JPH1140894A - Semiconductor optical element - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a semiconductor optical element whose production yield is not lowered, which is low-cost and which can obtain a low polarization dependence. SOLUTION: Signal light which is inputted from a light input end 101 is amplified by a light amplifier 102 in a light amplifying part 1. Then, the amplified signal light is incident on a main waveguide 104, which constitutes a directional coupler 103 in a gain control unit 2. At this time, TE-polarized light out of the signal light is shifted to a coupling part 103a having a coupling wavelength of 3000 μm in the directional coupler 103, so as to be radiated from a light output end 107. In addition, the TE-polarized light out of the signal light advances in the main waveguide 104 as it is, so as to be radiated from a light output end 106.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a semiconductor optical device used for optical communication and optical information processing.

[0002]

2. Description of the Related Art Semiconductor optical amplifiers utilizing the optical amplification effect of the stimulated emission phenomenon of light have been studied for their application as optical functional elements such as optical switches and wavelength converters in addition to level reproduction of optical signals. I have. Also, the semiconductor optical amplifying element
There is an advantage that integration with other semiconductor optical function elements is possible. A major problem of the semiconductor optical amplifying device is the polarization dependence of its amplification characteristics. TE
Of the polarized light and the TM polarized light, the TE polarized light usually tends to be amplified more. If the balance between the two polarized lights is largely shifted, the light receiving element cannot receive the light. For this reason, various methods for eliminating the polarization dependence have been tried. A typical example is a method using a plurality of optical amplifiers or a hybrid configuration of optical amplifiers and bulk optical components.

In the conventional hybrid configuration, as shown in FIG. 8A, an optical amplifier 801 and an optical amplifier 802 rotated 90 ° with respect to the optical amplifier 801 are arranged in series. This allows the optical amplifier 80
1 to the optical amplifier 8
02 to realize amplification characteristics with low polarization dependence as a whole. As another hybrid configuration, as shown in FIG. 8B, the signal light is split into two paths for each polarization, the TE polarization is amplified by the optical amplifier 801, and the TM polarization is amplified by the optical amplifier 802. There is also a configuration for amplification. By doing so, the two polarized lights are amplified respectively, and low polarization dependency is realized as described above.

[0004]

However, conventionally, since a plurality of optical amplifying elements are individually arranged as described above, the number of steps for mounting these optical elements increases, and a hybrid configuration is adopted. There is a problem that the production yield of the optical amplifier is greatly reduced. In addition, in the above-described configuration, since a plurality of optical amplifying elements are required, an optical isolator is required for each of the optical amplifying elements. Therefore, when viewed as one module component, the cost is very high.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems, and it is possible to obtain a state with low polarization dependence at low cost without lowering the production yield. The purpose is to:

[0006]

A semiconductor optical device according to the present invention comprises a lower clad layer, a first core layer formed in a predetermined area on the lower clad layer, and a first core layer formed on the first core layer. An upper clad layer, a first electrode formed on the upper clad layer, a first waveguide having a light input end formed at one end of the waveguide including the first core layer and formed perpendicular to the surface of the lower clad layer; A region having a band gap larger than that of the first core layer and wider than the first core layer, and formed on the lower clad layer surface so as to be continuous with the other end of the first core layer. A second core layer, a second upper cladding layer at least partially formed on one end of the first upper cladding layer, and a second electrode formed on the second core layer; The optical signal formed perpendicular to the layer surface and input from the first core layer It was configured from and a second region with a force to the light output end. As shown above,
First, by applying a current to the first electrode, the signal light input from the optical input terminal is amplified in the first region. Also,
Since the width of the second core layer is wider than the width of the first core layer, it is possible to provide two or more waveguides in the second region. Further, in the semiconductor optical device of the present invention, in addition to the above, the second upper cladding layer includes a main cladding layer formed continuously with the first upper cladding layer, and a part of the main cladding layer. A main cladding layer and a branch cladding layer having the same width as the main cladding layer, and a light output end is defined by the main cladding layer. The main light output end from which light is output from the waveguide and the sub-light output end from which light is output from the branch waveguide defined by the branch cladding layer. As described above, in the second region, the main waveguide including the main cladding layer continuous with the first core layer, and the branch waveguide including the branch cladding layer provided near the coupling portion to the main waveguide are provided. As a result, one of the polarized lights constituting the signal light amplified in the first region travels through the main waveguide, and the polarized light perpendicular thereto moves to the branch waveguide at the coupling portion.

[0007]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a plan view showing a semiconductor optical device according to an embodiment of the present invention. This semiconductor optical device is roughly divided into two regions. One is
The light amplification unit 1 has a length of, for example, 600 μm and amplifies incident light. The other is a gain control unit 2 for compensating the polarization dependence of the gain in the optical amplification unit 1. In this embodiment, a directional coupler is used for gain control. In this semiconductor optical device, first, as shown in FIG. 1, a signal light input from an optical input terminal 101 is applied to an optical amplifier 1 of an optical amplifier 1.
02 is amplified. Then, the amplified signal light is
The light enters the main waveguide 104 constituting the directional coupler 103 of the gain control unit 2. Here, the coupling section 103a of the directional coupler 103 having a coupling length of 3000 μm generates TE
The polarized light shifts to the branch waveguide 105 and is emitted from the light output end 107. The TM polarized light of the signal light travels through the main waveguide 104 as it is and is emitted from the light output end 106.

By controlling the electric field applied to the main waveguide electrode 104a and the branch waveguide electrode 105a,
The rate at which the TE-polarized light shifts in the optical coupling section 103a is controlled, and the output of the TE-polarized light traveling through the main waveguide without shifting is controlled. The main waveguide electrode 104a is connected to the wiring 1
It is connected to the electrode pad 104c through the electrode pad 104c. Similarly, the branch waveguide electrode 105a is connected to the electrode pad 105c via the wiring 105b. As described above, the output of the TM polarized light and the output of the TE polarized light at the optical output terminal 105 are made the same, whereby the polarization dependence is eliminated in the semiconductor optical device of this embodiment. That is, in the semiconductor optical device according to the present embodiment, the signal light is amplified by the optical amplifier 1 and the polarization dependence thereof is eliminated by the gain controller 2.

Next, the structure of the optical amplifier 102 and the directional coupler 103 constituting the above-described semiconductor optical device will be described in more detail. FIG. 2A shows an AA ′ cross section of the directional coupler 103 in FIG. FIG.
FIG. 2B shows a cross section taken along line BB ′ of the optical amplifier 102 in FIG. As shown in FIG. 2A, the directional coupler 103 constituting the gain control unit 2 is composed of a 1.45 μm composition InGaAlA on a substrate 201 made of n-type InP.
s / InAlAs MQW, 0.4 μm thick and 1 width
A core layer 202 of 00 μm is formed. On top of this, a thickness of 5 μm made of InGaAsP having a composition of 1.3 μm.
A 0 nm buffer layer 203 is formed.

Then, on the buffer layer 203, a p-type I
clad layer 2 of 1.5 μm in thickness and 2 μm in width made of nP
04 are arranged in two rows. The cladding layer 204 forms the first and branch waveguides 104 and 105 shown in FIG. In addition, the distance between the two cladding layers 204 in these coupling portions 103a (FIG. 1), that is, in FIG. 2A, is 2.5 μm. This interval is
The wavelength may be about the wavelength of the signal light to be amplified. A 0.3 μm thick contact layer 205 made of p-type InGaAs is formed on each clad layer 204. Further, on the contact layer 205, the p-side electrode 207 is formed.
Is formed, and an n-side electrode 208 is formed on the back surface of the substrate 201.

On the other hand, as shown in FIG. 2B, an optical amplifier 102 has InGaAs / InGaAs on a substrate 201.
A core layer 209 made of sPMQW and having a thickness of 0.15 μm and a width of 1.5 μm is formed, a cladding layer 204 is formed thereon with a buffer layer 203 interposed therebetween, and a contact layer 205 is formed.
Is formed, and a p-side electrode 207 is formed thereon. That is, the optical amplifier 102 and the directional coupler 103
Have the same substrate 201 and the same cladding layer 204. However, the core layer 202 has a larger band gap than the core layer 209, and the core layer 202 is formed to be sufficiently wider than the core layer 209. For example, the width of the core layer 202 is three times or more the wavelength of the target signal light.

In the optical amplifier 102, a part of the substrate 201 having a ridge structure, the core layer 209, the clad layer 204 on the core layer 209, and the core layer 20 are formed.
9 is buried with a buried layer 206. The p-side electrode 207 is in contact with the contact layer 205. Note that the p-side electrode 207 extends on the buried layer 206 via the insulating film 210 made of SiO ₂ in the optical amplification unit 1. An n-side electrode 208 is formed on the back surface of the substrate 201.

Next, a method of manufacturing the above-described semiconductor optical integrated device will be described. First, on the substrate 201, the MOVP
According to the E method, InGaAs and InGaAsP are alternately grown to form an MQW layer. Next, after forming an SiO ₂ film over the entire area of the MQW layer, the Si ₂ film is formed by known photolithography and dry etching (RIE).
The O ₂ film is patterned. That is, an SiO ₂ pattern in which the SiO ₂ film is left in a desired stripe shape is formed. Then, using the SiO ₂ pattern as a mask, MQ
The core layer 209 is formed by selectively etching the W layer. This selective etching is performed by wet etching using a sulfuric acid-based etchant. Next, M
This time, InGaAs is formed on the substrate 201 by the OVPE method.
An MQW layer serving as the core layer 202 is formed by alternately growing lAs and InAlAs. Then, this MQW layer is joined to the already formed core layer 209 by butt joint. Here, since the SiO ₂ pattern remains on the core layer 209, the MQW layer does not grow thereon.

Next, the SiO 2 remaining on the core layer 209
After removing the _two patterns, p-type InP serving as the cladding layer 204 is grown by MOVPE, and subsequently p-type InGaAs serving as the contact layer 205 is grown.
Thereafter, a mask pattern of SiO ₂ is formed on the p-type InGaAs layer. This mask pattern has a stripe shape with a width of 1.5 μm in the optical amplification unit and a width of 100 μm in the gain control unit. Then, using this mask pattern as a mask, the cladding layer 204 and the contact layer 205 are formed by dry etching using a halogen-based gas.

Next, after processing damage due to dry etching is removed by wet etching, the Fe-doped I
The nP is grown by the MOVPE method, and the buried layer 206 is formed.
To form Here, the Fe-doped InP is also SiO
Does not grow on the mask pattern consisting of _2, the buried layer 206 is clad layer 204 and contact layer 205
Are formed so as to bury them. Then, after removing the mask pattern, an insulating film 210 is formed.
A window is opened at a predetermined position of 10, and a PIe-side electrode 207 is formed.

Thereafter, the cladding layer 204 (FIG. 2) of the gain control unit 2 (FIG. 1) is selectively etched by halogen-based dry etching and chlorine-based wet etching so as to leave a desired region. ) Is formed. As a result, in the gain control unit 2, as shown in FIG.
Two cladding layers 204 having a thickness of 5 μm and constituting two waveguides are formed. Finally, n
By forming the side electrode 208, the basic configuration of the semiconductor optical integrated device in this embodiment is completed.

Hereinafter, the operation characteristics of the semiconductor optical integrated device according to this embodiment will be described in more detail. FIG.
(A) shows the relationship between the coupling length of the directional coupler 103 constituting the gain control unit 2 (FIGS. 1 and 2 (a)) and the transition ratio of the optical power. As shown in FIG.
When it is set to 00 μm, when no electric field is applied to the directional coupler 103, in the case of TE polarization, the power transfer ratio becomes 100% (cross state). In contrast, T
For M-polarized light, the power transfer ratio is 0% (bar state).
As is apparent from FIG. 3A, generally, the TM-polarized light has a longer coupling length than the TE-polarized light. In other words, in FIG. 1, of the signal light amplified by the optical amplifier 102, T
The M-polarized light travels through the main waveguide 104 and exits from the light output end 106. On the other hand, the TE polarized light travels to the branch waveguide 105 and exits from the light output end 107. Therefore, in this state, the gain control unit 2 is a polarization splitter.

Therefore, if an electric field is applied to the branch waveguide electrode 105a to reduce the power transfer ratio of the TE polarized light, and the TE polarized light also travels through the main waveguide 104, the TE polarized light is emitted from the light output end 106. The polarization dependence of the signal light can be reduced. This power transfer ratio changes by applying an electric field to the coupling portion 103a. That is, although the refractive index of the waveguide changes due to the electric field, the change in the propagation constant of TE polarized light is large with respect to the change in the refractive index. FIG. 3B shows the relationship between the phase difference between waveguides caused by the application of an electric field and the coupling length. FIG.
In (b), the cross state is indicated by a black circle, and the bar state is indicated by a white circle.

As shown in FIG. 3B, ΔβL / π = 0
The state that was in the cross state A at this time is changed to the bar state B when ΔβL / π = 3. In the case of TM polarization, Δ
When βL / π = 0, the bar state is C, but ΔβL / π = 3
At the time of, it does not reach the cross state. Here, β is a propagation constant, and L is a coupling length. As described above, the ratio of transition of the TE polarized light in the coupling portion 103a is determined by the
By controlling with the electric field applied to 3a, the polarization dependence of the gain in the optical amplifier 1 (FIGS. 1 and 2B) can be compensated.

For example, as shown in FIG. 4, the relationship between the injection current and the gain in the optical amplification unit 1 having the configuration shown in FIGS.
There is a gain difference of about 7 dB between the polarizations. And this difference in gain is not always constant. This is because the signal light incident on the optical amplifier 1 takes various polarization states with time. In the signal light obtained from the optical amplifying unit 1, the TE polarized light travels to the branch waveguide 105 in the coupling unit 103 a of the gain control unit 2 and exits from the output terminal 107. On the other hand, of the signal light obtained from the optical amplifier 1, T
The M-polarized light propagates through the main waveguide 104 without exiting to the branch waveguide 105 in the coupling section 103a of the gain control section 2 and exits from the output end 106.

Here, as shown in FIG. 5, the relationship between the voltage applied to the branch waveguide electrode 105a (FIG. 1) and the normalized light output of each of the TE polarized light and the TM polarized light from the output end 106 is considered. When the applied voltage is around 2 V, the two light outputs match. That is, by applying a voltage to the branch waveguide electrode 105a, the transition of the TE polarized light at the coupling portion 103a decreases, and the output of the TE polarized light from the output end 106 increases.
Therefore, by applying a predetermined voltage to the branch waveguide electrode 105a, the optical output of the TE polarized light and the TM polarized light from the output end 106 can be matched. By changing the applied electric field, the gain control unit becomes a variable polarization splitter.

The use of the main waveguide electrode 104a shown in FIG. 1 is the same as the case where the above-mentioned branch waveguide electrode 105a is used. That is, by applying a voltage to the main waveguide electrode 104a, the transition of the TE polarized light in the coupling portion 103a decreases, and the output of the TE polarized light from the output end 106 increases. Here, when a voltage is applied to the main waveguide electrode 104a, light absorption also occurs. Generally, in the case of an optical device in which the operating wavelength is set near the absorption edge using MQW, FIG.
As shown in (2), the absorption becomes larger when a voltage is applied in the TE deflection than in the TM polarization.

More specifically, as shown in FIG. 6, when a voltage is applied to the main waveguide electrode 104a, the TE polarized light is applied until the applied voltage reaches about 4 V, as in the case shown in FIG. In the connecting portion 103a decreases,
The output of the TE polarization from output end 106 increases. However, when a voltage is further applied, the TM polarization does not change, but the light output from the output end 106 of the TE polarization decreases. When the voltage applied to the main waveguide electrode 104a is about 6.5 V, the optical output of the TE polarized light and the TM polarized light from the output end 106 becomes the same again. Finally, when the voltage applied to the main waveguide electrode 104a is close to 8 V, the optical output of the TE polarized light from the output terminal 106 becomes zero.

Here, FIG. 7 shows an example of an optical integrated circuit using the semiconductor optical device in this embodiment described above.
As shown in FIG. 7, the semiconductor optical device includes an optical amplifier 81 and a gain controller 82. The signal light incident on the optical amplifier 81 is split by an optical splitter 83, and a part thereof is polarized. It is guided to the splitter 84. The signal light guided to the polarization splitter 84 is separated here into TE polarized light and TM polarized light, each of which is photoelectrically converted by a light receiver 85, and the photoelectrically converted signal is input to a differential amplifier 86. Then, the ratio between the TE polarized light and the TM polarized light separated by the polarization splitter 84 is converted into a voltage difference by the differential amplifier 86.
By operating the drive circuit 87 with this voltage difference and controlling the voltage applied to the coupling unit of the gain control unit 82 using the directional coupler, the polarization dependence of the signal light output from the gain control unit 82 Can be eliminated.

[0025]

As described above, according to the present invention, the lower clad layer, the first core layer formed in a predetermined region of the surface, and the first upper clad formed on the first core layer are formed. A first region comprising a first electrode formed thereon, a first region having a light input end formed perpendicular to the surface of the lower cladding layer at one end of the waveguide including the first core layer; A second cladding layer formed on the surface of the lower cladding layer continuously to the other end of the first core layer in a state having a band gap larger than that of the first core layer and wider than the first core layer. A second upper cladding layer formed at least partially on one end of the first upper cladding layer,
A second region formed on the second electrode and formed perpendicular to the surface of the lower cladding layer and having an optical output end for outputting an optical signal input from the first core layer; To be configured.

As described above, first, by applying a current to the first electrode, in the first region, the signal light input from the light input terminal is amplified. In addition, since the width of the second core layer is wider than the width of the first core layer, it is possible to provide two or more waveguides in the second region. Therefore, according to the present invention, the second
Can be composed of a portion continuous with the first cladding layer and a portion disposed close to the continuous portion. The two portions of the second cladding layer allow the signal light amplified in the first region to be in a state without polarization dependence.

[0027] In addition to the above, in the semiconductor optical device of the present invention, the second upper cladding layer includes a main cladding layer formed continuously with the first upper cladding layer, And a branch cladding layer formed with the same width as the main cladding layer, a part of which is disposed close to part of the main cladding layer, and the light output end is defined by the main cladding layer. A main light output end from which light is output from the main waveguide and a sub light output end from which light is output from the branch waveguide defined by the branch cladding layer.

As described above, in the second area,
Since the main waveguide including the main cladding layer continuous to the first core layer and the branch waveguide including the branch cladding layer provided near the main waveguide at the coupling portion are provided, the main waveguide is amplified in the first region. One of the polarized lights constituting the signal light travels through the main waveguide, and the polarized light perpendicular thereto moves to the branch waveguide at the coupling portion. Here, by applying a voltage to the coupling portion, it is possible to make the amount of different polarized light in the main waveguide the same by reducing the amount transferred to the branch waveguide,
As a result, it is possible to eliminate the polarization dependence of the optical signal output from the main optical output terminal.

Further, since the main waveguide and the branch waveguide in the second region are defined by the main cladding layer and the branch cladding layer, in the second region, the main waveguide and the branch waveguide are formed without processing the second core layer. Waveguides and branch waveguides can be formed. Therefore, an increase in the number of steps due to processing the second core layer can be suppressed, and an increase in light propagation loss due to processing damage can be prevented.

[Brief description of the drawings]

FIG. 1 is a plan view showing a semiconductor optical device according to an embodiment of the present invention.

FIG. 2A is a diagram illustrating a directional coupler 103 of FIG.
1A shows a cross section, and FIG. 1B shows a cross section of the optical amplifier 102 of FIG.
FIG. 3 is a configuration diagram showing a section B ′.

FIG. 3 is a characteristic diagram showing a relationship between a coupling length of a directional coupler 103 included in the gain control unit 2 (FIGS. 1 and 2A) and a transition ratio of optical power.

FIG. 4 is a characteristic diagram showing a relationship between an injection current and a gain in the optical amplifier 1 shown in FIGS.

FIG. 5 is a characteristic diagram showing a relationship between a voltage applied to a branch waveguide electrode 105a (FIG. 1) and standardized optical outputs of TE polarized light and TM polarized light from an output terminal 106.

FIG. 6 is a correlation diagram showing the applied voltage dependence of the absorption of TE polarization and TM polarization in an optical element in which the operating wavelength is set near the absorption edge using MQW.

FIG. 7 is a configuration diagram illustrating an example of an optical integrated circuit using a semiconductor optical device according to an embodiment of the present invention.

FIG. 8 is a perspective view showing a configuration of a conventional example for eliminating polarization dependence.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Optical amplification part, 2 ... Gain control part, 101 ... Optical input terminal, 1
02: optical amplifier, 103: directional coupler, 104: main waveguide, 105: branch waveguide, 106, 107: optical output terminal, 201: substrate, 202: core layer, 203: buffer layer,
204: cladding layer, 205: contact layer, 206:
Embedded layer, 207: p-side electrode, 208: n-side electrode, 2
09: core layer, 210: insulating film.

Claims (6)

[Claims] 1. A lower clad layer, a first core layer formed on a predetermined region of the surface, a first upper clad layer formed on the first core layer, and a first upper clad layer formed on the first core layer. A first region having a light input end formed perpendicular to the lower clad layer surface at one end of a waveguide including the first core layer; and A second core layer having a large band gap and being formed on the surface of the lower cladding layer continuously from the other end of the first core layer in a state where the width is wider than the first core layer; At least part of the first
A second upper cladding layer formed continuously on one end of the upper cladding layer, a second electrode formed thereon, and formed perpendicular to the surface of the lower cladding layer;
A second region having an optical output end for outputting an optical signal input from the first core layer. 2. The semiconductor optical device according to claim 1, wherein the second upper cladding layer is formed continuously with the first upper cladding layer, and a part of the main cladding layer. And a branch cladding layer formed with the same width as the main cladding layer, wherein the light output end is defined by the main cladding layer. A main light output end from which light is output from the divided main waveguide, and a sub light output end from which light is output from the branch waveguide defined by the branch cladding layer. element. 3. The semiconductor optical device according to claim 1, wherein said first electrode is provided on said first upper cladding layer.
The second electrode is formed on the second upper cladding layer through a second contact layer.
A semiconductor optical element formed via the contact layer of (1). 4. The semiconductor optical device according to claim 2, wherein the second electrode is provided on a main waveguide electrode formed on the main cladding layer of the coupling portion, and on the branch cladding layer of the coupling portion. A semiconductor optical device comprising: a formed branch waveguide electrode. 5. The semiconductor optical device according to claim 1, wherein said first upper cladding layer and said second upper cladding layer are formed of the same layer. Semiconductor optical device. 6. The semiconductor optical device according to claim 1, wherein, in the first region, the first core layer, the first upper clad layer, and the first clad layer constituting a waveguide are provided. A semiconductor optical device, wherein a part of a side surface of the lower cladding layer below a first core layer is buried with an insulating layer.