JP2013229568A - Semiconductor optical device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a semiconductor optical element having a structure by which an optical field is unevenly distributed to an n-type clad layer side and that is less likely to generate crystal defects.SOLUTION: A mesa stripe structure having a lamination structure obtained by holding an active layer between an n-type clad layer and a p-type clad layer is arranged on a substrate made of InP. A current blocking region is buried on a substrate on both sides of the mesa stripe structure. A pair of electrodes injecting carriers to the active layer are formed. The n-type clad layer includes a high-refractive index layer made of AlGaInAs lattice-matching with the InP substrate or AlInAs lattice-matching with the InP substrate. For one example, the current blocking region includes an n-type InP-based material layer and a p-type InP-based material layer. A vertical relation between the n-type InP-based material layer and the p-type InP-based material layer in the current blocking region is reverse to that between the n-type clad layer and the p-type clad layer of the mesa stripe structure.

Description

  The present invention relates to a semiconductor optical device including an active layer in a mesa stripe structure.

  A buried heterostructure (BH structure) in which a GaInAsP-based active layer is buried with an InP clad is employed in a high-power laser element of a band of 1.2 μm to 1.6 μm on an InP substrate. One of the causes of optical loss of a semiconductor laser is absorption between valence bands. Since the magnitude of the valence band absorption is proportional to the hole density, the valence band absorption in the p-type cladding layer has a great influence on the optical loss.

  Patent Document 1 listed below discloses a semiconductor laser in which an optical field is unevenly distributed on the n-type clad side. In this semiconductor laser, a clad made of n-type InP and p-type InP is used. An optical field control layer made of n-type GaInAsP is inserted into a part of the n-type cladding layer. In the semiconductor laser disclosed in Patent Document 2 below, the n-type cladding layer is formed of GaInAsP. Since the refractive index of GaInAsP is larger than that of InP, the light field is unevenly distributed toward the n-type cladding layer.

  By making the optical field unevenly distributed from the p-type cladding layer to the n-type cladding layer, it is possible to reduce optical loss due to valence band absorption in the p-type cladding layer.

  In the simulation described in this specification, the physical constants of compound semiconductors disclosed in Non-Patent Document 1 and Non-Patent Document 2 below were used.

JP 2000-174394 A Japanese Patent Laid-Open No. 2004-153212

Jpn. J. Appl. Phys., Vol.19, No.2, pp.L137-L140, (1980), K. Utaka et.al. IEEE Photonics Technology Letters, vol. 4, No. 6, pp. 627-630 (1992), M. J. Mondry et al.

  If the composition is such that the refractive index of the cladding layer made of n-type GaInAsP is increased under the condition of lattice matching with InP, the band gap energy is reduced. The band gap energy of the n-type cladding layer is preferably close to the band gap energy of InP in order to confine carriers in the active layer. When the band gap energy of GaInAsP is made closer to the band gap energy of InP, the difference in refractive index between InP and GaInAsP is also reduced. If the refractive index difference between the n-type cladding layer and the p-type cladding layer is small, the n-type cladding layer must be made sufficiently thick in order to make the optical field unevenly distributed on the n-type cladding layer side. For example, the thickness of the n-type cladding layer of the semiconductor laser disclosed in Patent Document 2 is 7.5 μm.

When GaInAsP is formed in such a thickness, the composition of each element tends to deviate from the target composition. When the composition is shifted, the lattice constant varies. When the region where the lattice constant fluctuates exceeds the critical film thickness, crystal defects are likely to occur.

  An object of the present invention is to provide a semiconductor optical device having a structure in which an optical field is unevenly distributed on the n-type cladding layer side and crystal defects are hardly generated.

According to one aspect of the invention,
A substrate made of InP;
A mesa stripe structure including a stacked structure in which an active layer is sandwiched between an n-type cladding layer and a p-type cladding layer, and disposed on the substrate;
A current blocking region disposed on the substrate on both sides of the mesa stripe structure;
A pair of electrodes for injecting carriers into the active layer;
The n-type cladding layer is provided with a semiconductor optical device including an AlGaInAs lattice-matched to the substrate made of InP or a high refractive index layer made of AlInAs lattice-matched to the substrate made of InP.

  As an example, the thickness of the high refractive index layer made of AlGaInAs is 160 nm or more, and the thickness of the high refractive index layer made of AlInAs is 1200 nm or more.

  Furthermore, as an example, the current blocking region includes at least an n-type InP-based material layer and a p-type InP-based material layer. The vertical relationship between the n-type InP-based material layer and the p-type InP-based material layer is opposite to the vertical relationship between the n-type cladding layer and the p-type cladding layer having the mesa stripe structure. Here, the InP-based material refers to a material containing In and P, such as InP, GaInAsP, and AlGaInAs, which can be grown on an InP substrate.

  As another example, a semi-insulating InP-based material layer may be used in the current blocking region. Examples of the semi-insulating InP-based material include materials such as Fe-doped InP.

  Since the n-type cladding layer includes the high refractive index layer, the optical field is unevenly distributed from the p-type cladding layer to the n-type cladding layer. For this reason, the optical loss resulting from the absorption between valence bands in a p-type cladding layer can be suppressed.

1A is a plan view of a semiconductor optical device according to Example 1, and FIG. 1B is a cross-sectional view taken along one-dot chain line 1B-1B in FIG. 1A. FIG. 2 is a cross-sectional view of the core layer of the semiconductor optical device according to the first embodiment. 3A and 3B are cross-sectional views in the course of manufacturing the semiconductor optical device according to the first embodiment. 3C and 3D are cross-sectional views of the semiconductor optical device according to Example 1 in the course of manufacturing. FIG. 4 is a graph showing the relationship between the band gap energy and the refractive index in the wavelength region of 3600 nm when the composition ratio of AlGaInAs and GaInAsP is changed under conditions of lattice matching with InP. FIG. 5 is a graph showing the relationship between the band gap energy when the composition ratio of AlGaInAs and GaInAsP is changed under the condition of lattice matching with InP and the refractive index in the wavelength region of 1480 nm. FIG. 6A is a cross-sectional view of Sample A to Sample C to be simulated, and FIG. 6B is a graph showing a simulation result. FIG. 7 is a cross-sectional view of a semiconductor optical device according to a modification of the first embodiment. FIG. 8 is a sectional view of a semiconductor optical device according to the second embodiment. FIG. 9 is a cross-sectional view of the semiconductor optical device according to the third embodiment. FIG. 10 is a sectional view of a semiconductor optical device according to the fourth embodiment. FIG. 11 is a cross-sectional view of a semiconductor optical device according to the fifth embodiment. FIG. 12 is a sectional view of a semiconductor optical device according to Example 6. FIG. 13A is a cross-sectional view of the semiconductor optical device according to Example 7 in the middle of manufacturing, and FIG. 13B is a cross-sectional view of the semiconductor optical device according to Example 7. 14A is a cross-sectional view of the semiconductor optical device according to Example 8 in the middle of manufacturing, and FIG. 14B is a cross-sectional view of the semiconductor optical device according to Example 8. 15A is a cross-sectional view of the semiconductor optical device according to Example 9 in the middle of manufacturing, and FIG. 15B is a cross-sectional view of the semiconductor optical device according to Example 9. FIG. 16 is a schematic sectional view of an excitation light source according to the tenth embodiment. 17A to 17E are cross-sectional views of the semiconductor optical device and the heat sink when the semiconductor optical device according to the first to fifth embodiments is mounted on a heat sink.

[Example 1]
FIG. 1A shows a plan view of a semiconductor laser as an example of the semiconductor optical device 20 according to the first embodiment. A mesa stripe structure 30 that is long in one direction is formed on a substrate 21 having a rectangular planar shape. The substrate 21 and the mesa stripe structure 30 have a pair of end faces that intersect the longitudinal direction of the mesa stripe structure 30. A high reflection film 31 is formed on one end face, and a low reflection film 32 is formed on the other end face. A laser beam generated by stimulated emission in the mesa stripe structure 30 is emitted from the low reflection film 32 to the outside. The semiconductor optical device can be operated as a distributed feedback (DFB) laser by forming the low reflection film 32 with a non-reflection coating. As another example, it is possible to operate the semiconductor optical device as a semiconductor optical amplifier (SOA) by using anti-reflection coatings at both ends.

  FIG. 1B is a cross-sectional view taken along one-dot chain line 1B-1B in FIG. 1A. A mesa stripe structure 30 is formed on a substrate 21 made of n-type InP. The plane orientation of the substrate 21 is (100). The mesa stripe structure 30 includes a laminated structure in which the active layer 24 is sandwiched between the n-type cladding layer 22 and the p-type cladding layer 26. In Example 1, the n-type cladding layer 22 is disposed on the substrate side from the active layer 24.

  An n-side isolation and confinement heterostructure (SCH) layer 23 is disposed between the active layer 24 and the n-type cladding layer 22, and a p-side isolation and closure is provided between the active layer 24 and the p-type cladding layer 26. A buried heterostructure (SCH) layer 25 is disposed. The n-type cladding layer 22 is made of n-type AlGaInAs, and the p-type cladding layer 26 is made of p-type InP. The composition ratio of each element of AlGaInAs forming the n-type cladding layer 22 is selected so as to lattice match with the InP substrate 21. In this specification, a laminated structure including the n-side SCH layer 23, the active layer 24, and the p-side SCH layer 25 is referred to as a “core layer” 35. The detailed structure of the core layer 35 will be described later with reference to FIG. A portion of the n-type cladding layer 22 on the substrate side extends to the regions on both sides of the mesa stripe structure 30 and covers the entire surface of the substrate 21.

  Current blocking regions 40 are respectively disposed on the substrates on both sides of the mesa stripe structure 30. The current blocking region 40 includes a p-type InP layer 41 and an n-type InP layer 42. The vertical relationship between the n-type InP layer 42 and the p-type InP layer 41 is opposite to the vertical relationship between the n-type cladding layer 22 and the p-type cladding layer 26 of the mesa stripe structure 30. In the first embodiment, the p-type InP layer 41 is disposed under the n-type InP layer 42.

  Instead of the p-type InP layer 41 and the n-type InP layer 42, a p-type InP-based material layer and an n-type InP-based material layer may be used, respectively. Here, the “InP-based material” means a compound semiconductor material that contains In as a group III element and P as a group V element and can be epitaxially grown on an InP substrate. Examples of InP-based materials include InP and GaInAsP.

  A second p-type cladding layer 27 is disposed on the p-type cladding layer 26 and the current blocking region 40. The second p-type cladding layer 27 is made of p-type InP. A p-type contact layer 33 made of p-type GaInAsP or p-type InGaAs is disposed on the second p-type cladding layer 27. A p-side electrode 28 is formed on the p-type contact layer 33. An n-side electrode 29 is formed on the back surface of the substrate 21. The p-side electrode 28 is in ohmic contact with the p-type contact layer 33, and the n-side electrode 29 is in ohmic contact with the substrate 21.

  A DC voltage is applied to the n-side electrode 29 and the p-side electrode 28 so that a forward bias is applied in the thickness direction of the mesa stripe structure 30. At this time, since a reverse bias is applied to the pn junction of the current blocking region 40, almost no current flows in the current blocking region 40. For this reason, current concentrates on the mesa stripe structure 30, and the efficiency of carrier injection into the active layer 24 can be increased.

  The refractive index of the n-type cladding layer 22 is higher than the refractive index of the p-type cladding layer 26. For this reason, the light field 45 is unevenly distributed from the p-type cladding layer 26 toward the n-type cladding layer 22.

  FIG. 2 shows a cross-sectional view of the core layer 35. The active layer 24 is sandwiched between the n-side SCH layer 23 and the p-side SCH layer 25. The active layer 24 has a multiple quantum well structure including a barrier layer 24B and a well layer 24W. The barrier layers 24B and the well layers 24W are alternately stacked, and the barrier layers 24B are disposed at both ends in the stacking direction. The barrier layer 24B and the well layer 24W are made of undoped AlGaInAs. The composition ratio of the constituent elements is selected so that the band gap energy of the barrier layer 24B is larger than the band gap energy of the well layer 24W. The barrier layer 24B may be formed of undoped AlInAs.

  The n-side SCH layer 23 is composed of three undoped AlGaInAs layers 23A to 23C having different Al composition ratios. The Al composition ratio in the n-side SCH layer 23 decreases stepwise from the n-type cladding layer 22 toward the active layer 24. Four or more AlGaInAs layers having different Al composition ratios may be used. Alternatively, the n-side SCH layer 23 may be configured such that the Al composition ratio decreases continuously toward the active layer 24. The p-side SCH layer 25 has a structure in which the n-side SCH layer 23 is turned upside down. That is, the p-side SCH layer 25 also includes three undoped AlGaInAs layers 25A to 25C having different Al composition ratios. The Al composition ratio in the p-side SCH layer 25 also decreases stepwise or continuously from the p-type cladding layer 26 toward the active layer 24.

  The n-side SCH layer 23 and the p-side SCH layer 25 have band gap energy larger than that of the active layer 24 and confine carriers in the active layer 24. The refractive index of the n-side SCH layer 23 and the p-side SCH layer 25 is smaller than the refractive index of the active layer 24. However, since the n-side SCH layer 23 and the p-side SCH layer 25 do not have a sufficient thickness for confining light, the light confinement function of these layers is small. The optical confinement function is realized mainly by the n-type cladding layer 22 and the p-type cladding layer 26.

In Example 1, the n-side SCH layer 23, the active layer 24, and the p-side SCH layer 25 are formed of AlGaIn.
Although an example using As will be described, GaInAsP may be used for the n-side SCH layer 23, the active layer 24, and the p-side SCH layer 25.

  Next, with reference to FIGS. 3A to 3D, a method for manufacturing a semiconductor optical device according to Example 1 will be described.

  As shown in FIG. 3A, an n-type cladding layer 22, an n-side SCH layer 23, an active layer 24, a p-side SCH layer 25, and a p-type cladding layer 26 are sequentially deposited on a substrate 21 made of n-type InP. Let For the deposition of these films, for example, a metal organic chemical vapor deposition (MOCVD) method or the like can be applied. A mask pattern 46 is formed on the p-type cladding layer 26. The mask pattern 46 has a planar shape that matches the mesa stripe structure 30 (FIG. 1A). In the following description, the n-side SCH layer 23, the active layer 24, and the p-side SCH layer 25 are collectively referred to as a core layer 35.

  As shown in FIG. 3B, etching is performed from the p-type cladding layer 26 to the middle of the n-type cladding layer 22 in the thickness direction using the mask pattern 46 as an etching mask. For this etching, dry etching or wet etching can be applied. By this etching, the mesa stripe structure 30 including the n-type cladding layer 22, the core layer 35, and the p-type cladding layer 26 is formed. The side surface of the mesa stripe structure 30 is inclined such that the bottom of the mesa stripe structure 30 is wider than the top. The mesa stripe structure 30 has a so-called “forward mesa structure”. A portion of the n-type cladding layer 22 on the substrate side remains on the entire surface of the substrate 21.

  As shown in FIG. 3C, a p-type InP layer 41 and an n-type InP layer 42 are formed on the n-type cladding layer 22 on both sides of the mesa stripe structure 30. Thereby, current blocking regions 40 embedded on both sides of the mesa stripe structure 30 are formed. For the formation of the p-type InP layer 41 and the n-type InP layer 42, selective growth using the mask pattern 46 as a mask is applied. After forming the current blocking region 40, the mask pattern 46 is removed.

  As shown in FIG. 3D, the p-type cladding layer 27 and the p-type contact layer 33 are formed on the mesa stripe structure 30 and the current blocking region 40. For example, the MOCVD method is applied to the formation of the p-type cladding layer 27 and the p-type contact layer 33. Thereafter, as shown in FIG. 1B, a p-side electrode 28 and an n-side electrode 29 are formed. Cleaving in a direction perpendicular to the mesa stripe structure 30 is made to be a laser bar in which a plurality of laser chips are arranged in the horizontal direction. By this cleavage, the end face of the laser chip is exposed. As shown in FIG. 1A, the high reflection film 31 and the low reflection film 32 are formed on the exposed end face of the laser bar, and the laminated structure from the p-side electrode 28 to the n-side electrode 29 is divided into laser chips.

  The effect of using AlGaInAs for the n-type cladding layer 22 will be described with reference to FIGS. FIG. 4 shows the relationship between the band gap energy when the composition ratio of AlGaInAs and GaInAsP is changed under the condition of lattice matching with InP and the refractive index in the wavelength region of about 3600 nm. FIG. 5 shows the relationship between the band gap energy when the composition ratio of AlGaInAs and GaInAsP is changed under the condition of lattice matching with InP and the refractive index in the wavelength region of 1480 nm. As the physical property values of AlGaInAs and GaInAsP in the wavelength region of 1480 nm, those disclosed in Non-Patent Document 1 and Non-Patent Document 2 described above were used. A laser light source having a wavelength of 1480 nm is used as, for example, a pump laser light source for an erbium-doped fiber amplifier.

4 and 5, the horizontal axis represents the band gap energy in the unit “eV”, and the vertical axis represents the refractive index. The solid line shown in FIGS. 4 and 5 shows the relationship between the band gap energy and the refractive index when the composition ratio of AlGaInAs is changed, and the broken line shows the band gap energy when the composition ratio of GaInAsP is changed. The relationship with the refractive index is shown.

  In FIG. 4, the upper left end points of the solid line and the broken line indicate the band gap energy and refractive index of InGaAs. 4 and 5, the lower right end point of the solid line shows the band gap energy and refractive index of AlInAs, and the lower right end point of the broken line shows the band gap energy and refractive index of InP. The effective refractive index and band gap energy of the active layer 24 are values within the range of the region 24A. That is, the refractive index of the active layer 24 is about 3.35 to 3.4. This band gap energy and refractive index can be realized in any configuration using AlGaInAs and GaInAsP in the active layer. Under the condition that the band gap energy of the n-type cladding layer 22 made of AlGaInAs and the band gap energy of the p-type cladding layer 27 made of InP are substantially the same, the composition ratio of AlGaInAs is selected from within the region 22A. . At this time, the refractive index in the wavelength region of 3600 nm of the n-type cladding layer 22 is about 3.24 (FIG. 4), and the refractive index in the wavelength region of 1480 nm is about 3.245 (FIG. 5).

  As a comparative example, consider the case where the n-type cladding layer 22 is formed of GaInAsP lattice-matched to InP. In the comparative example, the band gap energy of the n-type cladding layer 22 cannot be made the same as the band gap energy of the p-type cladding layer 26 made of InP. When the band gap energy of the n-type cladding layer 22 made of GaInAsP is set to a value close to the band gap energy of InP, the refractive index difference between GaInAsP and InP becomes small. When the refractive index difference between the two is small, a sufficient effect of unevenly distributing the optical field from the p-type cladding layer 26 toward the n-type cladding layer 22 cannot be obtained.

  When the refractive index of the n-type cladding layer 22 made of GaInAsP is set to the same level as that of the first embodiment (3.24 in the condition of FIG. 4 and 3.245 in the condition of FIG. 5), the band gap energy is as shown in FIG. Under the condition of 4, the voltage is narrowed to about 1.05 eV, and under the condition of FIG. Thus, when the n-type cladding layer 22 is formed of AlGaInAs, the band gap energy of the n-type cladding layer 22 is close to the band gap energy of InP as compared to the case of forming it with GaInAsP. The refractive index of the n-type cladding layer 22 can be made sufficiently larger than the refractive index of the p-type cladding layer 26.

  When the refractive index of the n-type cladding layer 22 is sufficiently higher than the refractive index of the p-type cladding layer 26, the optical field is more unevenly distributed toward the n-type cladding layer 22. In order to use GaInAsP for the n-type cladding layer 22 so that the optical field is unevenly distributed to the same extent as in the first embodiment, the n-type cladding layer 22 must be thickened. When AlGaInAs is used for the n-type cladding layer 22, it is possible to make the n-type cladding layer 22 thinner while maintaining the effect of unevenly distributing the optical field as compared with the case of using GaInAsP.

  In the first embodiment, the n-type cladding layer 22 is formed of a single layer made of n-type AlGaInAs, but may be formed of a plurality of mixed crystal semiconductor layers having different compositions. At this time, at least one layer is a high refractive index layer containing AlGaInAs. This high refractive index layer has the same function as the n-type cladding layer 22 made of AlGaInAs shown in FIG. 1B. That is, the refractive index of the high refractive index layer is higher than the refractive index of the p-type cladding layer 26. In Example 1, the n-type cladding layer 22 functions as a high refractive index layer.

Next, with reference to FIG. 6A and FIG. 6B, the effect of Example 1 is demonstrated concretely. The optical field distribution of sample A (FIG. 6A) using a GaInAsP-based semiconductor for the active layer 24 and sample B (FIG. 6A) using an AlGaInAs-based semiconductor for the active layer 24 was obtained by simulation. Hereinafter, simulation results will be described.

  FIG. 6A shows a cross-sectional view of a sample A and a sample B to be simulated. The basic laminated structure from the substrate 21 to the p-type cladding layer 26 is the same as the laminated structure of the semiconductor optical device according to the first embodiment (FIG. 1B). Sample C will be referred to in the description of Examples later.

  In the sample A, the n-type cladding layer 22, the n-side SCH layer 23, the active layer 24, and the p-side SCH layer 25 are formed of a GaInAsP-based semiconductor. The thickness of the well layer 24W of the active layer 24 is 4 nm, and the thickness of the barrier layer 24B is 10 nm. The composition wavelength of the barrier layer 24B is 1.2 μm. Here, the “composition wavelength” represents the band gap energy of the mixed crystal semiconductor in terms of the wavelength of light, and corresponds to the wavelength at which the light intensity of the photoluminescence output shows the maximum value. The oscillation wavelength is adjusted to 1480 nm.

In order from the substrate side, the n-side SCH layer 23 is a GaInAsP layer having a composition wavelength of 1.10 μm and a thickness of 7 nm, a GaInAsP layer having a composition wavelength of 1.15 μm and a thickness of 6 nm, and a composition wavelength of 1.2 μm. And a three-layer structure in which GaInAsP layers having a thickness of 6 nm are stacked. The p-side SCH layer 25 has a three-layer structure in which the stacking order of the three layers constituting the n-side SCH layer 23 is reversed. The p-type cladding layer 26 is made of p-type InP having an impurity concentration of 7.0 × 10 ¹⁷ cm ⁻³ and has a thickness of 500 nm.

  An n-type cladding layer 22 made of n-type GaInAsP is inserted between the substrate 21 and the n-side SCH layer 23. The n-type cladding layer 22 functions as a high refractive index layer that unevenly distributes the optical field.

  In sample B, the n-side SCH layer 23, the active layer 24, and the p-side SCH layer 25 are formed of an AlGaInAs-based semiconductor. The film thickness and composition wavelength of each layer are the same as the film thickness and composition wavelength of the corresponding layer of Sample A. An n-type cladding layer 22 made of AlGaInAs is inserted between the substrate 21 and the n-side SCH layer 23. The n-type cladding layer 22 functions as a high refractive index layer that unevenly distributes the optical field.

  FIG. 6B shows the simulation result. The horizontal axis represents the thickness of the high refractive index layer in the unit “μm”, and the vertical axis represents the ratio of the light oozing out to the p-type cladding layer 26 in the unit “%”. Here, the “permeation ratio of light to the p-type cladding layer 26” is a graph representing a one-dimensional light intensity distribution in the thickness direction, and the p-type cladding layer out of the total area of the light intensity distribution. The ratio of the area of the part in 26 is meant.

  If the sample A has a configuration in which the high refractive index layer (n-type cladding layer 22) is not disposed (the point where the thickness of the high refractive index layer is 0 in FIG. 6B), the rate of light leakage into the p-type cladding layer 26 Is about 47%. As an example, if the thickness of the high refractive index layer is 7.5 μm, the rate of light leakage into the p-type cladding layer 26 is about 22.5%. Even if the thickness of the high refractive index layer is greater than 7.5 μm, the rate of light leakage to the p-type cladding layer 26 hardly decreases. As can be seen from the simulation results, when a GaInAsP-based semiconductor is used, it is difficult to make the rate of light leakage into the p-type cladding layer 26 22.5% or less.

In the case of using an AlGaInAs-based semiconductor as in Sample B, when the thickness of the high refractive index layer is 500 nm, the rate of light leakage into the p-type cladding layer 26 is about 22.5%. . If the high refractive index layer is thicker than 500 nm, the rate of light leakage into the p-type cladding layer 26 further decreases. As described above, in the sample B, by making the thickness of the high refractive index layer thicker than 500 nm, it is difficult to realize the rate of light leakage into the p-type cladding layer 26 with the configuration of the sample A using a GaInAsP-based semiconductor. It can be lowered to the value that was. By setting the thickness of the high refractive index layer made of AlGaInAs to 500 nm or more, a special effect that cannot be obtained with a configuration using GaInAsP as the high refractive index layer can be obtained.

  In Sample A to Sample C (FIG. 6A), the oscillation wavelength was adjusted to 1480 nm. Next, consider a sample whose oscillation wavelength is adjusted to 1600 nm. The laminated structure of the sample other than the active layer is the same as that of the sample B. The active layer includes a well layer made of InGaAs and a barrier layer made of AlGaInAs lattice-matched to InP. The thickness of the well layer is 6 nm, and the thickness of the barrier layer is 10 nm. The composition wavelength of the barrier layer is 1.573 μm.

  As the n-type cladding layer 22, AlGaInAs having the same composition as the barrier layer is used. When a simulation was performed on this sample, when the thickness of the n-type cladding layer 22 was 160 nm, the rate of light leakage into the p-type cladding layer 26 was 22.5%. By setting the thickness of the n-type cladding layer 22 to 160 nm or more, a special effect that cannot be obtained with a configuration using GaInAsP as the high refractive index layer can be obtained.

  Since GaInAsP includes two types of As and P as group V elements, the in-plane variation in the composition ratio of the constituent elements tends to increase due to the variation in the growth temperature. In addition, since it is easy to deviate from the lattice matching condition during growth, the uniformity of the composition also decreases in the thickness direction when a thick film that functions as a cladding layer is formed. Furthermore, GaInAsP has an immiscible region in the relationship between the growth temperature and the composition. For this reason, it is difficult to perform stable growth.

  On the other hand, AlGaInAs contains only As as a group V element. For this reason, in-plane variation in composition is unlikely to occur, and it is easy to ensure composition uniformity in the thickness direction. Furthermore, by using AlGaInAs, the influence of the immiscible region in the relationship between the growth temperature and the composition can be reduced. For this reason, it is possible to perform stable growth.

  When a semiconductor layer containing As is grown on a base surface containing P, defects are likely to occur at a growth interface (hereinafter referred to as an As / P interface). This defect absorbs light and causes light loss. In Example 1, the interface between the substrate 21 and the n-type cladding layer 22 is an As / P interface. On the other hand, when GaInAsP is used for the n-type cladding layer 22, the interface between the n-type cladding layer 22 and the core layer 35 is an As / P interface. In the semiconductor optical device according to the first embodiment, the As / P interface is farther from the active layer 24 than when GaInAsP is used for the n-type cladding layer 22. For this reason, the optical loss resulting from the defect of an As / P interface can be reduced.

  The semiconductor optical device according to Example 1 is used as an excitation light source for an erbium-doped fiber amplifier, for example. When used in this application, the composition ratio of the constituent elements of AlGaInAs in the barrier layer 24B of the active layer 24 and the well layer 24W (FIG. 2) is such that the oscillation wavelength is a wavelength of about 1400 nm or more and less than 1500 nm, for example, 1480 nm. Selected. In addition, the semiconductor optical device according to Example 1 can be applied to a signal light source such as a full-band tunable laser module such as a 1.5 μm band. When the semiconductor optical device according to Example 1 is used as a signal light source, the composition ratio of the active layer 24 is selected so that the oscillation wavelength is 1250 nm or more and less than 1620 nm.

Furthermore, in Example 1, the n-type cladding layer 22 made of AlGaInAs having a higher refractive index than InP spreads on both sides of the mesa stripe structure 30 (FIG. 1B). For this reason, in the lateral direction, the difference in refractive index between the region in the mesa stripe structure 30 and the region outside it is small. As a result, the cut-off mesa width of the high-order transverse mode is widened. For example, even when the width of the top portion of the mesa stripe structure 30 is 3 μm or more, generation of a high-order transverse mode can be suppressed and oscillation can be performed in a single mode. By increasing the mesa width, the element resistance can be reduced.

  FIG. 7 shows a cross-sectional view of a semiconductor optical device according to a modification of the first embodiment. In this modification, a single-layer semi-insulating InP-based material layer is used for the current blocking region 40. Examples of the semi-insulating InP-based material include Fe-doped InP.

[Example 2]
FIG. 8 is a sectional view of a semiconductor optical device according to the second embodiment. Hereinafter, differences from the first embodiment will be described, and description of the same configuration will be omitted. In Example 1, as shown in FIG. 1B, a part of the n-type cladding layer 22 on the substrate side was disposed over the entire surface of the substrate 21. In Example 2, the bottom surface of the n-type cladding layer 22 is disposed at a position higher than the bottom surfaces of the current blocking regions 40 on both sides of the mesa stripe structure 30. That is, the n-type cladding layer 22 does not spread on both sides of the mesa stripe structure 30.

  The composition ratio of the constituent elements of the n-type cladding layer 22 is selected so that the band gap energy is about 1.1 eV (the value in the region 22B in FIGS. 4 and 5). When the band gap energy is 1.1 eV, the refractive index of AlGaInAs is about 3.3 in the wavelength region of 3600 nm and 3.33 in the wavelength region of 1480 nm. The refractive index of the n-type cladding layer 22 of Example 1 was about 3.24 in the wavelength region of 3600 nm and 3.245 in the wavelength region of 1480 nm. Since the refractive index of the n-type cladding layer 22 of Example 2 is higher than the refractive index of the n-type cladding layer 22 of Example 1, even if the n-type cladding layer 22 is made thin, a sufficient effect of unevenly distributing the optical field is obtained. Can be obtained.

  In Example 1, the thickness of the n-type cladding layer 22 was about 2.5 μm. In the second embodiment, when the refractive index of the n-type cladding layer 22 is 3.33, the light field can be unevenly distributed to the same extent as in the first embodiment even if the thickness is reduced to 1.5 μm.

  Further, in Example 1, the n-type clad layer 22 containing Al is exposed to the entire surface of the substrate during the manufacturing process shown in FIG. 3B. Since there are many active bonds in the (100) plane of AlGaInAs, the exposed surface is easily oxidized. For this reason, it is preferable to perform a sufficient cleaning process on the surface of the n-type clad layer 22 which is the base of the growth before the buried growth of the current blocking region 40 shown in FIG. 3C. In Example 2, the substrate 21 made of n-type InP is exposed as a base when the current blocking region 40 is regrown. For this reason, high-quality buried growth can be performed.

  The n-type cladding layer 22 made of AlGaInAs is exposed on the side surface of the mesa stripe structure 30, but this exposed surface has fewer active bonds than the (100) plane. Therefore, it is possible to maintain a sufficient quality of buried growth. Furthermore, since impurities are hardly mixed, the characteristics and reliability of the semiconductor optical device can be improved.

  Further, as shown in FIG. 2, the Al composition ratio of the n-side SCH layer 23 and the p-side SCH layer 25 decreases from the cladding layer toward the active layer 24. Therefore, oxidation in the vicinity of the active layer 24 is suppressed in the region exposed on the side surface of the mesa stripe structure 30 shown in FIG. Since defects due to oxidation hardly occur in the vicinity of the active layer 24, light loss can be reduced.

[Example 3]
FIG. 9 shows a cross-sectional view of a semiconductor optical device according to the third embodiment. Hereinafter, differences from the second embodiment shown in FIG. 8 will be described, and description of the same configuration will be omitted. In Example 2, the current blocking region 40 is composed of two layers of the p-type InP layer 41 and the n-type InP layer 42. In Example 3, the current blocking region 40 includes an AlGaInAs layer 43 in addition to the two layers. FIG. 9 shows an example in which the n-type AlGaInAs layer 43 is disposed between the p-type InP layer 41 and the n-type InP layer 42. The AlGaInAs layer 43 may be inserted at another position in the current blocking region 40. For example, it may be inserted between the p-type InP layer 41 and the substrate 21. In the case of this configuration, the conductivity type of AlGaInAs is p-type. Further, the n-type AlGaInAs layer 43 may be disposed on the n-type InP layer 42. The AlGaInAs layer 43 is thinner than the n-type cladding layer 22 made of n-type AlGaInAs in the mesa stripe structure 30 or has the same thickness as the n-type cladding layer 22.

  In Example 3, the AlGaInAs layers 43 having a higher refractive index than InP are disposed on both sides of the mesa stripe structure 30. For this reason, compared with Example 2, the difference in refractive index in the lateral direction is reduced. For this reason, the cut-off mesa width of the high-order transverse mode is widened. That is, even if the mesa width W at the top of the mesa stripe structure 30 is widened, it is possible to oscillate in a single mode. Compared to the second embodiment, the mesa width W can be increased, so that the element resistance can be reduced.

[Example 4]
FIG. 10 is a sectional view of a semiconductor optical device according to the fourth embodiment. Hereinafter, differences from the second embodiment shown in FIG. 8 will be described, and description of the same configuration will be omitted. In the second embodiment, n-type InP is used for the substrate 21, but in the fourth embodiment, p-type InP is used for the substrate 21. The substrate 21 made of p-type InP also serves as a p-type cladding layer. An n-type cladding layer 50 made of n-type AlGaInAs is disposed on the core layer 35. The mesa stripe structure 30 includes a part of the substrate 21 made of p-type InP, a core layer 35, and an n-type cladding layer 50. The vertical relationship between the n-type InP layer and the p-type InP layer constituting the current blocking region 40 is the vertical relationship between the n-type InP layer 42 and the p-type InP layer 41 in the current blocking region 40 (FIG. 8) of the second embodiment. The opposite is true.

  A second n-type cladding layer 51 made of n-type InP is disposed on the n-type cladding layer 50 and the current blocking region 40. An n-type contact layer 54 made of n-type GaInAsP or n-type GaInAs is formed on the n-type cladding layer 51. An n-side electrode 52 is formed on the n-type contact layer 54. A p-side electrode 53 is formed on the back surface of the substrate 21.

  Also in Example 4, the refractive index of the n-type cladding layer 50 is higher than the refractive index of the p-type cladding layer (the substrate 21 made of p-type InP). For this reason, the optical field can be unevenly distributed from the p-type cladding layer (substrate 21) to the n-type cladding layer 50 (away from the substrate 21).

[Example 5]
FIG. 11 is a sectional view of a semiconductor optical device according to the fifth embodiment. Hereinafter, differences from the fourth embodiment shown in FIG. 10 will be described, and description of the same configuration will be omitted. In the fifth embodiment, a p-type AlGaInAs layer 55 is inserted in the current blocking region 40 as in the third embodiment shown in FIG. As an example, a p-type AlGaInAs layer 55 is disposed between an n-type InP layer on the substrate side and an upper p-type InP layer.

  In the fifth embodiment, similarly to the third embodiment, the cut-off mesa width in the high-order transverse mode is widened. Compared to the fourth embodiment, the mesa width W can be increased, so that the element resistance can be reduced.

[Example 6]
A semiconductor optical device according to Example 6 will be described with reference to FIG. Hereinafter, differences from the first embodiment will be described, and description of the same configuration will be omitted.

  FIG. 12 is a sectional view of a semiconductor optical device according to the sixth embodiment. In Example 1, the n-type cladding layer 22 was formed of n-type AlGaInAs, but in Example 6, the n-type cladding layer 22 was formed of n-type AlInAs. The n-type cladding layer 22 made of n-type AlInAs functions as a high refractive index layer for unevenly distributing the optical field 45 from the p-type cladding layer 26 to the n-type cladding layer 22 side.

  When AlInAs is used for the n-type cladding layer 22, the refractive index difference between the n-type cladding layer 22 and the p-type cladding layer 26 made of InP is smaller than that in Example 1 in which AlGaInAs is used for the n-type cladding layer 22. (See FIGS. 4 and 5). For this reason, the effect of reducing the rate of light leakage to the p-type cladding layer 26 is lower than that of the first embodiment.

  With reference to FIGS. 6A and 6B, effects of the semiconductor optical device according to Example 6 will be described. A simulation was performed for Sample C corresponding to Example 6. In the sample C, the material of the n-type cladding layer 22 of the sample B is changed from n-type AlGaInAs to n-type AlInAs. As shown in FIG. 6B, even when n-type AlInAs is used for the n-type cladding layer 22, it is possible to reduce the rate of light leakage to the p-type cladding layer 26 compared to the sample A.

  In the case of sample C, the thickness of the high refractive index layer (n-type clad layer 22) is made thicker than 1200 nm, so that the rate of light leakage into the p-type clad layer 26 is determined using a sample using a GaInAsP-based semiconductor. The value can be lowered to a value that is difficult to realize with the configuration of A.

  AlInAs, like AlGaInAs, contains only one type of As as a group V element. For this reason, as in the case of Example 1, in-plane variation in the composition ratio of the n-type cladding layer 22 can be reduced as compared with the case where GaInAsP is used for the n-type cladding layer 22. AlGaInAs contains three types of elements as group III elements, but AlInAs contains only two types of elements as group III elements. For this reason, in Example 6, the in-plane variation in the composition ratio of the n-type cladding layer 22 can be further reduced as compared with the case where AlGaInAs is used for the n-type cladding layer 22.

  In Example 6 in which AlInAs is used for the n-type cladding layer 22, the n-type cladding layer 22, that is, the high refractive index layer is thickened in order to obtain the same effect as in Example 1 in which AlGaInAs is used for the n-type cladding layer 22. Must. However, when AlInAs is used, the in-plane variation of the composition ratio can be reduced, so that a thick film can be easily formed as compared with AlGaInAs. For this reason, making the n-type cladding layer 22 made of AlInAs thicker than the n-type cladding layer 22 made of AlGaInAs in Example 1 does not hinder the adoption of the configuration of Example 6.

[Example 7]
A semiconductor optical device according to Example 7 will be described with reference to FIGS. 13A and 13B. Hereinafter, differences from the first embodiment will be described, and description of the same configuration will be omitted.

FIG. 13A shows a cross-sectional view of the semiconductor optical device corresponding to the end of the etching process shown in FIG. 3B of the first embodiment. In Example 1, the n-type cladding layer 22 was etched halfway in the thickness direction on the side of the mesa stripe structure 30. On the other hand, in Example 7, the core layer 35 is etched to the bottom surface on the side of the mesa stripe structure 30, and the etching is stopped on the upper surface of the n-type cladding layer 22.

  FIG. 13B shows a sectional view of the semiconductor optical device according to the seventh embodiment. In the seventh embodiment, the upper surface of the n-type cladding layer 22 immediately below the core layer 35 and the upper surface of the n-type cladding layer 22 below the current blocking region 40 have substantially the same height. Also in the seventh embodiment, since the n-type cladding layer 22 having a higher refractive index than the p-type cladding layer 26 is disposed immediately below the core layer 35, the p-type cladding layer 26 is exposed to the p-type cladding layer 26 as in the first embodiment. The ratio of light oozing can be reduced.

  Also in Example 7, as in Example 6, the n-type cladding layer 22 may be formed of n-type AlInAs.

  In the said Example 1- Example 8, n-type AlGaInAs or n-type AlInAs was used as a high refractive index layer. As another configuration example, the high refractive index layer is a stacked structure in which n-type AlGaInAs layers and n-type AlInAs are alternately stacked, a stacked structure in which n-type InP layers and n-type AlGaInAs layers are alternately stacked, or A stacked structure in which n-type InP layers and n-type AlInAs layers are alternately stacked may be employed. The thickness of each layer constituting the high refractive index layer and the number of stacked layers can be arbitrarily set based on the effective refractive index and thickness required for the high refractive index layer.

[Example 8]
A semiconductor optical device according to Example 8 will be described with reference to FIGS. 14A and 14B. Hereinafter, differences from the seventh embodiment will be described, and description of the same configuration will be omitted.

  FIG. 14A shows a cross-sectional view of the semiconductor optical device corresponding to the end of the etching step shown in FIG. In Example 8, an intermediate layer 36 made of n-type InP is formed on the n-type cladding layer 22. The core layer 35 is disposed on the intermediate layer 36. The thickness of the intermediate layer 36 is, for example, 3 nm to 20 nm.

  The core layer 35 is etched under the condition that the etching rate of the core layer 35 is higher than the etching rate of the intermediate layer 36. When etching is performed under these etching conditions, the intermediate layer 36 functions as an etching stopper. For this reason, variation in etching depth can be reduced.

  FIG. 14B shows a sectional view of a semiconductor optical device according to Example 8. A current blocking region 40 is formed on the intermediate layer 36 on both sides of the mesa stripe structure 30. At the end of etching of the core layer 35, the n-type cladding layer 22 made of n-type AlGaInAs is protected by the intermediate layer 36 and is not exposed. In general, a semiconductor layer containing Al is easily oxidized. In Example 8, since the n-type cladding layer 22 containing Al is not exposed even after the etching of the core layer 35, the oxidation of the n-type cladding layer 22 can be prevented. As the n-type cladding layer 22, n-type AlGaInAs may be used as in the first embodiment, or n-type AlInAs may be used as in the sixth embodiment.

  Furthermore, in Example 8, the current blocking region 40 made of InP is regrown on the intermediate layer 36 made of InP. Since the base surface when the current blocking region 40 is regrown is the same as the composition of the current blocking region 40, a good regrowth interface can be obtained. Thereby, the reliability and yield of the semiconductor optical device are improved.

  In the eighth embodiment, the intermediate layer 36 is formed of n-type InP, but may be formed of undoped InP. When the intermediate layer 36 is undoped, the thickness of the intermediate layer 36 is preferably several nm or less in order to suppress an increase in device resistance.

[Example 9]
A semiconductor optical device according to Example 9 will be described with reference to FIGS. 15A and 15B. Hereinafter, differences from the eighth embodiment will be described, and description of the same configuration will be omitted. In Example 9, after the etching process shown in FIG. 14A of Example 8, the intermediate layer 36 exposed on both sides of the mesa stripe structure 30 is etched.

  FIG. 15A shows a cross-sectional view of the semiconductor optical device after the intermediate layer 36 has been etched. The intermediate layer 36 remains under the core layer 35, and the n-type cladding layer 22 is exposed on both sides of the mesa stripe structure 30. The etching of the intermediate layer 36 is performed under conditions that allow InP to be selectively etched with respect to the core layer 35 made of AlGaInAs. For example, dry etching is applied. At the time of this etching, the side surface of the p-type cladding layer 26 is exposed to the etching atmosphere. However, since the intermediate layer 36 is sufficiently thinner than the width of the mesa stripe structure 30, the etching of the side surface of the p-type cladding layer 26 is negligible. It is.

  FIG. 15B shows a cross-sectional view of the semiconductor optical device according to the eighth embodiment. A current blocking region 40 is formed on the n-type cladding layer 22 exposed on both sides of the mesa stripe structure 30. In the manufacturing method according to Example 9, the surface of the n-type cladding layer 22 is not exposed to the atmosphere from the start of etching of the intermediate layer 36 shown in FIG. 15A until the end of the formation of the current blocking region 40. For this reason, oxidation of the n-type cladding layer 22 can be prevented.

Hereinafter, an example of a method for handling a semiconductor optical device in the middle of manufacturing in a state where the surface of the n-type cladding layer 22 is not exposed to the air will be described. The substrate with the intermediate layer 36 exposed on both sides of the mesa stripe structure 30 is loaded into the chamber of a metal organic chemical vapor deposition (MOCVD) apparatus for forming the current blocking region 40. The intermediate layer 36 is selectively etched by introducing an etching gas such as CBr ₄ into the chamber. As etching gas for the intermediate layer 36, in addition CBr _4, it is also possible to use the available gas to selectively etch the InP against AlGaInAs. After the intermediate layer 36 is etched, the current blocking region 40 is selectively grown in the same chamber.

  An apparatus in which the chamber for etching the intermediate layer 36 and the chamber for film formation of the current blocking region 40 are separately arranged and connected via a transfer chamber that can be evacuated may be used.

  In the ninth embodiment, since oxidation of the base surface for re-growing the current blocking region 40 is prevented, good crystal growth of the current blocking region 40 is possible.

[Example 10]
FIG. 16 is a sectional view of a semiconductor device according to the tenth embodiment. The semiconductor device according to Example 10 is used, for example, as a pumping light source for an erbium-doped optical fiber amplifier. A temperature adjustment module 61 is mounted on the bottom surface of the ceramic package 60. For the temperature adjustment module 61, for example, a Peltier element is used. A base 62 is fixed on the temperature adjustment module 61.

  A heat sink 63, a lens 66, an isolator 67, and a photodetector 70 are fixed to the base 62. A semiconductor optical device 64 and a thermistor 65 are mounted on the heat sink 63. As the semiconductor optical device 64, any one of the semiconductor optical devices according to the first to ninth embodiments is used. A through hole is formed in the side surface of the package 60, and the lens 68 and the incident end of the optical fiber 69 are fixed in the through hole.

The temperature adjustment module 61 adjusts the temperature of the semiconductor optical device 64 via the heat sink 63. In general, the semiconductor optical device 64 is cooled by the temperature adjustment module 61. Note that the temperature adjustment module 61 may be omitted, and the semiconductor optical element 64 may be cooled by natural heat dissipation from the heat sink 63. The heat sink 63 is not limited to one having an outer shape provided with heat radiation fins. For example, a metal plate having high thermal conductivity can be used as the heat sink 63. The laser beam emitted from the semiconductor optical element 64 is collimated by the lens 66. The collimated laser beam is incident on the lens 68 via the isolator 67. The lens 68 condenses the incident laser beam on the incident end face of the optical fiber 69. The thermistor 65 measures the temperature during operation of the semiconductor optical device 64 via the heat sink 63. The laser beam leaking from the highly reflective film of the semiconductor optical device 64 is detected by the photodetector 70.

  17A to 17E are sectional views of the heat sink 63 and the semiconductor optical device 64 when the semiconductor optical device 64 of the first to ninth embodiments is mounted on the heat sink 63, respectively. The mounting form of the semiconductor optical device 64 of Examples 1 to 3 and Examples 6 to 9 is junction down. On the other hand, the mounting form of the semiconductor optical device 64 of Example 4 and Example 5 is junction-up.

  In any case, the semiconductor optical device 64 is mounted on the heat sink 63 with the n-type cladding layers 22 and 50 positioned on the opposite side of the heat sink 63 when viewed from the core layer 35. Usually, a ternary or quaternary mixed crystal semiconductor has lower thermal conductivity than a binary compound semiconductor such as InP. 17A to 17E, the ternary or quaternary mixed crystal semiconductor n-type cladding layers 22 and 50 are not disposed between the active layer 24 serving as a heat source and the heat sink 63. That is, of the pair of electrodes, the electrode opposite to the n-type cladding layers 22 and 50 when viewed from the active layer 24 is thermally coupled to the temperature adjustment module 61 (FIG. 16). With this configuration, the active layer 24 can be efficiently cooled.

  Although the present invention has been described with reference to the embodiments, the present invention is not limited thereto. It will be apparent to those skilled in the art that various modifications, improvements, combinations, and the like can be made.

20 Semiconductor Optical Device 21 Substrate 22 n-type Clad Layer 23 n-side Separated Confinement Heterostructure Layer (n-side SCH Layer)
24 active layer 24B barrier layer 24W well layer 25 p-side isolation confinement heterostructure layer (p-side SCH layer)
26 p-type cladding layer 27 second p-type cladding layer 28 p-side electrode 29 n-side electrode 30 mesa stripe structure 31 high reflection film 32 low reflection film 33 p-type contact layer 35 core layer 36 intermediate layer 40 current blocking region 41 p-type InP layer 42 n-type InP layer 43 AlGaInAs layer 45 optical field 46 mask pattern 50 n-type cladding layer 51 second n-type cladding layer 52 n-side electrode 53 p-side electrode 55 AlGaInAs layer 54 n-type contact layer 60 package 61 Temperature adjustment module 62 Base 63 Heat sink 64 Semiconductor optical element 65 Thermistor 66 Lens 67 Isolator 68 Lens 69 Optical fiber 70 Photo detector

Claims (19)

A substrate made of InP;
A mesa stripe structure including a stacked structure in which an active layer having a multiple quantum well structure is sandwiched between an n-type cladding layer and a p-type cladding layer, and disposed on the substrate;
A current blocking region disposed on the substrate on both sides of the mesa stripe structure;
A pair of electrodes for injecting carriers into the active layer;
The semiconductor optical device, wherein the n-type cladding layer includes an AlGaInAs lattice-matched to the substrate made of InP, or a high refractive index layer made of AlInAs lattice-matched to the substrate made of InP.   The semiconductor optical device according to claim 1, wherein the high refractive index layer is made of AlGaInAs and has a thickness of 160 nm or more.   The semiconductor optical device according to claim 1, wherein the high refractive index layer is made of AlInAs and has a thickness of 1200 nm or more.   The current blocking region includes at least an n-type InP-based material layer and a p-type InP-based material layer, and the vertical relationship between the n-type InP-based material layer and the p-type InP-based material layer is the mesa stripe structure. 4. The semiconductor optical device according to claim 1, wherein a vertical relationship between the n-type cladding layer and the p-type cladding layer is reversed. 5.   5. The semiconductor optical device according to claim 1, wherein a refractive index of the high refractive index layer is higher than a refractive index of the p-type cladding layer. 6.   The semiconductor optical device according to claim 1, wherein a bottom surface of the high refractive index layer is disposed at a position higher than a bottom surface of the current blocking region.   The semiconductor optical device according to claim 1, wherein the active layer has a multiple quantum well structure including at least one of an AlGaInAs layer, an AlInAs layer, and a GaInAsP layer. further,
An n-side separate confinement heterolayer disposed between the active layer and the n-type cladding layer;
A p-side isolation and confinement heterolayer disposed between the active layer and the p-type cladding layer;
Each of the n-side isolation and confinement heterolayer and the p-side isolation and confinement heterolayer contains AlGaInAs, and the Al composition ratio decreases continuously or stepwise toward the active layer. The semiconductor optical device according to any one of the above.   The semiconductor optical device according to claim 1, wherein the current blocking region further includes an AlGaInAs layer.   The semiconductor optical device according to claim 9, wherein a thickness of the AlGaInAs layer included in the current blocking region is equal to or less than a thickness of the high refractive index layer included in the n-type cladding layer.   The semiconductor optical device according to claim 1, further comprising an intermediate layer made of InP disposed between the active layer and the high refractive index layer.   The semiconductor optical device according to claim 11, wherein the intermediate layer is also disposed on the high refractive index layer on both sides of the mesa stripe structure.   The semiconductor optical device according to claim 1, wherein the active layer having the mesa stripe structure has a width of 3 μm or more. In addition, it has a heat sink,
One of the pair of electrodes is formed on the back surface of the substrate, and the other is formed on the mesa stripe structure,
14. The semiconductor according to claim 1, wherein, of the pair of electrodes, an electrode opposite to the n-type cladding layer as viewed from the active layer is thermally coupled to the heat sink. Optical device. The substrate and the mesa stripe structure include a pair of end faces intersecting with a longitudinal direction of the mesa stripe structure,
further,
An optical fiber having an incident end face;
The semiconductor optical device according to claim 1, further comprising a lens that couples a laser beam emitted from one of the pair of end faces to the incident end face of the optical fiber. Forming a laminated structure in which an active layer is sandwiched between an n-type cladding layer and a p-type cladding layer on a substrate made of InP;
Etching a portion of the stacked structure to at least the bottom surface of the active layer to leave a mesa stripe structure including the active layer;
Forming a current blocking region on the substrate on both sides of the mesa stripe structure;
Forming a pair of electrodes for injecting carriers into the active layer,
The method of manufacturing a semiconductor optical device, wherein the n-type clad layer includes AlGaInAs lattice-matched to the substrate made of InP or AlInAs lattice-matched to the substrate made of InP.   The method of manufacturing a semiconductor optical device according to claim 16, wherein the high refractive index layer is made of AlGaInAs and has a thickness of 160 nm or more.   The method of manufacturing a semiconductor optical device according to claim 16, wherein the high refractive index layer is made of AlInAs and has a thickness of 1200 nm or more. The step of forming the laminated structure includes a step of forming an intermediate layer made of InP between the active layer and the layer disposed on the substrate side of the n-type cladding layer and the p-type cladding layer. ,
In the step of leaving the mesa stripe structure, the active layer is etched until the intermediate layer is exposed,
Etching the intermediate layer exposed on both sides of the mesa stripe structure after the step of leaving the mesa stripe structure and before the step of forming the current blocking region,
The method of manufacturing a semiconductor optical device according to claim 16, wherein the current blocking region is formed without exposing the exposed surface to the atmosphere after etching the intermediate layer.

Images