JP2017156646A - Semiconductor Mach-Zehnder optical modulator - Google Patents

Abstract

PROBLEM TO BE SOLVED: To broaden a bandwidth and reduce size and power consumption simultaneously, which has been deemed impossible to do just by improving electrode structures as bandwidth broadening, miniaturization, and reduction of power consumption are in a trade-off relationship.SOLUTION: A semiconductor Mach-Zehnder optical modulator has a multiple quantum well structure comprising well layers containing In, Ga, Al, and As, and barrier layers containing In, Al, and As, the well layers of the multiple quantum well structure having tensile strain in a range of approximately 0.1-0.5%. Such an arrangement realizes enhanced modulation efficiency and allows for simultaneously achieving compactness, broadband, and low power consumption.SELECTED DRAWING: Figure 7

Description

  The present invention relates to an InP-based semiconductor Mach-Zehnder optical modulator that is an optical modulator used for optical communication and the like.

  In order to respond to the increasing demand for communication traffic, researches to increase the multiplicity of optical signals are being actively conducted. As a specific method for increasing the multiplicity of optical signals, a four-value phase modulation method (QPSK) that doubles the transmission capacity by assigning a binary value (multiplicity 2) to one symbol, and four values (multi-value) for one symbol. There are known multilevel optical modulation schemes such as a 16-value quadrature amplitude modulation scheme (16QAM) that assigns the severity 4) to quadruple the transmission capacity. A method of doubling the transmission capacity by optical polarization multiplexing is also known.

  When performing such multilevel optical modulation, an I / Q modulator is used as the optical modulator. The I / Q modulator, also called an orthogonal modulator, is a modulator that can independently generate orthogonal optical electric field components (I channel, Q channel), and is connected in parallel with a Mach Zehnder modulator. Has a special configuration.

In recent years, from the Mach-Zehnder optical modulator made of lithium niobate (LN: LiNbO ₃ ) to the miniaturization and lower power consumption of the optical transmitter module, research on semiconductor Mach-Zehnder optical modulators including InP has been carried out. Development is attracting attention.

  Although the semiconductor Mach-Zehnder optical modulator is smaller than the conventional LN modulator and has a feature of low driving voltage, it is further reduced in size, bandwidth, and power consumption according to the trend of the optical transmitter module. Realization is required. In order to reduce the size, increase the bandwidth, and reduce the power consumption, it is known as an effective technique to devise an electrode structure and length (see, for example, Non-Patent Documents 1 and 2).

  FIG. 1 shows a first example of a conventional semiconductor Mach-Zehnder optical modulator. As shown in FIG. 1A, the semiconductor Mach-Zehnder optical modulator includes an optical demultiplexer 102 connected to the input waveguide 101, an optical multiplexer 104 connected to the output waveguide 105, and an optical demultiplexer. A Mach-Zehnder interferometer including two arm waveguides 103a and 103b connecting between the wave multiplier 102 and the optical multiplexer 104 is provided. As shown in FIG. 1B, traveling wave electrodes 111a and 111b are formed above the arm waveguides 103a and 103b, and a coplanar line is formed together with the ground electrodes 112a to 112c. As shown in FIG. 1C, by connecting the modulation signal sources 121a and 121b and the termination resistors 122a and 122b to the traveling wave electrodes 111a and 111b, respectively, and applying a high frequency signal as a drive voltage, The input optical signal can be modulated.

  Each optical waveguide uses a multiple quantum well layer as a core layer through which light propagates. The quantum multiple well layer includes an unstrained well layer and an unstrained barrier layer. For example, although the element size of Non-Patent Document 1 is a size of 7.5 mm × 1.3 mm and a wide band of 80 Gb / s is realized at a driving voltage of 3 V, it is smaller and driven at a lower voltage (for example, 2 V or less). ) Is required.

  For example, downsizing and widening of the band can be realized by shortening the electrode length of the phase modulation unit, but shortening the electrode length means that the region contributing to phase modulation is shortened. However, the driving voltage becomes high. Conversely, a low drive voltage can be realized by increasing the electrode length of the phase modulation section, but the wiring loss of the electrode increases, so the band is degraded and the element size is increased. .

  FIG. 2 shows a second example of a conventional semiconductor Mach-Zehnder optical modulator. As a technique for widening the bandwidth, a capacity-loaded structure capable of reducing electrode loss is known (for example, see Non-Patent Document 3). The semiconductor Mach-Zehnder optical modulator includes a Y-branch 202 connected to the input waveguide 201, an MMI (Multi-Mode Interferometer) 204 connected to the output waveguides 205 a and 205 b, and the Y-branch 202 and the MMI 204. It has a Mach-Zehnder interferometer composed of two arm waveguides 203a and 203b that connect them. Traveling wave electrodes 211a and 211b are formed along the arm waveguides 203a and 203b, and a plurality of small electrodes branched from the traveling wave electrodes are provided on the arm waveguide. By connecting a modulation signal source 221 and a termination resistor 222 to each of the traveling wave electrodes 211a and 211b, the device operates as a capacitively loaded semiconductor Mach-Zehnder optical modulator.

  In the second example, a portion that does not contribute to the phase modulation region is generated on the arm waveguide, so that the electrode length is increased and the element size is increased as compared with the structure of the first example.

JP-A-8-064904

N. Kikuchi et al., “80 Gb / s low driving-voltage InP DQPSK modulator with an n-p-i-n structure,” IEEE Photonics Tecl1nol.Lett., Vol. 21, no. 12, pp. 787-789, 2009 Naoya Kono et al., “Compact and Low Power DP-QPSK Modulator Module with InP-Based Modulator and Driver ICs,” OFC2013, OW1G.2 H.N.Klein et al., “1.55μm Mach-Zehnder Modulators on InP for optical 40/80 Gbit / s transmission networks,” OFC2006, pp. 171-173

  As described above, widening, downsizing, and low power consumption are in a trade-off relationship with each other. Therefore, all of the downsizing, widening, and low power consumption can be achieved at the same time by using only the electrical structure of the electrode structure. It was impossible to realize.

  An object of the present invention is to optimize a multi-quantum well (MQW) structure that forms the core of a semiconductor Mach-Zehnder optical modulator, and to increase the refractive index variation due to the quantum confined Stark effect (QCSE). It is an object of the present invention to provide a semiconductor Mach-Zehnder optical modulator that simultaneously realizes a reduction in size, a wider band, and a lower drive voltage by improving modulation efficiency.

  In order to achieve such an object, the present invention provides an optical demultiplexer connected to an input waveguide, an optical multiplexer connected to an output waveguide, and the optical demultiplexer. And a Mach-Zehnder interferometer comprising two arm waveguides connecting between the optical multiplexer and the optical multiplexer, and applying the drive voltage to the electrodes formed on the arm waveguides, the input light A semiconductor Mach-Zehnder optical modulator for modulating a signal, wherein a core layer of the arm waveguide is made of In, Ga, Al, and As, a well layer (343) having tensile strain, In, Al, and As And having a multiple quantum well structure including a barrier layer (344) made of

  As described above, according to the present invention, by improving the modulation efficiency, the semiconductor Mach-Zehnder optical modulator can be downsized, widened, and driven at the same time.

It is a figure which shows the 1st example of the conventional semiconductor Mach-Zehnder optical modulator. It is a figure which shows the 2nd example of the conventional semiconductor Mach-Zehnder optical modulator. It is a figure for demonstrating the lattice matching of the multiple quantum well structure on a board | substrate. It is a figure which shows the band structure in a distortion material. 1 is a cross-sectional view of a semiconductor Mach-Zehnder optical modulator according to a first embodiment of the present invention. It is a figure which shows the manufacturing method of the semiconductor Mach-Zehnder optical modulator concerning 1st Embodiment. It is a figure which shows the multiple quantum well structure of the semiconductor Mach-Zehnder optical modulator concerning 1st Embodiment. It is a figure for demonstrating the change of the light absorption by distortion. It is a figure which shows the absorption loss change (V- (DELTA) (alpha)) characteristic with respect to an applied voltage, and the phase change (V- (DELTA) (phi)) characteristic with respect to an applied voltage. It is a figure which shows (DELTA) phi / deltaalpha with respect to the amount of tensile strain. It is sectional drawing of the semiconductor Mach-Zehnder optical modulator concerning the 2nd Embodiment of this invention. It is a figure which shows the multiple quantum well structure of the semiconductor Mach-Zehnder optical modulator concerning 2nd Embodiment.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the semiconductor Mach-Zehnder optical modulator of this embodiment, tensile strain (elongation strain) in a well layer (well) having a multiple quantum well (MQW) structure is used (Patent Document 1).

A semiconductor element such as a semiconductor Mach-Zehnder optical modulator is formed on a semiconductor substrate such as InP. When the lattice constant of the semiconductor substrate is a _sub and the lattice constant of the strained material (well layer here) is a _strain , the strain amount ε is given by the following equation (1).
ε = (a _strain −a _sub ) / a _sub (1)

  With reference to FIG. 3, the lattice matching of the multiple quantum well structure on the substrate will be described. When the amount of lattice matching distortion in the in-plane direction is the horizontal right direction on the plane of the paper, and the amount of lattice matching distortion is in the film thickness direction on the plane of the paper, the lattice matching is correct as shown in FIG. System), there is no strain, and the strain amount ε is zero (ε = 0). Further, in the case of an extension strain system as shown in FIG. 3A, the strain amount ε is smaller than zero by compressing in the film thickness direction as much as extending in the in-plane direction between the substrate and the strain material (ε <0). Further, as shown in FIG. 3C, in the case of a compression strain system, the amount of strain ε is less than zero as a result of the compression in the in-plane direction between the substrate and the strain material and the expansion in the film thickness direction. Larger (ε> 0).

FIG. 4 shows a band structure in the strained material. The upper part is a conduction band, the lower part is a valence band, E _hh is a heavy hole (heavy hole band), E _lh is a light hole (light hole band), and E _so is a spin orbit. K _XY represents a valence band in the in-plane direction of the well, and K _Z represents a valence band in the thickness direction of the well.

The tensile strain and the compressive strain are used for a semiconductor laser or a semiconductor laser type optical amplifying element. As shown in FIG. 4, in the first active layer having compressive strain, the energy E _{hh of the} heavy hole is in the conduction band side above the energy E _{lh of the} light hole in both the in-plane direction and the film thickness direction. The gain for TE polarized input light is larger than that of the matching system. On the other hand, in the second active layer having a tensile strain, the energy E _{hh of the} heavy hole is on the valence band side lower than the energy E _{lh of the} light hole in both the in-plane direction and the film thickness direction. In comparison, the gain for the TM polarized input light is increased.

(First embodiment)
FIG. 5 is a cross-sectional view of the semiconductor Mach-Zehnder optical modulator according to the first embodiment of the present invention. The configuration of the semiconductor Mach-Zehnder optical modulator of the first embodiment is the same as the configuration shown in FIG. 1, and FIG. 5 shows a cross section of one arm waveguide at VV ′ in FIG. The semiconductor layer of the arm waveguide includes an n-type contact layer 302, an n-type cladding layer 303 (first conductive semiconductor cladding layer), a non-doped cladding / core layer 304, and a p-type cladding layer on a semi-insulating InP substrate 301. It has a high-mesa waveguide structure in which 305 (second conductive semiconductor clad layer) and p-type contact layer 306 are stacked in this order. Dielectrics 113 a and 113 b are formed on both side surfaces of the waveguide structure, and the progress is formed on the ground electrodes 112 a and 112 b formed on the n-type contact layer 302 and the p-type contact layer 306. A high-frequency electrode is formed by the wave electrode 111.

The n-type contact layer 302 is InGaAsP with a carrier concentration of 5 × 10 ¹⁸ cm ⁻³ , and the n-type cladding layer 303 is InP with a carrier concentration of 1 × 10 ¹⁸ cm ⁻³ . The carrier concentration of the p-type cladding layer 305 is 5 × 10 ¹⁷ cm ^{−3 to} 1 × 10 ¹⁸ cm ⁻³ of InP in view of the light absorption coefficient and the electrical resistivity. The p-type contact layer 306 uses InGaAs having a carrier concentration of 5 × 10 ¹⁸ cm ⁻³ or more in order to reduce contact resistance.

  FIG. 6 shows a method for manufacturing the semiconductor Mach-Zehnder optical modulator according to the first embodiment. On the semi-insulating InP substrate 301, an n-type contact layer 302, an n-type cladding layer 303, a non-doped cladding / core layer 304, a p-type cladding layer 305, and a p-type contact layer 306 are sequentially grown as semiconductor layers ( FIG. 6 (a)). Next, the optical waveguide having a high mesa structure is formed by processing to a depth that leaves a part of the n-type cladding layer 303 by a semiconductor processing technique (FIG. 6B).

  Part of the n-type cladding layer 303 is removed, and ground electrodes 112a and 112b are deposited (FIG. 6C). Thereafter, the dielectrics 113a and 113b are applied and processed (FIG. 6D), and the traveling wave electrode 111 and the ground electrodes 112a and 112b are formed by plating or vapor deposition (FIG. 6E). The height of the high mesa structure and the height of the dielectrics 113a and 113b are desirably equal as shown in FIG.

  FIG. 7 shows a multiple quantum well structure of the semiconductor Mach-Zehnder optical modulator according to the first embodiment. The non-doped clad core layer 304 has a non-doped core layer (non-doped semiconductor core layer) between non-doped clad layers 341 and 342 (first and second non-doped semiconductor clad layers) made of InP and having the same thickness. The non-doped core layer has a multiple quantum well structure in which a plurality of tensile strained InGaAlAs well layers 343 and unstrained InAlAs barrier layers 344 are alternately stacked.

  The band gap wavelength of the non-doped core layer has little absorption at the operating light wavelength, and the QCSE (Quantum Confined Stark Effect) is effectively operated with high efficiency. Therefore, when the wavelength of the input light is in the 1.55 μm band In this case, the band gap wavelength of the core layer is set to 1.35 to 1.45 μm. The difference between the input wavelength and the band gap wavelength of the core layer is referred to as a detuning amount, which is 100 to 200 nm in the case of a Mach-Zehnder optical modulator.

  In the multiple quantum well of the InP-based semiconductor Mach-Zehnder optical modulator, it is ideal that the phase change in the Mach-Zehnder interferometer is increased. Therefore, it is preferable that the absorption is as small as possible and the refractive index change due to QCSE is large. On the other hand, in an electroabsorption (EA) modulator having a multiple quantum well structure, the absorption technique is increased and the chirp parameter is regarded as important (detuning amount is 50 nm or less), so the design method of the multiple quantum well is different. . That is, in the design of the InP-based semiconductor Mach-Zehnder optical modulator, the increase in loss due to bias is small in the V-Δα (change in absorption loss with respect to applied voltage) characteristic, and the bias in the V-Δφ (phase change with respect to applied voltage) characteristic. A larger amount of phase modulation is better.

  For the multiple quantum well structure of the InP-based semiconductor Mach-Zehnder optical modulator, a P-based semiconductor material (InGaAsP-based material) and an Al-based semiconductor material (InGaAlAs-based material) can be used. The Al-based semiconductor material has a larger conduction band offset value (ΔEc), a sharp absorption spectrum, and a higher modulation efficiency when a quantum well is produced than the P-based semiconductor material. Therefore, in order to realize a highly efficient semiconductor Mach-Zehnder optical modulator, it is necessary to use an Al-based semiconductor material.

  With reference to FIG. 8, the change of the light absorption by distortion is demonstrated. As shown in FIG. 8B, when an optimal tensile strain is applied to the InGaAlAs well layer, the separation between the light hole (LH) and the heavy hole (HH) is reduced, and the absorption spectrum can be made steeper. Thereby, the refractive index change by QCSE can be enlarged. When the amount of distortion is smaller than the optimum amount of distortion (FIG. 8 (a)) or larger than the amount of optimum distortion (FIG. 8 (c)), the separation of the light hole and the heavy hole is large, and the refractive index change due to QCSE is optimum. It becomes smaller than the case of a strain amount of a large tensile strain.

  FIG. 9 shows the absorption loss change (V−Δα) characteristic with respect to the applied voltage and the phase change (V−Δφ) characteristic with respect to the applied voltage. From V-Δα shown in FIG. 9 (a), when tensile strain is applied, the absorption loss at the same bias voltage is reduced. From V−Δφ shown in FIG. 9B, by applying a tensile strain, the amount of phase change at the same bias becomes the maximum when a desired tensile strain is applied. On the other hand, if the amount of tensile strain is excessively added, the amount of phase change is significantly reduced. As shown in FIG. 9C, when the phase change (Δα−Δφ) with respect to the absorption loss change is obtained, the amount of phase change at the same loss is large and high efficiency is obtained by applying a tensile strain in a desired range. You can see that

FIG. 10 shows Δφ / Δα with respect to the amount of tensile strain, and the larger Δφ / Δα, the higher the efficiency. Here, if the allowable efficiency is the maximum value of 0.5 × (1 / e), the required strain amount of the tensile strain is 0.1% to 0.5%. The thickness of one layer of the InGaAlAs well layer is preferably about 5.0 nm to 15.0 nm in order to make the absorption spectrum steep, and from FIG. 10, the tensile strain of about 0.1% to 0.5% is obtained. In terms of strain and composition, it is desirable that In _1-xy Ga _x Al _y As (0.35 ≦ x ≦ 0.53, 0.01 ≦ y ≦ 0.12). The thicker the well layer, the larger the optimum tensile strain amount, and the smaller the required tensile strain amount, the thinner. In addition, when the tensile strain amount is too large or too small than the optimum value, the modulation efficiency is deteriorated.

  Further, the thickness of one layer of the InAlAs barrier layer is preferably about 5.0 nm to 10.0 nm, and the film thickness of the multiple quantum well structure including the InGaAlAs well layer and the InAlAs barrier layer constitutes a Mach-Zehnder interferometer. The high mesa waveguide to be used is preferably about 0.4 μm to 0.75 μm.

  As described above, the higher the ΔEc, the higher the efficiency of the semiconductor Mach-Zehnder optical modulator that can be realized. Therefore, the InAlAs barrier layer is preferably unstrained. This is because ΔEc becomes small when compressive strain opposite to that of the InGaAlAs well layer is applied to the InAlAs barrier layer in order to obtain a strain-compensated multiple quantum well structure.

  As described above, ideally, the most efficient semiconductor Mach-Zehnder optical modulator can be realized because the InAlAs barrier layer is unstrained. On the other hand, although the modulation efficiency is lowered, compressive strain can be applied to the InAlAs barrier layer in order to increase the reliability and quality in epitaxial growth. In this case, compression distortion that completely realizes distortion compensation may be added, but it is desirable that the amount of change in ΔEc is small and the compression distortion is such that distortion can be compensated to some extent.

ΔEc is about 400 meV when both the well layer and the barrier layer are unstrained. In view of the modulation efficiency, the Mach-Zehnder optical modulator requires about 200 meV, which is about half of this. Therefore, it is desirable that the InAlAs barrier layer be an In _1-x Al _x As layer (0.32 ≦ x ≦ 0.48) in terms of composition with a compressive strain of about 0% to 1.0%.

  According to the present embodiment, the modulation efficiency can be increased as compared with the conventional semiconductor Mach-Zehnder optical modulator. For example, assuming that 1.5 times higher efficiency can be realized, the driving voltage can be reduced to 2/3 as compared with the conventional case if the electrode length is the same. If the driving voltage is the same, the electrode length can be reduced to 2/3 compared to the conventional case, and a reduction in size and a wider band can be realized. In addition, the effect of 1.5 times higher efficiency is realized by allocating the drive voltage, downsizing, and wide band effectively by designing the electrode length, thereby realizing wide band and low drive voltage at the same time. can do.

  Note that one of the upper cladding layer and the lower cladding layer may be an n-type semiconductor and the other may be a p-type semiconductor. On the other hand, both the upper clad layer and the lower clad layer are n-type semiconductors, and a third p-type clad layer is inserted between the upper clad layer and the core layer, or between the lower clad layer and the core layer. It can also be.

  In the first embodiment, the configuration of the semiconductor Mach-Zehnder optical modulator shown in FIG. 1 is equivalent to the configuration of the semiconductor Mach-Zehnder optical modulator shown in FIG. There may be.

(Second Embodiment)
FIG. 11 is a cross-sectional view of a semiconductor Mach-Zehnder optical modulator according to the second embodiment of the present invention. The configuration of the semiconductor Mach-Zehnder optical modulator of the second embodiment is the same as the configuration shown in FIG. 1, and FIG. 11 shows a cross section of one arm waveguide at VV ′ in FIG. The semiconductor layer of the arm waveguide is composed of an n-type contact layer 402, an n-type cladding layer 403, an undoped cladding / core layer 404, and a p-type carrier block layer 405 (third conductive semiconductor cladding) on a semi-insulating InP substrate 401. Layer), an n-type cladding layer 406, and an n-type contact layer 407 are stacked in this order. Dielectrics 113a and 113b are formed on both side surfaces of the waveguide structure, and the progress formed on the ground electrodes 112a and 112b formed on the n-type contact layer 402 and the p-type contact layer 407. A high-frequency electrode is formed by the wave electrode 111.

The n-type contact layer 402 is InGaAsP with a carrier concentration of 5 × 10 ¹⁸ cm ⁻³ , and the n-type cladding layer 403 is InP with a carrier concentration of 1 × 10 ¹⁸ cm ⁻³ . In the p-type carrier stop layer 405, InAlAs having a large band gap is doped with 1 × 10 ¹⁸ cm ⁻³ hole carriers so as to provide a sufficient barrier against electrons. The n-type cladding layer 406 is made of InP with a carrier concentration of 1 × 10 ¹⁸ cm ⁻³ , and the n-type contact layer 407 is made of InGaAs with a carrier concentration of 5 × 10 ¹⁸ cm ⁻³ .

  FIG. 12 shows a multiple quantum well structure of a semiconductor Mach-Zehnder optical modulator according to the second embodiment. The non-doped clad core layer 404 has a non-doped core layer between non-doped clad layers 441 and 442 made of InP and having the same thickness. The non-doped core layer has a multiple quantum well structure in which a plurality of tensile strained InGaAlAs well layers 443 and unstrained InAlAs barrier layers 444 are alternately stacked.

  The band gap wavelength of the non-doped core layer is QCSE that effectively operates at the operating light wavelength. Therefore, in the case of an input wavelength of 1.55 μm, the band gap wavelength of the core layer is 1.35 to 1.45 μm. To do. The configuration of the multiple quantum well structure is the same as that of the first embodiment described above.

  In the second embodiment, the configuration is the same as that of the semiconductor Mach-Zehnder optical modulator shown in FIG. 1 as an example, but the capacitance-loading configuration shown in FIG. There may be.

101, 201 Input waveguide 102 Optical demultiplexer 103, 203 Arm waveguide 104 Optical multiplexer 105, 205 Output waveguide 111, 211 Traveling wave electrode 112, 212 Ground electrode 121, 221 Signal source 122, 222 Termination resistance 202 Y-branch 204 MMI

Claims (10)

An optical demultiplexer connected to the input waveguide, an optical multiplexer connected to the output waveguide, and the two arm waveguides connecting the optical demultiplexer and the optical multiplexer A semiconductor Mach-Zehnder optical modulator that modulates an input optical signal by applying a driving voltage to an electrode formed on the arm waveguide,
The core layer of the arm waveguide is
A well layer composed of In, Ga, Al, and As and having tensile strain;
A semiconductor Mach-Zehnder optical modulator having a multiple quantum well structure including a barrier layer made of In, Al, and As.   2. The semiconductor Mach-Zehnder optical modulator according to claim 1, wherein a strain amount of the tensile strain is 0.1% or more and 0.5% or less. The well layer _{is, In 1-xy Ga x Al} y As (0.35 ≦ x ≦ 0.53,0.01 ≦ y ≦ 0.12) as claimed in claim 1 or 2, characterized in that it is Semiconductor Mach-Zehnder optical modulator. 4. The semiconductor Mach-Zehnder optical modulator according to claim 1, wherein the barrier layer is an In _1-x Al _x As layer (0.32 ≦ x ≦ 0.48).   5. The semiconductor Mach-Zehnder optical modulator according to claim 1, wherein a thickness of the well layer is 5.0 nm or more and 15.0 nm or less.   6. The semiconductor Mach-Zehnder optical modulator according to claim 1, wherein a thickness of the barrier layer is 5.0 nm or more and 10.0 nm or less.   The thickness of the multiple quantum well structure is 0.4 μm or more and 0.75 μm or less, and a band gap wavelength is 1.35 μm or more and 1.45 μm or less. Semiconductor Mach-Zehnder optical modulator. The arm waveguide is on a semiconductor substrate,
A first conductive semiconductor cladding layer;
A first non-doped semiconductor cladding layer;
A non-doped semiconductor core layer;
A second non-doped semiconductor cladding layer of the same material and thickness as the first non-doped semiconductor cladding layer;
8. The semiconductor Mach-Zehnder optical modulator according to claim 1, wherein the second conductive semiconductor clad layer is sequentially laminated.   The one of the first conductive semiconductor clad layer and the second conductive semiconductor clad layer is an n-type semiconductor and the other is a p-type semiconductor. Semiconductor Mach-Zehnder optical modulator.   Both the first conductive semiconductor clad layer and the second conductive semiconductor clad layer are n-type semiconductors, the non-doped semiconductor core layer, the first conductive semiconductor clad layer, or the second conductive semiconductor clad layer. 10. The semiconductor Mach-Zehnder optical modulator according to claim 8, wherein a p-type third conductive semiconductor cladding layer is inserted between at least one of the conductive semiconductor cladding layers.

Images