showing the relationship between C concentration and growth temperature, FIG10 is a graph showing the relationship between C concentration and growth pressure, and FIG11 is a graph showing the relationship between C concentration and V/III ratio. As shown in FIG9, it can be seen that the higher the growth temperature, the lower the C concentration. Also, as shown in FIG10, it can be seen that the higher the growth pressure, the lower the C concentration. Also, as shown in FIG11, it can be seen that the higher the V/III ratio, the lower the C concentration. Thus, it can be seen that the C concentration can be controlled by the growth temperature, growth pressure, and V/III ratio, and can be controlled to be below 1×10 ¹⁷ /cm ³ .

(Experiment 2)

The light-emitting element of Example 1 was produced by changing the V/III ratio when forming the electron blocking layer 14 and the first p-contact layer 15A, and the output P0 and the forward voltage Vf were measured.

Fig. 5 is a graph showing the relationship between the output P0 and the V/III ratio of the light emitting element. As shown in Fig. 5, the larger the V/III ratio, the higher the output P0, and in particular, the output P0 is greatly improved when the V/III ratio is above 2000. As shown in Fig. 3, due to the increase in the V/III ratio, H/Mg becomes larger, and it is believed that the carrier concentration of the electron blocking layer 14 and the first p-contact layer 15A increases, so the output P0 increases.

Fig. 6 is a graph showing the relationship between the forward voltage Vf and the V/III ratio of the light emitting element. As shown in Fig. 6, the larger the V/III ratio, the smaller the forward voltage Vf, confirming the lower resistance caused by the increase in the carrier concentration of the electron blocking layer 14 and the first p-contact layer 15A.

(Variation Example)

Example 1 achieves an increase in the carrier concentration of p-AlGaN, but the present invention is not only applicable to AlGaN, but also to AlN, AlInN, and AlGaInN. That is, it is applicable to any III-group nitride semiconductor that has Al as an essential component. In addition, as long as the present invention has Al as an essential component, the value of the Al composition is not limited, but if the Al composition is 20 mol% to 40 mol% or 70 mol% to 90 mol%, it is easier to increase H/Mg and the carrier concentration.

Industrial Applicability

The invention is suitable for manufacturing ultraviolet light emitting elements, and is particularly suitable for the case where the wavelength is 250 to 300 nm.

Claims (10)

1. A method for producing a p-type Group III nitride semiconductor, characterized in that, in a method for producing a Mg-doped p-type Group III nitride semiconductor having an Al composition of 60 mol % to 90 mol %, Before the heat treatment for converting to p-type, a Mg-doped Group III nitride semiconductor having an Al composition of 60 mol % to 90 mol % is formed so that the ratio of the H concentration to the Mg concentration in the crystal becomes 50 to 100 %. 2 . The method for producing a p-type Group III nitride semiconductor according to claim 1 , wherein a growth temperature of the Group III nitride semiconductor is 1100° C. or higher. 3 . The method for producing a p-type Group III nitride semiconductor according to claim 1 , wherein a V/III ratio of the Group III nitride semiconductor is 2000 to 10000. 4 . The method for producing a p-type Group III nitride semiconductor according to claim 2 , wherein a V/III ratio of the Group III nitride semiconductor is 2000 to 10000. 5 . The method for producing a p-type Group III nitride semiconductor according to claim 1 , wherein the Group III nitride semiconductor has a Mg concentration of 1×10 ¹⁸ to 2×10 ²⁰ /cm ³ . 6 . The method for producing a p-type Group III nitride semiconductor according to claim 1 , wherein the Group III nitride semiconductor has a Mg concentration of 7×10 ¹⁸ to 7×10 ¹⁹ /cm ³ . 7 . The method for producing a p-type Group III nitride semiconductor according to claim 1 , wherein an Al composition of the Group III nitride semiconductor is 70 mol % to 90 mol %. 8 . The method for producing a p-type Group III nitride semiconductor according to claim 1 , wherein a ratio of a peak intensity at 3 to 3.8 eV to a peak intensity at 4.37 to 4.6 eV in a CL spectrum is 3 or less. 9 . The method for producing a p-type Group III nitride semiconductor according to claim 1 , wherein a growth pressure of the Group III nitride semiconductor is 10 to 60 kPa. 10 . The method for producing a p-type Group III nitride semiconductor according to claim 1 , wherein a C concentration of the Group III nitride semiconductor is 1×10 ¹⁷ /cm ³ or less.