JPH11231273A - Optical semiconductor device - Google Patents

Abstract

(57) [Summary] [PROBLEMS] In an optical semiconductor device utilizing the Franz-Keldysh effect, even when the incident light intensity is large and the applied voltage is high, good between the incident light intensity and the output light intensity. Provided is an optical semiconductor device capable of obtaining excellent linear characteristics. The optical semiconductor device includes a first semiconductor layer 2 made of an n-type or p-type first semiconductor and a second semiconductor layer 4 made of a second semiconductor having a conductivity type opposite to that of the first semiconductor. Has a double hetero structure H in which an intrinsic semiconductor layer 3 is interposed, and the thickness of the intrinsic semiconductor layer 3 gradually decreases in the light traveling direction.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an optical semiconductor device, and more particularly, to an optical semiconductor device which operates by utilizing the Franz-Keldysh effect when the intensity of incident light is large and the applied reverse bias voltage is high. Even so, the present invention relates to an optical semiconductor device capable of obtaining good linear characteristics between the intensity of incident light and the intensity of output light.

[0002]

2. Description of the Related Art An electro-absorption type optical modulator made of a semiconductor generally has an intrinsic (i-type) semiconductor layer formed of a p-type semiconductor having a band gap energy larger than that of the i-type semiconductor and an n-type semiconductor. It has a double hetero structure of a pin structure or a nip structure sandwiched between semiconductor layers.

In the case of this double hetero structure, an optical waveguide is formed based on the refractive index difference between the i-type semiconductor layer and the p-type and n-type semiconductor layers which are joined above and below the i-type semiconductor layer. It is possible to propagate light to the light. When a reverse bias voltage is applied to this double heterostructure, a so-called Franz-Keldysh effect is exerted, and the light absorption coefficient of the i-type semiconductor changes, thereby modulating the intensity of light propagating therethrough.

Further, such an optical modulator is operated in a state arranged in front of a waveguide type light receiving element, and the intensity of light incident on the light receiving element is attenuated, so that an effective variable sensitivity operation of the light receiving element is achieved. There is also known an optical attenuator integrated waveguide type light receiving element for performing the following.

[0005]

By the way, in the case of the Franz-Keldysh effect when a reverse bias voltage is applied to a pin structure, in an i-type semiconductor, electrons and holes are generated by absorbed light. Then, an electric field is formed by these electrons and holes, and a phenomenon that a reverse bias voltage applied from the outside is canceled out by this electric field, that is, a so-called cleaning phenomenon appears. Such a phenomenon reduces the effect of the applied reverse bias voltage and is not preferable for the operation of the device.

The generated holes are formed between the i-type semiconductor layer and the p-type semiconductor layer.
It accumulates at the hetero interface of the type semiconductor layer, so that a reverse bias voltage applied from the outside is concentratedly applied to this hetero interface, so that a so-called hole pile-up effect is also exhibited. The above two problems are more prominent as the concentration of electrons and holes generated in the i-type semiconductor layer is higher. That is, the intensity of light incident on the device is large,
In addition, the above-described problem becomes more prominent as the reverse bias voltage applied to the device is larger and the light absorption in the i-type semiconductor layer is larger.

For this reason, in the case of an electroabsorption type optical modulator utilizing the Franz-Keldysh effect, when the applied voltage increases, a problem arises in that the intensity of the emitted light is not proportional to the intensity of the incident light. It shows a so-called non-linear characteristic. In order to solve such a problem, it has been proposed to introduce a layer made of a semiconductor having a band gap energy intermediate between the i-type semiconductor layer and the p-type semiconductor layer. (M. Suz
uki et al. Electronics Letters, vol. 25, p88-89,
1989).

However, even in the case of the optical modulator having the above-mentioned layer structure, the above-mentioned nonlinear characteristics cannot be completely solved. In addition, in the case of the above-mentioned layer structure, it is conceivable that when a high intensity optical signal is modulated, the modulated optical signal is distorted to cause deterioration of transmission characteristics. Also, in the case of an optical attenuator integrated waveguide type light receiving element in which a waveguide type light receiving element is optically coupled to the above light modulator, when the intensity of the optical signal modulated by the optical modulator increases, the light receiving current obtained by the light receiving element May be distorted.

According to the present invention, the above-mentioned problem in the optical semiconductor device utilizing the Franz-Keldysh effect, that is, when the applied reverse bias voltage is increased, a non-linear characteristic is generated between the output light intensity and the incident light intensity. It aims to provide a novel optical semiconductor device that solves the problem and obtains linear characteristics between the intensity of emitted light and the intensity of incident light even when a high voltage is applied, and an optical attenuator integrated waveguide type photodetector incorporating the same. I do.

[0010]

In order to achieve the above-mentioned object, according to the present invention, a first semiconductor layer made of an n-type or p-type first semiconductor is provided, and a first semiconductor layer having a reverse conductivity type to the first semiconductor is provided. It has a double hetero structure in which a layer of an intrinsic semiconductor is interposed between a second semiconductor layer made of a second semiconductor and the thickness of the layer of the intrinsic semiconductor gradually decreases in a light traveling direction. An optical semiconductor device (hereinafter referred to as a first device) is provided. This is used as an optical modulator utilizing the Franz-Keldysh effect.

Also, in the present invention, an optical semiconductor device in which a waveguide type light receiving element is optically coupled to a light emitting end of the optical semiconductor device (referred to as a second device), a so-called optical attenuator integrated waveguide type light receiving element. Is provided.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The apparatus of the present invention will be described below in detail with reference to the drawings. FIG. 1 is a cross-sectional view showing a cross-sectional structure of a first device A functioning as an electro-absorption type optical modulator, and FIG. 2 is a cross-sectional view taken along line II-II of FIG. In the case of this device A, for example, a buffer layer 2 also made of n-InP is laminated on a substrate 1 made of n-InP, and further, on the n-InP buffer layer 2, i-GaInAs
A light absorbing layer 3a made of P and a layer 3b made of i-InP are stacked in this order to form an i-type semiconductor layer 3, and i-type semiconductor layer 3 is formed.
On the InP layer 3b, a clad layer 4 made of p-InP and a contact layer 5 made of p-GaInAs are sequentially laminated.

The p-InP cladding layer 4 and the p-G
A part of the aInAs contact layer 5 is removed by etching to form a ridge waveguide, and its side and surface are covered with a dielectric insulating film 6, and an upper electrode 7a is loaded on the p-GaInAs contact layer 5, A lower electrode 7b is loaded on the back surface of the substrate 1. Antireflection films 8a and 8b made of, for example, SiNx are formed on both end surfaces of the device A, and a light incident end surface S1 and a light output end surface S2 are formed.

In this device A, the n-InP buffer layer 2 is an n-type first semiconductor layer, and the i-GaI
The nAsP layer 3a functions as a waveguide layer, and the i-
The i-type semiconductor layer 3 is constituted together with the InP layer 3b,
Further, the p-InP clad layer 4 thereover is a p-type second semiconductor layer, and a double hetero structure H having a pin structure is formed by these layers.

The greatest feature of the device A is that the thickness of the i-type semiconductor layer 3 gradually decreases from the light incident end face S1 to the light emitting end face S2, that is, along the light traveling direction. In FIG. 1, the i-InP layer 3b
Is gradually reduced, so that the entire i-type semiconductor layer 3 is in the above-described state. This device A has an incident end surface S
While an optical signal having a predetermined intensity from 1 is incident, the upper electrode 7a
It operates by applying a reverse bias voltage between the lower electrode 7b and the lower electrode 7b. At this time, since the thickness of the i-type semiconductor layer 3 gradually decreases in the traveling direction of the optical signal, between the intensity of the incident light from the incident end face S1 and the intensity of the emitted light from the exit end face S2, when the applied voltage is large. Even so, good linear characteristics can be obtained.

This is considered to be based on the following reasons. Since the thickness of the i-type semiconductor layer 3 is gradually reduced in the light traveling direction, the electric field strength generated in the waveguide layer i-GaInAsP layer 3a by the application of the reverse bias voltage is in the light traveling direction. Along with it. Therefore, the Franz-Keldysh effect is small near the incident end face S1 where the thickness of the i-type semiconductor layer 3 is large, and the emission end face S where the thickness of the i-type semiconductor layer 3 is small.
It becomes larger in the vicinity of 2. Therefore, the optical signal propagating through the i-type semiconductor layer 3 has a light absorption coefficient of the incident end face S
1 and gradually increases as it proceeds toward the emission end face S2.

On the other hand, the light intensity of the incident light is highest at the incident end face S1, and is gradually absorbed in the process of traveling toward the output end face S2, and the light intensity is reduced. Therefore, device A
In the i-type semiconductor layer 3 described above, the electric field intensity is low in the region where the light intensity is high, and the electric field intensity is higher in the region where the light intensity is lower. Therefore, it is considered that the electric field screening phenomenon and the hole pile-up effect, which are remarkably exhibited when the light intensity is high and the electric field intensity is high, are suppressed, and the propagating light is considered to exhibit good linear characteristics.

Next, a specific example of the device A and its characteristics will be described. First, by MOCVD, a large n-In
On the P substrate 1, an n-InP buffer layer 2 having a thickness of 0.5 μm and an i-GaInAsP light absorbing layer (band gap wavelength 1.2 μm) 3a having a thickness of 0.1 μm were formed. Next, as shown in FIG. 3, a mask pattern forming one unit mask 9 by a pair of masks 9a and 9b is formed on the i-GaInAsP light absorbing layer 3a by, for example, Si.
After film formation with Nx, i-InP is selectively grown here. Here, the unit mask 9 has a narrow opening 9.
c and the opening 9d that gradually widens, the i-InP layer formed on the i-GaInAsP light absorption layer 3a is thick in the narrow opening 9c and wide in the narrow opening 9c. In the portion 9d, the thickness is gradually reduced as the distance from the narrow opening 9c increases. Specifically, in the region where the width of the masks 9a and 9c is the widest, the thickness of the i-InP layer is 0.2 μm, and the thickness of the i-InP layer in the narrow opening 9c is 0.6 μm. The pattern was adjusted for selective growth.

After the selective growth of i-InP, the unit mask 9 is completely removed, and p-InP is crystal-grown on the entire surface to form a 2 μm-thick p-InP cladding layer 4. Then, a 0.3 μm-thick p-GaInAs contact layer (band gap wavelength 1.67 μm) 5 was formed. Therefore, at this point, the n-InP buffer layer 2 is used as the first semiconductor layer, the p-InP cladding layer 4 is used as the second semiconductor layer, and the i-type semiconductor layer 3 whose thickness is gradually reduced is interposed therebetween. A double heterostructure H is formed.

Next, the p-GaInAs contact layer 5
Of the surface of the unit mask 9, which was the narrow opening 9c of the unit mask 9, a 10 μm-wide striped etching mask was formed of, for example, SiNx, and then an etching process was performed to form a p-GaInAs contact layer and a p-GaInAs contact layer. In
A part of the P clad layer was removed to form a ridge waveguide as shown in FIG.

Then, after forming a dielectric insulating film 6 made of, for example, SiNx on the entire surface, the top portion of the ridge waveguide is removed to expose the p-GaInAs contact layer 5, where, for example, Ti / The upper electrode 7a is loaded by depositing Pt / Au. The back surface of the substrate 1 is polished to reduce the overall thickness to about 100 μm, and then, for example, Au
GeNi / Au is deposited to load the lower electrode 7b.

Next, the substrate is cleaved at the locations C0 and C1 shown in FIG. 3 to produce a plurality of laminated structures, and the reflectance of light having a wavelength of 1.3 μm is set to 1 on both of the cleavage surfaces using, for example, SiNx. %, The device A shown in FIG. 1 is obtained. here,
The end face cleaved at the point C1 in FIG. 3 is the incident end face S1, and the end face cleaved at the point C0 is the output end face.

The modulation characteristic of the device A is indicated by a circle in FIG. FIG. 4 shows the results when a semiconductor laser having an oscillation wavelength of 1.3 μm is used as the light source and the light intensity of the incident light is 1 mW. For comparison, the i-GaInAsP light absorbing layer 3
a, the unit mask 9 is not formed on the i-InP layer 3
An optical modulator was manufactured under the same conditions as in the apparatus A except that the i-InP layer 3b was formed to have a uniform thickness of 0.5 μm without performing selective growth when forming the film b. The modulation characteristics are also shown by the mark ● in FIG.

As is apparent from FIG. 4, in the device A of the present invention and the comparative example device, the emission light intensity was reduced by about 10 dB when the reverse bias voltage of 10 V was applied, and as a good optical modulator. It is working. Further, the reverse bias voltage applied in each of the apparatus A and the comparative example apparatus was changed, and the relationship between the incident light intensity and the output light intensity at that time was measured. The result in the case of the apparatus A is shown in FIG. 5, and the result in the case of the comparative example apparatus is shown in FIG.

As is clear from FIG. 5, the apparatus A of the present invention
In this case, the outgoing light intensity and the incident light intensity are in a proportional relationship up to an applied voltage of 10 V, and excellent linear characteristics are obtained.
On the other hand, in the case of the comparative example device, when the applied voltage is 5 V or more, the proportional relationship between the intensity of the emitted light and the intensity of the incident light is broken, and the device becomes non-linear. Next, another apparatus B of the present invention will be described.

FIG. 7 is a sectional view showing a sectional structure of another apparatus B of the present invention, FIG. 8 is a sectional view taken along line VIII-VIII of FIG. 7, and FIG. 9 is IX-IX of FIG. It is sectional drawing which follows a line. This device B is an optical modulator A having a cross-sectional structure shown in FIGS.
And the waveguide type light receiving element B1 are optically coupled and integrated.

In this device B, the incident light from the incident end face S1 becomes an outgoing light having a good linear characteristic as described above in the process of propagating through the i-type semiconductor layer 3, and becomes a waveguide to be described later. The light enters the light receiving element B1, where it is detected as a light receiving current. At this time, since the linear characteristic of the light emitted from the optical modulator A is good and not distorted, the light receiving current in the waveguide type light receiving element B1 is not distorted.

Here, the light receiving element B1 has, for example, the following layer structure. That is, a buffer layer 2 made of n-InP is laminated on a substrate 1 made of n-InP, and i-GaIn
Light absorption layer 3a made of AsP, spacer layer 3c made of i-InP, light absorption layer 3d made of i-GaInAsP
Are stacked in this order to form an i-type semiconductor layer 3, and p-In is further formed on the i-GaInAsP light absorbing layer 3d.
It has a laminated structure in which a cladding layer 4 'made of P and a contact layer 5 made of p-GaInAs are sequentially stacked. Then, the laminated structure is etched away at least until it reaches the n-InP buffer layer 2 to form a ridge waveguide, and this ridge waveguide is made of Fe-doped InP.
Buried with a layer 10, the surface is covered with a dielectric insulating film 6,
On the p-GaInAs contact layer 5, an upper electrode 7c for detecting a received light current is loaded. The lower electrode of the light receiving element B1 is common to the lower electrode 7b of the optical modulator A.

Next, a specific example of the device B and its characteristics will be described. First, the layer structure of the waveguide type light receiving element B1 is formed. That is, as in the case of the apparatus A, a 0.5 μm thick n-InP buffer 2 and a 0.1 μm thick i-type
-GaInAsP light absorbing layer (bandgap wavelength 1.2
μm) 3a is formed into a film, followed by a thickness of 0.01 μm
m very thin i-InP spacer layer 3c, thickness 0.1
3 μm i-GaInAsP light absorption layer (bandgap wavelength 1.45 μm), thin p-InP 0.1 μm thick
Cladding layer 4 ', p-G having a thickness of 0.05 μm (not shown)
aInAs etching stop layer (band gap wavelength 1.
67 μm).

Next, after covering the surface of a portion corresponding to the light receiving element B1 to be formed with a dielectric film such as SiNx, selective etching is performed to form i-GaInAsP.
The layered structure on the light absorbing layer 3a is removed. Specifically, the p-GaInAs etching stop layer and the i-GaIn
A mixture of sulfuric acid, hydrogen peroxide, and water is used as the etchant A when etching the AsP light absorbing layer 3d, and a mixture of hydrochloric acid and phosphoric acid is used when etching the p-InP cladding layer 4 '. Used as etchant B. At this time, the i-InP spacer layer 3c becomes the i-GaInAsP light absorbing layer 3 during the etching process.
It functions as a buffer layer for leaving a.

Then, after removing the dielectric film, a mask pattern for selective growth similar to that shown in FIG. 3 for manufacturing the device A is formed on the entire surface where the optical modulator A is to be formed. Then, i-InP is selectively grown so that the thickness of the flat portion becomes 0.2 μm. On the i-GaInAsP light absorbing layer 3a (or the i-InP spacer layer 3c thereon) at the position where the optical modulator is to be formed, there is formed an i-InP layer 3b whose thickness gradually decreases along the traveling direction of light. An i-InP layer having a substantially uniform thickness is formed at a place where the light receiving element B1 is to be formed.

Next, after removing the mask pattern for selective growth and covering a portion corresponding to the optical modulator A to be formed with a dielectric film, the etchant B and the etchant A are formed.
Are sequentially performed. As a result, the i- film formed at a position corresponding to the light receiving element B1 to be formed is formed.
The InP layer is removed by etching, and p-G (not shown)
The aInAs etching stop layer is also etched away, exposing a 0.1 μm thick p-InP cladding layer 4 ′.

Next, the dielectric film covering the portion of the optical modulator A to be formed is removed, and the i-InP layer 3 in that portion is removed.
After exposing b, the entire surface is covered and a 2-μm-thick p-I is formed at a position corresponding to the optical modulator A and the light receiving element B1.
nP cladding layer 4, 0.3 μm thick p-GaInAs
A contact layer (bad gap wavelength 1.67 μm) 5 is sequentially formed.

Next, a mask having a stripe shape with a width of about 10 μm on the entire surface and having an opening with a width of 3 μm at the boundary between the optical modulator A and the light receiving element B1 to be formed is made of, for example, SiNx. After the formation, an etching process is performed, and at least a portion until reaching the n-InP buffer layer 2 is removed by etching to form a ridge waveguide in which the boundary portion is a transverse groove.

Then, while leaving the mask,
The crystal of Fe-doped InP is grown, and the side surface of the ridge waveguide and the transverse groove are buried with the Fe-doped InP layer 10. Thereafter, the mask is once removed, and a dielectric insulating film 6 is newly formed on the entire surface. Then, the portions of the dielectric insulating film corresponding to the upper surfaces of the optical modulator A and the light receiving element B1 to be formed are removed. After removal, the p-GaInAs contact layers 6 and 6 are exposed.
a, 7c are loaded.

The back surface of the substrate 1 is polished to a total thickness of about 100 μm, a lower electrode 7b is loaded thereon, and the front end of the optical modulator A to be formed is cleaved. An incident end face S1 is formed by forming a non-reflective film 8a such that the reflectivity of 0.3 μm light becomes 1% or less. Then, by separating the whole, the device B1 of FIG. 7 in which the optical modulator A and the waveguide type light receiving element B1 are optically coupled via the Fe-doped InP layer 12 in the light traveling direction is obtained.

FIG. 10 shows the light receiving sensitivity characteristics of the device B. FIG. 10 shows a state in which a reverse bias voltage of 5 V is applied to the light receiving element B1 and the incident light intensity and the light receiving element B at that time using the reverse bias voltage applied to the optical modulator A as a parameter.
1 shows the relationship with the light receiving current. As is clear from FIG. 10, even when a reverse bias voltage of 10 V is applied to the optical modulator A, the incident light intensity and the received light current are in a proportional relationship, and good linear characteristics are obtained. This is because even if the optical modulator A, which is an optical attenuator, has a large incident light intensity and a large applied voltage, it shows good linear characteristics when the incident light intensity and the outgoing light intensity are open, and therefore, from the exit end face. This is because the light incident on the light receiving element B1 is not distorted.

In each of the apparatuses A and B, the intermediate layer, which is a conventional technique, is not used. However, since the layer structure and the intermediate layer of the present invention exhibit an effect independent of each other, both are used. Various effects can be exhibited by combining them. Further, device A, device B
Is an example of operating with light having a wavelength of 1.3 μm, but the technical idea of the device of the present invention is that if the device utilizes the Franz-Keldysh effect, the wavelength is 1.55 μm or 0.8.
It is clear that a similar effect can be exerted even for light of 5 μm or the like.

[0039]

As is clear from the above description, when the optical semiconductor device of the present invention is used as an optical modulator, the incident light intensity is high and the incident light intensity is high even when the applied voltage is high. Good linear characteristics can be obtained between the intensity and the output light intensity. Also, in the case of an optical attenuator integrated waveguide type light receiving element in which the light receiving element is integrated, a light receiving current proportional to the incident light intensity can be obtained.

This is an effect that the thickness of the i-type semiconductor layer is gradually reduced along the light traveling direction in the double heterostructure of the pin structure or the nip structure.

[Brief description of the drawings]

FIG. 1 is a sectional view showing a sectional structure of a device A of the present invention.

FIG. 2 is a cross-sectional view taken along the line II-II of FIG.

FIG. 3 is a partial perspective view showing an example of a mask pattern when selectively growing i-InP.

FIG. 4 is a graph showing a modulation characteristic of the device A;

FIG. 5 is a graph showing a relationship between an incident light intensity and an outgoing light intensity in the device A.

FIG. 6 is a graph showing a relationship between an incident light intensity and an outgoing light intensity in the comparative example device.

FIG. 7 is a sectional view showing a sectional structure of the device B of the present invention.

8 is a sectional view taken along the line VIII-VIII in FIG.

FIG. 9 is a sectional view taken along line IX-IX in FIG. 7;

FIG. 10 is a graph showing the light receiving sensitivity characteristics of the device B.

[Explanation of symbols]

 Reference Signs List 1 n-InP substrate 2 n-InP buffer layer (first semiconductor layer) 3 i-type semiconductor layer 3 a i-GaInAsP light absorbing layer 3 bi-InP layer 3 ci i-InP spacer layer 3 di-GaInAs light absorbing layer 4, 4 'p-InP cladding layer (second semiconductor layer) 5 p-GaInAs contact layer 6 dielectric insulating film 7a, 7c upper electrode 7b lower electrode 8a, 8b antireflection film 9 unit mask 9a, 9b mask 10 Fe-doped InP layer

Claims (2)

[Claims] An intrinsic semiconductor layer is interposed between a first semiconductor layer made of an n-type or p-type first semiconductor and a second semiconductor layer made of a second semiconductor having an opposite conductivity type to the first semiconductor. An optical semiconductor device having an interposed double hetero structure, wherein the thickness of the intrinsic semiconductor layer gradually decreases in the light traveling direction. 2. An optical semiconductor device according to claim 1, wherein a waveguide type light receiving element is optically coupled to the light emitting end of the optical semiconductor device according to claim 1.