WO2014088030A1 - Nitride semiconductor light-emitting element and method for manufacturing nitride semiconductor light-emitting element - Google Patents

Abstract

Provided is a nitride semiconductor light-emitting element having a structure capable of reducing the fluctuation in forward voltage (Vf) associated with energization. Electrons from an n-type cladding layer (23) are supplied to an active layer (15). When carrier overflow occurs, electrons overflow from the active layer (15) and propagate through a second InGaN layer (25a) of a center semiconductor region (19) to reach a second hetero interface (HJ2). Misfit dislocation is formed in a second group III nitride semiconductor region (17) on the second hetero interface (HJ2). The energy generated by non-radiative recombination is large enough to generate a reaction that causes activated p-type dopants to recombine with residual hydrogen, and free residual hydrogen combines with the already-activated p-type dopants to work as acceptor killers. However, the overflow electrons which have reached the second hetero interface (HJ2) disappear due to the non-radiative recombination at misfit dislocation.

Description

Nitride semiconductor light emitting device and method for manufacturing nitride semiconductor light emitting device

The present invention relates to a nitride semiconductor light emitting device and a method for manufacturing a nitride semiconductor light emitting device.

Patent Document 1 discloses a nitride semiconductor light emitting device that can operate at a low operating current or operating voltage.

JP 2011-159771 A

According to the knowledge of the inventors, when continuous energization is performed on a nitride semiconductor light emitting device using a semipolar plane, the forward voltage Vf of the nitride semiconductor light emitting device increases with energization. The inventors have studied the cause of this forward voltage Vf. One of the causes of the increase in the forward voltage Vf is a change in contact characteristics between the electrode and the p-type semiconductor. However, the inventors have not found any variation in contact characteristics corresponding to the variation in forward voltage. As a result of further investigations, the inventors have found that the characteristics of the semiconductor region between the active layer and the electrode of the nitride semiconductor light emitting element change with energization.

In Patent Document 1, the light emitting element forms an n-type layer, a p-type layer, and an active layer on the nonpolar main surface of the substrate. The active layer includes a barrier layer and a quantum well layer, the barrier layer closest to the substrate is an undoped layer, and at least one of the remaining barrier layers is an n-doped layer. In this light emitting element, electrons overflowing from the active layer disappear due to non-radiative recombination in a semiconductor region where hydrogen taken in during growth remains.

An object of one aspect of the present invention is to provide a nitride semiconductor light emitting device having a structure capable of reducing fluctuations in the forward voltage Vf due to energization, and another aspect of the present invention is the nitride semiconductor light emitting device. It is an object of the present invention to provide a method for manufacturing the above.

A nitride semiconductor light emitting device according to an aspect of the present invention has a semipolar main surface made of a gallium nitride based semiconductor, and includes a first group III nitride semiconductor region including an n-type cladding layer, and the first group III nitride. A second group III nitride semiconductor region including a p-type cladding layer and a plurality of InGaN layers provided on the semipolar principal surface of the semiconductor semiconductor region, and a center provided on the first group III nitride semiconductor region A semiconductor region, wherein the center semiconductor region is provided between the semipolar main surface of the first group III nitride semiconductor region and the second group III nitride semiconductor region, and the center semiconductor region is formed of GaN The center semiconductor region includes an active layer, a first InGaN layer, and a second InGaN layer, and the active layer includes the first gap. The active layer includes one or a plurality of InGaN well layers, and the plurality of InGaN layers includes the InGaN well layer, the first InGaN layer, and the second InGaN layer, provided between the nGaN layer and the second InGaN layer. Each of the plurality of InGaN layers in the center semiconductor region has an indium composition and a layer thickness, the unit in the layer thickness is nanometer, and the product of the indium composition and the layer thickness of each InGaN layer The total value over the center semiconductor region is 8 or more, and the second group III nitride semiconductor region is made of a group III nitride semiconductor having a band gap greater than or equal to that of GaN, and the second group III nitride semiconductor The thickness of the region is 550 nm or more, and the first InGaN layer of the center semiconductor region is formed of the first group III nitride half layer. A first heterointerface is formed in contact with the body region, and the second InGaN layer in the center semiconductor region forms a second heterointerface in contact with the second group III nitride semiconductor region, The group III nitride semiconductor region includes misfit dislocations at the second heterointerface, and the center semiconductor region is grown without supplying a p-type dopant and is substantially free of the p-type dopant.

According to another aspect of the present invention, there is provided a method for manufacturing a nitride semiconductor light emitting device, comprising: a center semiconductor region including a plurality of InGaN layers on a semipolar main surface of a group III nitride semiconductor region including an n-type cladding layer; Forming a second group III nitride semiconductor region including a p-type cladding layer on the semipolar main surface of the center semiconductor region and the first group III nitride semiconductor region, and A center semiconductor region is provided between the semipolar main surface of the first group III nitride semiconductor region and the second group III nitride semiconductor region, and the center semiconductor region has a band gap less than or equal to a band gap of GaN. The center semiconductor region includes an active layer, a first InGaN layer, and a second InGaN layer, and the active layer includes the first InGaN layer and the second InGaN layer. The active layer includes one or a plurality of InGaN well layers, and the plurality of InGaN layers include the InGaN well layer, the first InGaN layer, and the second InGaN layer, and the center semiconductor. In the growth of the region, the second InGaN layer is grown without supplying a p-type dopant, and each of the plurality of InGaN layers in the center semiconductor region has an indium composition and a layer thickness, and the unit in the layer thickness is The sum of the product of the indium composition and the layer thickness of each InGaN layer over the center semiconductor region is 8 or more, and the second group III nitride semiconductor region has a band gap of GaN or more. A group III nitride semiconductor having a band gap, and the thickness of the second group III nitride semiconductor region is 550 nm or more; The first InGaN layer in the center semiconductor region is in contact with the first group III nitride semiconductor region to form a first heterointerface, and the second InGaN layer in the center semiconductor region is the second group III nitride. A second heterointerface is formed in contact with the semiconductor region, the second group III nitride semiconductor region includes misfit dislocations at the second heterointerface, and the center semiconductor region supplies a p-type dopant. Grown substantially free of the p-type dopant.

A method for fabricating a nitride semiconductor light emitting device according to another aspect of the present invention includes a step of preparing a plurality of substrates, and a group III nitridation by changing the supply start timing of the p-type dopant on the plurality of substrates. Forming an epitaxial substrate comprising a first group III nitride semiconductor region comprising an n-type clad layer, a center semiconductor region comprising a plurality of InGaN layers, and a second group III nitride semiconductor region comprising a p-type clad layer Processing the epitaxial substrate to form electrodes, forming a plurality of substrate products, forming a plurality of nitride semiconductor light emitting devices from the substrate products, and the plurality of nitrides A step of conducting an energization test for each of the semiconductor light emitting elements to estimate a difference in forward voltage before and after energization, and a p-type dopant based on the estimate A step of determining a supply start timing, and a step of fabricating a nitride semiconductor light emitting device using the determined timing, wherein the center semiconductor region has a band gap less than or equal to that of GaN. The center semiconductor region includes an active layer, a first InGaN layer, and a second InGaN layer, and the active layer is provided between the first InGaN layer and the second InGaN layer, and the active layer is one or A plurality of InGaN well layers, the plurality of InGaN layers including the InGaN well layer, the first InGaN layer, and the second InGaN layer; and the first InGaN layer in the center semiconductor region includes the first group III nitride semiconductor. A first heterointerface is formed in contact with the region, and the second InGa in the center semiconductor region Layer, the second hetero interface constitutes a form of contact with the first 2III nitride semiconductor region, the second 2III nitride semiconductor region comprises a misfit dislocations in the second hetero-interface.

A nitride semiconductor light emitting device according to one aspect of the present invention has a semipolar main surface made of a gallium nitride based semiconductor, and includes a first group III nitride semiconductor region including an n-type cladding layer, and the first group III nitride. A second group III nitride semiconductor region including a p-type cladding layer and a center semiconductor region including a plurality of InGaN layers and provided on the first group III nitride semiconductor region. The center semiconductor region is provided between the semipolar main surface of the first group III nitride semiconductor region and the second group III nitride semiconductor region, and the center semiconductor region has a band less than or equal to a band gap of GaN. The center semiconductor region includes an active layer, a first InGaN layer, a second InGaN layer, and a third InGaN layer, and the active layer includes the first GaN-based semiconductor having a gap. The active layer includes one or a plurality of InGaN well layers, the third InGaN layer is provided between the second InGaN layer and the active layer, and is provided between the nGaN layer and the second InGaN layer. The indium composition of the 2InGaN layer is larger than the indium composition of the third InGaN layer and smaller than the indium composition of the InGaN well layer, and the plurality of InGaN layers include the InGaN well layer, the first InGaN layer, the second InGaN layer, A third InGaN layer, wherein each of the plurality of InGaN layers in the center semiconductor region has an indium composition and a layer thickness, the unit of the layer thickness being nanometers, and the indium composition and the layer of each InGaN layer The sum of the product with the thickness over the center semiconductor region is 8 or more. The group III nitride semiconductor region is made of a group III nitride semiconductor having a band gap greater than or equal to the band gap of GaN, and the thickness of the group III nitride semiconductor region is 550 nm or more. The first InGaN layer is in contact with the first group III nitride semiconductor region to form a first heterointerface, and the second InGaN layer in the center semiconductor region is in contact with the second group III nitride semiconductor region. Forming a second heterointerface, wherein the second group III nitride semiconductor region includes misfit dislocations at the second heterointerface, and the center semiconductor region is grown without supplying a p-type dopant, In particular, it does not contain the p-type dopant.

The above object and other objects, features, and advantages of the present invention will be more readily apparent from the following detailed description of embodiments of the present invention that proceeds with reference to the accompanying drawings.

As described above, according to one aspect of the present invention, it is possible to provide a nitride semiconductor light emitting device having a structure that can reduce fluctuations in the forward voltage Vf due to energization, and according to another aspect of the present invention, A method of manufacturing this nitride semiconductor light emitting device can be provided.

FIG. 1 is a drawing schematically showing a structure according to a nitride semiconductor light emitting device according to the present embodiment. FIG. 2 shows the structure of the nitride semiconductor laser LC. FIG. 3 is a diagram showing characteristics obtained by measuring the forward voltage change over a long period of time by continuously energizing the nitride semiconductor laser LC. FIG. 4 is a drawing schematically showing the mechanism of resistance variation of the p-type semiconductor layer. FIG. 5 is a diagram showing a structure related to region separation that separates a non-radiative recombination region of overflowed electrons from a region having a high residual hydrogen concentration. FIG. 6 is a drawing showing a structure related to region separation that separates a non-radiative recombination region of overflowed electrons from a region having a high residual hydrogen concentration. FIG. 7 is a drawing showing the magnesium concentration and the hydrogen concentration in a gallium nitride based semiconductor. FIG. 8 is a drawing showing a ridge-type nitride semiconductor laser fabricated on a semipolar surface. FIG. 9 is a drawing showing main steps in a method of manufacturing a nitride semiconductor laser. FIG. 10 is a diagram schematically illustrating the structure I, the structure J, and the structure K according to the embodiment. FIG. 11 is a diagram illustrating a structure L and a structure M according to the embodiment. FIG. 12 is a diagram illustrating a structure P according to the embodiment. FIG. 13 is a diagram illustrating a structure Q according to the embodiment. FIG. 14 is a drawing showing an example relating to the sum of products of In composition and film thickness. FIG. 15 is a drawing showing main steps in a method of manufacturing a nitride semiconductor laser.

A nitride semiconductor light emitting device according to an embodiment has (a) a semipolar main surface made of a gallium nitride based semiconductor and includes a first group III nitride semiconductor region including an n-type cladding layer; A second group III nitride semiconductor region including a p-type cladding layer provided on the first group III nitride semiconductor region; and (c) a plurality of InGaN layers provided on the first group III nitride semiconductor region. A center semiconductor region. The center semiconductor region is provided between the semipolar main surface of the first group III nitride semiconductor region and the second group III nitride semiconductor region, and the center semiconductor region has a band gap less than or equal to a band gap of GaN. The center semiconductor region includes an active layer, a first InGaN layer, and a second InGaN layer, and the active layer is provided between the first InGaN layer and the second InGaN layer, The active layer includes one or a plurality of InGaN well layers, the plurality of InGaN layers include the InGaN well layer, the first InGaN layer, and the second InGaN layer, and each of the plurality of InGaN layers in the center semiconductor region includes Having an indium composition and a layer thickness, the unit of the layer thickness being nanometers, and The total sum of the product of the rhodium composition and the layer thickness over the center semiconductor region is 8 or more, and the second group III nitride semiconductor region is made of a group III nitride semiconductor having a band gap greater than or equal to the band gap of GaN. The thickness of the second group III nitride semiconductor region is 550 nm or more, and the first InGaN layer in the center semiconductor region forms a first heterointerface by making contact with the first group III nitride semiconductor region The second InGaN layer in the center semiconductor region is in contact with the second group III nitride semiconductor region to form a second heterointerface, and the second group III nitride semiconductor region is in the second heterointerface Including misfit dislocations, the center semiconductor region is grown without supplying a p-type dopant and is substantially free of the p-type dopant.

A nitride semiconductor light emitting device according to an embodiment has (a) a semipolar main surface made of a gallium nitride based semiconductor and includes a first group III nitride semiconductor region including an n-type cladding layer; A second group III nitride semiconductor region including a p-type cladding layer provided on the first group III nitride semiconductor region; and (c) a plurality of InGaN layers provided on the first group III nitride semiconductor region. A center semiconductor region. The center semiconductor region is provided between the semipolar main surface of the first group III nitride semiconductor region and the second group III nitride semiconductor region, and the center semiconductor region has a band gap less than or equal to a band gap of GaN. The center semiconductor region includes an active layer, a first InGaN layer, a second InGaN layer, and a third InGaN layer, and the active layer is between the first InGaN layer and the second InGaN layer. The active layer includes one or a plurality of InGaN well layers, and the third InGaN layer is provided between the second InGaN layer and the active layer. The indium composition of the second InGaN layer is larger than the indium composition of the third InGaN layer and smaller than the indium composition of the InGaN well layer. The plurality of InGaN layers include the InGaN well layer, the first InGaN layer, the second InGaN layer, and the third InGaN layer, and each of the plurality of InGaN layers in the center semiconductor region has an indium composition and a layer thickness. The unit in the layer thickness is nanometer. The sum of the product of the indium composition and the layer thickness of each InGaN layer over the center semiconductor region is 8 or more, and the second group III nitride semiconductor region has a bandgap greater than or equal to the bandgap of GaN. The second group III nitride semiconductor region has a thickness of 550 nm or more, and the first InGaN layer of the center semiconductor region is in contact with the first group III nitride semiconductor region. 1 hetero interface is formed, the second InGaN layer of the center semiconductor region is in contact with the second group III nitride semiconductor region to form a second hetero interface, and the second group III nitride semiconductor region is Including a misfit dislocation at the second heterointerface, and the center semiconductor region is grown without supplying a p-type dopant, substantially forming the p-type dopant. It does not contain.

According to the nitride semiconductor light emitting device described above, the center semiconductor region includes a plurality of InGaN layers, and the gallium nitride based semiconductor in the center semiconductor region has a band gap less than or equal to the band gap of GaN, while the group III nitride semiconductor The region is made of a gallium nitride-based semiconductor having a band gap greater than or equal to that of GaN.

When the sum of the product of the indium composition and the layer thickness in each of the plurality of InGaN layers in the center semiconductor region over the center semiconductor region is 8 or more and the thickness of the group III nitride semiconductor region is 550 nm or more, In the second heterointerface where the second InGaN layer in the center semiconductor region is in contact with the second group III nitride semiconductor region, the second group III nitride semiconductor region includes misfit dislocations.

Electrons from the n-type cladding layer are supplied to the active layer. When carrier overflow occurs, electrons overflow from the active layer, and the electrons propagate through the second InGaN layer in the center semiconductor region and reach the second heterointerface. Since misfit dislocations are formed in the second group III nitride semiconductor region at the second hetero interface, the overflow electrons that have reached the second hetero interface disappear due to non-radiative recombination via the misfit dislocations.

Energy is released in the non-radiative recombination process of overflow electrons. According to the inventors' knowledge, the energy generated in at least some of the many non-emissive recombination is of a magnitude that can generate a reaction in which the activated p-type dopant recombines with residual hydrogen. For this reason, the residual hydrogen liberated in the Group III nitride semiconductor region is bonded to the already activated p-type dopant and acts as an acceptor killer. Since the center semiconductor region is grown without supplying the p-type dopant, the center semiconductor region is substantially free of the p-type dopant. On the other hand, part or all of the Group III nitride semiconductor region contains both the activated p-type dopant and residual hydrogen. However, in this nitride semiconductor light emitting device, since hydrogen does not remain on both sides of the second hetero interface, the possibility that the residual hydrogen receives energy resulting from recombination at the second hetero interface can be reduced. Moreover, since overflow electrons recombine non-radiatively through misfit dislocations, the ratio of non-radiative recombination in the group III nitride semiconductor region is reduced. Therefore, the rate at which residual hydrogen in the Group III nitride semiconductor region receives energy resulting from non-radiative recombination can be reduced.

Therefore, it is possible to reduce the recombination of the activated p-type dopant with the released residual hydrogen, and it is possible to reduce the increase in the specific resistance of the Group III nitride semiconductor region due to the recombination.

In the nitride semiconductor light emitting device according to one embodiment, the semipolar principal surface of the first group III nitride semiconductor region is 40 with respect to the c-axis of the gallium nitride semiconductor of the first group III nitride semiconductor region. It can be inclined at an angle in the range of not less than 80 degrees and not more than 80 degrees or in the range of not less than 100 degrees and not more than 170 degrees.

According to this nitride semiconductor light emitting device, the semipolar main surface is in the range of 40 degrees or more and 80 degrees or less, or in the range of 100 degrees or more and 170 degrees or less with respect to the c-axis of the gallium nitride semiconductor in the group III nitride semiconductor region. When inclined at the inner angle, hydrogen taken in during the growth of the Group III nitride semiconductor region is difficult to escape. For this reason, it is not easy to reduce the total amount of hydrogen remaining in the nitride semiconductor light emitting device.

In the nitride semiconductor light emitting device according to one embodiment, the density of the misfit dislocations is preferably 5 × 10 ³ cm ⁻¹ or more.

According to this nitride semiconductor light emitting device, the misfit dislocation density at the heterointerface between the group III nitride semiconductor region and the center semiconductor region is 5 × 10 ³ cm ⁻¹ or more. The misfit dislocation at the heterointerface causes the electrons overflowing from the active layer to disappear through a non-luminescent process.

In the nitride semiconductor light emitting device according to one embodiment, the misfit dislocation density is preferably 5 × 10 ⁵ cm ⁻¹ or less.

According to this nitride semiconductor light emitting device, the misfit dislocation density at the heterointerface related to the Group III nitride semiconductor region is in an allowable range with respect to device characteristics. In addition, a significant decrease in the concentration of holes reaching the light emitting layer can be avoided.

In the nitride semiconductor light emitting device according to one embodiment, the indium composition of the second InGaN layer is 0.015 or more and preferably 0.055 or less. At this time, it becomes easy to satisfy the conditions for introducing misfit dislocations into the Group III nitride semiconductor at the second heterointerface. In the nitride semiconductor light emitting device according to the present invention, the hydrogen concentration in the second InGaN layer may be 1 × 10 ¹⁷ cm ⁻³ or less. According to this nitride semiconductor light emitting device, when the second InGaN layer is grown as an undoped layer, the hydrogen concentration in the second InGaN layer is 1 × 10 ¹⁷ cm ⁻³ or less.

In the nitride semiconductor light emitting device according to one embodiment, the second InGaN layer in the center semiconductor region is in contact with the first gallium nitride based semiconductor layer in the second group III nitride semiconductor region to form the second heterointerface. And the first gallium nitride based semiconductor layer in the second group III nitride semiconductor region may be GaN or AlGaN.

According to this nitride semiconductor light emitting device, the second heterointerface is composed of a first gallium nitride based semiconductor layer such as GaN or AlGaN and a second InGaN layer. Since the strain at the second heterointerface increases, it becomes easy to control the location where the misfit dislocation is introduced to the second heterointerface.

In the nitride semiconductor light emitting device according to an embodiment, the emission wavelength of the active layer may be in the wavelength range of 500 nm or more and 570 nm or less.

According to this nitride semiconductor light emitting device, the active layer that requires a low emission wavelength in the wavelength range of 500 nm or more and 570 nm or less requires an InGaN well layer having a large indium composition.

A nitride semiconductor light emitting device according to an embodiment includes the first group III nitride semiconductor region, the center semiconductor region, and the second group III nitride semiconductor region, and is semipolar made of a group III nitride semiconductor A substrate having a main surface can be further provided. The semipolar principal surface of the substrate may be inclined with respect to a reference plane orthogonal to the c-axis of the group III nitride semiconductor, and the second heterointerface may be inclined with respect to the reference plane.

According to this nitride semiconductor light emitting device, the inclination of the second heterointerface is caused by the inclination of the semipolar main surface of the substrate on which the first group III nitride semiconductor region, the center semiconductor region, and the second group III nitride semiconductor region are mounted. Adjusted.

In the nitride semiconductor light emitting device according to one embodiment, the group III nitride semiconductor of the substrate is preferably made of GaN. According to this nitride semiconductor light emitting device, when the group III nitride semiconductor of the substrate is made of GaN, the generation of misfit dislocations can be easily controlled.

In the nitride semiconductor light emitting device according to an embodiment, the semipolar principal surface of the group III nitride semiconductor of the substrate is 40 degrees or more and 80 degrees or less with respect to the c-plane of the gallium nitride semiconductor, or 100 The c-plane is inclined in a direction from the c-axis of the group III nitride semiconductor of the substrate toward the m-axis or a-axis of the group III nitride semiconductor. Can be made.

According to the nitride semiconductor light emitting device, the semipolar main surface of the substrate has an angle within a range of 40 degrees to 80 degrees or a range of 100 degrees to 170 degrees with respect to the c-axis of the group III nitride semiconductor of the substrate. Incline at. The use of the semipolar main surface of the substrate facilitates not only the generation of misfit dislocations at the second heterointerface, but also the control of misfit dislocation generation at the first heterointerface. Misfit dislocations are generated by the c-plane acting as a slip plane, and the shear stress applied to the c-plane increases at an inclination angle of 45 degrees.

In the nitride semiconductor light emitting device according to one embodiment, the semipolar principal surface of the first group III nitride semiconductor region is inclined in a range of not less than 63 degrees and not more than 80 degrees with reference to the c-plane of the gallium nitride semiconductor. Can be tilted at a corner.

According to this nitride semiconductor light emitting device, when the semipolar main surface of the group III nitride semiconductor region is inclined at an inclination angle in the range of not less than 63 degrees and not more than 80 degrees with respect to the c-plane of the gallium nitride semiconductor. InGaN grown on this semipolar main surface has excellent In incorporation and is excellent in composition uniformity. Also, in this angular range, InGaN can have a low point defect density.

In the nitride semiconductor light emitting device according to an embodiment, the second InGaN layer in the center semiconductor region is in contact with the GaN layer in the second group III nitride semiconductor region to form the second heterointerface. good.

According to this nitride semiconductor light emitting device, the second heterointerface related to misfit dislocation is formed by contact between the second InGaN layer in the center semiconductor region and the GaN layer in the second group III nitride semiconductor region.

In the nitride semiconductor light emitting device according to an embodiment, the second InGaN layer in the center semiconductor region is in contact with the AlGaN layer in the second group III nitride semiconductor region to form the second heterointerface. Can do.

According to this nitride semiconductor light emitting device, the second heterointerface related to misfit dislocation is formed by contact between the second InGaN layer in the center semiconductor region and the AlGaN layer in the second group III nitride semiconductor region.

In the nitride semiconductor light emitting device according to one embodiment, the second group III nitride semiconductor region includes a semiconductor layer that forms a junction with the second InGaN layer of the center semiconductor region at the second hetero interface, and the semiconductor layer includes: , Grown without supplying a p-type dopant and substantially free of the p-type dopant.

According to this nitride semiconductor light emitting device, the semiconductor layer in the group III nitride semiconductor region related to the second heterointerface is grown without supplying a p-type dopant, and is grown as an undoped layer. Does not contain the p-type dopant. At this time, since the residual hydrogen in the semiconductor layer of the second group III nitride semiconductor is reduced, the proportion of residual hydrogen that receives energy due to recombination at the second hetero interface can be reduced, and the second group III nitride semiconductor by energization can be reduced. An increase in specific resistance can be reduced.

In the nitride semiconductor light emitting device according to one embodiment, a p-type dopant may be added throughout the second group III nitride semiconductor region. According to this nitride semiconductor light emitting device, since the p-type dopant is added throughout the second group III nitride semiconductor region, the specific resistance of the second group III nitride semiconductor region can be reduced.

In the nitride semiconductor light emitting device according to one embodiment, the second group III nitride semiconductor region includes a first semiconductor layer and a second semiconductor layer, and the second semiconductor layer of the second group III nitride semiconductor region includes: The first semiconductor layer is provided between the first semiconductor layer and the p-type cladding layer, and the first semiconductor layer forms a junction with the second InGaN layer in the center semiconductor region at the second heterointerface, The layer may be grown without adding a p-type dopant, and the second semiconductor layer may be grown while adding a p-type dopant.

According to the nitride semiconductor light emitting device, in the growth of the first semiconductor layer and the second semiconductor layer in the second group III nitride semiconductor region, the first semiconductor layer is grown without adding a p-type dopant and the first semiconductor layer is grown. The two semiconductor layers become p-type dopant doping layers. Therefore, the semiconductor region grown while adding the p-type dopant can be separated from the second heterointerface.

In the nitride semiconductor light emitting device according to one embodiment, the second semiconductor layer has a band gap smaller than that of the first semiconductor layer and the p-type cladding layer, and the thickness of the first semiconductor layer is The thickness of the second semiconductor layer may be smaller.

According to this nitride semiconductor light emitting device, since the first semiconductor layer forms a junction with the second InGaN layer at the second hetero interface and the band gap of the first semiconductor layer is larger than the band gap of the second semiconductor layer, the second InGaN layer It can function as a barrier against light and block electrons overflowing from the active layer. Since non-radiative recombination of overflow electrons in the group III nitride semiconductor is reduced, it is possible to suppress recombination of residual hydrogen in the group III nitride semiconductor with the p-type dopant. Further, by making the thickness of the second semiconductor layer thinner than the thickness of the first semiconductor layer, an increase in resistance and a decrease in crystal quality can be suppressed even if the function of the barrier is enhanced.

In the nitride semiconductor light emitting device according to one embodiment, the first semiconductor layer is made of AlGaN, the aluminum composition of the AlGaN is in the range of 0.02 to 0.06, and the film thickness of the first semiconductor layer is It can be 5 nm or more and 30 nm or less.

According to this nitride semiconductor light emitting device, when the aluminum composition of AlGaN is 0.02 or more, the first semiconductor layer is effective in controlling the misfit dislocation density. Moreover, when the film thickness of the first semiconductor layer is 5 nm or more, the first semiconductor layer is effective in controlling the misfit dislocation density. When the aluminum composition of AlGaN is 0.06 or less, the deterioration of the crystal quality of the first semiconductor layer is suppressed. Moreover, when the film thickness of the first semiconductor layer is 30 nm or less, the deterioration of the crystal quality of the first semiconductor layer is suppressed. The increase in Al composition may cause a decrease in p-type characteristics due to an increase in oxygen concentration for the semipolar surface of the present case. This decrease in p-type characteristics causes a new cause for an increase in electron overflow, so the Al composition is preferably 0.06 or less.

In the nitride semiconductor light emitting device according to an embodiment, the center semiconductor region includes a third InGaN layer between the active layer and the second InGaN layer, and a band gap of the second InGaN layer is a band gap of the InGaN well layer. The indium composition of the second InGaN layer may be between the indium composition of the InGaN well layer and the indium composition of the third InGaN layer.

According to this nitride semiconductor light emitting device, since the band gap of the second InGaN layer is smaller than the band gap of the third InGaN layer and the second group III nitride semiconductor region, the second InGaN layer can capture overflow electrons. Since non-radiative recombination of overflow electrons in the group III nitride semiconductor is reduced, it is possible to suppress recombination of residual hydrogen in the group III nitride semiconductor with the p-type dopant.

In the nitride semiconductor light emitting device according to one embodiment, the In composition of the second InGaN layer is in the range of 0.05 to 0.1, and the film thickness of the second InGaN layer is in the range of 2 nm to 10 nm. Can do.

According to this nitride semiconductor light-emitting device, when the In composition of the second InGaN layer is 0.05 or more, electron capture and misfit dislocation density can be controlled effectively. In addition, when the thickness of the second InGaN layer is 2 nm or more, electron capture and misfit dislocation density can be effectively controlled. When the In composition of the second InGaN layer exceeds 0.1, the crystallinity of the InGaN layer may be deteriorated, and it is not easy to control the misfit dislocation density within a desired range. Further, when the thickness of the second InGaN layer exceeds 10 nm, the crystallinity of the InGaN layer may be deteriorated, and it is not easy to control the misfit dislocation density within a desired range. The second InGaN layer may be grown while adding a p-type dopant. At this time, there is a possibility that the second InGaN layer is increased in resistance by energization. Since a film thickness that is too thick may degrade the electrical characteristics of the device, the film thickness of the second InGaN layer is preferably 10 nm or less.

In the nitride semiconductor light emitting device according to one embodiment, the active layer may include an InGaN and GaN barrier layer having a smaller In composition than the well layer. According to this nitride semiconductor light emitting device, the barrier layer of GaN or InGaN can contribute to the formation of a good quantum well structure.

In the nitride semiconductor light emitting device according to an embodiment, the barrier layer of the active layer may include GaN having a thickness of 20 nm or less.

According to this nitride semiconductor light emitting device, the barrier layer of GaN (or InGaN) having a thickness of 20 nm or less does not substantially affect the ease of control of misfit dislocations.

Another embodiment relates to a method of fabricating a nitride semiconductor light emitting device. The method includes: (a) forming a center semiconductor region including a plurality of InGaN layers on a semipolar main surface of a first group III nitride semiconductor region including an n-type cladding layer; and (b) the center semiconductor region. And forming a second group III nitride semiconductor region including a p-type cladding layer on the semipolar main surface of the first group III nitride semiconductor region, wherein the center semiconductor region has a band gap of GaN or less. The center semiconductor region includes an active layer, a first InGaN layer, and a second InGaN layer, and the active layer is provided between the first InGaN layer and the second InGaN layer. The active layer includes one or a plurality of InGaN well layers, and the plurality of InGaN layers include the InGaN well layer, the first InGaN layer, and the first InGaN layer. In the growth of the center semiconductor region, the second InGaN layer is grown without supplying a p-type dopant, and each of the plurality of InGaN layers in the center semiconductor region has an indium composition and a layer thickness. The unit in the layer thickness is nanometer, and the sum of the product of the indium composition and the layer thickness of each InGaN layer over the center semiconductor region is 8 or more, and the second group III nitride semiconductor region Is made of a group III nitride semiconductor having a band gap greater than or equal to the band gap of GaN, the thickness of the second group III nitride semiconductor region is 550 nm or more, and the second InGaN layer of the center semiconductor region is A second heterointerface is formed by making contact with the group III nitride semiconductor region, and the group III nitride semiconductor region is formed in the second heterointerface; Includes misfit dislocations at the center semiconductor region is grown without supplying a p-type dopant, substantially free of the p-type dopant.

According to the method for manufacturing the nitride semiconductor light emitting device (hereinafter referred to as “manufacturing method”), the center semiconductor region includes a plurality of InGaN layers, and the gallium nitride semiconductor in the center semiconductor region has a band gap of GaN or less. While having a band gap, the Group III nitride semiconductor region is made of a gallium nitride based semiconductor having a band gap greater than or equal to that of GaN.

When the sum of the product of the indium composition and the layer thickness in each of the plurality of InGaN layers in the center semiconductor region over the center semiconductor region is 8 or more and the thickness of the group III nitride semiconductor region is 550 nm or more, In the second heterointerface where the second InGaN layer in the center semiconductor region is in contact with the second group III nitride semiconductor region, the second group III nitride semiconductor region includes misfit dislocations.

Electrons from the n-type cladding layer are supplied to the active layer. When carrier overflow occurs, electrons overflow from the active layer, and the electrons propagate through the second InGaN layer in the center semiconductor region and reach the second heterointerface. Since misfit dislocations are formed in the second group III nitride semiconductor region at the second hetero interface, the overflow electrons that have reached the second hetero interface disappear due to non-radiative recombination via the misfit dislocations.

Overflow electrons emit energy through the process of non-radiative recombination. According to the inventors' knowledge, the energy generated in at least some of the non-radiative recombination that occurs in large numbers is large enough to generate a reaction in which the activated p-type dopant recombines with residual hydrogen. For this reason, the residual hydrogen liberated in the Group III nitride semiconductor region is bonded to the already activated p-type dopant and acts as an acceptor killer. Since the center semiconductor region is grown without supplying the p-type dopant, the center semiconductor region is substantially free of the p-type dopant. On the other hand, part or all of the Group III nitride semiconductor region contains both the activated p-type dopant and residual hydrogen. However, in this nitride semiconductor light emitting device, since a large amount of hydrogen does not remain on both sides of the second hetero interface, the possibility that the residual hydrogen receives energy resulting from recombination at the second hetero interface can be reduced. . Moreover, since overflow electrons recombine non-radiatively through misfit dislocations, the ratio of non-radiative recombination in the group III nitride semiconductor region is reduced. Therefore, the rate at which residual hydrogen in the Group III nitride semiconductor region receives energy resulting from non-radiative recombination can be reduced.

Therefore, it is possible to reduce the recombination of the activated p-type dopant with the released residual hydrogen, and it is possible to reduce the increase in the specific resistance of the Group III nitride semiconductor region due to the recombination.

In the manufacturing method according to another embodiment, the semipolar principal surface of the first group III nitride semiconductor region is 40 degrees or more with respect to the c-axis of the gallium nitride semiconductor of the first group III nitride semiconductor region. It can be inclined at an angle in the range of 80 degrees or less or in the range of 100 degrees to 170 degrees.

According to this manufacturing method, the semipolar main surface has an angle in the range of 40 degrees to 80 degrees or in the range of 100 degrees to 170 degrees with respect to the c-axis of the gallium nitride semiconductor in the group III nitride semiconductor region When it is inclined at, it is difficult for hydrogen taken in during the growth of the Group III nitride semiconductor region to escape. For this reason, it is not easy to reduce the total amount of hydrogen remaining in the nitride semiconductor light emitting device.

In the manufacturing method according to another embodiment, the density of the misfit dislocations may be 5 × 10 ³ cm ⁻¹ or more. According to this manufacturing method, when the misfit dislocation density at the heterointerface between the group III nitride semiconductor region and the center semiconductor region is 5 × 10 ³ cm ⁻¹ or more, the misfit dislocation at the heterointerface is The electrons overflowing from the light are lost through a non-light-emitting process.

In the manufacturing method according to another embodiment, the density of the misfit dislocations may be 5 × 10 ⁵ cm ⁻¹ or less. According to this fabrication method, the formation of misfit dislocations at the heterointerface related to the Group III nitride semiconductor region is acceptable with respect to device characteristics. In addition, a significant decrease in the concentration of holes reaching the light emitting layer can be avoided.

In the manufacturing method according to another embodiment, the indium composition of the second InGaN layer may be 0.015 or more and 0.055 or less. At this time, it becomes easy to satisfy the conditions for introducing misfit dislocations into the Group III nitride semiconductor at the second heterointerface. In the manufacturing method according to the present invention, the hydrogen concentration in the second InGaN layer may be 1 × 10 ¹⁷ cm ⁻³ or less. According to this fabrication method, the second InGaN layer is grown without providing a p-type dopant.

In the manufacturing method according to another embodiment, the p-type cladding layer in the second group III nitride semiconductor region can be bonded to the second InGaN layer in the center semiconductor region. According to this manufacturing method, the p-type cladding layer is grown so as to form a junction with the second InGaN layer.

In the manufacturing method according to another embodiment, in the growth of the second group III nitride semiconductor region, the first semiconductor region is grown from the start of the growth of the second group III nitride semiconductor region. A second semiconductor region is grown after the growth, and the second semiconductor region of the second group III nitride semiconductor region is provided between the first semiconductor region and the p-type cladding layer, and the first semiconductor region Forms a junction with the second InGaN layer of the center semiconductor region at the second heterointerface, and the second semiconductor region has a band gap smaller than the band gap of the first semiconductor region and the p-type cladding layer. be able to.

According to this manufacturing method, since the band gap of the first semiconductor region and the p-type cladding layer is larger than the band gap of the second semiconductor region, it can be easily introduced by controlling misfit dislocations at the second hetero interface. Become.

In the manufacturing method according to another embodiment, in the growth of the center semiconductor region, the second InGaN layer is grown without supplying a p-type dopant, and in the growth of the second group III nitride semiconductor region, the second group III group is grown. The third semiconductor region is grown without supplying the p-type dopant from the start of the growth of the nitride semiconductor region, and the fourth semiconductor region is grown while supplying the p-type dopant after the growth of the third semiconductor region, The third semiconductor region can form a junction with the second InGaN layer of the center semiconductor region at the second heterointerface.

According to this manufacturing method, in the growth of the second group III nitride semiconductor region, the p-type dopant is grown after the third semiconductor region is grown without supplying the p-type dopant from the start of the growth of the second group III nitride semiconductor region. Since the fourth semiconductor region is grown while supplying the p-type dopant, the semiconductor region grown while adding the p-type dopant can be separated from the second heterointerface.

In the manufacturing method according to another embodiment, in the growth of the second group III nitride semiconductor region, the second group III nitride semiconductor region is supplied while supplying a p-type dopant from the start of the growth of the second group III nitride semiconductor region. Can grow. According to this manufacturing method, the p-type dopant can be added over the entire group III nitride semiconductor region.

Yet another embodiment relates to a method of fabricating a nitride semiconductor light emitting device. In this method, (a) a step of preparing a plurality of substrates, and (b) an n-type cladding layer made of a group III nitride semiconductor is formed on the plurality of substrates by changing the supply start timing of the p-type dopant. Forming an epitaxial substrate comprising a first group III nitride semiconductor region including a center semiconductor region including a plurality of InGaN layers, and a second group III nitride semiconductor region including a p-type cladding layer; and (c) an electrode. Processing the epitaxial substrate to form a plurality of substrate products; (d) producing a plurality of nitride semiconductor light emitting devices from the substrate products; and (e) the plurality of the plurality of substrate products. A step of conducting an energization test for each of the nitride semiconductor light-emitting elements and evaluating the difference in forward voltage before and after energization; and (f) a supply start timing of p-type dopant based on the evaluation And a step of determining, the step of, fabricating a nitride semiconductor light emitting device using the timing determined (d). The center semiconductor region is made of a gallium nitride based semiconductor having a band gap equal to or less than that of GaN. The center semiconductor region includes an active layer, a first InGaN layer, and a second InGaN layer, and the active layer includes the first InGaN layer. The active layer includes one or a plurality of InGaN well layers, and the plurality of InGaN layers include the InGaN well layer, the first InGaN layer, and the second InGaN layer. The first InGaN layer in the center semiconductor region is in contact with the first group III nitride semiconductor region to form a first heterointerface, and the second InGaN layer in the center semiconductor region is the second group III nitride. A second heterointerface is formed in contact with the oxide semiconductor region, and the second group III nitride semiconductor region is in the second heter Including the misfit dislocations at the interface.

The knowledge of the present invention can be easily understood by considering the following detailed description with reference to the accompanying drawings shown as examples. Subsequently, embodiments of the present invention relating to a nitride semiconductor light emitting device and a method for manufacturing the nitride semiconductor light emitting device will be described with reference to the accompanying drawings. Where possible, the same parts are denoted by the same reference numerals.

FIG. 1 is a drawing showing a structure related to a nitride semiconductor light emitting device according to the present embodiment. FIG. 1 shows an XYZ coordinate system S and a crystal coordinate system CR. The crystal coordinate system CR has a c-axis, a-axis, and m-axis. Referring to FIG. 1, the nitride semiconductor light emitting device 11 has a gain waveguide structure, but may also have a ridge structure and other structures.

The nitride semiconductor light emitting device 11 includes a first group III nitride semiconductor region 13, a center semiconductor region 19, and a second group III nitride semiconductor region 17. The center semiconductor region 19 includes an active layer 15, a first InGaN layer 21a, and a second InGaN layer 25a, and the active layer 15 is provided between the first InGaN layer 21a and the second InGaN layer 25a. In one embodiment, the first InGaN layer 21a is in contact with the active layer 15, and the active layer 15 is in contact with the second InGaN layer 25a. The center semiconductor region 19 is provided on the first group III nitride semiconductor region 13. The second group III nitride semiconductor region 17 is provided on the center semiconductor region 19. The first group III nitride semiconductor region 13 has a semipolar main surface 13a made of a gallium nitride semiconductor. The active layer 15 has a semipolar main surface 15a made of a gallium nitride semiconductor. The second group III nitride semiconductor region 17 has a semipolar main surface 17a made of a gallium nitride based semiconductor.
The center semiconductor region 19 is provided between the semipolar main surface 13a of the first group III nitride semiconductor region 13 and the second group III nitride semiconductor region 17, and the center semiconductor region 19 has a band gap less than or equal to the band gap of GaN. It consists of a gallium nitride based semiconductor having The second group III nitride semiconductor region is made of a group III nitride semiconductor having a band gap equal to or larger than that of GaN. The center semiconductor region 19 includes a plurality of InGaN layers provided on the first group III nitride semiconductor region 13.

The first group III nitride semiconductor region 13 includes a first inner semiconductor layer 21 b and an n-type cladding layer 23. The first inner semiconductor layer 21 b is provided on the n-type cladding layer 23. In one embodiment, the n-type cladding layer 23 is in contact with the first inner semiconductor layer 21b. The center semiconductor region 19 is provided on the first inner semiconductor layer 21b. In one embodiment, the center semiconductor region 19 is in contact with the first group III nitride semiconductor region 13 (first inner semiconductor layer 21b). The semiconductor region from the first InGaN layer 21a to the first inner semiconductor layer 21b constitutes an n-side light guide layer. In the embodiment shown in FIG. 1, the light guide layer includes a first InGaN layer 21a and a first inner semiconductor layer 21b, and the first inner semiconductor layer 21b forms a junction with the first InGaN layer 21a. The first inner semiconductor layer 21b can be made of, for example, GaN.

The second group III nitride semiconductor region 17 includes a second inner semiconductor layer 25 b and a p-type cladding layer 27. The p-type cladding layer 27 is provided on the second inner semiconductor layer 25b. In one embodiment, the second inner semiconductor layer 25 b is in contact with the p-type cladding layer 27. The second inner semiconductor layer 25 b is provided on the center semiconductor region 19. In one embodiment, the center semiconductor region 19 is in contact with the second group III nitride semiconductor region 17 (second inner semiconductor layer 25b). The semiconductor region from the second InGaN layer 25a to the second inner semiconductor layer 25b constitutes a p-side light guide layer. In the embodiment shown in FIG. 1, the light guide layer includes a second InGaN layer 25a and a second inner semiconductor layer 25b, and the second inner semiconductor layer 25b forms a heterojunction with the second InGaN layer 25a. The second inner semiconductor layer 25b can be made of, for example, GaN, AlGaN, or the like.

The first inner semiconductor layer 21 b is provided between the active layer 15 and the n-type cladding layer 23. The first InGaN layer 21a is provided between the active layer 15 and the first inner semiconductor layer 21b. The second inner semiconductor layer 25 b is provided between the active layer 15 and the p-type cladding layer 27. The second InGaN layer 25a is provided between the active layer 15 and the second inner semiconductor layer 25b. The electrode 41 is provided on the second group III nitride semiconductor region 17 and makes contact with the surface of the second group III nitride semiconductor region 17. The first group III nitride semiconductor region 13, the center semiconductor region 19, and the second group III nitride semiconductor region 17 are sequentially arranged along the stacking axis Ax (the Z-axis method of the coordinate system S).

From the viewpoint of optical confinement, the first InGaN layer 21a, the first inner semiconductor layer 21b, the active layer 15, the second InGaN layer 25a, and the second inner semiconductor layer 25b constitute a core region 31, and the core region 31 is an n-type cladding layer. 23 and the p-type cladding layer 27. The n-type cladding layer 23, the core region 31 and the p-type cladding layer 27 constitute an optical waveguide structure.

The first inner semiconductor layer 21b constitutes the first heterojunction HJ1 with the first InGaN layer 21a. The n-type cladding layer 23 is made of a group III nitride semiconductor, and the first heterojunction HJ1 is zero from the reference plane Sc extending along the c-plane of the group III nitride semiconductor of the n-type cladding layer 23. It inclines with a large inclination angle Angle. In FIG. 1, the reference plane in the n-type cladding layer 23 is orthogonal to the axis indicating the c-axis direction of the crystal coordinate system CR (the axis indicated by the vector VC).

The second inner semiconductor layer 25b constitutes the second heterojunction HJ2 with the second InGaN layer 25a. The n-type cladding layer 23 is made of a group III nitride semiconductor, and the second heterojunction HJ2 is zero from the reference plane Sc extending along the c-plane of the group III nitride semiconductor of the n-type cladding layer 23. It inclines with a large inclination angle Angle.

The active layer 15 includes one or a plurality of well layers 33a. The well layers 33a can be made of, for example, a gallium nitride based semiconductor, and the well layers 33a can include, for example, an InGaN layer. The well layer 33a contains compressive strain.

The active layer 15 can include a plurality of well layers 33a and at least one barrier layer 33b, if necessary. A barrier layer 33b is provided between adjacent well layers 33a. The outermost layer of the active layer 15 can be a well layer. The barrier layer 33b is made of, for example, a gallium nitride based semiconductor, and the barrier layer 33b can include, for example, a GaN layer or an InGaN layer. The well layer 33a closest to the n-type cladding layer 23 in the active layer 15 forms a heterojunction with the first InGaN layer 21a. The well layer 33a closest to the p-type cladding layer 27 in the active layer 15 forms a heterojunction with the second InGaN layer 25a.

When the barrier layer 33b includes a GaN or InGaN barrier layer, the barrier layer 33b can contribute to the formation of a good quantum well structure. The barrier layer 33b can include GaN having a thickness of 20 nm or less as a total. A GaN layer having a thickness of 20 nm or less does not substantially affect the ease of controlling misfit dislocations.

As already described, the second group III nitride semiconductor region 27 is made of a group III nitride semiconductor having a band gap greater than or equal to the band gap of GaN. The thickness D17 of the second group III nitride semiconductor region 17 is 550 nm or more. In the present embodiment, the Group III nitride semiconductor region 17 is drawn so as to include the contact layer 29. However, when the contact layer 29 has a band gap equal to or larger than GaN, the contact layer 29 has the Group III nitride. It is included in the film thickness of the semiconductor region 17.

As already described, the center semiconductor region 19 includes the active layer 15, the first InGaN layer 21a, and the second InGaN layer 25a. The active layer 15 is provided between the first InGaN layer 21a and the second InGaN layer 25a. The active layer 15 includes one or a plurality of InGaN well layers. These InGaN layers include an InGaN well layer 33a, a first InGaN layer 21a, and a second InGaN layer 25a. When the barrier layer 33b is made of InGaN, the center semiconductor region 19 includes one or a plurality of InGaN barrier layers. Thus, the center semiconductor region 19 includes a plurality of InGaN layers provided on the first group III nitride semiconductor region 13. Each of the plurality of InGaN layers in the center semiconductor region 19 has an indium composition and a layer thickness. The unit in the layer thickness is nanometer, and the sum of the product of the indium composition and the layer thickness of each InGaN layer over the center semiconductor region 19 is 8 or more.

The first InGaN layer 21a in the center semiconductor region 19 is in contact with the first group III nitride semiconductor region 13 to form the first heterointerface HJ1. The second InGaN layer 25a in the center semiconductor region 19 is in contact with the second group III nitride semiconductor region 17 to form the second heterointerface HJ2. The group III nitride semiconductor region 17 includes misfit dislocations at the second heterointerface HJ2. The center semiconductor region 19 is grown without supplying a p-type dopant and is substantially free of the p-type dopant.

According to the nitride semiconductor light emitting device 11, the center semiconductor region 19 includes a plurality of InGaN layers 21a, 25a, and 33a, and the gallium nitride semiconductor in the center semiconductor region 19 has a band gap equal to or less than that of GaN. The group III nitride semiconductor region 17 is made of a gallium nitride based semiconductor having a band gap greater than or equal to that of GaN.

The sum of the product of the indium composition and the layer thickness in each of all InGaN layers (for example, InGaN layers 21a, 25a, 33a) in the center semiconductor region 19 is 8 or more over the entire center semiconductor region 19, and the second group III When the thickness of the nitride semiconductor region 17 is 550 nm or more, the second group III nitride is formed at the second hetero interface HJ2 in which the second InGaN layer 25a of the center semiconductor region 19 is in contact with the second group III nitride semiconductor region 17. The semiconductor region 17 includes misfit dislocations.

Electrons from the n-type cladding layer 23 are supplied to the active layer 15. When carrier overflow occurs, electrons overflow from the active layer 15, and the electrons propagate through the second InGaN layer 25 a in the center semiconductor region 19 and reach the second hetero interface HJ 2. Since misfit dislocations are formed in the second group III nitride semiconductor region 17 at the second heterointerface HJ2, the overflow electrons that have reached the second heterointerface HJ2 disappear due to non-radiative recombination via the misfit dislocations. To do.

Overflow electrons emit energy through the process of non-radiative recombination. According to the inventors' knowledge, the energy produced in at least a portion of non-radiative recombination is of a magnitude that can produce a reaction in which the activated p-type dopant recombines with residual hydrogen. For this reason, the residual hydrogen liberated in the second group III nitride semiconductor region 17 is bonded to the already activated p-type dopant and acts as an acceptor killer. Since the center semiconductor region 19 is grown without supplying the p-type dopant, the center semiconductor region 19 does not substantially contain the p-type dopant. On the other hand, part or all of the second group III nitride semiconductor region 17 includes both the activated p-type dopant and residual hydrogen. However, in this nitride semiconductor light emitting device 11, since a large amount of hydrogen does not remain on both sides of the second hetero interface HJ2, the residual hydrogen may receive energy due to recombination at the second hetero interface HJ2. Can be reduced. Further, since overflow electrons recombine non-radiatively through misfit dislocations, the ratio of non-radiative recombination in the group III nitride semiconductor region 17 is reduced. Therefore, the rate at which the residual hydrogen in the second group III nitride semiconductor region 17 receives energy due to non-radiative recombination can be reduced.

Therefore, the occurrence of recombination between the activated p-type dopant and the liberated residual hydrogen can be reduced, and the increase in the specific resistance of the Group III nitride semiconductor region due to the recombination can be reduced.

The nitride semiconductor light emitting device 11 can further include a substrate 39. The substrate 39 has a semipolar main surface 39a made of a group III nitride semiconductor. Semipolar principal surface 39a is inclined with respect to reference plane Sc perpendicular to the axis (axis Cx indicated by vector VC) extending in the c-axis direction of the group III nitride semiconductor. The inclination of the second heterointerface HJ2 is defined by the inclination of the semipolar main surface 39a of the substrate 39 on which the first group III nitride semiconductor region 13, the center semiconductor region 19, and the second group III nitride semiconductor region 17 are mounted. An electrode 43 is provided on the back surface 39 b of the substrate 39.

The angle formed by the semipolar main surface 39a and the reference surface Sc (substantially equal to the angle Angle) can be in the range of 40 degrees to 80 degrees or 110 degrees to 170 degrees. These angular ranges make it possible to distinguish the semipolar plane from the c-plane.

The first group III nitride semiconductor region 13, the center semiconductor region 19, and the second group III nitride semiconductor region 17 are mounted on the semipolar main surface 39a. The InGaN layer epitaxially grown on the substrate 39 forms a heterojunction with GaN, AlGaN or the like. The substrate 39 can be made of GaN. The InGaN layer epitaxially grown on the GaN substrate contains compressive strain. When the group III nitride semiconductor of the substrate 39 is made of GaN, the generation of misfit dislocations can be easily controlled.

In addition, the angle of inclination Angle can be in the range of 40 degrees to 80 degrees, or 100 degrees to 170 degrees with respect to the c-plane of the gallium nitride semiconductor. Further, the c-plane can be inclined in a direction from the c-axis of the group III nitride semiconductor of the substrate 39 toward the m-axis or a-axis of the group III nitride semiconductor. The use of the semipolar main surface of the substrate 39 facilitates not only the generation of misfit dislocations at the second heterointerface HJ2, but also the control of misfit dislocation generation at the first heterointerface HJ1. Misfit dislocations are generated by the c-plane acting as a slip plane, and the shear stress applied to the c-plane increases at an inclination angle of 45 degrees.

Furthermore, the semipolar main surface 39a of the substrate 39 is inclined at an inclination angle Angle in the range of not less than 63 degrees and not more than 80 degrees with respect to the c-plane of the group III nitride semiconductor. The inclination angle Angle can be in the range of 63 degrees to 80 degrees. The semipolar surface 39a having the inclination angle Angle makes it possible to grow homogeneous In incorporation and high In composition gallium nitride semiconductors. Further, when the semipolar principal surface 13a of the first group III nitride semiconductor region 13 is inclined at an inclination angle in the range of not less than 63 degrees and not more than 80 degrees with respect to the c-plane of the gallium nitride semiconductor, this semipolar principal surface InGaN grown on 13a has excellent In uptake and excellent compositional uniformity. Also, in this angular range, InGaN can have a low point defect density.

The active layer 15 can be provided so as to generate an emission spectrum having a peak wavelength within a range of 500 nm or more and 570 nm or less. An active layer that requires a low emission wavelength in the wavelength range of 500 nm or more and 570 nm or less requires an InGaN well layer having a large indium composition. The active layer 15 that generates an emission spectrum having a peak wavelength in the range of 500 nm or more and 570 nm or less is manufactured using a semipolar plane.

In the nitride semiconductor light emitting device 11, the semipolar principal surface 13 a of the first group III nitride semiconductor region 13 is 40 degrees or more and 80 degrees or less with respect to the c-axis of the gallium nitride semiconductor of the first group III nitride semiconductor region 13. Or an angle in the range of 100 degrees to 170 degrees. In the range of this inclination angle, hydrogen taken in during the growth of the second group III nitride semiconductor region 17 is difficult to escape. For this reason, it is not easy to reduce the total amount of hydrogen remaining in the nitride semiconductor light emitting device 11.

The density of misfit dislocations in the heterojunction HJ2 is preferably 5 × 10 ³ cm ⁻¹ or more. When the misfit dislocation density at the heterointerface HJ2 between the group III nitride semiconductor region 17 and the center semiconductor region 19 is 5 × 10 ³ cm ⁻¹ or more, the misfit dislocation at the heterointerface HJ2 overflows from the active layer 15 The emitted electrons can be lost through a non-light emitting process.

The density of misfit dislocations in the heterojunction HJ2 is preferably 5 × 10 ⁵ cm ⁻¹ or less. Formation of misfit dislocations at the heterointerface HJ2 related to the Group III nitride semiconductor region 17 is acceptable with respect to device characteristics. In addition, a significant decrease in the concentration of holes reaching the active layer 15 from the second group III nitride semiconductor region 17 can be avoided.

The hydrogen concentration in the second InGaN layer 25a can be 1 × 10 ¹⁷ cm ⁻³ or less. When the second InGaN layer 25a is grown undoped, the hydrogen concentration in the second InGaN layer 25a becomes 1 × 10 ¹⁷ cm ⁻³ or less.

Regarding the indium composition of the second InGaN layer 25a, when the indium composition of the second InGaN layer 25a is 0.015 or more, introduction of misfit dislocations at the heterointerface HJ2 between the group III nitride semiconductor region 17 and the center semiconductor region 19 And control becomes easier. Further, when the indium composition of the second InGaN layer 25a is 0.055 or less, it is possible to avoid the introduction of excessive misfit dislocations and the deterioration of the crystal quality of the second InGaN layer 25a. Further, in this range of indium composition, light confinement can be improved.

The density of misfit dislocations in the heterojunction HJ1 related to the group III nitride semiconductor region 13 is preferably 5 × 10 ³ cm ⁻¹ or more. When the misfit dislocation density at the heterointerface HJ1 between the group III nitride semiconductor region 13 and the center semiconductor region 19 is 5 × 10 ³ cm ⁻¹ or more, the group III nitride semiconductor region 17 and the center semiconductor region 19 The density of misfit dislocations at the heterointerface HJ2 can be controlled within a good range.

The density of misfit dislocations in the heterojunction HJ1 is preferably 5 × 10 ⁵ cm ⁻¹ or less. Formation of misfit dislocations at the heterointerface HJ1 related to the group III nitride semiconductor region 13 is acceptable with respect to device characteristics. In addition, a significant decrease in the concentration of electrons reaching the active layer 15 from the first group III nitride semiconductor region 13 can be avoided.

The second InGaN layer 25a in the center semiconductor region 19 is in contact with the first gallium nitride based semiconductor layer in the second group III nitride semiconductor region 17 to form the second heterointerface HJ2. The first gallium nitride based semiconductor layer of the second group III nitride semiconductor region 17 may be GaN or AlGaN.

Since the p-type dopant is added throughout the second group III nitride semiconductor region 17, the specific resistance of the second group III nitride semiconductor region 17 can be reduced. For example, the p-type dopant concentration of the p-type GaN guide layer is, for example, about 5 × 10 ¹⁷ to 5 × 10 ¹⁸ cm ⁻³ .

For the n-side light guide layer 21, the indium composition of the first InGaN layer 21a is 0.015 or more and preferably 0.055 or less. When the indium composition of the first InGaN layer 21a is 0.015 or more, the misfit dislocation density at the heterointerface HJ1 between the first group III nitride semiconductor region 13 and the center semiconductor region 19 and thus the second group III nitride semiconductor region 17 are increased. It becomes easy to introduce and control the misfit dislocation density at the hetero interface HJ2 between the center semiconductor region 19 and the center semiconductor region 19. Further, when the indium composition of the first InGaN layer 21a is 0.055 or less, it is possible to avoid the introduction of excessive misfit dislocations and the deterioration of the crystal quality of the first InGaN layer 21a. Further, in this range of indium composition, light confinement can be improved.

Referring to FIG. 1, structures A1, A2, A3, and A4 applicable to the nitride semiconductor light emitting device 11 are illustrated.
Structure A1.
The second inner semiconductor layer 25b of the second group III nitride semiconductor region 17 is composed of a GaN layer, and the GaN layer of the second inner semiconductor layer 25b is in contact with the second InGaN layer 25a of the center semiconductor region 19 to form a second heterogeneous layer. The interface HJ2 is configured.

According to this structure A1, the second terror interface related to misfit dislocations is formed by contact between the second InGaN layer 25a in the center semiconductor region and the GaN layer in the second group III nitride semiconductor region. GaN functions as a light guide layer.
The second inner semiconductor layer 25b includes a p-type dopant, and the second InGaN layer 25a is grown as an undoped layer with the second heterointerface HJ2 as a boundary.
Structure A2.
Similar to the structure A1, the GaN layer of the second inner semiconductor layer 25b is in contact with the second InGaN layer 25a to form the second heterointerface HJ2. According to this structure A2, the second terror interface related to misfit dislocations is formed by contact between the second InGaN layer 25a in the center semiconductor region 19 and the GaN layer in the second group III nitride semiconductor region 17.

From the viewpoint of the p-type dopant profile, the second InGaN layer 25a is grown as an undoped layer with the second heterointerface HJ2 as a boundary. Alternatively, a part of the second group III nitride semiconductor region 17 (for example, part or all of the second inner semiconductor layer 25b) is grown as an undoped layer, and the rest of the second group III nitride semiconductor region 17 (for example, a p-type cladding). Layer 27 and p-type contact layer 29) contain a p-type dopant and exhibit p conductivity. In this structure and other structures, in the embodiment in which the second group III nitride semiconductor region 17 includes an undoped layer, the thickness of the undoped layer is preferably 3 nm or more and 20 nm or less.

The second group III nitride semiconductor region 17 includes a substantially undoped first semiconductor layer and a p-type doped second semiconductor layer. The second semiconductor layer may be the same material as the first semiconductor layer or a different material. The first semiconductor layer is thinner than the second semiconductor layer. Since the first semiconductor layer forms a junction with the second InGaN layer 25a at the second heterointerface HJ2, and the band gap of the first semiconductor layer is larger than the band gap of the second semiconductor layer, the first semiconductor layer functions as a barrier to the second InGaN layer 25a and is active. The electrons overflowing from the layer 15 can be blocked. Since non-radiative recombination of overflow electrons in the second group III nitride semiconductor region 17 is reduced, it is possible to suppress residual hydrogen in the second group III nitride semiconductor region 17 from recombining with the p-type dopant. When the group III nitride semiconductor region 17 includes the first semiconductor layer as an undoped layer, the thickness of the undoped layer is preferably 3 nm or more and 20 nm or less. The hydrogen concentration in the vicinity of the second heterointerface HJ2 can be reduced without causing significant deterioration in carrier injection efficiency and resistance.
Structure A3.
The second inner semiconductor layer 25b of the second group III nitride semiconductor region 17 is composed of an AlGaN layer, and the AlGaN layer of the second inner semiconductor layer 25b is in contact with the second InGaN layer 25a of the center semiconductor region 19 to form a second heterojunction. The interface HJ2 is configured.

According to this nitride semiconductor light emitting device, the second terror interface related to misfit dislocation is formed by contact between the second InGaN layer 25a in the center semiconductor region 19 and the AlGaN layer in the second group III nitride semiconductor region 17. AlGaN acts as an electron blocking layer. The second group III nitride semiconductor region 17 may include a third inner semiconductor layer 25c in addition to the second inner semiconductor layer 25b. The third inner semiconductor layer 25c can include, for example, a GaN layer. The second inner semiconductor layer 25b is located between the second InGaN layer 25a and the third inner semiconductor layer 25c, and makes contact with the second InGaN layer 25a and the third inner semiconductor layer 25c. In this embodiment, since the band gap of the third inner semiconductor layer 25c is smaller than the cladding layer 27, the third inner semiconductor layer 25c is included in the light guide layer.

The aluminum composition of the AlGaN layer is preferably in the range of 0.02 to 0.06. The film thickness of the AlGaN layer can be 5 nm or more and 30 nm or less. When the aluminum composition of the AlGaN layer is 0.02 or more, this AlGaN layer is effective in controlling the misfit dislocation density. Further, when the thickness of the AlGaN layer is 5 nm or more, the AlGaN layer is effective in controlling the misfit dislocation density. When the aluminum composition of AlGaN is 0.06 or less, the deterioration of the crystal quality of the AlGaN layer is suppressed. Moreover, when the film thickness of the AlGaN layer is 30 nm or less, the deterioration of the crystal quality of the AlGaN layer is suppressed. The increase in Al composition may cause a decrease in p-type characteristics due to an increase in oxygen concentration for the semipolar surface of the present case. This decrease in the p-type characteristics causes a new cause for an increase in electron overflow.

From the viewpoint of the p-type dopant profile, the second InGaN layer 25a can be grown without adding the p-type dopant, and the second inner semiconductor layer 25b can be grown while adding the p-type dopant.

Alternatively, the second group III nitride semiconductor region 17 may include a substantially undoped first semiconductor layer and a p-type doped second semiconductor layer. Here, when the second group III nitride semiconductor region 17 includes the first semiconductor layer as the undoped layer, the thickness of the undoped layer is preferably 3 nm or more and 20 nm or less. The second semiconductor layer is provided between the first semiconductor layer and the p-type contact layer or cladding layer 27. The first semiconductor layer (for example, the second inner semiconductor layer 25b) forms a junction with the second InGaN layer 25a in the center semiconductor region 19 at the second heterointerface HJ2. The first semiconductor layer can be grown without adding a p-type dopant, and the second semiconductor layer can be grown while adding a p-type dopant. Therefore, the semiconductor grown while adding the p-type dopant can be separated from the second hetero interface, and the hydrogen concentration in the vicinity of the second hetero interface HJ2 can be reduced.

For example, the second InGaN layer 25a is undoped, and part or all of the second inner semiconductor layer 25b (for example, an AlGaN layer) is grown as undoped. When all of the second inner semiconductor layer 25b is grown as undoped, a part or all of the third inner semiconductor layer (for example, GaN layer) 25c can be grown as undoped.

Structure A4.
The center semiconductor region 19 can include a third InGaN layer 25d, and the third InGaN layer 25d is provided between the active layer 15 and the second InGaN layer 25a. The band gap of the third InGaN layer 25d is between the band gap of the InGaN well layer 33a and the band gap of the second InGaN layer 25a. The indium composition of the third InGaN layer 25d can be between the indium composition of the InGaN well layer 33a and the indium composition of the second InGaN layer 25a.

In this structure A4, since the band gap of the second InGaN layer 25a is smaller than that of the third InGaN layer 25d, the second InGaN layer 25a can capture overflow electrons. In addition, the second InGaN layer 25a having a small band gap can promote electron recombination.

Since the indium composition of the second InGaN layer 25a is larger than the indium composition of the third InGaN layer 25d, it is easy to introduce misfit dislocations at the second heterointerface HJ2. An appropriate amount of misfit dislocation density can provide a technical benefit by promoting electron recombination.

The In composition of the second InGaN layer 25a can be in the range of 0.05 to 0.1, and the film thickness of the second InGaN layer 25a can be in the range of 2 nm to 10 nm. When the In composition of the second InGaN layer 25a is 0.05 or more, electron capture and misfit dislocation density can be effectively controlled. In addition, when the thickness of the second InGaN layer 25a is 2 nm or more, electron capture and misfit dislocation density can be effectively controlled. When the In composition of the second InGaN layer 25a exceeds 0.1, the crystallinity of the InGaN layer may be deteriorated, and it is not easy to control the misfit dislocation density within a desired range. Further, when the thickness of the second InGaN layer 25a exceeds 10 nm, the crystallinity of the InGaN layer may be deteriorated, and it is not easy to control the misfit dislocation density in a desired range.

From the viewpoint of the p-type dopant profile, the third InGaN layer 25d can be grown without adding the p-type dopant, and the second inner semiconductor layer 25b can be grown while adding the p-type dopant. At this time, since the second InGaN layer 25a may be increased in resistance by energization, an excessively thick film thickness may deteriorate the electrical characteristics of the device, and it is preferable to use an InGaN layer of 10 nm or less.

Alternatively, the second group III nitride semiconductor region 17 includes a substantially undoped first semiconductor layer and a p-type doped second semiconductor layer. Here, the second semiconductor layer is provided between the first semiconductor layer and the p-type contact layer 29 or the cladding layer 27. The first semiconductor layer (for example, the second inner semiconductor layer 25b) forms a junction with the second InGaN layer 25a in the center semiconductor region 19 at the second heterointerface HJ2. The first semiconductor layer can be grown without adding a p-type dopant, and the second semiconductor layer can be grown while adding a p-type dopant. Therefore, the semiconductor grown while adding the p-type dopant can be separated from the second hetero interface, and the hydrogen concentration in the vicinity of the second hetero interface HJ2 can be reduced.

For example, the second InGaN layer 25a is undoped, and part or all of the second inner semiconductor layer 25b (for example, GaN layer) is grown as undoped.

The second InGaN layer 25a having a small band gap can be grown without adding a p-type dopant. However, the second InGaN layer 25a may be grown while adding a p-type dopant.

(Example 1)
FIG. 2 shows the structure of the nitride semiconductor laser LC. The nitride semiconductor laser LC uses a plane orientation (20-21) plane GaN substrate (hereinafter referred to as “m75 degrees off”). Referring to the nitride semiconductor laser LC, a p-type dopant is added partway through the p-side InGaN optical guide layer. Therefore, the hetero interface between the p-side InGaN optical guide layer and the p-type GaN guide layer is completely included in the p-type region, and a large concentration of p-type dopant and hydrogen exist on both sides of the hetero interface. In the nitride semiconductor laser LC, in the p-side InGaN optical guide layer containing the p-type dopant, electrons overflowed from the active layer recombine with holes by non-radiative recombination via defects in the InGaN optical guide layer.

FIG. 3 is a diagram showing characteristics obtained by measuring a change in forward voltage over a long period of time by continuously energizing the nitride semiconductor laser LC. In the nitride semiconductor laser LC shown in FIG. 2 using an m75 degree off-GaN substrate, as shown in FIG. 3, the forward voltage Vf is increased by energization. The inventors' experiments indicate that the increase in the forward voltage Vf is not a deterioration of the ohmic electrode on the anode side. Further experiments have shown that the resistance of the p-type semiconductor layer varies to increase. Such a variation in specific resistance has not been reported so far in the nitride semiconductor laser using the c-plane GaN substrate as far as the inventors know. From this point, it is considered that the phenomenon is unique to the semipolar plane.

FIG. 4 is a drawing schematically showing the mechanism of resistance variation of the p-type semiconductor layer. The p layer and the n layer form a junction, which shows the basic structure of the diode. As shown in part (a) of FIG. 4, in as-grown, the p-type dopant Mg is inactivated by hydrogen. In addition to this hydrogen, there are also hydrogen bonded to vacancies in the crystal and hydrogen located between lattices.

As shown in part (b) of FIG. 4, hydrogen is released from the p-type semiconductor region by some activation method, and the p-type semiconductor region becomes p-conductive and shows a low resistance. . The nitride semiconductor laser LC manufactured in this way generates green laser light when energized.

As shown in part (c) of FIG. 4, the electrons overflowing from the active layer to the p-type semiconductor region during energization disappear due to non-radiative recombination. At the time of disappearance, energy is given to hydrogen remaining in the p-type semiconductor region (for example, hydrogen bonded to vacancies or hydrogen located between lattices). Part of the hydrogen that has gained energy binds to Mg and deactivates the p-type dopant Mg again, and this deactivation increases the resistance of the p-type semiconductor region (high resistance).

(Example 2)
Based on the mechanism described above, a structure that can reduce the overflow of electrons, a structure of a p-type semiconductor region that does not deteriorate due to the overflow of electrons, and the like have been studied. From this study, in the nitride semiconductor laser using the semipolar plane, as shown in FIGS. 5 and 6, the region separation (the non-radiative recombination region of the overflowed electron and the region having a high residual hydrogen concentration are mutually separated. It was found that separation was good.

(A) of FIG. 5 is a drawing showing an example of the region separation structure E. The InGaN guide layer and the p-GaN guide layer form a heterointerface HJ, and misfit dislocations are formed at this interface. During the epitaxial growth, the p-type dopant is supplied from the heterointerface HJ. In the region separation structure E, since misfit dislocations are close to the p-type semiconductor layer, the decrease due to non-radiative recombination of hole (majority carrier on the p side) is very small.

(B) of FIG. 5 is a drawing showing an example of the region separation structure F. After the hetero interface HJ is formed, the supply of the p-type dopant is started from a position away from the hetero interface HJ. With this structure, the hetero interface HJ in which non-radiative recombination occurs can be separated from the p-type semiconductor layer having a large residual hydrogen concentration. A part of the holes from the p-type cladding layer passes through a region having a low p-type carrier concentration before being injected into the light-emitting layer, and is consumed for non-radiative recombination in this low carrier region. Therefore, it is better to control the misfit dislocation density at the heterointerface HJ.

FIG. 5C shows an example of the region separation structure G. The InGaN guide layer and the p-AlGaN layer form a heterointerface HJ, and misfit dislocations are formed at this interface. During the epitaxial growth, the p-type dopant is supplied from the heterointerface HJ. By using AlGaN, the misfit dislocation density can be easily controlled. In the region isolation structure G, it is preferable to lower the Mg concentration of the p-AlGaN layer (for example, the Mg concentration is 5 × 10 ¹⁷ cm ⁻³ or more and 5 × 10 ¹⁸ cm ⁻³ or less). The concentration can be reduced.

FIG. 5D shows an example of the region separation structure H.
The p-InGaN layer and the p-GaN guide layer constitute a heterointerface HJ, and misfit dislocations are formed at this interface. During epitaxial growth, the p-type dopant can be supplied from the heterointerface HJ. However, a p-type dopant may be added to the p-InGaN layer. In the region isolation structure H, recombination occurs both in the misfit dislocation of the heterointerface HJ and in the thin p-InGaN layer. In the form in which the p-type dopant is added to the p-InGaN layer, the p-InGaN layer itself may become high resistance due to the effect of Mg inactivation by hydrogen, but the band gap of this InGaN layer is the band on both sides. Since it is smaller than the gap and / or the p-InGaN layer is also thin, there is no variation that appears in the device characteristics.

FIG. 7 is a drawing showing a magnesium concentration (p-type dopant concentration) and a hydrogen concentration in a gallium nitride based semiconductor. The hydrogen concentration in the p-type semiconductor layer is approximately the same as the Mg concentration as-grown. For this reason, the hydrogen concentration after the activation treatment depends on the Mg concentration. Also, the hydrogen concentration after the treatment varies depending on the activation treatment method. As shown in the figure, hydrogen removal tends to be promoted when the annealing temperature is high or the processing atmosphere is vacuum.

(Example 3)
In a ridge type nitride semiconductor laser fabricated on the c-plane, it is not observed that the resistance of the p-type nitride semiconductor region increases when placed under a continuous energization state. On the other hand, in the ridge type nitride semiconductor laser 11a produced on the semipolar plane shown in FIG. 8, the resistance of the p type nitride semiconductor region increases when placed under a continuous energization state.

FIG. 9 is a drawing showing main steps in a method of manufacturing a nitride semiconductor laser. In step S101, a semipolar GaN substrate 10a is prepared. The main surface of this semipolar GaN substrate has a {20-21} plane. In the {20-21} plane, the GaN c-axis of the substrate is inclined at an angle of 75 degrees in the direction of the GaN m-axis. Crystal growth is performed by metal organic vapor phase epitaxy. In step S102, thermal cleaning of the GaN substrate is performed in a growth furnace. The thermal cleaning is performed in an atmosphere containing ammonia (NH ₃ ) and hydrogen (H ₂ ), and the heat treatment temperature is 1050 degrees Celsius. After this pretreatment, in step S103, the first group III nitride semiconductor region 10b is grown. First, in step S104, an n-type GaN layer is grown on the semipolar main surface of the GaN substrate. The growth temperature is 1050 degrees Celsius. In step S105, after the substrate temperature is lowered to 840 degrees Celsius, an n-type cladding layer is grown on the n-type GaN layer. In this embodiment, an n-type InAlGaN cladding layer having a thickness of 2 μm is grown as an n-type cladding layer. The n-type InAlGaN cladding layer has an In composition of 0.03 and an Al composition of 0.14. In step S106, an n-type GaN light guide layer is grown on the n-type InAlGaN cladding layer at a substrate temperature of 840 degrees Celsius.

Next, in step S107, the center semiconductor region 10c is grown. First, in step S108, an n-type InGaN light guide layer is grown so that the n-type GaN light guide layer forms a heterojunction. The In composition of the InGaN layer is 0.04. After forming the n-side inner semiconductor layer made of these light guide layers, an active layer is grown on the inner semiconductor layer in step S109. In this embodiment, as an active layer, an InGaN well layer is grown at a substrate temperature of 790 degrees Celsius in Step S110. The In composition of this InGaN layer is 0.30, and the thickness of the InGaN layer is 2.5 nm. Thereafter, in step S111, an InGaN layer is grown at a substrate temperature of 840 degrees Celsius. The In composition of the InGaN layer is 0.04, and the thickness of the InGaN layer is 2.5 nm. Next, an InGaN well layer is grown at a substrate temperature of 790 degrees Celsius. The In composition of this InGaN layer is 0.30, and the thickness of the InGaN layer is 2.5 nm. These growths complete the production of the active layer. For example, after the substrate temperature is raised to 840 degrees Celsius on the active layer, an undoped or Mg-doped InGaN light guide layer is grown on the active layer in Step S112. The In composition of this InGaN layer is 0.02.

Next, in step S113, the second group III nitride semiconductor region 10d is grown. In step S114, a p-type GaN light guide layer is grown so as to form a heterojunction with the undoped InGaN light guide layer. After forming the p-side inner semiconductor layer made of these light guide layers, in step S115, a p-type InAlGaN cladding layer having a thickness of 400 nm is grown on the inner semiconductor layer. The p-type InAlGaN cladding layer has an In composition of 0.02 and an Al composition of 0.07. After raising the substrate temperature to 1000 degrees Celsius, in step S116, a p-type GaN contact layer having a thickness of 50 nm is grown on the p-type InAlGaN cladding layer. An epitaxial substrate can be manufactured by these steps.

In step S117, a substrate product is produced from the epitaxial substrate. Photolithography, dry etching, and vacuum deposition are applied to the epitaxial substrate to fabricate a ridge-type gallium nitride semiconductor laser having a semiconductor ridge with a width of 2 μm and an optical resonator with a length of 600 μm.

In this fabrication, the second group III nitride semiconductor region is etched to form a semiconductor ridge. The semiconductor ridge is processed by dry etching.

In step S118, the upper surface and side surfaces of the semiconductor ridge are formed by processing by dry etching. After forming the semiconductor ridge, in step S119, an insulating film, for example, a silicon oxide film (specifically, SiO ₂ ) is formed. This insulating film covers the side surface of the semiconductor ridge and the surface of the light guide layer (surface formed by etching) and has an opening on the upper surface of the semiconductor ridge (contact surface showing semipolarity). An electrode is formed on the semiconductor ridge. In step S120, an anode electrode (for example, Ni / Au) is formed on the upper surface of the semiconductor ridge by vapor deposition. A pad electrode (for example, Ti / Au) is formed so as to cover the ohmic electrode. The back surface of the GaN substrate is polished to form a polished substrate having a substrate thickness of 80 μm. A cathode electrode (for example, Ti / Al) and a pad electrode (for example, Ti / Au) are formed over the entire polished surface of the GaN substrate. By these steps, a substrate product is produced.

After forming the electrodes, in step S121, the substrate product is cleaved to form an end face for the optical resonator (an end face different from the cleaved face). A dielectric multilayer film is formed on these end faces. In step S122, the dielectric multilayer film is made of SiO ₂ / TiO ₂ . In step S123, a semiconductor laser is manufactured by separating the laser bar. Through these steps, a semiconductor laser is fabricated on the semipolar GaN substrate {20-21} plane inclined at an angle of 75 degrees in the m-axis direction. This semiconductor laser can emit light in the 520 nm wavelength band.

Example 4
A laser structure that operates in the 520 nm wavelength band is fabricated on a GaN substrate having a semipolar principal surface inclined at an angle of 75 degrees in the m-axis direction (corresponding to the (20-21) plane). Pretreatment (thermal cleaning) is performed by holding the GaN substrate in an atmosphere of ammonia and hydrogen at a temperature of 1050 degrees Celsius for about 10 minutes. Thereafter, a Si-doped GaN layer is grown on the semipolar main surface of the GaN substrate at a temperature of 1050 degrees Celsius. After setting the substrate temperature to 840 degrees Celsius, a Si-doped InAlGaN cladding layer (In composition: 0.03, Al composition: 0.14) layer having a thickness of 2 μm is grown. A lower Si-doped GaN optical guide layer is grown at a substrate temperature of 840 degrees Celsius. Next, a lower Si-doped InGaN optical guide layer (In composition: 0.04) is grown. An active layer is grown on the light guide region. The active layer has a 2QW structure. The InGaN well layer is grown at a growth temperature of 790 degrees Celsius, its thickness is 3 nm, and its In composition is 0.3. The InGaN barrier layer is grown at a growth temperature of 840 degrees Celsius, its thickness is 2.5 nm, and its In composition is 0.04. A light guide region is grown on the active layer at a substrate temperature of 840 degrees Celsius. An upper undoped or Mg-doped InGaN light guide layer (In composition: 0.04) is grown. Then, an upper undoped or Mg-doped GaN light guide layer is grown. An Mg-doped InAlGaN cladding layer (In composition: 0.02, Al composition: 0.07) layer having a thickness of 400 nm is grown at a substrate temperature of 840 degrees Celsius. A 50 nm thick Mg-doped GaN contact layer is grown on the cladding layer at a temperature of 1000 degrees Celsius.

The Mg concentration in the p-side semiconductor region is as follows. Here, for example, the notation “2E + 18” indicates “2 × 10 ¹⁸ ”.
Upper GaN guide layer: 2E + 18 cm ⁻³ .
p-type InAlGaN cladding layer: 1E + 19 cm ⁻³ .
p-type GaN contact layer: 1E + 20 cm ⁻³ .
The hydrogen concentration of the undoped layer (InGaN or GaN) grown on the active layer including two InGaN well layers is a value below the detection limit in the evaluation of the secondary ion mass spectrometry (SIMS) method: [H] <7E + 16 cm ^{−. 3} .
Mg concentration of Mg-doped InGaN or GaN layer grown on an active layer including two InGaN well layers: 1.5E + 17 cm ⁻³ .
Evaluation of misfit dislocation density at the n-side InGaN / GaN interface by transmission electron microscope (TEM) image: 1.3E + 5 cm ⁻¹ .
Evaluation of misfit dislocation density at the p-side InGaN / GaN interface by transmission electron microscope (TEM) image: 8E + 4 cm ⁻¹ .
Misfit dislocations are introduced into the n-side heterojunction HJ1 as well as the p-side heterojunction HJ2 (approximately the same numerical range).

In the fabrication of the laser structure having the basic structure as described above, three types of laser structures that operate in the 520 nm band are fabricated by changing the timing of starting the supply of Mg dopant. In FIG. 10, the arrow indicates the Mg doping start position.

A ridge type laser diode having a width of 2 μm and a resonator length of 600 μm is manufactured by using photolithography, dry etching and vacuum deposition for the structures I, J and K. SiO2 is deposited on the side surface of the ridge processed by dry etching, and an ohmic electrode (Ni / Au) is deposited on the top surface of the ridge as a p-side electrode. A p-pad electrode (eg, Ti / Au) is deposited so as to cover them. After polishing the back surface of the GaN substrate to a thickness of 80 μm, the n-side electrode (for example, Ti / Au) is damaged, and an n-pad electrode (for example, Ti / Au) is deposited on the entire surface. By these steps, a substrate product is produced.

¡Laser bars are formed by separating substrate products. A dielectric multilayer film (SiO 2 / TiO 2) is formed on the cavity end face of the laser bar. On this laser bar, a laser chip is manufactured from the laser bar. When initial characteristics were evaluated by applying a pulse current to the laser diode manufactured in this way, the representative values in the measurement of the threshold value Ith of the structures I, J, and K were 60 mA, 60 mA, and 70 mA, respectively. It is. The threshold value Ith of structure K is slightly larger than structures I and J. The reason for this is considered that the Mg doping start position is away from the misfit dislocation site (position where non-radiative recombination occurs).

Next, an energization test for 100 hours is performed at an operating current of 200 mA and an environmental temperature Tc = 60 degrees Celsius. The forward voltage difference ΔV before and after the energization test of Structure I, Structure J, and Structure K is measured.
Structure, forward voltage difference ΔV.
Structure I, 0.4 volts.
Structure J, 0.08 volts.
Structure K, 0.06 volts.
Structure I (0.4 volts) is larger than the voltage values in Structure J and Structure K. This is presumably because the heterojunction in which non-radiative recombination occurs in misfit dislocation is the same as the Mg-added region with much residual hydrogen. The difference in forward voltage between the structures J and K is smaller than that in the structure I, which is considered to be caused by the heterojunction being the same as the Mg addition start position or away from the Mg addition region. As a good value, the distance from the misfit dislocation interface to the Mg doping start position is 20 nm or less. In this range, both the forward voltage change due to energization and the initial electrical characteristics can be satisfied.

(Example 5)
In the structure J in Example 4, the structure L and the structure M are manufactured by changing the Mg concentration of the p-side GaN guide layer. Structure L and structure M are shown in FIG.
Structure, Mg concentration [Mg], hydrogen concentration [H], forward voltage difference ΔV.
Structure J, 2E + 18 cm ⁻³ , 1.5E + 17 cm ⁻³ , 0.08 volts.
Structure L, 5E + 18 cm ⁻³ , 4E + 17 cm ⁻³ , 0.1 volts.
Structure M, 8E + 18 cm ⁻³ , 8E + 17 cm ⁻³ , 0.11 volts.
The H concentration also changes according to the Mg concentration.
Normally (for example, a laser diode on the c-plane), when magnesium (Mg) is increased in an appropriate range, the hole concentration is increased, so that the electron overflow is reduced. However, in the experiment related to the semipolar plane, the forward voltage difference ΔV after energization increases as the Mg concentration increases. The reason for this is considered to be that the increase in the hydrogen concentration, which is the source of the increase in the forward voltage after energization with the increase in the Mg concentration, is large.

From this result, it is preferable that the Mg concentration of the p-type semiconductor region near the misfit dislocation is in a range smaller than the Mg concentration considered necessary for suppressing overflow. The addition range of Mg concentration and the magnitude of Mg concentration are useful for overcoming the problem peculiar to the semipolarity of increasing the forward voltage after energization.

(Example 6)
In the structure J in Example 4, a structure P including an AlGaN layer (Al composition: 0.05) having a thickness of 10 nm provided between the p-side InGaN guide layer and the GaN guide layer is produced. Structure P is shown in FIG.
The p-type AlGaN layer is grown at a growth temperature of 840 degrees Celsius. The Mg concentration is 2 × 10 ¹⁸ cm ⁻³ . This Mg concentration is at a low level, and Mg concentrations in the range of 5 × 10 ¹⁷ cm ⁻³ to 5 × 10 ¹⁸ cm ⁻³ can be used.
The misfit dislocation density at the InGaN / AlGaN interface is 1 × 10 ⁵ cm ^−1, which is slightly increased compared to the structure J. This increase can be attributed to an increase in the degree of lattice mismatch.
When energization was performed under the same conditions as in Example 4 and the increase in forward voltage Vf was examined, the difference ΔV before and after energization was structure J (0.08 V) V, and the structure P difference ΔV was 0.06 V. Yes, smaller than structure J.
In addition to the AlGaN layer functioning as an electron blocking layer, it contributes to an increase in misfit dislocations at the interface. From these factors, it is considered that electrons are suppressing overflow.

(Example 7)
In the structure J in Example 4, a structure P including an InGaN layer (In composition: 0.07) having a thickness of 3 nm provided between the p-side InGaN guide layer and the GaN guide layer is produced. Structure Q is shown in FIG.
The p-type InGaN layer is grown at a growth temperature of 840 degrees Celsius. The Mg concentration is 2 × 10 ¹⁸ cm ⁻³ . As this Mg concentration, a Mg concentration in the range of 5 × 10 ¹⁷ cm ⁻³ to 5 × 10 ¹⁸ cm ⁻³ can be used.
The misfit dislocation density at the InGaN / AlGaN interface is 2 × 10 ⁴ cm ⁻¹ , which is almost equal to the value of structure J.
When energization was performed under the same conditions as in Example 4 and the rise in forward voltage Vf was examined, the difference ΔV before and after energization was structure J (0.08 V) V, and the structure P difference ΔV was 0.07 V. Yes, smaller than structure J.
Since electron trapping by a thin InGaN layer is used in addition to the heterointerface to cause a non-light emitting transition, it is considered that the overflow of electrons is more effectively suppressed.

(Example 8)
The center semiconductor region in the embodiment includes a plurality of InGaN layers (well layers, guide layers, and in some cases, barrier layers). These InGaN regions seem to be strained as a whole with respect to strain. Therefore, misfit dislocations occur in the center semiconductor region (InGaN region) and the group III nitride region (lattice constant region greater than GaN, hereinafter referred to as “non- Introduced at the interface with “InGaN region”. For this reason, in order to introduce misfit dislocations into the p-side semiconductor region (semiconductor region between the active layer and the anode electrode), the following method can be considered.
(1) The In composition and the film thickness of the InGaN region are set to a certain value or more.
(2) The film thickness of the lattice constant region of GaN or more is set to a certain value or more.
According to the inventors' experiments, the following conditions help to introduce the misfit dislocation at the desired position.
(A) The sum of products of (In composition) and (film thickness) is 8 or more in each layer of the InGaN region.
(B) The total film thickness of the non-InGaN region is 550 nm or more.
This condition can be slightly relaxed, and the center semiconductor region can include a GaN layer having a thickness of 20 nm or less. Under these conditions, a desired misfit dislocation can be introduced into a desired interface. The In composition is a numerical value in the range of 0 to 1, and indicates the molar ratio of In. The film thickness is expressed in nanometer units.
In the epi structure shown in FIG. 14, for condition (b), the total film thickness of the non-InGaN regions is estimated to be 650 nm. For condition (a), the product of each InGaN layer is shown.
n-side InGaN layer: 0.04 × 150 = 6.
InGaN well layer: 0.30 × 2.5 = 0.75.
InGaN barrier layer: 0.04 × 2.5 = 0.1.
p-side InGaN layer: 0.02 × 150 = 3.
The sum of these is 6 + 0.75 × 2 + 0.1 + 3 = 10.6.
Based on the above findings, the product sum of the In composition and the thickness is preferably 10 or more and 12.6 or less.

FIG. 15 is a drawing showing main steps in the method of manufacturing the nitride semiconductor light emitting device according to the present embodiment. In step S201, a plurality of substrates are prepared. This substrate can be, for example, a GaN substrate having a semipolar surface. In step S202, the supply start timing of the p-type dopant is changed on the plurality of substrates, the first group III nitride semiconductor region including the n-type cladding layer, the center semiconductor region including the plurality of InGaN layers, and the p-type An epitaxial substrate provided with a Group III nitride semiconductor region including a cladding layer is formed. In step S203, the epitaxial substrate is processed to form electrodes to form a plurality of substrate products. In step S204, a plurality of nitride semiconductor light emitting devices are manufactured from the substrate product. In step 205, an energization test is performed on each of the plurality of nitride semiconductor light emitting elements to estimate the difference in forward voltage before and after energization. In step S206, the supply start timing of the p-type dopant is determined based on the estimation. In step 207, a nitride semiconductor light emitting element is manufactured using the determined supply start timing. For this production, for example, the production method in the embodiment already described can be applied.
The center semiconductor region is made of a gallium nitride based semiconductor having a band gap less than that of GaN. The center semiconductor region includes an active layer, a first InGaN layer, and a second InGaN layer, the active layer is provided between the first InGaN layer and the second InGaN layer, and the active layer is one or a plurality of InGaN well layers Is provided. The plurality of InGaN layers include an InGaN well layer, a first InGaN layer, and a second InGaN layer. The first InGaN layer in the center semiconductor region is in contact with the first group III nitride semiconductor region to form a first heterointerface. The second InGaN layer in the center semiconductor region is in contact with the second group III nitride semiconductor region to form a second heterointerface. The Group III nitride semiconductor region includes misfit dislocations at the second heterointerface.

According to this manufacturing method, a nitride semiconductor light emitting device having a small difference in forward voltage before and after energization can be provided. The relationship between the second heterojunction and the p-type dopant profile can be determined by adjusting the supply start timing of the p-type dopant.

While the principles of the invention have been illustrated and described in certain embodiments, it will be appreciated by those skilled in the art that the invention may be modified in arrangement and detail without departing from such principles. The present invention is not limited to the specific configuration disclosed in the present embodiment. We therefore claim all modifications and changes that come within the scope and spirit of the following claims.

According to the present embodiment, it is possible to provide a nitride semiconductor light emitting device having a structure that can reduce fluctuations in the forward voltage Vf accompanying energization, and to provide a method for manufacturing this nitride semiconductor light emitting device.

DESCRIPTION OF SYMBOLS 11 ... Nitride semiconductor light emitting element, 13 ... 1st group III nitride semiconductor region, 15 ... Active layer, 17 ... 2nd group III nitride semiconductor region, 19 ... Center semiconductor region, 21b ... 1st inner side semiconductor layer, 23 ... n Type cladding layer, 25b ... second inner semiconductor layer, 27 ... p-type cladding layer, 29 ... p-type contact layer, Ax ... stack axis, 31 ... core region, HJ1, HJ2 ... heterojunction, 33a ... well layer, 33b ... Barrier layer, 39 ... substrate, 39a ... semipolar main surface, Angle ... tilt angle, Sc ... reference plane.

Claims (35)

A nitride semiconductor light emitting device,
A first group III nitride semiconductor region having a semipolar main surface made of a gallium nitride based semiconductor and including an n-type cladding layer;
A second group III nitride semiconductor region provided on the semipolar main surface of the first group III nitride semiconductor region and including a p-type cladding layer;
A center semiconductor region including a plurality of InGaN layers and provided on the first group III nitride semiconductor region;
With
The center semiconductor region is provided between the semipolar main surface of the first group III nitride semiconductor region and the second group III nitride semiconductor region, and the center semiconductor region has a band gap less than or equal to a band gap of GaN. A gallium nitride based semiconductor having
The center semiconductor region includes an active layer, a first InGaN layer, and a second InGaN layer, the active layer is provided between the first InGaN layer and the second InGaN layer, and the active layer includes one or a plurality of InGaN wells. With layers,
The plurality of InGaN layers include the InGaN well layer, the first InGaN layer, and the second InGaN layer,
Each of the plurality of InGaN layers in the center semiconductor region has an indium composition and a layer thickness, and the unit in the layer thickness is nanometers,
The sum of the product of the indium composition and the layer thickness of each InGaN layer over the center semiconductor region is 8 or more,
The second group III nitride semiconductor region is made of a group III nitride semiconductor having a band gap greater than or equal to that of GaN, and the thickness of the second group III nitride semiconductor region is 550 nm or more,
The first InGaN layer of the center semiconductor region is in contact with the first group III nitride semiconductor region to form a first heterointerface;
The second InGaN layer in the center semiconductor region is in contact with the second group III nitride semiconductor region to form a second heterointerface, and the second group III nitride semiconductor region is misfit at the second heterointerface. Including dislocations,
The nitride semiconductor light emitting device, wherein the center semiconductor region is grown without supplying a p-type dopant and substantially does not contain the p-type dopant. The semipolar principal surface of the first group III nitride semiconductor region is in a range of 40 degrees to 80 degrees or 100 degrees to 170 degrees with respect to the c-axis of the gallium nitride semiconductor of the first group III nitride semiconductor region. The nitride semiconductor light emitting device according to claim 1, wherein the nitride semiconductor light emitting device is inclined at an angle within a range of less than or equal to degrees. ^{3. The} nitride semiconductor light emitting device according to claim 1, wherein a density of the misfit dislocations is 5 × 10 ³ cm ⁻¹ or more. The nitride semiconductor light-emitting element according to any one of claims 1 to 3, wherein a density of the misfit dislocations is 5 × 10 ⁵ cm ^-1 or less. The nitride semiconductor light emitting device according to any one of claims 1 to 4, wherein an indium composition of the second InGaN layer is 0.015 or more and 0.055 or less. The nitride semiconductor light emitting device according to any one of claims 1 to 5, wherein a hydrogen concentration in the second InGaN layer is 1 × 10 ¹⁷ cm ^-3 or less. The second InGaN layer in the center semiconductor region is in contact with the first gallium nitride based semiconductor layer in the second group III nitride semiconductor region to form the second heterointerface;
The nitride semiconductor light emitting device according to any one of claims 1 to 6, wherein the first gallium nitride based semiconductor layer in the second group III nitride semiconductor region is GaN or AlGaN. The nitride semiconductor light-emitting element according to any one of claims 1 to 7, wherein an emission wavelength of the active layer is in a wavelength range of 500 nm or more and 570 nm or less. Mounting the first group III nitride semiconductor region, the center semiconductor region, and the second group III nitride semiconductor region, further comprising a substrate having a semipolar main surface made of a group III nitride semiconductor;
The semipolar principal surface of the substrate is inclined with respect to a reference plane orthogonal to the c-axis of the group III nitride semiconductor,
The nitride semiconductor light emitting device according to any one of claims 1 to 8, wherein the second hetero interface is inclined with respect to the reference plane. 10. The nitride semiconductor light emitting device according to claim 9, wherein the group III nitride semiconductor of the substrate is made of GaN. The semipolar principal surface of the group III nitride semiconductor of the substrate has an inclination angle in the range of 40 degrees to 80 degrees, or 100 degrees to 170 degrees, with respect to the c-plane of the group III nitride semiconductor. Tilt,
The inclination of the c-plane is formed in a direction from the c-axis of the group III nitride semiconductor of the substrate toward the m-axis or a-axis of the group III nitride semiconductor. Nitride semiconductor light emitting device. The semipolar main surface of the first group III nitride semiconductor region is inclined at an inclination angle in the range of not less than 63 degrees and not more than 80 degrees with respect to the c-plane of the gallium nitride semiconductor. The nitride semiconductor light-emitting device described in any one of the above. The second InGaN layer in the center semiconductor region is in contact with the GaN layer in the second group III nitride semiconductor region to form the second heterointerface;
The nitride semiconductor light emitting device according to any one of claims 1 to 12, wherein a p-type dopant concentration of the GaN layer is smaller than a p-type dopant concentration of the p-type cladding layer. The second InGaN layer in the center semiconductor region is in contact with the AlGaN layer in the second group III nitride semiconductor region to form the second heterointerface;
The nitride semiconductor light emitting device according to any one of claims 1 to 12, wherein a p-type dopant concentration of the AlGaN layer is smaller than a p-type dopant concentration of the p-type cladding layer. The second group III nitride semiconductor region includes a semiconductor layer that forms a junction with the second InGaN layer of the center semiconductor region at the second heterointerface,
The nitride semiconductor light emitting device according to any one of claims 1 to 13, wherein the semiconductor layer is grown without supplying a p-type dopant and substantially does not contain the p-type dopant. The nitride semiconductor light-emitting device according to any one of claims 1 to 13, wherein a p-type dopant is added to the entire group III nitride semiconductor region. The second group III nitride semiconductor region includes a first semiconductor layer and a second semiconductor layer,
The second semiconductor layer of the second group III nitride semiconductor region is provided between the first semiconductor layer and the p-type cladding layer;
The first semiconductor layer forms a junction with the second InGaN layer in the center semiconductor region at the second heterointerface,
13. The first semiconductor layer is grown without adding a p-type dopant, and the second semiconductor layer is grown while adding a p-type dopant. The nitride semiconductor light emitting device described in 1. The second group III nitride semiconductor region includes a first semiconductor layer and a second semiconductor layer,
The second semiconductor layer of the second group III nitride semiconductor region is provided between the first semiconductor layer and the p-type cladding layer;
The second semiconductor layer has a band gap smaller than that of the first semiconductor layer and the p-type cladding layer;
The nitride semiconductor light emitting device according to any one of claims 1 to 12, wherein a thickness of the first semiconductor layer is thinner than a thickness of the second semiconductor layer. The first semiconductor layer is made of AlGaN;
The aluminum composition of the AlGaN is in the range of 0.02 to 0.06,
The nitride semiconductor light emitting device according to claim 18, wherein the film thickness of the first semiconductor layer is not less than 5 nm and not more than 30 nm. The center semiconductor region includes a third InGaN layer between the active layer and the second InGaN layer,
The band gap of the second InGaN layer is between the band gap of the InGaN well layer and the band gap of the third InGaN layer,
The nitride semiconductor light emitting device according to any one of claims 1 to 12, wherein an indium composition of the second InGaN layer is between an indium composition of the InGaN well layer and an indium composition of the third InGaN layer. . The In composition of the second InGaN layer is in the range of 0.05 to 0.1,
21. The nitride semiconductor light emitting element according to claim 20, wherein a film thickness of the second InGaN layer is in a range of 2 nm to 10 nm. The nitride semiconductor light-emitting element according to any one of claims 1 to 21, wherein the active layer includes a barrier layer of GaN or InGaN. 23. The nitride semiconductor light emitting device according to claim 22, wherein the barrier layer of the active layer includes GaN having a thickness of 20 nm or less. A method for producing a nitride semiconductor light emitting device, comprising:
forming a center semiconductor region including a plurality of InGaN layers on a semipolar main surface of the first group III nitride semiconductor region including an n-type cladding layer;
Forming a second group III nitride semiconductor region including a p-type cladding layer on the semipolar main surface of the center semiconductor region and the first group III nitride semiconductor region;
With
The center semiconductor region is provided between the semipolar main surface of the first group III nitride semiconductor region and the second group III nitride semiconductor region, and the center semiconductor region has a band gap less than or equal to a band gap of GaN. A gallium nitride based semiconductor having
The center semiconductor region includes an active layer, a first InGaN layer, and a second InGaN layer, the active layer is provided between the first InGaN layer and the second InGaN layer, and the active layer includes one or a plurality of InGaN wells. With layers,
The plurality of InGaN layers include the InGaN well layer, the first InGaN layer, and the second InGaN layer,
In the growth of the center semiconductor region, the second InGaN layer is grown without supplying a p-type dopant,
Each of the plurality of InGaN layers in the center semiconductor region has an indium composition and a layer thickness, and the unit in the layer thickness is nanometers,
The sum of the product of the indium composition and the layer thickness of each InGaN layer over the center semiconductor region is 8 or more,
The second group III nitride semiconductor region is made of a group III nitride semiconductor having a band gap greater than or equal to that of GaN, and the thickness of the second group III nitride semiconductor region is 550 nm or more,
The first InGaN layer of the center semiconductor region is in contact with the first group III nitride semiconductor region to form a first heterointerface;
The second InGaN layer in the center semiconductor region is in contact with the second group III nitride semiconductor region to form a second heterointerface, and the second group III nitride semiconductor region is misfit at the second heterointerface. Including dislocations,
A method of fabricating a nitride semiconductor light emitting device, wherein the center semiconductor region is grown without supplying a p-type dopant and is substantially free of the p-type dopant. The semipolar principal surface of the first group III nitride semiconductor region is in a range of 40 degrees to 80 degrees or 100 degrees to 170 degrees with respect to the c-axis of the gallium nitride semiconductor of the first group III nitride semiconductor region. The method for producing a nitride semiconductor light emitting device according to claim 24, wherein the nitride semiconductor light emitting device is inclined at an angle within a range of less than or equal to degrees. The method for producing a nitride semiconductor light emitting device according to claim 24 or 25, wherein a density of the misfit dislocations is 5 × 10 ³ cm ^-1 or more. The method for producing a nitride semiconductor light emitting device according to any one of claims 24 to 26, wherein a density of the misfit dislocations is 5 × 10 ⁵ cm ^-1 or less. The method for producing a nitride semiconductor light emitting element according to any one of claims 24 to 27, wherein an indium composition of the second InGaN layer is 0.015 or more and 0.055 or less. The method for producing a nitride semiconductor light-emitting element according to any one of claims 24 to 28, wherein a hydrogen concentration in the second InGaN layer is 1 × 10 ¹⁷ cm ^-3 or less. The nitride semiconductor light emitting device according to any one of claims 24 to 29, wherein the p-type cladding layer of the second group III nitride semiconductor region forms a junction with the second InGaN layer of the center semiconductor region. How to make. In the growth of the second group III nitride semiconductor region, the first semiconductor region is grown from the start of the growth of the second group III nitride semiconductor region, and the second semiconductor region is grown after the growth of the first semiconductor region,
The second semiconductor region of the second group III nitride semiconductor region is provided between the first semiconductor region and the p-type cladding layer;
The first semiconductor region forms a junction with the second InGaN layer of the center semiconductor region at the second heterointerface,
The nitride semiconductor light-emitting element according to any one of claims 24 to 30, wherein the second semiconductor region has a band gap smaller than that of the first semiconductor region and the p-type cladding layer. How to make. In the growth of the center semiconductor region, the second InGaN layer is grown without supplying a p-type dopant,
In the growth of the second group III nitride semiconductor region, the third semiconductor region is grown without supplying a p-type dopant from the start of the growth of the second group III nitride semiconductor region, and after the growth of the third semiconductor region. growing a fourth semiconductor region while supplying a p-type dopant;
The nitride semiconductor light emitting device according to any one of claims 24 to 30, wherein the third semiconductor region forms a junction with the second InGaN layer of the center semiconductor region at the second hetero interface. how to. In the growth of the second group III nitride semiconductor region, the second group III nitride semiconductor region is grown while supplying a p-type dopant from the start of the growth of the second group III nitride semiconductor region. 31. A method for producing the nitride semiconductor light-emitting device according to any one of 31. A method for producing a nitride semiconductor light emitting device, comprising:
Preparing a plurality of substrates;
A center semiconductor region including a first group III nitride semiconductor region made of a group III nitride semiconductor and including an n-type cladding layer and a plurality of InGaN layers on the plurality of substrates by changing the supply start timing of the p-type dopant. And forming an epitaxial substrate comprising a Group III nitride semiconductor region including a p-type cladding layer;
Processing the epitaxial substrate to form electrodes to form a plurality of substrate products;
Producing a plurality of nitride semiconductor light emitting devices from the substrate product;
For each of the plurality of nitride semiconductor light emitting elements, conducting a current test to estimate the difference in forward voltage before and after power supply;
Determining the supply start timing of the p-type dopant based on the estimate;
Using the determined timing, producing a nitride semiconductor light emitting device;
With
The center semiconductor region is made of a gallium nitride based semiconductor having a band gap equal to or less than that of GaN,
The center semiconductor region includes an active layer, a first InGaN layer, and a second InGaN layer, the active layer is provided between the first InGaN layer and the second InGaN layer, and the active layer includes one or a plurality of InGaN wells. With layers,
The plurality of InGaN layers include the InGaN well layer, the first InGaN layer, and the second InGaN layer,
The first InGaN layer of the center semiconductor region is in contact with the first group III nitride semiconductor region to form a first heterointerface;
The second InGaN layer in the center semiconductor region is in contact with the second group III nitride semiconductor region to form a second heterointerface, and the second group III nitride semiconductor region is misfit at the second heterointerface. A method for manufacturing a nitride semiconductor light emitting device including dislocations. A nitride semiconductor light emitting device,
A first group III nitride semiconductor region having a semipolar main surface made of a gallium nitride based semiconductor and including an n-type cladding layer;
A second group III nitride semiconductor region provided on the first group III nitride semiconductor region and including a p-type cladding layer;
A center semiconductor region including a plurality of InGaN layers and provided on the first group III nitride semiconductor region;
With
The center semiconductor region is provided between the semipolar main surface of the first group III nitride semiconductor region and the second group III nitride semiconductor region, and the center semiconductor region has a band gap less than or equal to a band gap of GaN. A gallium nitride based semiconductor having
The center semiconductor region includes an active layer, a first InGaN layer, a second InGaN layer, and a third InGaN layer. The active layer is provided between the first InGaN layer and the second InGaN layer, and the active layer includes one or more active layers. InGaN well layer, the third InGaN layer is provided between the second InGaN layer and the active layer, the indium composition of the second InGaN layer is larger than the indium composition of the third InGaN layer, the InGaN well layer of Smaller than the indium composition,
The plurality of InGaN layers include the InGaN well layer, the first InGaN layer, the second InGaN layer, and the third InGaN layer,
Each of the plurality of InGaN layers in the center semiconductor region has an indium composition and a layer thickness, and the unit in the layer thickness is nanometers,
The sum of the product of the indium composition and the layer thickness of each InGaN layer over the center semiconductor region is 8 or more,
The second group III nitride semiconductor region is made of a group III nitride semiconductor having a band gap greater than or equal to that of GaN, and the thickness of the second group III nitride semiconductor region is 550 nm or more,
The first InGaN layer of the center semiconductor region is in contact with the first group III nitride semiconductor region to form a first heterointerface;
The second InGaN layer in the center semiconductor region is in contact with the second group III nitride semiconductor region to form a second heterointerface, and the second group III nitride semiconductor region is misfit at the second heterointerface. Including dislocations,
The nitride semiconductor light emitting device, wherein the center semiconductor region is grown without supplying a p-type dopant and substantially does not contain the p-type dopant.

Images