JPH043012A - Method for driving semiconductor optical device - Google Patents

Abstract

PURPOSE:To enhance the performance of the optical device by adjusting the width of semiconductor quantum well layers thereby setting the light absorption end by the excitons based on the light holes in the semiconductor quantum well layers near an operating wavelength. CONSTITUTION:The light absorption end by the excitons based on the light holes in the semiconductor quantum well layers 3 is set near the operating wavelength by adjusting the width of the quantum well layers. A change in the absorption coefft. of the semiconductor quantum well layers 3 arising from the impression of voltage between two electrodes 5 and 6 and the control of light by a change in refractive index is executed with respect to the light having the electric field component perpendicular to the semiconductor quantum well layers 3. Namely, this device is driven by utilizing the electric field shift effect at the exciton resonance absorption end by the light holes when TM polarized light is made incident. The performance of the semiconductor device is enhanced in this way.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to a method for driving semiconductor optical devices such as optical modulators and optical switches, and relates to a method for driving semiconductor optical devices such as optical modulators and optical switches. By inputting light and setting an exciton resonance absorption edge due to light holes in the vicinity of the operating wavelength, it is possible to improve the performance of the device, such as high modulation degree and low voltage operation.

[Conventional technology]

Electrons generated by light irradiation on the semiconductor and the regular tag combine by Coulomb attraction to form excitons.

When these excitons are formed within the quantum well, they create a strongly bound state due to the high potential of the barrier layer, and the excitons become stable even at room temperature. These room-temperature excitons have been used in the following ways. FIG. 6(a) is an absorption coefficient spectrum for TE polarized light having an electric field component parallel to the quantum well layer accompanying this exciton. This absorption coefficient spectrum is shifted to the lower energy side by an external electric field (referred to as the quantum confined Stark effect). For example, if the wavelength of the incident light is λ1 in the figure, if the device structure parameters are set so that the light absorption peak wavelength after applying an electric field matches λ1, a light intensity modulator that can control the light intensity by the electric field can be configured. can do. FIG. 6(b) shows the device structure, and FIG. 6(C) shows its modulation characteristics. That is, in this case, a large On10ff ratio (FIG. 6(C)) can be obtained by setting the optical absorption peak wavelength when an electric field is applied to the operating wavelength λ1 by adjusting the quantum well layer width. As shown in Figure 6 (al), the exciton absorption spectrum generally consists of two peaks. The peak on the lower energy side is the exciton (
The exciton on the high energy side corresponds to the exciton (Lt) (abbreviated as Heavy-Hole: HH), which is composed of light holes and electrons. The electric field change at the optical absorption edge caused by this has been utilized.The utilization of this HH absorption assumes that the light emitted from the semiconductor laser is TE polarized light.

(Problems to be Solved by the Invention) The conventional driving method using exciton absorption by a heavy tag based on the incidence of TE polarized light as described above has a high modulation degree of exciton-based optical devices such as optical modulators and optical switches. There was a problem that high performance such as low voltage operation could not be achieved sufficiently.In fact, there is no need to insist on the incidence of TE polarized light; instead, it is possible to input 7M polarized light with an electric field component perpendicular to the other quantum well layer. If it is possible to improve the performance of optical devices by using a driving method, the latter method will have higher utility value.

However, until now, there has been no proposal to improve the performance of exciton-based optical devices using 7M polarized light.

In order to solve the conventional problems, the present invention has developed a semiconductor quantum well structure by using a driving method that utilizes the electric field shift effect of the exciton resonance absorption edge due to light holes when 7M polarized light is incident. The purpose of this research is to improve the performance of semiconductor devices.

[Means for Solving the Problems] In order to achieve the above object, a method for driving a semiconductor optical device according to the present invention includes a semiconductor quantum well layer and two electrodes installed to apply an electric field to the semiconductor quantum well layer. By adjusting the quantum well layer width in a structure with
The light absorption edge due to excitons based on the light weight in the semiconductor quantum well layer is set near a given operating wavelength, and the absorption coefficient of the semiconductor quantum well layer changes with the application of voltage between the two electrodes. The present invention is characterized in that light having an electric field component perpendicular to the semiconductor quantum well layer is controlled by changing the refractive index.

[Principle and operation of the invention]

Near the absorption edge of exciton resonance absorption in a quantum well, 2
The two absorption peaks (HH and LH) are important when considering applications to optical modulators, optical switches, etc. From the theoretical study of the optical transition process, the optical absorption coefficient due to these excitons strongly depends on the polarization direction of the irradiated light through the transition matrix elements.

FIG. 1(a) shows the dependence of the square M2 of the transition matrix element on the polarization direction.

The angle θ is the angle that the polarization direction makes with the direction perpendicular to the quantum well. Mt is expressed with the HHO value of θ=906 as 1. In TE mode light (θ=90'), M2 of LH is H
It is 1/3 times that of H.

On the other hand, in 7M mode light (θ=0°), M2 of LH
increases by 4/3 times, and M2 of HH becomes 0. By the way, the absorption coefficient peak α9 is expressed by the following formula % where A is a constant specific to the material and a is the exciton radius (electron/
average distance between holes), is the energy linewidth of absorption.

Due to the θ dependence as shown in FIG. 1(a), the absorption coefficient spectrum for TE polarized light becomes as shown in FIG. 1(b).

That is, the HH observed for TE polarized light disappears for 7M polarized light, and instead, LH shows a peak value of absorption coefficient larger than HH.

Therefore, by combining this 7M polarized light and enhanced LH absorption, it becomes possible to improve the performance of optical devices. There are several advantages of using this combination (TM light and LH absorption) in addition to the above-mentioned enhancement of M2. This will be explained below.

As is clear from FIG. 1(b), the LH absorption edge is on the higher energy side than the HH absorption edge. Therefore, given a certain operating wavelength, for example an optical fiber communication wavelength of 1.55 μm,
When using the TM light/LH absorption edge (hereinafter referred to as TM operation), the TE light/f (when using the H absorption edge (TE
The device must have a quantum well structure that differs from its operation (operation).

This is achieved by setting the width of the quantum well layer to LTM>LTE and lowering the quantum level. Increasing the well layer width in this way for TM operation of the device
It brings about the following excellent effects ((1) and (ii)
)). Other advantages of TM operation compared to TE operation (including the M2 enhancement (iv) discussed above) are also described below.

(i) The larger the well layer width, the smaller the absorption energy line width F becomes, so the absorption edge becomes sharper, and at the same time, the absorption peak α2 increases due to 2(1). ``decrease in and α,
An increase in will result in a large on10ff ratio.

(ii) The larger the well layer width, the larger the shift in the peak position due to the application of an electric field, which enables low electric field operation.

(Mountain) The exciton radius a is aLH × aHH for LH and HH.
Therefore, α2 becomes larger when LH is used according to equation (1).

(iv) Increase in α2 due to enhancement of M2 during TM operation.

[Example:· Embodiments of the present invention will be described in detail below with reference to the drawings.

As shown in FIG. 2(a), the InGaAs of the first embodiment
A driving method for an optical intensity modulator using a /fnAIAs quantum well structure will be described. The optical intensity modulator shown in FIG. 2(a) is manufactured as follows. First, by molecular beam epitaxial method, n-type InG was deposited on an n-type 1nP substrate 1.
Grow aAs layer 2. On top of this, an undoped In with a well layer width of 101 layers and a barrier layer width of 100 layers as described above.
GaAs/InAlAs quantum well layer 3 is grown and then p°
After growing the type 1nGaAs layer 4, it is processed as shown in the figure and electrode metals 5 and 6 are attached. When a reverse bias is applied between the electrodes 5 and 6 of the device manufactured in this manner, an electric field is applied to the quantum well N3, and the absorption coefficient of this region changes as shown in the second [m(b)]. Here, regarding TE operation, when electric field F = 160 kV crrV' is applied, the HH absorption peak position is at the operating wavelength λ, = 1.5
The well layer width L-76 was determined to match 5 μm.

In the figure, the absorption coefficient spectrum α(F) for F=O and 160 kV −an−′ is indicated by a broken line.

For TM operation, electric field F=120kV -c+n
-' is applied, the LH absorption peak position is at the operating wavelength zI
Well layer width L = 10 to match = 1.55 μm
I decided on one person. In Fig. 2(b), solid lines indicate F=O and 1.
Absorption coefficient spectrum α (F
) is the notation. Comparing α□1(F) and α0(F) at the operating wavelength λ, α7M(120)=1°37X
10'cm-'> αtt(160) =0.67X1
0'cm-' and aTl, I(0) = 20c
m-'<αyE(0) = 35 cm, and the TM operation has a more advantageous result. Moreover, TM operation has a lower electric field (120
This is achieved with kVgo ``1).

However, TM operation has a larger well width L (101 people)
is used, so when comparing operating voltages, it is not possible to determine superiority or inferiority from this alone. Therefore, we introduce the following performance index.

ζ−(α (F) −α (0))/α (0)/(
F(L+W)) (2) Here, W is the barrier layer width. The term related to α in (2) represents the efficiency of change in the absorption coefficient due to the application of an electric field, and the larger this value, the higher the performance. F (L+W) corresponds to the voltage applied in one period, and it is desirable that this is small. If we estimate ζ assuming W = 100 people, we get 1
．． 4/ζ↑E = 4, 2, which makes the figure of merit approximately 4
It can be seen that the performance is doubled.

FIG. 2(c) shows the temporal change in the intensity of the emitted light 8 when the intensity of the TM incident light 7 with a wavelength of 1.55 μm is kept constant. It can be seen that the intensity of the emitted light is modulated depending on the applied electric field. Compared to the conventional TE operation type (well layer width of 76 people), the modulation efficiency of the TM operation type is η = (11-Io) /
It can be seen that l+ is approximately doubled.

FIG. 3(a) is a diagram illustrating a method for driving an optical phase modulator according to a second embodiment. The manufacturing method of the phase modulator shown in the figure is the same as that of the optical intensity modulator used in the first embodiment, but the well layer widths are 64 and 83 for the TE operation and TM operation, respectively. The device thus fabricated operates as follows. TM incident light 7 with a wavelength of 1.55 μm enters the quantum well region! When an electric field is applied, the phase changes due to changes in the refractive index caused by the application of an electric field. At this time, it is desirable that the refractive index change is large, but since light propagates in the medium before and after the electric field is applied, it is required that the absorption coefficient before and after the electric field is applied is very small. Since it is clear that the absorption coefficient is small when the electric field is 0, the value of the absorption coefficient after application of the electric field becomes a problem.

Therefore, when an electric field of 160 kV-cm-' is applied, the absorption coefficient at the operating wavelength λ+ = 1.55 μm is 100
The above well layer width (
64 and 83 for TE and TM operations, respectively.
person) was decided. FIG. 3(b) shows the refractive index change spectrum Δn (F) due to application of an electric field in a quantum well structure having such a well layer width. Since the refractive index is obtained by subjecting the absorption coefficient to Kramers-Kronig transformation, there is an effect of enhancing the refractive index as in FIG. 2(b).

As is clear from the figure, 1Δnter(16
0) l = 1.72X10-”>ΔnTE(16
0) l =0.737 x 10-'', the TM operation has a more advantageous result.

However, since the TM operation uses a larger well width L (83 wells), the superiority or inferiority of the operating voltage cannot be determined from this alone. Therefore, we introduce the following performance index.

η=1Δn (F)i/α(F)/(F(L+W)) (3) where W is the barrier layer width. 1Δn(F)1 in formula (3)
It can be said that the larger the value, the higher the performance. α(F) corresponds to the absorption coefficient after application of an electric field, and F (L+W) corresponds to the voltage applied in one cycle, and these are preferably small. W
Estimating η assuming = 100 people, η0.4/ηt
It can be seen that t=2.3, which means that the performance index is about twice as good.

FIG. 3(c) shows a change in the phase of the emitted light due to the application of an electric field. In the TM operation, a phase change approximately twice as large as that for the same driving voltage is observed, reflecting the results shown in FIG. 3(b).

FIG. 4(a) is a diagram illustrating a method for driving a crossed waveguide type optical switch according to the third embodiment. The optical switch shown in the figure is manufactured by the following procedure. First, by molecular beam epitaxial method, a well layer, which is a component of the active region, has a width of 113 layers and a barrier layer has a width of 1 layer on the entire surface of an n-type and nP-type substrate 11.
After crystal growth of the undoped InGaAs/InAlAs multiple quantum well layer 12a and the P-type InP layer 12b, mesa etching is performed leaving only the rectangular active region 12. Furthermore, another semiconductor material (n-type 1nP layer) which is a component of the passive region 14 is crystal-grown on the entire surface, and then this grown layer is mesa-etched in a cross-shaped manner so as to leave the passive region 14 with a restoring electrode. to complete the element.

The device thus fabricated operates as follows. TM incident light 7 incident from the PI is guided through the passive region 14-3 and reaches the active region 12a. When no voltage is applied to the active region 12a, light passes through this region and is guided through the passive region 14-2 to the light output boat P.
Reach 2. Bridge metal 13 provided in the active region 12.
When a voltage is applied to 13, the refractive index of this region changes,
The light undergoes total reflection at the interface between the active region 12 and the passive region 14, and is now guided through the passive region 14-1, reaching the light output boat P3. However, the crossing angle of the passive region 14 is determined so that total reflection occurs for the refractive index value after voltage application. That is, the light having a wavelength of 1.55 μm that is incident from the light input port 1-Pl is distributed to the light output port P2 and the light output port 1-P3 according to the application of an electric field, thereby forming an optical switch.

The change in the refractive index of the active region 12 due to the above electric field application will be explained using FIG. 5. Regarding TE operation, when applying an electric field F = 160 kV -cm-', HH
The peak position of the refractive index change due to the operating wavelength λ.

The well layer width L=88 was determined so as to match L=1.55 μm. In the figure, the broken line indicates F = 160 kV -am-'
The refractive index change spectrum Δn (F) for

For TM operation, the electric field F = sokv −clm−
The well layer width L=113 was determined so that the peak position of the refractive index change due to LH coincides with the operating wavelength λI=1.55 μm when ' is applied. In the figure, the solid line indicates F = 80kV
・Refractive index change spectrum Δn(F) for Cm-'
is the notation. ΔnTjl(F) at operating wavelength λ,
Comparing Δnti(F), 1Δn T, 4(80)
=0.343> lΔnt, (160) l =0.1
96 and TM operation have more advantageous results. Moreover, the TM operation achieves this with a lower electric field (80 to ν·cs −').

However, TM operation has a larger well width L (113 people)
is used, so when comparing operating voltages, superiority or inferiority cannot be determined from this alone. Therefore, we introduce the following performance index.

ζ=IΔn (F) l / (F (L+W))
(4) Here, W is the barrier layer width. 1Δ in formula (4)
It can be said that the larger n(F)l is, the higher the performance is. F (L+W
) corresponds to the voltage applied in one period, which is preferably small. When we estimate ζ assuming W=100 people, we get 4/ζ at=3.1, which means that the performance index is about 3 times better.

FIG. 4(b) shows the change in the intensity of the emitted light observed at the light emitting port P3 due to the electric field. When TM operation is compared with conventional TE operation, a large change in output light intensity is observed reflecting a large change in refractive index. Furthermore, since the refractive index change is large, the intersection angle between the two waveguides can be made larger than in the past, which has the advantage of increasing the convenience in manufacturing the device.

Although the above three embodiments have been described using InGaAs/InAlAs quantum wells as an example, the present invention can be similarly applied to other semiconductor material systems.

〔Effect of the invention〕

As explained above, in a semiconductor quantum well, the present invention is based on the TM polarization of the exciton resonance absorption edge due to light holes, rather than the conventional electric field effect on TE polarization of the exciton resonance absorption edge due to heavy holes. Since this is a driving method for semiconductor optical devices that utilizes the electric field effect on light holes, it is possible to obtain a large electric field shift in the absorption coefficient and refractive index due to the excellent effect unique to light hole excitons. There is an advantage that large modulation efficiency can be obtained in optical devices such as optical modulators and optical switches that utilize excitons.

[Brief explanation of the drawing]

FIGS. 1(a) and 1(b) are diagrams for explaining the present invention in detail, and FIG. 1(a) shows the square of the matrix element of optical transition (M
Figure 1(b) is a diagram showing the absorption coefficient spectrum for TE polarized light and TM polarized light. Figure 2(a), (b), and (c) are diagrams showing the polarization direction dependence of 2). FIG. 2(a) is a diagram showing the structure of the optical intensity modulator and its driving method, and FIG. 2(b) is a diagram showing the TE operation and TM operation.
Two quantum wells (each well layer width is 7 mm) correspond to the operation.
Figure 2 (c) is a diagram showing the modulation characteristics, Figure 3 (a) (b) (c) is a diagram showing the absorption coefficient spectrum for 6 people and 101 people) and how it changes due to the electric field. 3(a) is a diagram illustrating the structure of an optical phase modulator and its driving method, and FIG. 3(b) is a diagram illustrating the TE operation and TM operation.
Two quantum wells corresponding to the operation (each well layer width is 6
Figure 3 (c) is a diagram showing the modulation characteristics, Figures 4 (a) and (b), and Figure 5 are diagrams showing the change in refractive index due to the electric field for 4 people and 83 people. FIG. 4(a) is a diagram showing the structure of an optical switch and its driving method, FIG. 4(b) is a diagram showing switch characteristics, and FIG. 5 corresponds to TE operation and TM operation. Two quantum wells (well layer widths of 88 and 113, respectively)
Figures 6(a), 6(b), and 6(c) show the changes in the refractive index due to the electric field. Figures 6(a), 6(b), and 6(c) show the conventional technology of an optical intensity modulator using an absorption edge due to the exciton of a heavy hole and TE polarized light. FIG. 6(a) is a diagram showing the absorption coefficient spectrum for TE polarized light and its change due to electric field, and FIG. 6(b) is a diagram explaining the structure of the light intensity modulator and its driving method. This figure and FIG. 6(C) are diagrams showing control of transmitted light by an electric field. 1...n-1nP substrate 2--n''-1nGaAs 3...Undoped InGaAs/In^IAs
Quantum well layer 4 ---p''-1nGaAs 5... Electrode metal 6... Electrode metal 7... TM incident light 8... Outgoing light 9 ...TE incident light 10 ... Output light Pl ... Light input boat P2 ... Light output boat P3 ... Light output boat P4. ..... Optical boat 11 .....n-rnP substrate 12 ..... Active region 13 ..... Electrode metal 14 ..... Passive region

Claims (1)

[Claims] In a structure having a semiconductor quantum well layer and two electrodes installed to apply an electric field to the semiconductor quantum well layer, by adjusting the width of the semiconductor quantum well layer, light holes in the semiconductor quantum well layer can be The light absorption edge due to excitons based on the semiconductor quantum well layer is set in the vicinity of a given operating wavelength, and the light is controlled by a change in the absorption coefficient or a change in the refractive index of the semiconductor quantum well layer due to the application of a voltage between the two electrodes. A method for driving a semiconductor optical device characterized in that the method is performed using light having an electric field component perpendicular to a quantum well layer.