WO2023095261A1

WO2023095261A1 - Mach-zehnder modulator

Info

Publication number: WO2023095261A1
Application number: PCT/JP2021/043252
Authority: WO
Inventors: 崇柳楽; 鴻介木村
Original assignee: 三菱電機株式会社
Priority date: 2021-11-25
Filing date: 2021-11-25
Publication date: 2023-06-01

Abstract

A p-type semiconductor layer (12) is formed on a semi-insulating substrate (11). First and second arm waveguides (4a, 4b) are formed on the p-type semiconductor layer (12) and are spaced apart from one another. First and second traveling wave electrodes (5a, 5b) are formed on the first and second arm waveguides (4a, 4b), respectively. A metal electrode (18) which is not connected to a power source for DC bias is provided between the arm waveguide (4a) and the arm waveguide (4b) so as to be in contact with the p-type semiconductor layer (12).

Description

mach-zehnder modulator

The present disclosure relates to Mach-Zehnder modulators.

In order to cope with the recent increase in communication capacity, multi-level technology using digital coherent technology is progressing, and in multi-level optical modulators, the amplitude and phase of light can be controlled, and zero-chirp optical modulation signals are generated. A possible Mach-Zehnder modulator has been used (see, for example, US Pat. In addition, the response speed required for modulators continues to increase in order to increase the signal capacity per unit time. There is a demand for an optical modulator capable of inputting a modulated electrical signal of 64 GBaud or 96 GBaud or more with low loss and generating a high-speed modulated optical signal by electrical-optical interaction.

In order to realize the above optical modulator, a Mach-Zehnder modulator equipped with traveling wave electrodes is being vigorously developed. The traveling-wave electrodes have a high-frequency line structure optimized for being driven by a differential signal, making it possible to drive them with a differential driver with high power efficiency.

It is necessary to match the characteristic impedance or termination resistance of the driver and the impedance of the traveling wave electrode. If impedance matching is not obtained, reflection occurs when an electric signal is input from the driver to the traveling wave electrode, resulting in power loss. In addition, the electric signal reflected by the impedance mismatch at the terminating resistor travels backward in the traveling wave electrode, and light interacts with the trailing wave, resulting in deterioration of the modulated waveform.

In addition, it is necessary to match the propagation speed of the microwave propagating in the electrode and the light propagating in the waveguide installed near the electrode. If velocity matching is not obtained, a phase shift occurs with the modulated light that is modulated by the electric field amplitude while propagating with the microwave. In particular, the higher the wavelength of the microwave becomes, the more pronounced the phase shift appears, leading to deterioration of the modulation band.

Japanese Patent Application Laid-Open No. 2016-24409

After satisfying the impedance matching and velocity matching as described above, it is necessary to reduce the loss of microwaves propagating in the traveling wave electrode and improve the modulation band. Loss reduction can also be achieved by increasing the electro-optical interaction (refractive index change) per unit length and shortening the modulator length to obtain the necessary amount of phase rotation. However, it is more essentially achieved by reducing the resistance of the traveling wave electrodes and the resistance between the traveling wave electrodes. Particularly at high frequencies, losses due to semiconductor resistance and contact resistance between traveling wave electrodes are dominant, and reduction of these resistances is important for improving the modulation band.

The present disclosure has been made to solve the problems described above, and its object is to obtain a Mach-Zehnder modulator capable of improving the modulation band.

A Mach-Zehnder modulator according to the present disclosure includes a semi-insulating substrate, a p-type semiconductor layer formed on the semi-insulating substrate, and first and spaced apart first and second semiconductor layers formed on the p-type semiconductor layer. two arm waveguides, first and second traveling-wave electrodes respectively formed on the first and second arm waveguides, the first arm waveguide and the second arm waveguide. and a metal electrode that is in contact with the p-type semiconductor layer between and is not connected to a power supply for DC bias.

In the present disclosure, the metal electrode is in contact with the p-type semiconductor layer between the first arm waveguide and the second arm waveguide. Since the metal electrode is connected in parallel with the p-type semiconductor layer, the resistance value between the arms is greatly reduced. Therefore, the loss of microwaves propagating through the traveling wave electrode can be reduced, and the modulation band of electro-optical conversion can be greatly improved.

1 is a plan view showing a Mach-Zehnder modulator according to Embodiment 1; FIG. 1 is a cross-sectional view showing a Mach-Zehnder modulator according to Embodiment 1; FIG. 2 is an equivalent circuit diagram of the Mach-Zehnder modulator according to Embodiment 1. FIG. 5 is a graph showing calculation results of electro-optical modulation bands of Embodiment 1 and Comparative Example. FIG. 8 is a cross-sectional view showing a Mach-Zehnder modulator according to Embodiment 2; FIG. 11 is a cross-sectional view showing a Mach-Zehnder modulator according to Embodiment 3;

A Mach-Zehnder modulator according to an embodiment will be described with reference to the drawings. The same reference numerals are given to the same or corresponding components, and repetition of description may be omitted.

Embodiment 1.
FIG. 1 is a plan view showing a Mach-Zehnder modulator according to Embodiment 1. FIG. Although only one Mach-Zehnder modulator 1 is shown here, an actual IQ modulator has a configuration in which a plurality of similar Mach-Zehnder modulators are arranged.

The light input from the optical input waveguide 2 is demultiplexed into two by the MMI coupler 3 and guided to the

arm waveguides

4a and 4b constituting the Mach-Zehnder interferometer, respectively. Traveling

wave electrodes

5a and 5b are provided on the

arm waveguides

4a and 4b, respectively. Ground lines 6a and 6b are provided in parallel with the

traveling wave electrodes

5a and 5b, respectively.

The phase of the light guided through the

arm waveguides

4a and 4b is modulated by the differentially modulated electric field amplitude input to the

traveling wave electrodes

5a and 5b. Terminal resistors 7 matching the differential impedance of the

traveling wave electrodes

5a and 5b are connected to the ends of the

traveling wave electrodes

5a and 5b. The terminating resistor 7 suppresses reflection of the electric field amplitude propagated through the

traveling wave electrodes

5a and 5b. Before the phase-modulated light is combined by the MMI coupler 8, the optical phase is adjusted by the phase adjusters 9a and 9b so as to be coupled to the optical output waveguide 10. FIG.

FIG. 2 is a cross-sectional view showing the Mach-Zehnder modulator according to Embodiment 1. FIG. A p-type semiconductor layer 12 is formed on a semi-insulating substrate 11 .

Arm waveguides

4a and 4b are formed on p-type semiconductor layer 12 while being spaced apart from each other. Traveling

wave electrodes

5a and 5b are formed on

arm waveguides

4a and 4b, respectively.

The arm waveguide 4a has a p-type clad layer 13a, a waveguide layer 14a formed on the p-type clad layer 13a, and an n-type clad layer 15a formed on the waveguide layer 14a. The arm waveguide 4b has a p-type clad layer 13b, a waveguide layer 14b formed on the p-type clad layer 13b, and an n-type clad layer 15b formed on the waveguide layer 14b. The

waveguide layers

14a, 14b are composed of semiconductor quantum wells.

The side surface of the waveguide layer 14a is covered with a semi-insulating semiconductor embedded layer 16a and an insulating film 17a to be protected from the external environment. Similarly, the side surface of the waveguide layer 14b is covered with a semi-insulating semiconductor buried layer 16b and an insulating film 17b.

A metal electrode 18 is formed on the p-type semiconductor layer 12 between the

arm waveguides

4a and 4b. The bottom surface of the metal electrode 18 is in contact with the top surface of the p-type semiconductor layer 12 . Side surfaces of the metal electrode 18 are in contact with the

insulating films

17a and 17b. The metal electrode 18 is not electrically connected to any structure other than the p-type semiconductor layer 12 and is connected only to the p-type semiconductor layer 12 . In particular, the metal electrode 18 is not connected to a power source for DC biasing and is not a biasing electrode.

FIG. 3 is an equivalent circuit diagram of the Mach-Zehnder modulator according to Embodiment 1. FIG. It is necessary to satisfy the conditions of impedance matching of the

traveling wave electrodes

5a and 5b and propagation velocity matching with light. Therefore, the inductances La and Lb of the traveling-

wave electrodes

5a and 5b and the capacitances Ca and Cb of the

waveguide layers

14a and 14b sandwiched between the n-

type cladding layers

15a and 15b and the p-

type cladding layers

13a and 13b are specified. value should be controlled. The inductances La and Lb are controlled by the thickness, width and electrode spacing of the

traveling wave electrodes

5a and 5b. The capacitances Ca and Cb of the

waveguide layers

14a and 14b are controlled by the width of the

waveguide layers

14a and 14b and the thickness between the n-type

clad layers

15a and 15b and the p-

type clad layers

13a and 13b.

As an example, each parameter is shown when the differential impedance is 70 Ω and the microwave propagation velocity is matched with light guided through a waveguide with a refractive index of 3.5. The thickness of the traveling

wave electrodes

5a and 5b is about 2.5 μm, the width is about 3.5 μm, and the distance between the centers of the electrodes is about 15 μm. The width of the waveguide layers 14a and 14b is approximately 1.5 μm, and the thickness between the n-type clad

layers

15a and 15b and the p-type clad

layers

13a and 13b is approximately 0.8 μm. Under this condition, the above differential impedance and propagation velocity can be matched. The waveguide layers 14a and 14b and the semi-insulating semiconductor buried

layers

16a and 16b have a dielectric constant of about 14.

The parameters for matching the impedance and propagation velocity are not limited to the above examples, and the thickness, width, electrode spacing, etc. of the traveling

wave electrodes

5a and 5b have a degree of freedom. For example, when the electrode interval is widened, the thickness and width of the electrodes must be increased in order to obtain a desired value of inductance. pass characteristic in the relatively low frequency range.

When the metal electrode 18 is not provided, the resistance value between the arms is determined only by the resistance value R1 of the p-type semiconductor layer 12. As the distance between the traveling

wave electrodes

5a and 5b increases, the resistance value between the arms increases remarkably, and the transmission characteristics deteriorate in the high frequency range of 20 GHz or higher. On the other hand, when the low-resistance resistor R2 of the metal electrode 18 is connected in parallel to the p-type semiconductor layer 12, the increase in the resistance value between the arms due to the expansion of the electrode spacing is slight, and the pass characteristic deteriorates in the high frequency region. will be limited.

FIG. 4 is a graph showing calculation results of electro-optical modulation bands in Embodiment 1 and Comparative Example. The metal electrode 18 is not provided in the comparative example. In the high frequency range of 20 GHz or higher, the resistance value between the arms becomes a dominant factor of the pass characteristics. In Embodiment 1, since the resistance value between the arms is reduced by the metal electrode 18, the 3 dB band of electrical-to-optical conversion can be improved to 97 GHz.

The surface orientation of the semi-insulating substrate 11 is <100>. It is desirable that the extending direction of the

arm waveguides

4a and 4b is <011> due to restrictions on the regrowth shape after waveguide formation. In the case of this waveguide extending direction, by laminating a p-type clad layer, a waveguide layer, and an n-type clad layer on the semi-insulating substrate 11 in this order, an improvement in phase modulation efficiency due to the Pockels effect can be expected. In the case of such a laminated structure,

arm waveguides

4a and 4b are separately formed so as to form a Mach-Zehnder interferometer by dry-etching the waveguide layer and the n-type clad layer. When the traveling

wave electrodes

5a and 5b are formed on the p-type clad

layers

13a and 13b by forming the traveling-

wave electrodes

5a and 5b on the n-type clad

layers

15a and 15b of the

arm waveguides

4a and 4b, respectively. contact resistance can be greatly reduced compared to On the other hand, since the separately formed

arm waveguides

4a and 4b are connected by the p-type semiconductor layer 12, the resistance value between the arms of the Mach-Zehnder modulator is governed by the resistance value of the p-type semiconductor layer 12. FIG. The resistance value of the p-type semiconductor layer 12 can be controlled by the doping acceptor concentration and the distance between the arms. However, when the acceptor concentration is increased, the light absorption loss increases, and when the distance between the arms is shortened, the inductance decreases, which makes impedance matching difficult.

Therefore, in the present embodiment, the metal electrode 18 is brought into contact with the p-type semiconductor layer 12 between the

arm waveguides

4a and 4b. Since the metal electrode 18 is connected in parallel with the p-type semiconductor layer 12, the resistance value between the arms is greatly reduced. Therefore, the loss of microwaves propagating through the traveling

wave electrodes

5a and 5b can be reduced, and the modulation band of electro-optical conversion can be greatly improved.

Also, the metal electrode 18 is formed so as to be in contact with the p-type semiconductor layer 12 over the entire region from the insulating film 17a to the insulating film 17b. Thereby, the resistance value between the arms can be reduced to the maximum.

The p-type semiconductor layer 12 is made of InGaAs or the like, and the p-type clad

layers

13a and 13b are made of InP or the like. Therefore, the p-type semiconductor layer 12 has a smaller bandgap energy than the p-type clad

layers

13a and 13b. Therefore, the contact resistance between the metal electrode 18 and the p-type semiconductor layer 12 can be reduced.

Forming the metal electrode 18 reduces the induction component (inductance) at high frequencies. This lowers the impedance and increases the propagation speed of microwaves. Impedance matching and velocity matching must be performed in consideration of the influence of this metal electrode 18 .

Embodiment 2.
FIG. 5 is a cross-sectional view showing a Mach-Zehnder modulator according to Embodiment 2. FIG. In this embodiment, the p-type cladding layer 13 is not etched separately between the

arm waveguides

4a and 4b. A metal electrode 18 is in contact with the p-type clad layer 13 . As a result, the resistance value between the arms can be further reduced, and the modulation band can be further widened. Other configurations and effects are the same as those of the first embodiment.

However, as the material of the p-type cladding layer 13, InP or the like having a lower refractive index than the waveguide layers 14a and 14b is used. Therefore, the contact resistance with the metal electrode 18 is higher than that of materials such as InGaAs having a small bandgap energy. Further, since the p-type clad layer 13 is arranged near the waveguide layers 14a and 14b, the acceptor concentration of the p-type clad layer 13 is set to 1.0E+18 [cm] so as to suppress light absorption due to intervalence band absorption. ⁻³ ] or lower. Therefore, the contact resistance between the metal electrode 18 and the p-type clad layer 13 tends to increase. Therefore, in order to fully exhibit the effect of reducing the resistance value between the arms by forming the metal electrodes 18, it is necessary to appropriately perform sintering after the metal electrodes 18 are formed.

Embodiment 3.
FIG. 6 is a cross-sectional view showing a Mach-Zehnder modulator according to Embodiment 3. FIG. In the present embodiment, a p-type contact layer 19 having a bandgap energy smaller than that of the p-type cladding layer 13 is formed on the p-type cladding layer 13 of the second embodiment. Metal electrode 18 is in contact with p-type contact layer 19 . The p-type contact layer 19 with a small bandgap energy has a smaller contact resistance with the metal electrode 18 than the p-type cladding layer 13 . As a result, the resistance reduction effect of the metal electrode 18 can be sufficiently obtained while maintaining the thickness of the p-type semiconductor layer between the arms. Therefore, it is expected that the 3 dB band of electrical-to-optical conversion can be improved to 100 GHz or more. Other configurations and effects are the same as those of the second embodiment.

1 Mach-Zehnder modulator, 4a, 4b arm waveguides, 5a, 5b traveling wave electrodes, 11 semi-insulating substrate, 12 p-type semiconductor layers, 13a, 13b p-type cladding layers, 14a, 14b waveguide layers, 15a, 15b n type cladding layer, 17a, 17b insulation film, 18 metal electrode, 19 p-type contact layer

Claims

a semi-insulating substrate;
a p-type semiconductor layer formed on the semi-insulating substrate;
first and second arm waveguides formed on the p-type semiconductor layer and separated from each other;
first and second traveling wave electrodes respectively formed on the first and second arm waveguides;
A Mach-Zehnder modulator comprising a metal electrode that is in contact with the p-type semiconductor layer between the first arm waveguide and the second arm waveguide and that is not connected to a DC bias power supply. .
further comprising first and second insulating films covering side surfaces of the first and second arm waveguides, respectively;
2. The Mach-Zehnder according to claim 1, wherein said metal electrode is formed so as to be in contact with said p-type semiconductor layer over the entire region from said first insulating film to said second insulating film. modulator.
Each of the first and second arm waveguides includes a p-type clad layer, a waveguide layer formed on the p-type clad layer, and an n-type clad layer formed on the waveguide layer. and
3. The Mach-Zehnder modulator according to claim 1, wherein said p-type semiconductor layer has a bandgap energy smaller than that of said p-type cladding layer.
The p-type semiconductor layer has a p-type cladding layer,
each of the first and second arm waveguides has a waveguide layer formed on the p-type cladding layer and an n-type cladding layer formed on the waveguide layer;
3. The Mach-Zehnder modulator according to claim 1, wherein said metal electrode is in contact with said p-type cladding layer.
The p-type semiconductor layer has a p-type cladding layer and a p-type contact layer formed on the p-type cladding layer and having a bandgap energy smaller than that of the p-type cladding layer,
each of the first and second arm waveguides has a waveguide layer formed on the p-type cladding layer and an n-type cladding layer formed on the waveguide layer;
3. The Mach-Zehnder modulator according to claim 1, wherein said metal electrode is in contact with said p-type contact layer.
The plane orientation of the surface of the semi-insulating substrate is <100>,
6. The Mach-Zehnder modulator according to claim 1, wherein the extension direction of said first and second arm waveguides is <011>.