JP3228384B2 - Optical switch array - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an optical switch array driven and controlled by an optical signal.

[0002]

2. Description of the Related Art The development of an optical switch array as a key device for optical signal processing and optical information processing is highly desired. Conventionally, as this type of element, for example, a document (IEEE)
E PHOTONICS TECHNOLOGY LE
As shown in TTERS Vol. 1, p. 62 (1989)), a light receiving section composed of a photodiode, a load resistor and a constant voltage source, an amplification section composed of an FET, and a multiple quantum well (M
QW) An optical modulator composed of an optical modulator composed of a pin type optical modulator.
An element called “field-effect transistor seed (FET-SEED)” having a function of changing the transmitted light intensity of light applied to a pin optical modulator has been proposed. In this device, the transmitted light intensity of the light biased at a constant intensity can be controlled by the input light having the same wavelength as that of the light biased by the quantum confinement Stark effect. The configuration and characteristics will be described with reference to FIG.

[0003] A p-AlGaAs is formed on a p-GaAs substrate.
Layer, i-MQW layer, n ⁻ -GaAs layer and n ⁺ -Ga
FIG. 15 shows the epitaxial substrates stacked in the order of As.
It is processed as in (A). FET is MQW-
In the pin structure, an n ⁻ -GaAs layer is used as a channel. Assuming that light incident on the left MQW-pin structure is input light, light incident on the right MQW-pin structure is bias light, and transmitted light thereof is output light, FIG.
The characteristic shown in (B) appears.

[0004]

The above-described device has the following problems.

[0005] First, since an MQW-pin diode is used as an optical modulator, the extinction ratio is low and the loss is large.

Second, the optical system becomes complicated because bias light needs to be incident on the light modulation section.

Third, since the operating voltage of the light modulator is as large as about 10 V, the response speed is low.

Fourthly, the operating range of the optical modulator using the quantum confined Stark effect is limited to several nm, and the operating wavelength of the optical modulator fluctuates due to heat or the like. Is tough.

An object of the present invention is to solve the above problems,
An object is to realize an optical switch array having a high extinction ratio and a high response speed.

[0010]

[Means for Solving the Problems]

1) An optical switch array according to the present invention includes: a light receiving unit for converting an input signal into an electric signal; an amplifying unit for amplifying the electric signal; An optical switch comprising a laser output unit is monolithically formed on a substrate.

2) Further, in the optical switch array according to the present invention, in the element described in 1) above, the light receiving portion may be an MSM (M
et al-Semiconductor-Metal)
It is characterized by comprising a photodiode or a photoconductor.

3) The optical switch array according to the present invention comprises:
The element according to the above 1), wherein the light receiving section is composed of a plurality of light receiving elements.

[0013]

According to the present invention, the above-mentioned problems have been solved as follows in order to solve the above-mentioned problems.

That is, in the optical switch array according to the present invention, high contrast can be obtained because a surface emitting laser (SELD) is used as the light modulation section. In addition, an operation voltage of about 2 V is sufficient, so that high-speed operation can be realized. Further, since this device does not require bias light and operates only with an input light source, the wavelength of the light source is less limited, and the optical system is simplified. Further, when this element is configured in multiple stages and optical interconnection or the like is performed by performing optical connection such that output light from the previous stage is used as input light, SELD
The oscillation wavelength is very sensitive to the fluctuation of the film thickness and the control is difficult.
If an MSM photodiode or the like is used, almost uniform photosensitivity can be obtained over a wide wavelength range of 100 nm or more, so that there is no limitation on the oscillation wavelength of the preceding SELD, which is advantageous for multistage operation.

[0015]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below in more detail with reference to the accompanying drawings by way of embodiments, but it is needless to say that the present invention is not limited to these embodiments.

First, the structure of the device of the present invention will be described with reference to FIG. 1, and then the operation principle and characteristics of the device will be described with reference to FIGS. In the following, M
An example in which an SM photodiode is used as an amplifying section using an FET will be described. Alternatively, a pin diode and a photoconductor may be used as a light receiving section, and a bipolar transistor such as a heterojunction transistor may be used as an amplifying section. Further, specific examples of the layer configuration of the element and the optical switch characteristics will be described with reference to FIGS.

FIG. 1 shows the structure of the element.
FIG. 1A shows a basic configuration of the device. A surface emitting laser 12 as an optical modulation unit 10 and an F unit as an amplification unit 20 are shown.
The ET 22 is connected in series to the constant voltage source. At this time, the surface emitting laser 12 is connected so as to have a forward bias. 24, 26 and 28 are the source, gate and drain of the FET 22, respectively. M as the light receiving unit 30
The SM photodiode 32 and the load resistor 34 are also connected in series with the constant voltage source, and the contact 36 between the MSM photodiode 32 and the load resistor 34 is connected to the gate 26 of the FET. P _in and P _out are input light and output light, respectively.
Input light P _in is incident on MSM photodiode 32.
FIG. 1B shows an example in which the light receiving unit is composed of two MSM photodiodes 3.
It is a figure at the time of connecting and configuring 2,32 series. In this case, when an optical signal to be input to each of the MSM photodiode and P _A and P _B, only the gate voltage of the FET22 is changed if both P _A and P _B are entered simultaneously, as described below, V ₁ > V ₂ , NAND operation, and V ₁ <V ₂ , AND operation. FIG. 1C shows that the light receiving unit 30 is provided with two MSM photodiodes 32 and 32.
Are connected in parallel to each other. Assuming that the optical signals input to the respective MSM photodiodes are P _A and P _B , the gate voltage of the FET changes when either P _A or P _B is input, and V ₁ > V ₂ as described later. NOR operation when V ₁ <V ₂
Work.

Next, the operation principle of the device of the present invention will be described.

FIG. 2 shows an example of the relationship between current-voltage characteristics and current-output light intensity of a surface emitting laser. The threshold current is 1.5 mA, and the voltage at that time is 2.2 V. The light output light intensity increases as the voltage increases.
For example, when the current is 3.1 mA, the output light intensity is 1 mW.
It is.

FIG. 3A shows M in the configuration of FIG.
The change of the gate voltage with respect to the intensity of the incident light incident on the SM photodiode is shown. FET gate voltage V _G
Is varied slightly from V _G1 to V _G2 (V _G1 <V _G2 ) by input light, is slightly smaller than V ₁ (about 0.1 V), and V ₂
The greater than V _G2 (smaller than the withstand voltage of the MSM photodiode can be freely set) set. Therefore, the gate voltage increases as the input light intensity increases. Also, F
The gate voltage dependence of the drain current of ET is shown in FIG. As a result, the input light intensity dependence of the drain current is as shown in FIG.

FIG. 4 is a diagram for explaining the operating characteristics of the device shown in FIG. FIG. 3C shows the input light intensity dependency of the drain current and the characteristics when the current-voltage characteristics of the surface emitting laser are considered together with the power supply voltage. First, when the input light is P ₀ , the operating point is at a, and almost no current flows through the surface emitting laser. Therefore, the output light intensity is almost 0 from FIG. When an input light intensity P _2, the operating point moves to c, the current flowing through the surface-emitting laser is lasing threshold. When the input light intensity is further increased, the surface emitting laser starts oscillating, and the output light increases rapidly. Therefore, the output characteristics are as shown in FIG. At this time, when _Vb is small, even if the input light is increased, the output light is saturated at a certain current value. In this case, the output light intensity is saturated as shown in the figure.

In this device, the input / output characteristics can be inverted by inverting the voltages applied to V ₁ and V ₂ . In the case of FIG. 6 (A), slightly larger than the V ₁ V _G2 (about 0.1 V), the V ₂ less than V _G1, i.e. set to a negative polarity. Therefore, in this case, the gate voltage decreases as the input light intensity increases. As a result, the input light intensity dependence of the drain current is shown in FIG.
The drain current decreases as the input light intensity increases.

FIG. 7 is a diagram for explaining the operating characteristics of the device. Current-voltage characteristics are shown in the same manner as in FIG. First, when the input light is P ₀ , the operating point is at a, and a large current flows through the surface emitting laser, so that a large output light intensity can be obtained. Increasing the input light intensity, decreases gradually output light, current input light intensity flows through the surface emitting laser when P ₃ is equal to the threshold value. When the input light intensity is further increased, the surface emitting laser no longer oscillates and only emits spontaneous light. Therefore, the input / output characteristics are as shown in FIG. Also at this time, when _Vb is small, the current value does not decrease to a certain gate voltage even if the gate voltage of the FET becomes small, so that the output light intensity is constant in the region where the input light intensity is small as shown in the figure. Area.

(Specific Example) Next, FIG. 9 shows a specific example of the optical switch array of the present invention shown in FIG. FIG. 9 shows GaAs /
FIG. 3 is a cross-sectional view of an element using AlGaAs.

A light receiving section on a semi-insulating GaAs substrate 101,
A layer composed of an i-GaAs light absorption layer 102 (2 μm) n ⁻ -GaAs channel layer 103 (0.2 μm) n ⁺ -GaAs contact layer 104 (0.4 μm) as an amplifying section, and InGa as an etching stop layer
P layer 105 (10 nm) Further, n-AlAs (71.5 nm) / n-Al _0.15 Ga _0.85
N-DBR (Distributed Bragg) having a structure in which As (62.9 nm) layers are alternately stacked for 30 periods.
(Reflector) layer 106, i-AlGaAs active layer 107, p-AlAs (71.5 nm) / p-Al
_A layer composed of a surface emitting laser including a p-DBR layer 108 having a structure in which _0.15 Ga _0.85 As (62.9 nm) was alternately laminated for 25 periods was formed by a molecular beam epitaxial growth method. Be and Si are p-type and n-type dopants, respectively.
Was used.

The grown wafer was processed as shown in FIG. 9 to produce an optical switch. AuZn for the p-type electrode 110
Ni was used for the n-type electrode 111, Au / Ge / Ti / Pt / Au for the Schottky electrode 112, and Cr / Au for the inter-element wiring metal 113. in this case,
Although the tungsten silicide thin film is used as the load resistance 114, other active resistances using a metal thin film or FET may be used. Further, the steps are filled with polyimide 115. Further, an element can be formed without performing mesa separation by performing an ion implantation method or the like. In the case of arraying, the elements may be arranged two-dimensionally.

FIG. 10 shows the device characteristics of this embodiment. Figure 10 (A) is V _b and V ₂ 3V, the V ₁ -
This is the case where the voltage is set to 0.2V. When the input light is 10 μW or less, the output light is only spontaneous emission light of 1 μW or less. However, when the input light is further increased, the output light intensity increases,
When the input light intensity was 100 μW, the output light intensity was 1 mW. Figure 10 (B) and the V _b 3V, -3V and V _2, V
₁ is set to 0.2V. When there was no input light, the output light was 1 mW or more. However, when the input light was further increased, the output light intensity decreased. When the input light intensity was 100 μW, the output light intensity became 1 μW or less. In both cases, the response time of this element was 1 ns or less when the input light intensity was 100 μW and both the on-time and the off-time. Further, the characteristics did not change even when the wavelength of the input light was changed from 750 nm to 870 nm.

In the first embodiment shown in FIG. 9, by inserting an etching stop layer, the n ⁺
-Heading of the GaAs contact layer (0.4 μm) was performed. This is because the thickness of the n ⁻ -GaAs channel layer fluctuates if there is unevenness in the thickness of the layer remaining above the n ⁻ -GaAs channel layer when a recess is formed by etching the gate portion of the FET. This is to prevent fluctuations in the current-voltage characteristics of the FET. In order to further suppress this fluctuation, the channel layer thickness can be made uniform using HEMT or the like.

Further, as a method without using an etching stop layer in the second embodiment, there is a selective growth. In this case, MOCVD is used as a method for growing the surface emitting laser.
Method was used. The procedure is shown in FIG. First, a light receiving section and an amplifying section are grown with the same layer configuration as in the first embodiment. The growth method at this time is MOC epitaxial growth.
The VD method may be used. That is, the i-GaAs substrate 101,
An i-GaAs light absorption layer 102, an n ⁻ -GaAs channel layer 103, and an n ⁺ -GaAs contact layer 104 are sequentially stacked (FIG. 11A). Next, an insulating film 1 such as SiO ₂
Then, the insulating film is removed by etching only where the surface emitting laser is to be grown (FIG. 11).
(B)). Then, a surface emitting laser portion is grown by MOCVD (FIG. 11C). Finally, processing is performed as shown in FIG. The size of the window of the insulating film is 100 μm, and n
-The influence of the resistance of the DBR layer was reduced, and only the active layer portion was set to 10 μm. In this case, the p-DBR layer is made of a semiconductor, but the dielectric multilayer film 11 such as SiO ₂ / TiO _{2 is used.}
6 can also be used. This example is shown in FIG.

In addition to the above, there is a method in which an FET is formed by an ion implantation method without using an etching stop layer.

In this case, a layer composed of an i-GaAs light absorbing layer (2 μm) as a light receiving part and an amplifying part on a semi-insulating GaAs substrate is used as a light modulating part. N-AlAs (7.15 nm) / n-Al _0.15 Ga _0.85
N-DBR (Distributed Brag) having a structure in which As (62.9 nm) is alternately stacked for 30 periods.
g Reflector) layer, i-AlGaAs active layer, p-AlAs (71.5 nm) / p-Al _0.15 Ga
_A layer composed of a surface emitting laser including a p-DBR layer having a structure in which _0.85 As (62.9 nm) was alternately laminated for 25 periods was formed by a molecular beam epitaxial growth method. p-type,
Be and Si were used as the n-type dopants, respectively.

After a surface emitting laser is formed by performing mesa etching on the layer for the light modulation portion of the epitaxial substrate, Si is formed to form an n-channel layer on the i-GaAs light absorbing layer exposed on the surface. Ion implantation and FET
To form

In the above embodiments, the MSM photodiode and the FET are used as the light receiving section and the amplifying section.
n Diode and HBT (Hetero Bipolar)
FIG. 14 shows an example in which an optical switch array is configured by using a transistor.

A light receiving section on a semi-insulating GaAs substrate 101,
N ⁺ -GaAs contact layer 120 (0.5 μm) n-AlGaAs emitter layer 121 (0.2 μm) p-GaAs base layer 122 (0.1 μm) i-GaAs light absorption layer and collector layer 123 (1. 0
μm) A layer composed of the n ⁺ -GaAs contact layer 124 (0.2 μm) was grown. The surface emitting laser portion can constitute an element in the same manner as in the first and second embodiments. However, in this case, similarly to FIG. 5, only those having input / output characteristics in which the output light intensity increases as the input light intensity increases can be realized.

Although the embodiments described so far have been grown on a GaAs substrate, similar elements can be constructed on an AlGaAs substrate. In this case, since the substrate is transparent to the oscillation wavelength of the surface emitting laser, there is an advantage that light can be input from both sides.

In the above embodiment, GaAs / AlGa
Although the optical switch array is composed of As, it is not limited to this, and InGaAs / InP, InAlAs / In
It can be applied to other material systems such as GaAs and GaAs / InGaAs.

[0037]

As described above, the optical switch array according to the present invention has an optical amplifying function, has a large extinction ratio, and can operate at high speed with low input intensity.
Further, in the optical switch array according to the present invention, the FET
The one using as an amplifying unit has a feature that the operation can be inverted by the operating voltage. Due to these features, it is very promising as an optical switch array according to the present invention or a future optical information processing device.

[Brief description of the drawings]

1 (A), 1 (B) and 1 (C) are circuit diagrams showing the structure of the device of the present invention.

FIG. 2 is a characteristic diagram showing an example of a current-voltage characteristic and a current-output light intensity relationship of a surface emitting laser.

3 (A) is slightly smaller than the V ₁ V _G1 (0.1V
Degree), the V ₂ can be freely set smaller than the breakdown voltage of greater than V _G2 (MSM photodiode) characteristic diagram showing the change of the gate voltage with respect to the incident light intensity incident on the MSM photodiode in the case of setting, (B) Is a characteristic diagram showing the gate voltage dependence of the drain current of the FET, (C)
FIG. 4 is a characteristic diagram showing input light intensity dependence of a drain current.

FIG. 4 is a characteristic diagram illustrating operating characteristics of the device of FIG.

FIG. 5 is a characteristic diagram illustrating input / output characteristics of the device of FIG.

FIG. 6 (A) shows that V ₁ is slightly larger than V _G2 (0.1 V
Characteristic) showing the input light intensity dependency of the gate voltage when V ₂ is smaller than VG ₁ , that is, when the polarity is set to a negative polarity. FIG. 3B shows the drain current with respect to the drain current of the FET when the input light intensity changes. FIG. 4 is a characteristic diagram illustrating a relationship between voltages.

FIG. 7 is a characteristic diagram illustrating operation characteristics of the element.

FIG. 8 is a characteristic diagram showing input / output characteristics in the case of FIG. 7;

FIG. 9 is a cross-sectional view of an element using GaAs / AlGaAs.

FIGS. 10A and 10B are characteristic diagrams of the device of the present invention, respectively.

11A and 11B are process diagrams showing a procedure for selectively growing a surface emitting laser portion by using the MOCVD method, wherein FIG. 11A is a cross-sectional view showing a light receiving portion and an amplifying portion, and FIG. FIG. 4C is a cross-sectional view showing patterning, and FIG. 4C is a cross-sectional view showing surface emitting laser partial growth (selective growth by MOCVD).

FIG. 12 is a configuration diagram of an element in the case of FIG. 11;

FIG. 13 is a configuration diagram of an element when a dielectric multilayer film is used.

FIG. 14 is a cross-sectional view showing a configuration of an element using an HBT and a pin diode.

FIG. 15A is a configuration diagram of a conventional device, and FIG. 15B is a characteristic diagram thereof.

DESCRIPTION OF SYMBOLS 10 Light modulation unit 12 Surface emitting laser 20 Amplification unit 22 FET 24 Source 26 Gate 28 Drain 30 Light receiving unit 32 MSM photodiode 34 Load resistance 36 Contact 101 Semi-insulating AlGaAs substrate 102 i-GaAs layer 103 n ⁻ -GaAs channel layer 104 n ⁺ -GaAs contact layer 105 InGaP layer 106 n-DBR layer 107 i-AlGaAs active layer 108 p-DBR layer 110 p-type electrode 111 n-type electrode 112 Schottky electrode 113 Cr / Au 114 load resistance layer 115 polyimide layer 116 dielectric multi-layer film mirror 120 n ⁺ -GaAs contact layer 121 n-AlGaAs emitter layer 122 p-GaAs base layer 123 i-GaAs layer light-absorbing layer and the collector layer 124 n ⁺ -GaAs con Transfected layer P _in the input light P _out output light P _A, P _B light signal

──────────────────────────────────────────────────続 き Continued on the front page (72) Inventor Seiji Fukushima 1-1-6 Uchisaiwaicho, Chiyoda-ku, Tokyo Nippon Telegraph and Telephone Corporation (72) Inventor Taketaka Kohama 1-16-1 Uchisaiwaicho, Chiyoda-ku, Tokyo Nippon Telegraph and Telephone Corporation (72) Inventor Yoshitaka Oiso 1-1-6 Uchisaiwai-cho, Chiyoda-ku, Tokyo Nippon Telegraph and Telephone Corporation (56) References JP-A-62-14465 (JP, A) JP-A Sho JP-A-3-268352 (JP, A) JP-A-4-211172 (JP, A) JP-A-5-41509 (JP, A) JP-A-4-163986 (JP, A) A) (58) Fields surveyed (Int. Cl. ⁷ , DB name) H01L 27/15 H01S 5/00-5/50

Claims (3)

(57) [Claims] 1. An output unit comprising a light receiving unit for converting an input signal into an electric signal, an amplifying unit for amplifying the electric signal, and a vertical cavity surface emitting laser for oscillating output light by the amplified electric signal. An optical switch array comprising: an optical switch comprising: a monolithically structured optical switch on a substrate. 2. The method according to claim 1, wherein the light receiving unit is an MSM (Metal-Semi).
2. The optical switch array according to claim 1, wherein the optical switch array is made of a photo diode or a photoconductor. 3. The optical switch array according to claim 1, wherein the light receiving section comprises a plurality of light receiving elements.