Semiconductor optical modulator

Abstract

The present invention provides a semiconductor optical modulator capable of dealing with a broader operating wavelength band than conventional semiconductor optical modulators. A semiconductor optical modulator according to the present disclosure comprises an optical waveguide and a phase modulation unit that modulates the phase of light propagating through the optical waveguide, the optical waveguide being structured so as to have sequentially stacked on a semiconductor substrate: a lower cladding layer formed from at least one layer of an n-type semiconductor or a p-type semiconductor; a core layer formed from a non-doped semiconductor including an MQW layer; and an upper cladding layer formed from at least one layer of an n-type semiconductor or a p-type semiconductor. The semiconductor optical modulator is characterized in that: the phase modulation unit comprises a heater mechanism for heating the optical waveguide and an RF electrode connected to the optical waveguide; the optical waveguide has a ridge structure; and the heater mechanism is formed on the core layer of the optical waveguide.

Description

Technical Field

[0001] The present disclosure relates to a semiconductor optical modulator.

Background Art

[0002] In recent optical fiber communication systems, backbone network transmission technologies of 100 Gbps per wavelength have been established through the introduction of digital signal processing technologies including digital coherent. Currently, the speed of this backbone network transmission technology is being increased to 400-600 Gbps per wavelength, and is reaching the level of practical application.

[0003] Communication systems exceeding 400 Gbps require broader bandwidth for analog components (for example, a modulation bandwidth of 40 GHz or more). For the purpose of reducing Radio Frequency (hereinafter also referred to as RF) loss and achieving miniaturization, a configuration in which an RF driver (driver IC) and an optical modulator are mounted in an integrated package on the transmission side is attracting attention. A transmitter using such a mounting configuration has been standardized in the OIF (The Optical Internetworking Forum) under the name of High-Bandwidth Coherent Driver Modulator (HB-CDM: high-speed driver integrated optical modulator) (see, for example, Non-Patent Document 1). Further, in HB-CDM, an indium phosphide (InP) modulator is mainly used (see, for example, Non-Patent Document 2).

[0004] In recent years, in optical transmission devices, semiconductor-based optical modulators have attracted attention from the viewpoint of device miniaturization and cost reduction, in place of conventional lithium niobate (LN) optical modulators. In particular, compound semiconductors represented by InP are mainly used for higher-speed modulation operations, and research and development are concentrated on silicon (Si)-based optical devices in systems where further miniaturization and cost reduction are emphasized.

[0005] Semiconductor optical modulators also have material-specific advantages and disadvantages. For example, while InP modulators excel in high-speed modulation, temperature control of the optical modulator chip is considered essential during modulation to control band-edge absorption effects. On the other hand, LN modulators and Si modulators have the advantage of not requiring temperature control, making them advantageous for low power consumption.

[0006] When modularizing an InP modulator, the optical modulator chip needs to be mounted on a Peltier element for temperature control, which can contribute to increased power consumption. In addition, the placement of the Peltier element prevents miniaturization of the package size. Therefore, if the Peltier element can be eliminated or its area can be reduced, it will be possible to realize a high-speed, low-power, and low-profile optical module equipped with an InP modulator that excels in high-speed modulation operation.

[0007] Furthermore, because InP modulators utilize band-edge absorption effects, it is essential to modify the epitaxial structure (controlling absorption by PL wavelength, etc.) for C-band and L-band applications, which presents the challenge of sharing chips between the C-band and L-band. [Prior art documents] [Non-patent literature]

[0008] [Non-Patent Document 1] Implementation Agreement for the High Bandwidth Coherent Driver Modulator (HB-CDM) , OIF-HB-CDM-02.0 (2021)https: / / www.oiforum.com / wp-content / uploads / OIF-HB-CDM-02.0.pdf [Non-Patent Document 2] J. Ozaki, Y. Ogiso, Y. Hashizume, H. Yamazaki, K. Nagashima and M. Ishikawa, "Coherent Driver Modulator With Flexible Printed Circuit RF Interface for 128-Gbaud Operations", IEEE Photonics Technology Letters, vol. 34, no. 23, pp. 1289-1292 (2022) doi: 10.1109 / LPT.2022.3212678. [Overview of the Initiative]

[0009] This disclosure has been made in view of the above-mentioned problems, and its purpose is to provide a semiconductor optical modulator that can handle a wider operating wavelength band (e.g., C+L band) than conventional semiconductor optical modulators.

[0010] To address the above-mentioned problems, this disclosure provides a semiconductor optical modulator comprising: an optical waveguide having a structure in which a lower cladding layer formed of at least one n-type or p-type semiconductor, a core layer formed of an undoped semiconductor including an MQW layer, and an upper cladding layer formed of at least one n-type or p-type semiconductor are sequentially stacked on a semiconductor substrate; and a phase modulation unit that performs phase modulation of light propagating through the optical waveguide, wherein the phase modulation unit comprises a heater mechanism for heating the optical waveguide and an RF electrode connected to the optical waveguide, the optical waveguide has a ridge structure, and the heater mechanism is formed on the core layer of the optical waveguide. [Brief explanation of the drawing]

[0011] [Figure 1] This figure illustrates the structure of an InP modulator 100 according to the first embodiment of the present disclosure, where (a) is a top view, (b) is a cross-sectional view along the Ib-Ib cross-section, and (c) is a cross-sectional view along the Ic-Ic cross-section. [Figure 2]As another embodiment of the InP modulator 100 according to the first embodiment of this disclosure, the following are cross-sectional views in the width direction showing a different arrangement of the heater mechanism 108: (a) is a cross-sectional view along the Ib-Ib cross-section in Figure 1 when the InP optical waveguide 102 has a ridge structure; (b) is a cross-sectional view along the Ic-Ic cross-section in Figure 1 when the InP optical waveguide 102 has a ridge structure; (c) is a cross-sectional view along the Ib-Ib cross-section in Figure 1 when the InP optical waveguide 102 has a high mesa structure; and (d) is a cross-sectional view along the Ic-Ic cross-section in Figure 1 when the InP optical waveguide 102 has a high mesa structure. [Figure 3] The following are widthwise cross-sectional views showing another embodiment of the InP modulator 100 according to the first embodiment of the present disclosure, where (a) shows a cross-sectional view along the Ib-Ib cross-sectional line in Figure 1, and (b) shows a cross-sectional view along the Ic-Ic cross-sectional line in Figure 1. [Figure 4] The following are widthwise cross-sectional views showing another embodiment of the InP modulator 100 according to the first embodiment of the present disclosure, where (a) shows a cross-sectional view along the Ib-Ib cross-sectional line in Figure 1, and (b) shows a cross-sectional view along the Ic-Ic cross-sectional line in Figure 1. [Figure 5] Figure 1 shows another cross-sectional view along the Ib-Ib line, illustrating a different embodiment of the InP modulator 100 according to the first embodiment of the present disclosure, where (a) shows a configuration in which the heater mechanism 108 is installed on the n-InP 102a, and (b) shows a configuration in which one end of the heater mechanism 108 is on the n-InP 102a and the other end is on the substrate 110. [Figure 6] This is a longitudinal cross-sectional view illustrating the structure of an InP modulation device 600 according to a second embodiment of the present disclosure, where (a) shows a structure without the carrier 603 and (b) shows a structure including the carrier 603. [Modes for carrying out the invention]

[0012] Various embodiments of this disclosure are described below in detail with reference to the drawings. Identical or similar reference numerals indicate identical or similar elements, and redundant descriptions may be omitted. Materials and numerical values ​​are illustrative and are not intended to limit the technical scope of this disclosure. The following description is illustrative and some configurations may be omitted or modified, or implemented with additional configurations, without departing from the gist of one embodiment of this disclosure.

[0013] (First embodiment) A first embodiment of this disclosure will be described in detail below with reference to the drawings. A single Mach-Zehnder modulator is shown below as an example, but multiple such modulators may be arranged in a sequence, such as an IQ modulator or a Twin-IQ modulator. Furthermore, an InP modulator is used as an example in the description below, but semiconductor optical modulators of other materials such as GaAs may also be used.

[0014] Figure 1 illustrates the structure of an InP modulator 100 according to a first embodiment of the present disclosure, where (a) is a top view, (b) is a cross-sectional view along the Ib-Ib cross-section, and (c) is a cross-sectional view along the Ic-Ic cross-section. As shown in Figure 1, the InP modulator 100 includes a 1-input, 2-output (1×2) multimode interferometer (MMI) 101, an InP optical waveguide 102 through which light branched by the MMI 101 propagates, a phase modulation unit 103 that performs phase modulation of the propagating light, a phase adjustment electrode 104 installed on the output side of the phase modulation unit 103 to adjust the phase of the phase-modulated light, and a 2×1 MMI 105 installed on the output side of the phase adjustment electrode 104 that combines the branched light and outputs it to the outside. As shown in Figure 1, the InP modulator 100 is a Mach-Zehnder type optical modulator.

[0015] As shown in Figures 1(a) and (b), the phase modulation unit 103 further includes a substrate 110, an optical waveguide 102 formed on the substrate 110, an RF phase modulation electrode 106 that functions as an electrode for the applied RF, an RF electrode 107 which is an electrode for applying the high frequency supplied from the RF phase modulation electrode 106 to the optical waveguide 102, a heater mechanism 108 for heating the optical waveguide 102, and a dielectric layer 111 formed on the substrate 110 and covering the periphery of the optical waveguide 102 and the heater mechanism 108. The heater mechanism 108 is connected to a metal wiring 112, which is aggregated into one at PADs 109a and 109b and connected to an external power supply and GND. Although Figure 1 illustrates a configuration including the metal wiring 112, the portion of the metal wiring 112 may be the same metal wiring as the heater mechanism 108, or it may be a completely separate metal wiring. Furthermore, although Figure 1 shows the heater mechanism 108 as being thicker and the metal wiring 112 as being thinner, it is also possible for the heater mechanism 108 to be thinner and the metal wiring 112 to be thicker.

[0016] The optical waveguide 102 has a nipn layer structure in which a first cladding layer 102a formed of at least one n-type semiconductor layer formed on a substrate, a second cladding layer 102b formed of at least one p-type semiconductor layer formed on 102a, an undoped semiconductor core layer 102c formed on 102b, and a third cladding layer 102d formed of at least one n-type semiconductor layer formed on the semiconductor core layer 102c are stacked. The semiconductor core layer 102c includes a multi-quantum well (MQW) layer using, for example, an undoped material system such as InP, InGaAsP, or InGaAlAs. The bandgap wavelength of the MQW layer is arbitrarily set within a range in which the electro-optic effect is effective at the optical wavelength used and optical absorption is not a problem. Here, the semiconductor core layer 102c functions as a core through which light propagates. Such a nipn layer structure allows for a thinning of the p-layer, which has a significant impact on high-frequency loss, and enables the contact surface with the metal RF electrode 107 to be the n-layer. Therefore, it is possible to significantly reduce high-frequency loss compared to the conventional pin layer structure. However, this is for illustrative purposes only, and the optical waveguide 102 may also have a pin layer structure.

[0017] The RF phase modulation electrode 106 has a capacitive loading structure and includes a main line section 106a and T-shaped sections 106b that are periodically branched from the main line section 106a for applying a modulation signal to the InP optical waveguide 102. The InP modulator 100 is called a capacitive loading type because the T-shaped sections 106b have a structure that adds capacitance. In Figure 1, for simplification, only three T-shaped sections 106b are shown, but this is for illustrative purposes only, and more than three T-shaped sections 106b may be arranged, or fewer. Furthermore, although this embodiment shows a capacitive loading structure as an example, this patented structure is also effective for structures other than capacitive loading structures.

[0018] The heater mechanism 108 supplies heat to the MQW, which contributes to phase modulation and constitutes the undoped semiconductor core layer 102c of the InP modulator 100. This configuration allows for temperature control of the MQW without the use of a Peltier element. By controlling the temperature of the MQW with the heater mechanism 108, it is possible to appropriately adjust the amount of optical absorption and refractive index effect based on the quantum confined Stark effect (QCSE) of the MQW. For example, as the temperature of the MQW rises, the band edge shifts to the longer wave side, making it suitable for operation at longer wavelengths. The heater mechanism 108 is usually made of metal. Therefore, if the heater mechanism 108 is placed directly below the main transmission line 106a in the thickness direction (z direction), it may affect the high-frequency electromagnetic field distribution of the RF phase modulation electrode 106 and the RF electrode 107, and as a result, the high-frequency characteristics of the InP modulator 100 may deteriorate. Therefore, it is desirable that the heater mechanism 108 be positioned at a distance from the RF phase modulation electrode 106 and RF electrode 107 such that it does not interfere with them (does not affect the electromagnetic field distribution of each high frequency) (for example, as shown in Figures 1(a) and (b), so as to be positioned between the main line section 106a and the RF electrode 107 in the width direction (y direction)), and parallel to the longitudinal direction (x direction) of the RF phase modulation electrode 106 and RF electrode 107. Furthermore, since the heater mechanism 108 needs to heat all of the MQW layer in the portion that contributes to phase modulation, it is desirable that its length in the x direction be equal to or greater than the length of the RF phase modulation electrode 106 in the x direction. The resistance value of the heater mechanism 108 may be set arbitrarily according to the design. For example, the resistance of the heater mechanism 108 may be set higher only in the peripheral region where the RF electrode 107 connected to the T-junction 106b is installed, compared to other regions. However, since it is necessary for the optical modulator to have the desired amount of heat generated, it is desirable that the resistance value of the heater mechanism 108 be at least 300Ω.

[0019] It is desirable that the heater mechanism 108 be configured to uniformly heat each of the branched arms of the optical waveguide 102. For this reason, the heater mechanism 108 is provided for each of the branched optical waveguides 102. Furthermore, in order to supply equivalent heat to each of the branched optical waveguides 102, the heater mechanism 108 is desirably integrated into one PAD 109a, b and then connected to a power source (voltage source or current source) and GND. For example, when the InP modulator 100 is a Twin-IQ modulator or the like, the configuration has four Mach-Zehnder modulators integrated therein, so eight heater mechanisms 108 are required. When such a Twin-IQ modulator is used in HB-CDM, the number of required PADs and pins may increase, and control may also become difficult. Therefore, it is desirable that all eight heater mechanisms 108 be integrated into the same PAD 109a, b. Note that although the shapes of PAD 109a, b are drawn as circular shapes in FIG. 1, this is for illustrative purposes, and for example, they may be square shapes.

[0020] In the phase modulation section 103, the structure of the optical waveguide 102 is preferably a ridge structure from the viewpoint of directly heating the semiconductor core layer 102c. The structure may also be a high mesa structure, but in that case, the width (length in the y-direction) of the optical waveguide 102 will be processed to be the same width as the third cladding layer 102d, making it impossible to install the heater mechanism 108 on the semiconductor core layer 102c. Therefore, if the optical waveguide 102 has a high mesa structure, the heater mechanism 108 needs to be placed in another location (for example, on the substrate 110) as described later, and consequently the heating efficiency of the semiconductor core layer P102c decreases. On the other hand, if the structure of the optical waveguide 102 is a ridge structure, the heater mechanism 108 can be installed on the semiconductor core layer 102c, making it possible to directly heat the semiconductor core layer 102c. From this viewpoint, it can be said that a ridge structure is more effective than a high mesa structure for the optical waveguide 102. However, even if the heater mechanism 108 is installed in another location (for example, on the substrate 110), it is still possible to heat the semiconductor core layer 102c, so the heater mechanism 108 does not necessarily have to be installed on the semiconductor core layer 102c.

[0021] Figure 2 is a cross-sectional view in the width direction showing a different configuration of the arrangement of the heater mechanism 108 as another form of the InP modulator 100 according to the first embodiment of this disclosure, where (a) is a cross-sectional view along the Ib-Ib section in Figure 1 when the optical waveguide 102 has a ridge structure, (b) is a cross-sectional view along the Ic-Ic section in Figure 1 when the optical waveguide 102 has a ridge structure, (c) is a cross-sectional view along the Ib-Ib section in Figure 1 when the optical waveguide 102 has a high mesa structure, and (d) is a cross-sectional view along the Ic-Ic section in Figure 1 when the optical waveguide 102 has a high mesa structure. As shown in Figure 2, in the InP modulator 100, the heater mechanism 108 may be arranged on the substrate 110. In this configuration, the structure of the optical waveguide 102 may be a ridge structure or a high mesa structure. As described above, in the configuration in which the heater mechanism 108 is placed on the substrate 110, the efficiency in heating the semiconductor core layer 102c including the MQW is reduced compared to the configuration shown in Figure 1. However, this configuration has the advantage that the distance between the heater mechanism 108 and the RF phase modulation electrode 106 and RF electrode 107 is larger than in the configuration shown in Figure 1, thus suppressing degradation of high-frequency characteristics.

[0022] As described above, when the distance between the heater mechanism 108 and the RF phase modulation electrode 106 and RF electrode 107 is short, the heater mechanism 108, which is usually made of metal, normally affects the high-frequency electromagnetic field distribution of the RF phase modulation electrode 106 and RF electrode 107, and as a result, the high-frequency characteristics of the InP modulator 100 may be degraded. From such a point of view, in the InP modulator 100 according to the present disclosure, it is desirable that the heater mechanism 108 and the RF phase modulation electrode 106 are arranged such that the respective distance is as long as possible (separated from each other). For example, in a configuration where the heater mechanism 108 is disposed on the substrate 110 as shown in FIG. 2, the distance in the thickness direction (z-direction) between the RF phase modulation electrode 106 and RF electrode 107 and the heater mechanism 108 can be increased by, for example, making the dielectric layer 111 a thick film. As another example, as shown in FIG. 3, a via 301 may be provided between the T-shaped portion 106b and the RF electrode 107. Further, when the RF electrode 107 is embedded in the dielectric layer 111 as shown in FIG. 3, the RF electrode 107 is formed as a thick film, so that the distance in the thickness direction between the RF phase modulation electrode 106 and the heater mechanism 108 can be made longer. With such a configuration, the via 301 can also be omitted. As still another example, as shown in FIG. 4, a structure may be adopted in which only the region corresponding to the upper portion of the optical waveguide 102 in the dielectric layer 111 is thinned (in other words, only the region where the main line portion 106a is disposed has the dielectric layer 111 thickened). However, in the case of the embodiment as shown in FIG. 4, the distance between the T-shaped portion 106b and the heater mechanism 108 may be partially shortened.

[0023] Furthermore, the degradation of high-frequency characteristics described above may be suppressed by controlling the arrangement or dimensions of the heater mechanism 108. For example, by making the diameter (plane perpendicular to the xy plane) of the heater mechanism 108 smaller, the influence of the heater mechanism 108 on the high-frequency electromagnetic field distribution of the RF phase modulation electrode 106 can be suppressed. The width (length in the y direction) of the heater mechanism 108 is preferably 1 μm or less, and if it can be made 0.1 μm or less, the influence on high-frequency characteristics can be almost ignored. However, if the diameter becomes extremely small, there is a concern that the wire may break during heating, so it is necessary to ensure a diameter that does not cause such breakage. Also, the distance between the main line section 106a and the heater mechanism 108 in the thickness direction (z direction) and width direction (y direction) is preferably at least 5 μm. Considering these points, it is preferable that the diameter of the heater mechanism 108 is such that both the thickness and width are 1 μm or less.

[0024] On the other hand, the heater mechanism 108 needs to generate at least 100°C of heat, considering that a typical InP modulator requires a heat generation of 40-60°C or more, which is controlled by a Peltier element. Therefore, for example, it is desirable that the heater mechanism 108 be configured to have a resistance of at least 300Ω or more. As an additional example, from the viewpoint of efficiently supplying heat to the semiconductor core layer 102c, the third cladding layer 102b may be set to have a higher thermal resistance than the semiconductor cladding layer 102c in order to suppress heat from being released to the bottom of the semiconductor core layer 102c.

[0025] Furthermore, the heater mechanism 108 may be installed on the first cladding layer 102a, as shown in Figure 5. In addition, the heater mechanism 108 may be installed in different locations, for example, one end on the first cladding layer 102a and the other end on the substrate 110.

[0026] In the InP modulator 100 configured in this way according to this embodiment, the operating wavelength range of the InP modulator can be broadened by controlling the amount of heat generated by the heater mechanism 108 with either current or voltage according to the operating wavelength. Generally, InP modulators use the refractive index effect due to the change in band edge absorption called QCSE. Generally, the amount of light absorption of an optical semiconductor is large on the short wave side and small on the long wave side. Therefore, when used in a wide wavelength range, the modulation efficiency decreases on the long wave side, and it was necessary to change the PL wavelength of the MQW layer separately for modulators operating in the C band and modulators operating in the L band (for example, in conventional InP modulators, the PL wavelength for the L band had to be set to the longer wave side compared to the C band). In contrast, the InP modulator 100 according to this embodiment can, for example, increase the amount of heat generated by the heater mechanism 108 when operating on the long-wave side under the same temperature, thereby raising the MQW temperature and shifting the band edge to the long-wave side. Conversely, reduce the amount of heat generated on the short-wave side and lower the MQW temperature, thereby reducing the amount of shift of the band edge to the long-wave side. By adjusting the amount of change in the band edge according to the wavelength, it becomes possible to create an InP modulator that can operate in the C-band and L-band, which were not previously possible. As will be described later, similar effects can also be obtained by changing the temperature of the Peltier element according to the wavelength when the InP modulator is mounted on a Peltier element.

[0027] Furthermore, in the InP modulator 100 according to this embodiment, the temperature of the semiconductor core layer having an MQW layer can be controlled by the heater mechanism 108. Therefore, it is possible to eliminate the need for a Peltier element as in the conventional technology, enabling miniaturization and lower power consumption of the optical module using the InP modulator 100 according to this embodiment. On the other hand, it is also possible to use this structure in combination with a Peltier element (details will be described in the second embodiment below).

[0028] Furthermore, in configurations such as HB-CDM, where the driver is mounted so close to the modulator that it is affected by heat from the modulator's heater mechanism, if the driver IC has a temperature monitoring mechanism, it is possible to calculate the relative value of the i-InP102c(MQW) temperature by utilizing the driver IC's temperature monitoring mechanism. This allows for control such as heating the heater mechanism 108 to a desired temperature using the monitored temperature as the monitoring value. For example, when operating at the same wavelength, if the driver's monitored temperature is high, heating of the heater mechanism 108 can be suppressed, and if the monitored temperature is low, heating of the heater can be promoted.

[0029] Furthermore, when using the InP modulator 100 according to this embodiment in an HB-CDM, it is desirable that the operating temperature range of the phase modulation unit 103 be set to a different range than that of conventional semiconductor optical modulators. Typically, HB-CDMs are required to operate at ambient temperatures of at least -5 to 75°C. On the other hand, the heater mechanism 108 can only heat, not cool. Therefore, considering, for example, an arbitrary wavelength, the temperature of the semiconductor core layer 102c (MQW) needs to be controlled to a constant temperature by the heater mechanism 108 within the ambient temperature range of -5 to 75°C. Specifically, the operating temperature of the InP modulator 100 needs to be controlled by the heater mechanism 108 to be 75°C or higher than the ambient temperature. In conventional semiconductor optical modulators, temperature control is performed by a Peltier element within a range of approximately 40 to 60°C. However, if the appropriate operating temperature of the InP modulator 100 according to this embodiment is set to 40 to 60°C, as in the conventional technology, a problem arises where cooling cannot be performed when the ambient temperature reaches 75°C, and the desired temperature cannot be achieved. Therefore, in order to maintain the temperature of the semiconductor core layer 102c (MQW) within an appropriate range under ambient temperatures of -5 to 75°C during operation of any single wavelength, the InP modulator 100 according to this embodiment must be designed to operate properly at 75°C or higher. Furthermore, considering power consumption and stable operation of the heater mechanism 108, it is desirable to set the appropriate temperature to be within approximately 10°C of the maximum ambient temperature.

[0030] (Second embodiment) A second embodiment of the present disclosure will be described in detail below with reference to the drawings. This embodiment relates to an InP modulation apparatus including the InP modulator 100 and a Peltier element described in the first embodiment.

[0031] Figure 6 is a longitudinal cross-sectional view illustrating the structure of an InP modulation device 600 according to a second embodiment of the present disclosure, where (a) shows a structure without carriers 603 and (b) shows a structure including carriers 603. As shown in Figure 6, the InP modulation device 600 according to this embodiment has a structure in which the InP modulator 100 according to the first embodiment is mounted on a Peltier element 601 and a dielectric substrate 602. Furthermore, as shown in Figure 6(b), the InP modulation device 600 may further include carriers 603 on top of the Peltier element 601 and the dielectric substrate 602, and below the InP modulator 100.

[0032] To achieve the most efficient heat application (minimizing the power consumption of the Peltier element as much as possible), it is desirable that the Peltier element 601 be installed so that only the phase modulation unit 103 is mounted on the Peltier element 601, as shown in Figure 6. For further optimization, it is best to provide the Peltier element only on the lower part of the phase modulation arm. However, this is just one example, and the entire chip may be mounted on the Peltier element 601. In that case, the power consumption will increase compared to a normal modulator because an extra heater mechanism is provided. This is because the power consumption of the Peltier depends on the area of ​​the Peltier element. Therefore, considering the increase in power consumption of the Peltier element due to the heat generated by the heater mechanism, the size of the Peltier element needs to be made very small in an optical modulator with this heater mechanism in order to achieve the same power consumption as conventional devices.

[0033] The dielectric substrate 602 does not necessarily have to be included in the InP modulation device 600, but it is desirable to install it from the viewpoint of securing an area for mounting optical mounting components (not shown), such as lenses. The dielectric substrate 602 may be made of ceramics such as aluminum nitride (AlN) or metal, which have excellent heat dissipation properties.

[0034] From the viewpoint of thermally separating the RF phase modulation electrode 106 and the phase adjustment electrode 104, it is desirable that the carrier 603 further include a thermal separation groove 604. This is because, for example, in InP modulators, the phase adjustment electrode 104 is generally composed of a heater, and in that case, the heat from the heater of the phase adjustment electrode 104 flows into the Peltier element 601, increasing the power consumption of the Peltier element 601. Furthermore, the carrier 603 may be in a form that overhangs from the Peltier element 601, and an additional component (for example, a dielectric substrate 602) may be installed below the carrier 603. In addition, the carrier 603 may have a step (not shown) only in the area where optical mounting members such as lenses are mounted, if necessary.

[0035] In the InP modulation device 600 having such a configuration, since it includes a Peltier element 601, the operating wavelength range can be broadened by controlling the temperature of the Peltier element 601 according to the wavelength, similar to the effect of the heater mechanism 108 of the InP modulator 100. In the InP modulation device 600 according to this embodiment, the Peltier element 601 is installed only in the region corresponding to the RF phase modulation electrode 106. Therefore, compared to the case where the entire InP modulation device 600 is mounted on a Peltier element, the area occupied by the Peltier element 601 can be reduced, and power consumption can be suppressed.

[0036] On the other hand, conventional InP modulators do not switch the temperature of the Peltier element for each wavelength, making the above-described operation highly unlikely. In this case, the Peltier element is operated with a constant set temperature. In this case, the heater mechanism 108, which has a configuration similar to that of the first embodiment, can be used to apply a controlled amount of heat to that specific part, thereby expanding the operating wavelength and enabling C+L band operation, etc. However, if the Peltier element 601 can switch its temperature for each wavelength, it is possible to expand the operating wavelength without using the heater mechanism 108. [Industrial applicability]

[0037] As described above, the semiconductor optical modulator (InP modulator) according to this disclosure can expand the wavelength operating range compared to the prior art. Such a semiconductor optical modulator is expected to be applied to devices for optical fiber communication systems such as HB-CDM, as it can be applied to optical modulators with a wider bandwidth than the prior art.

Claims

1. A semiconductor optical modulator, An optical waveguide having a structure in which a lower cladding layer formed of at least one n-type or p-type semiconductor, a core layer formed of an undoped semiconductor including an MQW layer, and an upper cladding layer formed of at least one n-type or p-type semiconductor are sequentially stacked on a semiconductor substrate, A phase modulation unit that performs phase modulation of light propagating through the optical waveguide, Equipped with, The phase modulation unit is A heater mechanism for heating the optical waveguide, An RF electrode connected to the optical waveguide, Equipped with, The optical waveguide has a ridge structure, The heater mechanism is formed on the core layer of the optical waveguide and is arranged parallel to the longitudinal direction of the RF electrode. The RF phase modulation electrode further comprises a main transmission line and a plurality of T-junctions connecting the main transmission line and the RF electrode, wherein the distances between the main transmission line and the heater mechanism in the thickness and width directions are such that they do not affect the high-frequency electromagnetic field distribution of the RF electrode and the RF phase modulation electrode. Semiconductor optical modulator.

2. A semiconductor optical modulator, An optical waveguide having a structure in which a lower cladding layer formed of at least one n-type or p-type semiconductor, a core layer formed of an undoped semiconductor including an MQW layer, and an upper cladding layer formed of at least one n-type or p-type semiconductor are sequentially stacked on a semiconductor substrate, A phase modulation unit that performs phase modulation of light propagating through the optical waveguide, Equipped with, The phase modulation unit is A heater mechanism for heating the optical waveguide, An RF electrode connected to the optical waveguide, Equipped with, The optical waveguide has a ridge structure or a high mesa structure. The heater mechanism is formed on the semiconductor substrate and is arranged parallel to the longitudinal direction of the RF electrode. The RF phase modulation electrode further comprises a main transmission line and a plurality of T-junctions connecting the main transmission line and the RF electrode, wherein the distances between the main transmission line and the heater mechanism in the thickness and width directions are such that they do not affect the high-frequency electromagnetic field distribution of the RF electrode and the RF phase modulation electrode. Semiconductor optical modulator.

3. The semiconductor optical modulator according to claim 1 or 2, wherein the heater mechanism has a resistance of 300 Ω or more.

4. The semiconductor optical modulator according to claim 1 or 2, characterized in that the distance between the main line section and the heater mechanism in the thickness direction and the width direction is at least 5 μm or more.

5. The optical waveguide is formed on a semi-insulating substrate. A first n-type semiconductor layer, which is composed of at least one n-type semiconductor layer formed on the semi-insulating substrate, A p-type semiconductor layer, which is composed of at least one p-type semiconductor formed on the first n-type semiconductor layer, A core layer formed on the aforementioned p-type semiconductor layer and made of an undoped semiconductor including an MQW layer, A second n-type semiconductor layer, which is composed of at least one n-type semiconductor formed on the core layer, It has a nipn structure equipped with, The thermal resistance of the p-type semiconductor is set higher than that of the undoped semiconductor. A semiconductor optical modulator according to claim 1 or 2.

6. The semiconductor optical modulator according to claim 1 or 2, characterized in that, under the same ambient temperature, the temperature of the optical waveguide in the phase modulation section is adjusted for each wavelength, and the operating temperature of the optical waveguide is higher on the longer wavelength side.

7. The semiconductor optical modulator according to claim 6, characterized in that the design operating temperature of the semiconductor optical modulator is at least equal to or greater than the maximum ambient temperature at which the semiconductor optical modulator is required to operate, and the amount of heat generated by the heater mechanism is controlled by either current or voltage control according to the operating wavelength, and under the same temperature, the amount of heat generated by the heater is greater when operating at longer wavelengths.

8. A semiconductor modulation apparatus comprising a semiconductor optical modulator according to claim 1 or 2, It includes a driver IC that has a temperature monitoring mechanism and is mounted adjacent to the semiconductor optical modulator in the same package, A semiconductor modulation device characterized in that, by referring to the temperature around the driver IC monitored by the temperature monitoring mechanism of the driver IC, the amount of heat generated by the heater mechanism of the semiconductor optical modulator is controlled by either current or voltage control, and when compared at the same wavelength, the higher the temperature monitored by the temperature monitoring mechanism, the smaller the amount of heat generated by the heater mechanism.

Images