JP7560058B2 - Manufacturing method of semiconductor laser diode, semiconductor laser diode - Google Patents

Description

The present invention relates to a method for manufacturing a semiconductor laser diode.

There are various types of semiconductor laser diodes with different structures. One of them is the Fabry-Perot type semiconductor laser diode, which resonates and amplifies light generated in a resonance region between the end faces of the cavity.
There are two methods for forming the cavity facets: cutting along the cleavage plane of the crystal, and using an etching technique. The etching technique is generally used because it is applicable even when the cleavage planes of the substrate and the nitride semiconductor on the substrate are different, and is highly versatile.

Japanese Patent Application Laid-Open No. 2003-233993 and Japanese Patent Application Laid-Open No. 2003-233996 disclose a method for forming a resonator end face by using an etching technique.
Patent Document 1 describes that the cavity end faces formed by dry etching are damaged by ions and have defects, which causes a problem of light absorption at the cavity end faces. It also describes that in order to solve this problem, the surface specularity of the cavity end faces is improved by wet-etching the dry-etched end faces with an etching solution.

Patent document 2 describes a semiconductor laser diode having a base layer which is a first semiconductor layer, a resonance region formed on a part of the upper surface of the base layer and in which the first semiconductor layer, the first guide layer, the light emitting layer, the second guide layer, and the second semiconductor layer are stacked in this order, a first electrode formed in a part of the base layer where the resonance region is not formed, and a second electrode formed on the resonance region.

In the method of manufacturing a semiconductor laser diode described in Patent Document 2, a laminate including a first semiconductor layer, a first guide layer, a light emitting layer, a second guide layer, and a second semiconductor layer is formed on a substrate, and then the laminate is etched to separate the portion that will become the resonance region from the other portions. After that, a second electrode is formed on the second semiconductor layer by photolithography and etching, and a first electrode is formed on the portion of the base layer where the resonance region is not formed, and then a resist mask is formed with a width equal to the length of the resonator, and the resonator end faces are formed by photolithography and dry etching.

In the method described in Patent Document 2, after the second electrode is formed, the resonator end faces are formed by photolithography and dry etching. Therefore, in a semiconductor laser diode manufactured by this method, the resonator end faces and the end faces of the second electrode in the resonance direction of the resonance region are not in the same plane, and the resonator end faces are located outside the end faces of the electrodes in a planar view.

On the other hand, one of the challenges in realizing an ultraviolet semiconductor laser diode is to increase output, and one approach to achieving this is to reduce the threshold current density of laser oscillation.

Japanese Patent Application Publication No. 10-41585 Japanese Patent Application Publication No. 11-145566 JP 2008-124498 A JP 2009-295952 A JP 2020-47635 A

The objective of the present invention is to provide a Fabry-Perot type semiconductor laser diode that is obtained through a process of forming a resonator end face using an etching technique, and that has a reduced threshold current density for laser oscillation and improved durability.

In order to solve the above problems, a method for manufacturing a semiconductor laser diode according to a first aspect of the present invention has the following configurations (1) to (4).
(1) A method for manufacturing a semiconductor laser diode having an underlayer, a resonance region formed on a part of the upper surface of the underlayer and including a first semiconductor layer, a waveguide layer, and a second semiconductor layer stacked in this order, a first electrode formed on the first semiconductor layer near the resonance region, and a second electrode formed on the resonance region, the first semiconductor layer of the resonance region and the first semiconductor layer on which the first electrode is formed are continuous (both first semiconductor layers are connected without interruption). The waveguide layer includes a light emitting layer. The underlayer may be a substrate, or may be a first semiconductor layer formed on a substrate.
(2) The method includes a step of forming a plurality of the semiconductor laser diodes on a substrate and then dividing the substrate into individual semiconductor laser diodes.
(3) The method includes a step of forming a laminate including a first semiconductor layer, a waveguide layer, and a second semiconductor layer in this order on a substrate.
(4) An electrode layer forming step of forming a layer to become the second electrode on the second semiconductor layer of the laminate, between the resonator facet of the resonance region and the position to be divided in the dividing step, and a first etching step of removing a portion of the layer to become the second electrode formed at a position outside the resonator facet and forming the resonator facet, simultaneously or consecutively, after the electrode layer forming step.

A second aspect of the present invention is a semiconductor laser diode obtained by the manufacturing method of the first aspect, and has the following configurations (11) and (12).
(11) A semiconductor device having a base layer, a resonance region formed in a part of the base region that is a part of the upper surface of the base layer, the resonance region being formed with a first semiconductor layer, a waveguide layer, and a second semiconductor layer stacked in this order, a first electrode formed on the first semiconductor layer near the resonance region, a second electrode formed on the resonance region, and a convex portion protruding from both a surface that constitutes a resonator end face of the resonance region and the upper surface of the base region. The first semiconductor layer of the resonance region and the first semiconductor layer on which the first electrode is formed are continuous (both first semiconductor layers are connected without interruption).
(12) The convex portion transitions directly onto the upper surface of the base region, which is a flat surface, on the side away from the cavity end face.

The manufacturing method of the first aspect of the present invention is expected to produce a semiconductor laser diode with a reduced threshold current density for laser oscillation and improved durability. In addition, by having an etching step after the electrode layer formation step in which the removal of the portion of the layer that will become the second electrode, which is formed at a position outside the cavity end face, and the formation of the cavity end face are performed simultaneously or consecutively, it is possible to perform an inspection to check the characteristics of each semiconductor laser diode before the division step.

The semiconductor laser diode of the second aspect of the present invention is a semiconductor laser diode obtained by the manufacturing method of the first aspect, and the convex portions protruding from both the faces constituting the resonator end faces of the resonance region and the upper surface of the base region protect the areas where current and heat are concentrated, so that it is expected to obtain effects such as a reduction in leakage current when current is applied, an increase in light emission efficiency, and suppression of fluctuations in the irradiation pattern.
That is, the semiconductor laser diode according to the second aspect of the present invention is expected to have a reduced threshold current density for laser oscillation and improved durability.

1A to 1C are diagrams for explaining each step constituting a method for manufacturing a semiconductor laser diode according to one embodiment of the present invention, showing a plan view of a substrate after each step and a cross-sectional view thereof taken along the line AA thereof; FIG. 1 is a diagram for explaining each step constituting a method for manufacturing a semiconductor laser diode according to one embodiment of the present invention, in which (a) and (b) show a plan view of the substrate after each step and a cross-sectional view taken along the line A-A of the substrate, and (c) shows a plan view and a front view of the substrate after the step. FIG. 3 is a plan view showing an area of a substrate on which a plurality of nitride semiconductor laser diodes are to be formed, the plan view corresponding to the state shown in FIG. 2( a ). This is a cross-sectional view taken along line B-B of FIG. FIG. 2( c ) is a perspective view of FIG. FIG. 3 is a plan view showing an area of the substrate on which a plurality of nitride semiconductor laser diodes are to be formed, the plan view corresponding to the state shown in FIG. 2( c ). 1 is a cross-sectional view showing a semiconductor laser diode according to one embodiment of the present invention.

[Findings of the present inventors]
It was found that when the resonator end faces are formed using etching technology, the resonator end faces and the end faces of the second electrode in the resonance direction of the resonance region are not in the same plane, and the resonator end faces are located outside the end faces of the electrodes in a planar view, which is one of the factors that increases the threshold current density.
They also discovered that if an inspection can be performed before the dividing step in which a terminal connected to an external power source is brought into contact with the second electrode and the first electrode to check the characteristics of each semiconductor laser diode, it is possible to remove defective wafers and reduce manufacturing costs.

[Embodiment]
Hereinafter, the embodiments of the present invention will be described, but the present invention is not limited to the embodiments shown below. In the embodiments shown below, technically preferable limitations are imposed for carrying out the present invention, but these limitations are not essential requirements for the present invention.

[composition]
The method for manufacturing a Fabry-Perot type semiconductor laser diode of this embodiment includes a step of forming a plurality of nitride semiconductor laser diodes on a substrate and then dividing the substrate into individual nitride semiconductor laser diodes (hereinafter referred to as a "division step"). Note that Figures 1, 2, 4, and 5 show the substrate in an area in which one nitride semiconductor laser diode is formed. Figures 3 and 6 show the substrate in an area in which a plurality of nitride semiconductor laser diodes are formed.

1A, a laminate 2 including a first semiconductor layer 21, a first guide layer 22, a light emitting layer 23, a second guide layer 24, and a second semiconductor layer 25 in this order is formed on a substrate 1. That is, the first guide layer 22, the light emitting layer 23, and the second guide layer 24 are formed as waveguide layers.
A sapphire substrate is used as the substrate 1, an n-Al _x Ga _(1-x) N layer is formed as the first semiconductor layer 21, and an Al _x Ga _(1-x) N layer is formed as the first guide layer 22. An Al _x Ga _(1-x) N layer is formed as the light emitting layer 23, an Al _x Ga _(1-x) N layer is formed as the second guide layer 24, and a p-Al _x Ga _(1-x) N layer is formed as the second semiconductor layer 25. That is, the method of this embodiment produces a nitride semiconductor laser diode.

1(b), a mesa etching step is performed in which the laminate 2 is etched and divided into a mesa portion (a portion including a portion that becomes a resonance region) 31 and the other portions, a first portion 32 and a second portion 33. As a result, the first semiconductor layer 21 is divided in the film thickness direction into a portion (mesa side portion) 21a that constitutes the mesa portion 31 and the other portion 21b. FIG. 1(b) shows the resonance direction K of the semiconductor laser diode to be manufactured.
Next, an insulating layer 4, which is a _SiO2 layer, is formed on the entire surface of the substrate 1 in the state of Fig. 1(b), and then a photolithography etching process is performed to provide a through hole 41 for a second electrode in the insulating layer 4 on the second semiconductor layer 25, and at the same time, a through hole 42 for a first electrode in the insulating layer 4 on the first portion 32. Fig. 1(c) shows the state after this process.

Next, a resist pattern is formed on the substrate 1 in the state shown in Fig. 1(c), and then a Ni/Pt/Au layer is evaporated and a lift-off process is performed to remove the resist pattern, followed by a high-temperature heat treatment to form a patterned NiPtAu alloy layer. This results in the state shown in Fig. 2(a). Fig. 2(a) shows the resonant direction K of the semiconductor laser diode to be manufactured and the position T of a pair of resonator end faces.
When this state is shown in the substrate in the range where a plurality of nitride semiconductor laser diodes are to be formed, parting lines L1 and L2 are formed as shown in Fig. 3. The positions along each parting line L1 and L2 are the positions at which the substrate is divided in the above-mentioned dividing step, and among them, the position of parting line L2 perpendicular to the resonance direction K is indicated by symbol E in Fig. 3.

That is, in the above-mentioned lithography lift-off process, as shown in FIG. 2(a) and FIG. 3, the second alloy layer (layer to become the second electrode) 51 and the first alloy layer (layer to become the first electrode) 61 are formed in the resonance direction K up to the area between the resonator end face position T and the position of the division line L2 (position to be divided in the division process) E. At that time, a connector 52 that electrically connects the second alloy layer 51 and the second semiconductor layer 25 is formed in the through hole 41 for the second electrode of the insulating layer 4, and a connector 62 that electrically connects the first alloy layer 61 and the second semiconductor layer 25 is formed in the through hole 42 for the first electrode of the insulating layer 4.

Next, after forming a Ni vapor deposition film on the substrate 1 in the state shown in FIG. 2(a), a photolithography lift-off process is performed so that the portion between the pair of resonator end face positions T in the resonance direction K is covered with a Ni mask 7 as shown in FIG. 2(b). In this state, the second alloy layer 51 and the first alloy layer 61 are each covered with the Ni mask 7 except for the portions 51a and 61a outside the resonator end face position T in the resonance direction K.

Next, dry etching is performed on the substrate 1 in the state shown in FIG. 2(b). This dry etching is performed using chlorine gas under conditions in which the mesa portion 31 is etched up to the boundary position H between the first portion 32 and the second portion 33. In this single continuous dry etching process (first etching process), the outer portion 51a of the second alloy layer 51 and the outer portion 61a of the first alloy layer 61 are removed, and as shown in FIG. 2(c), the end face 5A of the second electrode 5 and the end face 6A of the first electrode 6 are generated (i.e., the second electrode 5 and the first electrode 6 are formed), and the portion of the mesa portion 31 that is not covered by the mask 7 is removed to a predetermined depth, forming the resonator end face 200. In addition, the portion of the first portion 32 that is not covered by the mask 7 is also removed to a predetermined depth. As a result, the resonance region 310 and the underlayer 320 are formed.

Next, the substrate 1 after the dry etching is immersed in an alkaline aqueous solution to perform wet etching, and then the mask 7 is removed to obtain the substrate 1 (element precursor 10) in the state shown in Fig. 2(c). This element precursor 10 is shown in a cross-sectional view taken along line B-B of Fig. 2(c) in Fig. 4 and in a perspective view in Fig. 5. When this state is shown in the substrate in the range where a plurality of nitride semiconductor laser diodes 10A are to be formed, parting lines L1 and L2 are formed as shown in Fig. 6.
Next, the substrate is divided along division lines L1 and L2 shown in FIG.

7, a reflective layer 9 which is a dielectric multilayer film of _SiO2 and _HfO2 is formed by sputtering on the cavity facet 200 of the element precursor 10, the facet 5A of the second electrode 5, and the facet 6A of the first electrode 6. This results in the nitride semiconductor laser diode 10A shown in FIG.
The reflective layer 9 is formed by placing a jig having a slope that inclines the substrate surface by 45° on a stand for placing the substrate in the film-forming device, attaching the back surface of the substrate 1 to the slope of the jig, making the resonator end face 200 of the element precursor 10, which is the film-forming surface, a 45° upward surface, and covering the center in the resonance direction K of the upper surface of the element precursor 10 with a cover. In this way, the reflective layer 9 is formed so that the film thickness on the end face 5A of the second electrode 5 and the end face 6A of the first electrode 6 is thicker than the film thickness on the resonator end face 200, and a conductive portion D is secured on the upper surfaces of the second electrode 5 and the first electrode 6.

As shown in Figures 2(c), 4, and 5, the cavity end face 200 of the element precursor 10 (i.e., the nitride semiconductor laser diode 10A) is composed of the surfaces of the first semiconductor layer 21aa (part of the mesa side portion 21a of the first semiconductor layer 2), the first guide layer 22, the light emitting layer 23, the second guide layer 24, and the second semiconductor layer 25. The end face 5A in the resonance direction K of the second electrode 5 and the cavity end face 200 are in the same plane. Furthermore, the end face 6A in the resonance direction K of the first electrode 6 and the end face 32a of the first portion (part other than the mesa portion) 32 of the first semiconductor layer 21 are also in the same plane as the cavity end face 200.

Furthermore, the nitride semiconductor laser diode 10A has protrusions 81 to 84 formed by dry etching the substrate 1 in the state shown in FIG. 2(b).
The protrusions 81 protrude from both the surface of the first semiconductor layer 21aa constituting the resonator end face 200 and a first upper surface 321 of the base layer 320. The protrusions 82 protrude from both the end face 32a of the first semiconductor layer 21 that is in the same plane as the end face 6A of the first electrode 6 in the resonance direction K and a second upper surface 322 of the base layer 320.

The protrusion 81 has a triangular prism shape, one side of which contacts the first semiconductor layer 21aa constituting the resonator facet 200, the other side of which contacts the first upper surface 321 of the base layer 320, and the remaining side is present upward as a slope. In other words, the protrusion dimension of the protrusion 81 from the upper surface of the base layer 320 is greatest at the position where the protrusion 81 protrudes from the surface of the first semiconductor layer 21aa constituting the resonator facet 200.
Furthermore, the dimension W81 of the protrusion 81 in the direction perpendicular to the resonance direction K is larger than the dimension W of the connecting body 52 in the direction perpendicular to the resonance direction K.

The protrusions 83 and 84 are the surfaces of all layers (layers constituting the mesa portion 31) constituting the resonator end face 200, and protrude from both the oblique line L1 indicating the slope of the mesa portion 31, the perpendicular line L2 extending from the apex of the mesa portion 31 to the first upper surface 321 of the underlayer 320, and the elongated right-angled triangular surface consisting of the sides along the first upper surface 321 (see the lower diagram of FIG. 2(c)), and the triangular surface of the first upper surface 321 of the underlayer 320 (see the upper diagram of FIG. 2(c)). In other words, the protrusions 83 and 84 have a triangular pyramid shape with elongated triangular sides.

In addition, the convex portions 81, 83, and 84 transition directly to the first upper surface 321 of the underlayer 320, which is a flat surface, on the side away from the resonator end surface. In other words, the nitride semiconductor laser diode 10A does not have the convex portions 81, 83, and 84 extending up to the end surface position in the resonance direction of the underlayer 320 (for example, FIG. 5 of Patent Document 3 and FIG. 2 of Patent Document 4), nor does it have the convex portions 81, 83, and 84 transition to the first upper surface 321 via a concave portion (the tips of the convex portions 81, 83, and 84 descend to a position lower than the first upper surface 321 and then transition to the first upper surface 321) (for example, FIG. 3 of Patent Document 5).

2(b) is formed due to the fact that the dry etching rate using chlorine gas differs for each material. For example, when the Al _x Ga.sub _.(1-x) N layer is etched at 2.828 nm/s, if the film formation is performed under conditions where the etching rate of the _SiO.sub.2 layer is 0.585 nm/s and the etching rate of the NiPtAu alloy layer is 0.507 nm/s, the protrusions 81-84 are formed.
Here, as shown in the lower diagram of Figure 2(b), the portion of the mesa portion 31 that is not covered by the mask 7 is considered to be divided into a central portion 31a which is located below the outer portion 51a of the second alloy layer 51, an inclined portion 31b consisting of a right-angled triangle with the inclined surface of the mesa portion 31 as its hypotenuse and the straight line indicating the boundary position H as its base, and an end portion 31c consisting of a rectangle between the two.

In other words, since the etching rate of mesa portion 31 which is a laminate of Al _x Ga _(1-x) N layers is five times or more the etching rate of the NiPtAu alloy layer, when end portion 31c which does not have portion 51a which is the NiPtAu alloy layer thereon is etched up to boundary position H, central portion 31a which has portion 51a which is the NiPtAu alloy layer thereon has been etched only up to a position above boundary position H at the position closest to mask 7 in resonance direction K. As a result, convex portion 81 is formed.

Furthermore, because insulating layer 4 is also formed on the slope of mesa portion 31, the etching of slope portion 31b is affected by the etching of insulating layer 4 on the slope. In other words, the portion of slope portion 31b that is etched after insulating layer 4 on the slope is etched increases toward the lower side of slope portion 31b. Furthermore, because the etching rate of mesa portion 31, which is a laminate of Al _x Ga _(1-x) N layers, is about five times the etching rate of insulating layer 4, which is a SiO ₂ layer, when end portion 31c is etched up to boundary position H, a part of slope portion 31b in resonance direction K remains unetched and remains as convex portions 83, 84.

Furthermore, the portion of first portion 32 that is not covered by mask 7 is divided into central portion 32b that is below outer portion 61a of first alloy layer 61 and end portions 32c on both sides of central portion 32b. Since the etching rate of the Al _x Ga.sub. _(1-x) N layer is five times or more the etching rate of the NiPtAu alloy layer, at the same point in time, the etching depth of end portion 32c that does not have portion 61a of the NiPtAu alloy layer thereon is deeper than the etching depth of central portion 32b that has portion 61a of the NiPtAu alloy layer thereon.

That is, when the end portion 31c of the mesa portion 31 is etched up to the boundary position H, the end portion 32c of the first portion 32 is etched up to the second upper surface 322 which is lower than the first upper surface 321 of the underlayer 320, but the etching of the central portion 32b of the first portion 32 is only performed up to a position above the second upper surface 322 at the position closest to the mask 7 in the resonance direction K. As a result, a convex portion 82 is formed.
The reason why the convex portions 81 to 84 have a triangular prism or pyramid shape is that the corners are etched and removed in a concentrated manner.

In the nitride semiconductor laser diode of this embodiment, the first alloy layer 61 and the second alloy layer 51 are formed of the same material, but the first alloy layer 61 and the second alloy layer 51 may be formed of different materials. For example, a similar structure is formed even when the first alloy layer 61 is a NiPtAu alloy layer and the second alloy layer 51 is a V/Al/Ti/Au layer.

[Action, effect]
As described above, the manufacturing method of the nitride semiconductor laser diode of this embodiment includes an electrode layer formation step of forming second alloy layer (layer that becomes the second electrode) 51 on second semiconductor layer 25 of laminate 2 up to between cavity facet position T of resonance region 310 and position E at which it is divided in the dividing step, and an etching step of successively removing portion 51 a of second alloy layer (layer that becomes the second electrode) 51 outside cavity facet position T in resonance direction K and forming cavity facet 200, after the electrode layer formation step.

Therefore, according to the manufacturing method of the nitride semiconductor laser diode of this embodiment, it is possible to obtain a nitride semiconductor laser diode 10A in which the resonator facet 200 and the facet 5A of the second electrode 5 in the resonance direction K are on the same plane. As a result, the nitride semiconductor laser diode 10A obtained by the manufacturing method of this embodiment is expected to have a reduced threshold current density for laser oscillation.
Moreover, in the manufacturing method of the nitride semiconductor laser diode of this embodiment, the above-mentioned etching step is performed by a dry etching method using mask 7, and then wet etching is performed with an alkaline aqueous solution without removing mask 7, thereby obtaining a nitride semiconductor laser diode having a highly flattened cavity facet 200.

Furthermore, in the manufacturing method of the nitride semiconductor laser diode of this embodiment, as shown in FIG. 6, before the substrate is divided at the division lines L1 and L2, both the second electrode 5 and the first electrode 6 are not continuous between adjacent element precursors, so that in the state shown in FIG. 6, a terminal connected to an external power source can be brought into contact with the second electrode 5 and the first electrode 6 to easily inspect the characteristics of each element. In contrast, if at least one of the second electrode 5 and the first electrode 6 is continuous between adjacent elements (for example, the method described in Patent Document 3), the above-mentioned inspection method cannot perform a correct inspection because multiple elements are driven simultaneously. By performing this inspection before division, it is possible to remove defective wafers and reduce manufacturing costs. In addition, if the elements are divided, the labor required to place each element in an inspection device during inspection increases significantly, increasing manufacturing costs. With elements in a wafer state, only one wafer needs to be placed, so costs can be significantly reduced.

The nitride semiconductor laser diode 10A of this embodiment also has a protrusion 81 that protrudes from both the surface of the first semiconductor layer 21aa that constitutes the resonator end facet 200 of the resonance region 310 and the first upper surface (underlying region) 321 of the underlayer 320. This protrusion 81 protects the area where current and heat are concentrated (the surface of the first semiconductor layer 21aa that constitutes the resonator end facet 200), making this area less susceptible to corrosion due to air, water, etc. As a result, it can be expected to achieve effects such as a reduction in leakage current when current is applied, an increase in light emission efficiency, and suppression of irradiation pattern fluctuations.

Furthermore, the protrusion 81 has a shape in which the protrusion dimension from the first upper surface 321 of the underlayer 320 is greatest at the protruding position from the surface of the first semiconductor layer 21aa that constitutes the resonator facet 200, and the protrusion dimension decreases to zero as the distance from this surface increases. Therefore, it is possible to suppress natural light emission and reflection and scattering of excess laser light while maintaining a high protective effect of the protrusion 81 on the surface of the first semiconductor layer 21aa that constitutes the resonator facet 200.
Furthermore, since the dimension W81 of the convex portion 81 in the direction perpendicular to the resonance direction K is larger than the dimension W5 of the connector 52 which corresponds to the width of the substantial light-emitting portion, the substantial light-emitting portion is reliably protected by the convex portion 81.

Moreover, the nitride semiconductor laser diode 10A of the present embodiment has a protrusion 82. This protrusion 82 protects the end face 32a of the first portion 32 of the first semiconductor layer 21 that has been damaged by etching, and makes the end face 32a less susceptible to corrosion due to air, water, and the like.
Furthermore, the nitride semiconductor laser diode 10A of this embodiment has protrusions 83 and 84. These protrusions 83 and 84 protect both ends in the width direction of each layer constituting the cavity end face that has been damaged by etching, making these portions less susceptible to corrosion due to air, water, etc. As a result, it can be expected that an effect similar to that of the protrusion 81 can be obtained.

Furthermore, since the convex portions 81, 83, and 84 transition directly to the first upper surface 321 of the base layer 320, which is a flat surface, on the side away from the cavity end surface, the following effects can be expected to be obtained.
In other words, since the convex portions 81, 83, and 84 do not extend to the end face position in the resonance direction of the underlayer 320, and the first upper surface 321 exists, the convex portions 81, 83, and 84 are protected from damage due to external physical impact. For example, when the side surface of the chip is sandwiched by a collet during assembly, the convex portions 81, 83, and 84 can be prevented from directly contacting the collet, thereby preventing damage to the convex portions 81, 83, and 84. The same can be said about the convex portion 82.

In addition, since the convex portions 81, 83, and 84 do not extend to the end face position in the resonance direction of the base layer 320, and the first upper surface 321 exists, it is possible to prevent the convex portions 81, 83, and 84 from being destroyed even if chipping or cracks occur on the end face of the element during the division process. In addition, since the element division line can be kept away from the light-emitting portion of the semiconductor, damage to the light-emitting portion due to heat and shock during division can be suppressed. The same can be said for the convex portion 82 when chipping or cracks occur on the end face of the element during the division process.

Furthermore, on the side away from the resonator end face, the heat diffusion performance and deterioration resistance performance are improved compared to when the convex portions 81, 83, and 84 transition to the first upper surface 321 via the concave portions. This is because heat and current are concentrated in the base near these convex portions, and the semiconductor is prone to deterioration due to reaction with the outside air (oxygen, carbon), but without the concave portions, heat is more easily diffused, and the surface area is smaller due to the absence of the concave portions, making it less likely to react with the outside air, and therefore less likely to deteriorate.
In addition, natural light from inside the semiconductor is also emitted from the underlayer 320, but if the first upper surface 321 of the underlayer 320 has irregularities, the light extraction efficiency of this natural light increases, resulting in noise in the laser light. Therefore, by transitioning the convex portions 81, 83, and 84 to the first upper surface 321 without passing through a concave portion, it is possible to reduce noise in the laser light more than when the transition is made to the first upper surface 321 through a concave portion.

In addition, in the nitride semiconductor laser diode 10A of this embodiment, a reflective layer 9 is formed on the cavity facet 200 and the facet 5A in the resonance direction K of the second electrode 5, and the film thickness of the reflective layer 9 is thicker on the facet 5A of the second electrode 5 than on the facet 200 of the cavity. In other words, a thick reflective layer that can impart a high corrosion suppression effect is formed on the facet 5A of the second electrode 5, and a reflective layer with a design film thickness (thinner than the reflective layer formed on the facet 5A) is formed on the facet 200 of the cavity, thereby improving the durability of the nitride semiconductor laser diode.

[Differences between the first and second aspects and the embodiment]
In the manufacturing method of the semiconductor laser diode of this embodiment, before the electrode layer forming step, a step of forming an insulating layer 4 on the second semiconductor layer 25 and a step of providing through holes 41, 42 at predetermined positions (positions corresponding to the positions where the electrode layers are formed) of the insulating layer 4 are performed. As a result, the first electrode 6 and the second electrode 5 are connected to the second semiconductor layer 25 by the connectors 62, 52, respectively, and by performing the etching step via the insulating layer 4, convex portions 83, 84 are generated in the obtained semiconductor laser diode 10A.
In the semiconductor laser diode 10A of this embodiment, the protrusions 83, 84 and the protrusions 81, 82 have the same protrusion dimension outward in the resonance direction K, but this protrusion dimension does not have to be the same; for example, the protrusion dimension of the protrusions 83, 84 may be larger than that of the protrusions 81, 82.

2B, the area covered by the mask 7 in the resonance direction K is the same for the mesa portion 31 and the first portion 32 and the second portion 33 other than the mesa portion, but the entire upper surface of the first portion 32 may be covered. In that case, the first electrode 6 is formed to have a longer dimension in the resonance direction K than the second electrode 5, and the second upper surface 322 and the convex portion 82 are not formed.
In the manufacturing method of the semiconductor laser diode of this embodiment, dry etching using a mask 7 is performed as an etching process for continuously removing portions 51 a, 61 a of the second alloy layer 51 and the first alloy layer 61 outside the cavity facet position T and forming the cavity facet 200. However, in the state of Figure 2(a), FIB processing may be performed without using a mask, in which a focused ion beam (FIB) is irradiated only outside the cavity facet position T in the resonance direction K.

The semiconductor laser diode 10A of this embodiment does not have a ridge semiconductor layer, but a ridge semiconductor layer may be provided above a portion of the second semiconductor layer 25. By providing a ridge semiconductor layer, light emission in the light-emitting layer can be controlled so that it occurs only in the region below the ridge semiconductor layer, so that the current density of the semiconductor laser diode can be increased and the laser oscillation threshold can be lowered.

In the manufacturing method of the semiconductor laser diode of this embodiment, the laminate 2 is mesa-etched to divide the first semiconductor layer 21 into a portion 21a including a portion that becomes a resonance region and the other portion 21b in the film thickness direction, and the first electrode 6 is formed on the upper surface of the other portion 21b created by this etching. That is, the formation surface of the first electrode 6 (the second upper surface 322 of the base layer made of the first semiconductor layer) is formed by mesa etching. After that, patterning is performed using an etching mask different from that used for mesa etching, and then etching is performed again until the upper surface 321 of the base region reaches the position of the upper surface of the mesa-etched end face 32a, so that the upper surface of the base region becomes the upper surface of the first semiconductor layer. The depth of this re-etching may not be aligned with the position of the upper surface of the end face 32a, but may be etched below the base layer 320 made of the first semiconductor layer, for example, down to the substrate 1, so that the upper surface of the base region becomes the upper surface of the substrate 1 rather than the upper surface of the first semiconductor layer.

[Supplementary information regarding the configurations of the first and second aspects]
(substrate)
Specific examples of materials for forming the substrate include Si, SiC, MgO, _Ga2O3 , _Al2O3 , ZnO, GaN, InN, AlN _, or mixed crystals thereof. For ease of assembly, the substrate is preferably a thin rectangular plate, but is not limited thereto. In addition, the off-angle of the substrate is preferably greater than 0 degrees and smaller than ₂ degrees from the viewpoint of growing high-quality crystals. The thickness of the substrate is not particularly limited as long as the purpose is to laminate an AlGaN layer on top, but is preferably 50 μm or more and 1 mm or less.

The substrate is used for the purpose of supporting the thin films formed as the layers constituting the semiconductor laser diode, improving the crystallinity, and dissipating heat to the outside. For this reason, it is preferable to use an AlN substrate, which is a material with high thermal conductivity and allows AlGaN to be grown with high quality.
There is no particular restriction on the crystal quality of the substrate, but in order to form an element thin film having high luminous efficiency, the threading dislocation density is preferably 1×10 ⁹ cm ⁻² or less, and more preferably 1×10 ⁸ cm ⁻² or less. The growth surface of the substrate is preferably the commonly used +c-plane AlN because of its low cost, but may be −c-plane AlN, a semipolar plane substrate, or a nonpolar plane substrate. When growing III-composition-gradient AlGaN, which is widely used as a p-type semiconductor, +c-plane AlN is preferable from the viewpoint of increasing the effect of polarization doping.

(First Semiconductor Layer)
The thickness of the first semiconductor layer is not particularly limited. For example, the thickness may be 100 nm or more in order to reduce the resistance of the first semiconductor layer, and may be 10 μm or less in order to suppress the occurrence of cracks during the formation of the first semiconductor layer.
An example of a material for forming the first semiconductor layer is Al _x1 Ga _(1-x1) N (0≦x1≦1). The material for forming the first semiconductor layer may contain impurities such as a group V element other than N, such as P, As, or Sb, a group III element, such as In or B, or C, H, F, O, Si, Cd, Zn, or Be.

The Al composition ratio x of Al _x Ga _(1-x) N can be specified by exposing a cross section along the a-plane of AlGaN using a focused ion beam (FIB) device, observing the cross section with a transmission electron microscope, and performing energy dispersive X-ray analysis (EDX) of the cross section. The Al composition ratio x can be defined as the ratio of the number of moles of Al to the sum of the number of moles of Al and Ga, and can be defined using the values of the number of moles of Al and Ga quantified by EDX.

When the first semiconductor layer is an n-type semiconductor, it can be made n-type by doping with, for example, 1×10 ¹⁹ cm ⁻³ of Si. When the first semiconductor layer is a p-type semiconductor, it can be made p-type by doping with, for example, 3×10 ¹⁹ cm ⁻³ of Mg. The impurity concentration may be uniform or non-uniform throughout the layer, or may be non-uniform only in the film thickness direction or only in the horizontal direction of the substrate.
In addition, an intermediate layer such as an AlN layer or an AlGaN layer may be formed between the substrate 1 and the first semiconductor layer 21aa. In this case, this intermediate layer may be used as the base layer 320.

(Waveguide layer)
The waveguide layer is a layer including a light-emitting layer, and is, for example, a layer in which a first guide layer, a light-emitting layer, and a second guide layer are laminated in this order, but may not include either or both of the first guide layer and the second guide layer.

(First Guide Layer)
The first guide layer is formed on the first semiconductor layer of the first semiconductor layer. The first guide layer has a refractive index different from that of the first semiconductor layer in order to confine the light emitted from the light emitting layer in the light emitting section (the section composed of the first guide layer, the light emitting layer, and the second guide layer). Examples of materials for forming the first guide layer include mixed crystals of AlN and GaN.
A specific example of the material for forming the first guide layer is Al _x2 Ga _(1-x2) N (0<x2<1).
The Al composition ratio x2 of Al _x2 Ga _(1-x2) N forming the first guide layer may be smaller than the Al composition ratio x1 of Al _x1 Ga _(1-x1) N forming the first semiconductor layer. This makes it possible for the first guide layer to have a larger refractive index than the first semiconductor layer, and to confine the light emitted from the light emitting layer in the light emitting section. The material forming the first guide layer may contain impurities such as a Group V element other than N, such as P, As, or Sb, a Group III element, such as In or B, or C, H, F, O, Si, Cd, Zn, or Be.

When the first guide layer is an n-type semiconductor, it can be made n-type by doping it with, for example, 1×10 ¹⁹ cm ⁻³ of Si. When the first guide layer is a p-type semiconductor, it can be made p-type by doping it with, for example, 3×10 ¹⁹ cm ⁻³ of Mg. The first guide layer may have a dopant only in a portion in the thickness direction. That is, a portion in the thickness direction of the first guide layer may be a combination of an n-type semiconductor and an undoped layer, or a p-type semiconductor and an undoped layer. The first guide layer may be an undoped layer.

The first guide layer may have a structure with a graded composition. For example, the first guide layer may have a layer structure in which the Al composition ratio x2 of Al _x2 Ga _(1-x2) N changes continuously or stepwise from 0.6 to 0.5. The thickness of the first guide layer is not particularly limited. The thickness of the first guide layer may be 10 nm or more in order to efficiently confine the light emitted from the light emitting layer in the light emitting portion, and the thickness of the first guide layer may be 2 μm or less in order to reduce the resistance of the first guide layer.
The first guide layer may have an AlGaN layer that serves as a blocking layer within the range in which the purpose of confining light in the light emitting portion is maintained.

(Light Emitting Layer)
Examples of materials for forming the light-emitting layer include AlN, GaN, and mixed crystals thereof. A specific example of a material for forming the light-emitting layer is Al _x3 Ga _(1-x3) N (0≦x3≦1). The Al composition ratio x of Al _x3 Ga _(1-x3) N of the light-emitting layer may be smaller than the Al composition ratio x2 of Al _x2 Ga _(1-x2) N of the first guide layer in order to efficiently confine carriers injected from the first electrode and the second electrode in the light-emitting portion. For example, the light-emitting layer may be formed of Al _x3 Ga _(1-x3) N whose Al composition ratio x3 satisfies the relationship 0.2≦x3<1.
In addition, the material forming the light emitting layer may contain impurities such as a Group V element other than N, such as P, As, or Sb, a Group III element, such as In or B, or C, H, F, O, Si, Cd, Zn, or Be.

When the light emitting layer is an n-type semiconductor, it can be made n-type by doping with, for example, 1×10 ¹⁹ cm ⁻³ of Si. When the light emitting layer is a p-type semiconductor, it can be made p-type by doping with, for example, 3×10 ¹⁹ cm ⁻³ of Mg. The light emitting layer may be an undoped layer.
The light-emitting layer may have a well layer made of, for example, Al _x5 Ga _(1-x5) N and a barrier layer made of, for example, Al _x4 Ga _(1-x4) N adjacent to the well layer, and may have a multiple quantum well (MQW) structure in which the well layers and barrier layers are alternately stacked one by one.

The Al composition ratio x5 of the well layer is smaller than the Al composition ratio x4 of each of the first guide layer and the second guide layer. Also, the Al composition ratio x5 of the well layer is smaller than the Al composition ratio x4 of the barrier layer. The Al composition ratio x4 of the barrier layer may be the same as the Al composition ratio x4 of each of the first guide layer and the second guide layer, or may be higher or lower than the Al composition ratio x4 of each of the first guide layer and the second guide layer.
The average Al composition ratio of the well layers and the barrier layers is the Al composition ratio x of the light emitting layer. By providing a light emitting layer with a multiple quantum well structure, the light emitting efficiency and light intensity of the light emitting layer can be improved.

The light-emitting layer may have a double quantum well structure, for example, "barrier layer/well layer/barrier layer/well layer/barrier layer." The thickness of each of these well layers may be, for example, 4 nm, the thickness of each of these barrier layers may be, for example, 8 nm, and the thickness of the light-emitting layer may be 32 nm. The number of quantum wells in the multiple quantum well layer may be one (i.e., a single quantum well instead of a multiple quantum well), two, three, four, or five. A single well layer is preferable for the purpose of increasing the carrier density in one well layer.

(Second Guide Layer)
The second guide layer is formed on the light emitting layer. The second guide layer has a refractive index different from that of the second semiconductor layer in order to confine the light emitted by the light emitting layer in the light emitting portion. Examples of materials for forming the second guide layer include AlN, GaN, and mixed crystals thereof. A specific example of a material for forming the second guide layer is Al _x6 Ga _(1-x6) N (0≦x6≦1).
The Al composition ratio x6 of Al _x6 Ga _(1-x6) N of the second guide layer may be greater than the Al composition ratio x5 of Al _x5 Ga _(1-x5) N of the well layer. This makes it possible to confine carriers in the light emitting layer. The material forming the second guide layer may contain impurities such as a group V element other than N, such as P, As, or Sb, a group III element, such as In or B, or C, H, F, O, Si, Cd, Zn, or Be.

When the second guide layer is an n-type semiconductor, it can be made n-type by doping, for example, with Si at 1×10 ¹⁹ cm ⁻³ . When the second guide layer is a p-type semiconductor, it can be made p-type by doping, for example, with Mg at 3×10 ¹⁹ cm ⁻³ . The second guide layer may be an undoped layer. The second guide layer may have a structure in which the Al composition ratio x6 of Al _x6 Ga _(1-x6) N is graded. For example, the second guide layer may have a layer structure in which the Al composition ratio x6 of Al _x6 Ga _(1-x6) N changes continuously or stepwise from 0.5 to 0.6.

The thickness of the second guide layer is not particularly limited. The thickness of the second guide layer may be 10 nm or more in order to efficiently confine the light emitted from the light emitting layer in the light emitting portion. The thickness of the second guide layer may be 2 μm or less in order to reduce the resistance of the second guide layer.
The Al composition ratio x6 of Al _x6 Ga.sub. _(1-x6) N (0≦x6≦1) of the second guide layer and the Al composition ratio x4 of Al _x4 Ga.sub. _(1-x4) N (0≦x4≦1) of the first guide layer may be the same value or different values.

(Second Semiconductor Layer)
When the second semiconductor layer is an n-type semiconductor, it can be made n-type by doping with, for example, 1×10 ¹⁹ cm ⁻³ of Si. When the second semiconductor layer is a p-type semiconductor, it can be made p-type by doping with, for example, 3×10 ¹⁹ cm ⁻³ of Mg. The dopant concentration may be constant or non-uniform in the vertical direction of the substrate. It may be constant or non-uniform in the in-plane direction of the substrate.
The second semiconductor layer may have a structure in which the Al composition ratio of AlGaN is graded. When the second semiconductor layer has a layered structure, the second semiconductor layer may be an undoped layer that does not contain a dopant. The second semiconductor layer may have a laminated structure further including a layer with a high doping concentration on the top layer.

The second semiconductor layer may have a laminated structure of two or more layers. In that case, it is preferable that the Al composition ratio decreases toward the upper layer in order to efficiently transport carriers to the light emitting layer.
The Al composition ratio x7 of Al _x7 Ga _(1-x7) N, which is the material for forming the second semiconductor layer, may be 0. By using p-type (p-) GaN in which the Al composition ratio x7 of Al _x7 Ga _(1-x7) N is 0, as the uppermost layer of the second semiconductor layer, it is possible to reduce the contact resistance of the second electrode disposed on the second semiconductor layer.

(First Electrode)
When the first electrode is an n-type electrode, the material forming the first electrode can be a material corresponding to an n-type electrode of a general nitride semiconductor light-emitting element, so long as the first electrode is used for the purpose of injecting electrons into the first semiconductor layer. For example, when the first electrode is an n-type electrode, Ti, Al, Ni, Au, Cr, V, Zr, Hf, Nb, Ta, Mo, W and alloys thereof, ITO, etc. are applied as the forming material.

When the first electrode is a p-type electrode, the material forming the first electrode can be the same as the p-type electrode layer of a general nitride semiconductor light-emitting device, so long as the first electrode is used for the purpose of injecting holes into the nitride semiconductor light-emitting device. For example, when the first electrode is a p-type electrode, Ni, Au, Pt, Ag, Rh, Pd, Cu and alloys thereof, ITO, etc. are used as the forming material. When the first electrode is a p-type electrode, Ni, Au or alloys thereof, or ITO, which have a small contact resistance between the first electrode and the ridge semiconductor layer 17, may be used.

The first electrode may have a pad electrode on the top for the purpose of uniformly diffusing the current over the entire area of the first electrode. Examples of materials for forming the pad electrode include Au, Al, Cu, Ag, and W. From the viewpoint of electrical conductivity, the pad electrode may be formed of Au, which has high electrical conductivity among these materials. Specifically, the structure of the first electrode may include a structure in which a first contact electrode made of an alloy of Ni and Au is formed on the first semiconductor layer, and a first pad electrode made of Au is formed on the first contact electrode.
The first electrode is formed to a thickness of, for example, 240 nm.

(Second electrode)
When the second electrode is an n-type electrode, the material for forming the second electrode can be a material corresponding to the n-type electrode of a general semiconductor laser diode, if the second electrode is used for the purpose of injecting electrons into the first semiconductor layer. For example, when the second electrode is an n-type electrode, Ti, Al, Ni, Au, Cr, V, Zr, Hf, Nb, Ta, Mo, W and their alloys, or ITO, etc., can be used as the material for forming the second electrode.

When the second electrode is a p-type electrode, the material forming the second electrode can be the same as the p-type electrode layer of a general nitride semiconductor light-emitting device, so long as the second electrode is used for the purpose of injecting holes into the nitride semiconductor light-emitting device. For example, when the second electrode is a p-type electrode, Ni, Au, Pt, Ag, Rh, Pd, Cu and alloys thereof, ITO, etc. are used as the forming material. When the second electrode is a p-type electrode, Ni, Au or alloys thereof, or ITO, which have a small contact resistance between the second electrode and the second semiconductor layer, may be used.

The second electrode may have a pad electrode on the upper part for the purpose of uniformly diffusing the current over the entire area of the second electrode. Examples of materials for forming the pad electrode include Au, Al, Cu, Ag, and W. From the viewpoint of electrical conductivity, the pad electrode may be formed of Au, which has high electrical conductivity among these materials. Specifically, the structure of the second electrode may include a structure in which a second contact electrode formed of an alloy of a material selected from Ti, Al, Ni, and Au is formed on the second semiconductor layer, and a second pad electrode formed of Au is formed on the second contact electrode.
The second electrode is formed to a thickness of, for example, 60 nm. The second electrode and the first electrode may be formed to different thicknesses, or may be formed to the same thickness as the first electrode.

REFERENCE SIGNS LIST 1 Substrate 2 Laminate 21 First semiconductor layer 21a Mesa side portion of the first semiconductor layer (portion constituting the mesa portion of the first semiconductor layer)
21aa: A portion of the first semiconductor layer that becomes a resonant region (the first semiconductor layer that constitutes the resonator end face)
21b: A portion that becomes an underlayer of the first semiconductor layer; 22: A first guide layer (waveguide layer);
23 Light emitting layer (waveguide layer)
24 Second guide layer (waveguide layer)
25 Second semiconductor layer 31 Mesa portion (portion including a portion that becomes a resonance region)
32 First portion (portion other than the mesa portion)
33 Second portion (portion other than the mesa portion)
4: Insulating layer 5: Second electrode 5A: End face of second electrode in resonance direction 51: Second alloy layer (layer to become second electrode)
51a: Portion of second alloy layer formed outside the resonator end face position 52: Connector 6: First electrode 6A: End face of first electrode in resonance direction 61: First alloy layer (layer to become first electrode)
61a: Portion of the first alloy layer formed outside the cavity end face position 62: Connector 7: Mask 81: Convex portion 82: Convex portion 83: Convex portion 84: Convex portion 9: Reflective layer 200: Cavity end face 310: Resonance region 320: Underlayer 321: First upper surface of underlayer 322: Second upper surface of underlayer K: Resonance direction

Claims (8)

A method for manufacturing a semiconductor laser diode comprising: an underlayer; a resonance region formed on a part of an upper surface of the underlayer, the resonance region being formed by laminating a first semiconductor layer, a waveguide layer, and a second semiconductor layer in this order; a first electrode formed on the first semiconductor layer in the vicinity of the resonance region; and a second electrode formed on the resonance region, the first semiconductor layer of the resonance region being continuous with the first semiconductor layer on which the first electrode is formed, the method comprising the steps of:
forming a plurality of the semiconductor laser diodes on a substrate and then dividing the substrate into individual semiconductor laser diodes;
forming a stack including a first semiconductor layer, a waveguide layer, and a second semiconductor layer in this order on the substrate;
an electrode layer forming step of forming a layer to become the second electrode on the second semiconductor layer of the laminate, between the resonator end face of the resonance region and the position to be divided in the dividing step;
a first etching step for simultaneously or consecutively removing a portion of the layer to be the second electrode formed at a position outside the cavity end face and forming the cavity end face after the electrode layer forming step;
A method for manufacturing a semiconductor laser diode having the structure described above. The method further includes a mesa etching step of etching the laminate to divide the first semiconductor layer into a portion including a portion that becomes the resonance region and the other portion in a thickness direction,
2. The method for manufacturing a semiconductor laser diode according to claim 1, wherein said first electrode is formed on the upper surface of said other portion produced by said etching. The first etching step is performed by a dry etching method using a mask,
3. The method for producing a semiconductor laser diode according to claim 1, further comprising the step of: after said first etching step, performing wet etching with an alkaline aqueous solution without removing said mask. A base layer and
a resonance region formed in a part of an underlying region that is a part of an upper surface of the underlying layer, the resonance region being formed by laminating a first semiconductor layer, a waveguide layer, and a second semiconductor layer in this order;
a first electrode formed on the first semiconductor layer in the vicinity of the resonance region;
A second electrode formed on the resonance region;
a protrusion protruding from both a surface of the resonance region that constitutes a resonator end face and an upper surface of the base region;
having
the first semiconductor layer of the resonance region and the first semiconductor layer on which the first electrode is formed are continuous;
the upper surface of the base region is a flat surface on a side of the protrusion away from the cavity end surface,
The semiconductor laser diode has no recess between the tip of the protrusion and the upper surface . The semiconductor laser diode according to claim 4, wherein the resonator end face and the end face of the second electrode in the resonance direction of the resonance region are in the same plane. The semiconductor laser diode according to claim 4 or 5, wherein the protrusion has a maximum protruding dimension from the upper surface of the base layer at a protruding position from the surface that constitutes the resonator end face. the second electrode is formed on the second semiconductor layer via an insulating layer;
a connector that electrically connects the second electrode and the second semiconductor layer is formed in the through hole of the insulating layer;
the protrusion has a triangular prism shape, one side of the triangular prism is in contact with the first semiconductor layer constituting the cavity end face, the other side is in contact with the upper surface of the base layer, and the remaining side is present upward as a slope,
7. The semiconductor laser diode according to claim 4, wherein a dimension of the protrusion in a direction perpendicular to the resonance direction of the resonance region in a plan view is larger than that of the connector. The semiconductor laser diode according to any one of claims 4 to 7, wherein a reflective layer is formed on the end face of the resonator and the end face of the second electrode in the resonance direction of the resonance region, and the film thickness of the reflective layer on the end face of the second electrode is thicker than that of the resonator end face.

Images