JP5437289B2 - Semiconductor optical modulator - Google Patents

Description

  The present invention relates to a semiconductor optical modulator.

  In recent years, high-speed modulators of 10 GHz or higher have begun to spread, and in particular, those housed in small packages are becoming widespread. Among these high-speed modulators, there is a configuration in which intensity of laser light continuously oscillated from a fixed-wavelength distributed feedback laser (DFB laser) is modulated by an electroabsorption optical modulator (EA modulator) or a DFB laser. Some are directly modulated and used. In these cases, since the DFB laser has a fixed wavelength, it is necessary to prepare an EA modulator and a DFB laser or a DFB laser having different wavelengths for each channel as a stock in preparation for failure.

  Therefore, it is conceivable to use a tunable laser for the laser. This is because the channels within the variable range can be covered by preparing only one laser within the variable wavelength range of the tunable laser. For this reason, it is also possible to facilitate inventory management for failures. However, in the EA modulator, the characteristics of electroabsorption change depending on the operating wavelength and the detuning amount of the electroabsorber (the difference between the bandgap wavelength of the electroabsorber and the operating wavelength), and a delay occurs.

  Therefore, it is conceivable to use a Mach-Zehnder modulator (MZ modulator). Since the band of the MZ modulator can be set wide, when considering a combination with a variable wavelength laser, a small and high performance modulator can be realized. Further, since the problem of signal degradation due to chirping is alleviated, the transmission distance can be extended.

  On the other hand, a semiconductor Mach-Zehnder circuit is an optical circuit using two-beam interference, and it is difficult to produce an operating point as designed due to a phase error that occurs during manufacturing in terms of performance immediately after manufacturing. Therefore, in many cases, a phase adjustment unit for adjusting the operating point is added in addition to the modulation unit for modulating the signal at high speed. Here, what has become a problem in the past is the voltage level of the phase adjustment electrode of the phase adjustment unit.

There are two types of phase adjustment methods.
One is a method of adjusting the phase by performing a current injection and utilizing a refractive index change due to the injection. In this case, although phase adjustment is possible with a relatively short phase adjustment electrode, there is a problem that power is consumed because current is supplied for phase adjustment. In addition, since it is a current injection type device, there is a problem that long-term reliability tends to be low.

  The other is a method of adjusting the phase with an electric field. This is because the refractive index is changed by an electric field, so that almost no current flows, so that power consumption can be kept small. In addition, since it is not a device through which current flows, high reliability is easily obtained for long-term reliability. However, since the modulation efficiency is low, there is a problem that the length of the phase adjustment electrode becomes longer than that of current injection.

FIG. 7 is a schematic diagram showing an example of the configuration of a conventional linear MZ modulator.
As shown in FIG. 7, in a conventional linear MZ modulator, an input waveguide 100 is connected to an inlet-side coupler 101, and light is branched into two arm waveguides 102a and 102b. Thereafter, the light passes through the optical waveguide in which the modulation electrodes 103a and 103b as the modulation unit 103 are formed, and then passes through the phase adjustment unit 104 in which the phase adjustment electrodes 104a and 104b are formed.

  Then, after modulation and phase adjustment, the light that has interfered with the exit-side coupler 105 is output again from the output waveguides 106 a and 106 b, and the output changes depending on the phase difference given by the modulation unit 103. . In FIG. 7, the case where the phase adjustment unit 104 is arranged after the modulation unit 103 is described as an example, but the same operation and effect can be obtained even in the opposite arrangement.

Here, the operation of the conventional MZ modulator will be briefly described.
FIG. 8 is a graph showing the characteristics of the normalized transmission intensity with respect to the drive voltage to the modulation arm when the push-pull drive is performed in the conventional MZ modulator.
In FIG. 8, the vertical axis represents normalized transmission intensity [dB], and the horizontal axis represents drive voltage [V] to the modulation arm when push-pull driving is performed. Further, for convenience, the horizontal axis plots the case where the upper arm is driven in the negative direction in the negative part on the left side in FIG. 8, and the voltage obtained by driving the lower arm in the positive part on the right side in FIG. Multiplied by -1. For example, when the bias voltage is 3V, the horizontal axis + 2V means that the upper arm is 2V higher than the bias voltage by -1V and the lower arm is 2V lower by -4V.

  In FIG. 8, the broken line shows the values when the bias voltage is 2.5V, the dashed line is the bias voltage 3.5V, the dashed line is 4.5V, and the solid line is 5.5V. In addition, the wavelength of the light to be measured is 1560 nm, the length of the phase adjustment electrodes 104a and 104b is 1.2 mm, and the phase adjustment electrodes 104a and 104b are driven in a normally OFF state. The half-wave voltage corresponds to a voltage between the peaks of the half-wave voltage.

As shown in FIG. 8, it is understood that the half-wave voltage can be controlled by applying a bias voltage as a feature of the semiconductor. Many devices that drive high-frequency modulators can output an amplitude voltage of 2.5 V to 3 V. Therefore, when driving with OOK (ON-OFF-Keying), the period from ON to OFF in FIG. It is only necessary to make a transition at twice the driver amplitude. When the bias voltage is 4.5 V, driving with an amplitude of 0.85 V is possible.
When a driver with an amplitude of 2.5 V is assumed, the length can be further shortened. Therefore, the phase adjustment electrodes 104a and 104b can be driven even when the length is 600 μm, which is half of the length.

Here, the characteristics of the phase adjustment electrode in the phase adjustment unit will be described.
FIG. 9 is a graph showing normalized transmission intensity characteristics with respect to the drive voltage to the modulation arm when the conventional MZ modulator is driven by one arm by a 500 μm phase adjustment electrode.
As shown in FIG. 9, when driving one arm at a time by a 500 μm phase adjustment electrode, the half-wave voltage reaches -8V. For this reason, a high voltage is required for phase adjustment. Experimentally, phase adjustment is possible even with a half-wave voltage of −8V, but there are limitations due to the specifications of the power supply voltage prepared when operating as an actual device.

  In addition, when used in an actual device, this length is determined by the specifications of the power supply voltage of the drive circuit. However, a higher voltage output requires a larger transformer circuit, and it costs much more. The half-wave voltage is desirably small. Further, although it is desirable that it can be driven at a low voltage in terms of long-term reliability, there is a problem that when the voltage is lowered, the length of the phase adjustment electrodes 104a and 104b becomes long.

  For example, if the specification of the power supply voltage is 5 V or less, the phase adjustment electrodes 104a and 104b are insufficient at 500 μm, and a length of at least 1 mm is necessary. In actual design based on these values, the length of the modulation electrodes 103a and 103b that perform the main function of the MZ modulator may be about 0.6 mm, whereas the phase adjustment electrode that simply adjusts the phase. This means that electrodes of 1 mm or more are required for 104a and 104b.

  In recent years, transmission in a variety of formats has been attempted in order to realize a larger capacity communication, for example, differential quadrature phase shift keying (DQPSK). DQPSK is one of digital signal modulation methods, and is a method capable of allocating 2-bit data to each of four modulated phases (for example, see Non-Patent Document 1 below).

  In the case of this DQPSK, it is necessary to give a 90-degree phase difference between the signals that have passed through the two MZ modulators after each MZ modulator. This portion also requires the phase adjustment electrodes 104a and 104b for phase adjustment, and there is a problem that a longer length is required.

Nobuhiro Kikuchi, 7 others, “80-Gb / s Low-Driving-Voltage InP DQPSK Modulator With an n-p-i Structure”, IEEE PHOTOTONICS TECHNOLOGY TECHNOLOGY 21, NO. 12, June 15, 2009, p. 787-789

  As described above, the phase adjustment unit 104 having the phase adjustment electrodes 104a and 104b is based on a certain power supply voltage specification so that the phase adjustment can be performed in any state. Then, a corresponding length is required. For this reason, the phase adjustment electrodes 104a and 104b are originally longer than the modulation electrodes 103a and 103b that perform the main functions of the MZ modulator, and accordingly, the overall length of the MZ modulator is also increased. There was a problem.

  In addition, when the MZ modulator is housed together in the same small package as a separately manufactured laser, it is necessary to arrange a lens or the like between them, so that there is some space in the width direction of the package. is there. However, since there is no room in the package length direction, there has been a strong demand for shortening the length of the MZ modulator.

  In light of the above, an object of the present invention is to provide a semiconductor optical modulator that can be shortened in the length direction.

A semiconductor optical modulator according to a first invention for solving the above-described problem is
Two couplers and two arm waveguides connecting them are provided, and a modulation electrode that modulates by applying a high frequency on the same arm waveguide and a phase adjustment electrode that adjusts the operating point by applying a DC voltage are provided A Mach-Zehnder type semiconductor optical modulator made of a semiconductor,
An input side coupler for branching light from the input waveguide;
A development part connected after the input side coupler for widening the distance between the two arm waveguides;
A modulation electrode that is connected after the expanding portion and modulates light;
A delay unit connected after the modulation electrode to reduce the interval between the two arm waveguides and equalize the propagation distance of the two arm waveguides;
A folded portion made of a curved waveguide connected after the delay portion and converting the propagation direction of light by 180 degrees;
A phase adjusting electrode for adjusting the phase of light connected after the folded portion;
An output-side coupler connected after the phase adjustment electrode to interfere with each light;
An output waveguide formed on the same surface as the surface on which an input waveguide that outputs light that is connected and interfered after the output-side coupler is formed ;
The delay unit is by reducing the distance between the two arm waveguides in the asymmetric, characterized by equal to Rukoto the propagation distance of the two arm waveguides.

A semiconductor optical modulator according to a second invention for solving the above-mentioned problems is as follows.
Two couplers and two arm waveguides connecting them are provided, and a modulation electrode that modulates by applying a high frequency on the same arm waveguide and a phase adjustment electrode that adjusts the operating point by applying a DC voltage are provided A Mach-Zehnder type semiconductor optical modulator made of a semiconductor,
An input side coupler for branching light from the input waveguide;
A development part connected after the input side coupler for widening the distance between the two arm waveguides;
A modulation electrode that is connected after the expanding portion and modulates light;
A delay unit connected after the modulation electrode to reduce the interval between the two arm waveguides and equalize the propagation distance of the two arm waveguides;
A first folded part comprising a curved waveguide connected after the delay part and converting the propagation direction of light by 180 degrees;
A first phase adjustment electrode connected after the first folded portion and performing phase adjustment of light;
A second folded portion comprising a curved waveguide connected after the first phase adjusting electrode and converting the propagation direction of light again by 180 degrees;
A second phase adjustment electrode connected after the second folded portion and performing phase adjustment of light;
An output-side coupler connected after the second phase adjustment electrode to interfere with each light;
An output waveguide formed on a surface different from a surface on which an input waveguide that outputs light that is connected and interfered after the output coupler is formed ; and
The delay unit is by reducing the distance between the two arm waveguides in the asymmetric, characterized by equal to Rukoto the propagation distance of the two arm waveguides.

A semiconductor optical modulator according to a third invention for solving the above-described problem is
Two couplers and two arm waveguides connecting them are provided, and a modulation electrode that modulates by applying a high frequency on the same arm waveguide and a phase adjustment electrode that adjusts the operating point by applying a DC voltage are provided A Mach-Zehnder type semiconductor optical modulator made of a semiconductor,
An input side coupler for branching light from the input waveguide;
A development part connected after the input side coupler for widening the distance between the two arm waveguides;
A modulation electrode that is connected after the expanding portion and modulates light;
A delay unit connected after the modulation electrode to reduce the interval between the two arm waveguides and equalize the propagation distance of the two arm waveguides;
A folded portion made of a curved waveguide connected after the delay portion and converting the propagation direction of light by 180 degrees;
A phase adjusting electrode for adjusting the phase of light connected after the folded portion;
A second folded portion comprising a curved waveguide connected after the phase adjustment electrode and converting the propagation direction of light again by 180 degrees;
An output-side coupler that interferes with each light from the second folded portion;
An output waveguide connected after the output coupler and outputting the interfered light ,
The delay unit is by reducing the distance between the two arm waveguides in the asymmetric, characterized by equal to Rukoto the propagation distance of the two arm waveguides.

A semiconductor optical modulator according to a fourth invention for solving the above-described problems is as follows.
Two semiconductor optical modulators according to the third invention;
An input-side coupler that splits light into the two semiconductor optical modulators;
A phase difference adjusting unit that is connected to the output waveguides of the two semiconductor optical modulators and gives a phase difference of 90 degrees between the output lights;
An output side coupler that is connected after the phase difference adjusting unit and interferes with each light is configured.

A semiconductor optical modulator according to a fifth aspect of the present invention for solving the above problems is the semiconductor optical modulator according to any one of the first to fourth aspects of the invention,
When the distance between the two arm waveguides having the modulation electrode is D1 and the distance between the two arm waveguides having the phase adjustment electrode is D2, “D1> D2” is set.

  An object of the present invention is to provide a semiconductor optical modulator that can be shortened in the length direction.

It is the schematic diagram which showed the structure of the semiconductor optical modulator based on a 1st Example. It is the schematic diagram which showed the structure of the delay part in the semiconductor optical modulator based on a 1st Example. It is the schematic diagram which showed the mounting method to the package of the semiconductor optical modulator which concerns on a 1st Example. It is the schematic diagram which showed the structure of the semiconductor optical modulator based on a 2nd Example. It is the schematic diagram which showed the structure of the semiconductor optical modulator based on a 3rd Example. It is the schematic diagram which showed the manufacturing method of the semiconductor optical modulator based on the 1st-3rd Example. It is the schematic diagram which showed an example of the structure of the conventional linear MZ modulator. It is the figure which showed the characteristic of the transmission intensity normalized with respect to the drive voltage to the modulation arm at the time of push-pull drive in the conventional MZ modulator. It is the figure which showed the characteristic of the transmission intensity normalized with respect to the drive voltage to the modulation arm at the time of driving one arm at a time by the 500-micrometer phase adjustment electrode in the conventional MZ modulator.

Hereinafter, embodiments for implementing a semiconductor optical modulator according to the present invention will be described.
With respect to the folded structure in the semiconductor optical modulator according to the present invention, a conventionally known lithium niobate optical modulator (LN modulator) has been tried. However, in the case of an LN modulator, the difference in refractive index between the waveguides is relatively small, and loss occurs unless the waveguide is bent with a certain bending radius. For this reason, a mirror structure or the like is adopted for folding.

  On the other hand, in the semiconductor optical modulator according to the present invention, when the semiconductor optical modulator is formed of a semiconductor, the advantage of being able to reduce the bending radius is that the direction change of 180 degrees is realized by the curved waveguide. It is fundamentally different from the folding structure of the vessel.

  Conventionally, although a loop by a circuit in which a quartz-based planar lightwave circuit (PLC) and an LN modulator are bonded is proposed, phase adjustment is possible in the PLC unit, but there is a problem that high-speed modulation cannot be performed. That is, the high-speed modulation unit and the phase adjustment unit cannot be created in parallel, and the interval between the phase adjustment units is not taken into consideration in order to make the size smaller in consideration of electrical crosstalk of the high-speed modulation unit.

  The half-wave voltage of a modulator such as an LN modulator is determined by the element structure and the physical property value used mainly, and the length is also determined accordingly. On the other hand, in the semiconductor modulator, the half-wave voltage is variable by the bias voltage in addition to the element structure and the physical property value. Therefore, the half-wave voltage can be reduced by applying a bias voltage, and the length of the modulation electrode is also short.

  The semiconductor optical modulator according to the present invention solves the problem that a phase adjustment unit having a phase adjustment electrode unique to a semiconductor is longer than the modulation unit, and the problem to be solved is in a semiconductor modulator. This is not a simple folding circuit structure as seen in an LN modulator.

A first embodiment of a semiconductor optical modulator according to the present invention will be described below with reference to the drawings.
In the semiconductor optical modulator according to the present embodiment, a Mach-Zehnder type semiconductor optical modulator having a folded portion by a so-called one-sided pigtail curved waveguide is manufactured.
First, a method for manufacturing a semiconductor optical modulator according to this embodiment will be described.
FIG. 6 is a schematic view showing a method for manufacturing the semiconductor optical modulator according to the present embodiment.
First, as shown in FIG. 6A, a first n-type electrode layer 51 (n + -InP) is grown on a semi-insulating (SI) -InP substrate 50, and a first n-type electrode layer 51 (n + -InP) is grown thereon. 1 n-type cladding layer 52 (n-InP) is formed, and on the first n-type cladding layer 52, a first intermediate layer 53 (i-InGaAsP), a multiple quantum well (MQW) core layer 54, A second intermediate layer 55 (i-InGaAsP) is formed.

  After the first low-concentration cladding 56 (i-InP) is formed on the second intermediate layer 55 (i-InGaAsP), an electron barrier is formed on the first low-concentration cladding 56 (i-InP). As a result, a p-type cladding layer 57 (p-InP) is formed. A second n-type cladding layer 58 (n-InP) is formed on the p-type cladding layer 57, and a second n-type electrode layer 59 (n + -InP) is sequentially stacked thereon. Yes.

  Here, the multiple quantum well core layer 54 is configured so that the electro-optic effect works effectively at the operating light wavelength. For example, in the case of a 1.5 μm band device, a layer in which the Ga / Al composition of InGaAlAs is changed is used. The multi-quantum well structure can be a quantum well layer and a quantum barrier layer, respectively. Further, the first intermediate layer 53 functions as a connection layer for preventing carriers generated by light absorption from being trapped at the heterointerface.

  In order to manufacture the semiconductor optical modulator according to the present embodiment, first, in order to electrically separate the two arm waveguides (see the arm waveguides 102a and 102b shown in FIG. 7), the arm portion of the waveguide is used. A separation groove is formed. In the case of a Mach-Zehnder structure in which a modulation electrode and a phase adjustment electrode, which will be described later, are separated, a separation groove is provided between them to perform electrical separation. This is because the voltage applied to one arm for modulation affects the modulation of the other arm unless electrical separation is performed. The isolation groove is formed by a portion of the second n-type electrode layer 59 to the p-type cladding layer 57 functioning as an electron barrier, which is patterned by standard photolithography and patterning, and wet etching technology is used. It is formed by removing as a groove with a width of several microns.

  In this embodiment, the electrical separation is performed by the separation groove. However, the portion other than the periphery of the modulation portion in contact with the electrode functions as an electron barrier from the second n-type electrode layer 59 using a quartz hard mask. After removing up to the mold cladding layer 57, it may be grown again with semi-insulating InP and replaced to perform electrical separation.

  Next, as shown in FIG. 6B, the layers from the second n-type electrode layer 59 (n + -InP) to the first n-type cladding layer intermediate 52 are etched using a dry etching technique. A high-mesa waveguide structure is formed. Then, the first n-type cladding layer 52 is etched to expose the first n-type electrode layer 51.

  Finally, as shown in FIG. 6C, a first n-type electrode 60 to be a modulation electrode and a phase adjustment electrode, which will be described later, is placed on the second n-type electrode layer 59 and a second n-type to be a ground electrode. A mold electrode 61 is formed on each first n-type electrode layer 51. If necessary, a passivation film may be deposited to protect the mesa surface, or the high mesa structure may be protected using a polymer or the like.

Next, the configuration of the semiconductor optical modulator according to the present embodiment will be described.
FIG. 1 is a schematic diagram illustrating a configuration of a semiconductor optical modulator according to the present embodiment. FIG. 1 schematically shows the semiconductor optical modulator according to this embodiment as viewed from above.
As shown in FIG. 1, the semiconductor optical modulator 1 according to the present embodiment includes two couplers 11 and 17 and an arm waveguide connecting them, and performs modulation by applying a high frequency on the same arm waveguide. A Mach-Zehnder type semiconductor optical modulation comprising a semiconductor comprising a modulation section 13 having modulation electrodes 13a and 13b, which will be described later, and a phase adjustment section 16 having phase adjustment electrodes 16a and 16b for adjusting the operating point by applying a DC voltage. It is a vessel.

  The semiconductor optical modulator 1 according to the present embodiment includes an input-side coupler 11 that branches light from the input waveguide 10 and a developing portion 12a that is connected after the input-side coupler 11 and widens the interval between the two arm waveguides. , 12b, modulation electrodes 13a, 13b that are connected after the expanding portions 12a, 12b and modulate the light, and are connected after the modulation electrodes 13a, 13b, and the distance between the two arm waveguides is reduced, and the two arm guides are connected. Delay sections 14a and 14b for equalizing the propagation distances of the waveguides, folded sections 15a and 15b composed of curved waveguides connected after the delay sections 14a and 14b and converting the propagation direction of light by 180 degrees, and the folded sections 15a and 15b. Phase adjustment electrodes 16a and 16b that are connected later to adjust the phase of light, and output side caps that are connected after the phase adjustment electrodes 16a and 16b and interfere with the respective lights. A puller 17, the output waveguides 18a to the input waveguide 10 to output the connected light to interfere after the output side coupler 17 is formed on the same surface as the surface to be formed, is constituted by the 18b.

  In this embodiment, the minimum bending radius of the folded portions 15a and 15b is 150 μm. The width of the arm waveguide was 1.6 μm for both the straight part and the curved part. Although parameters other than these may be set, it is preferable to reduce the minimum bending radius because the size of the semiconductor optical modulator 1 can be reduced.

  Here, in order to compare the sizes of the semiconductor optical modulator according to the present embodiment and the conventional linear MZ modulator of the related art, the conventional linear MZ modulation having the modulation unit and the phase adjustment unit of the same length. Assuming that the modulator is designed and the length of the modulation electrode is 800 μm and the length of the phase adjustment electrode is 1.1 mm, the conventional linear MZ modulator has a length of 3.7 mm and a width of Although the design was possible with 800 μm, the semiconductor optical modulator according to this example can be designed with a length of 2.4 mm and a width of 1 mm by using a folded structure.

As described above, the semiconductor optical modulator according to the present embodiment can be reduced by 64% in length as compared with the conventional linear MZ modulator. The occupied area is 2.4 mm ² in the conventional linear MZ modulator, but the semiconductor optical modulator according to the present embodiment can be laid out with the same occupied area 2.4 mm ^2. . That is, according to the semiconductor optical modulator according to the present embodiment, the total length can be shortened without reducing the number of chips that can be collected from the wafer.

Here, the configuration of the delay unit in the semiconductor optical modulator according to the present embodiment will be described.
FIG. 2 is a schematic diagram illustrating a configuration of a delay unit in the semiconductor optical modulator according to the present embodiment.
Normally, when the output waveguide 18 is formed on the same surface as the surface on which the input waveguide 10 is formed, the optical path length differs between the arm waveguide positioned inside and the arm waveguide positioned outside.

  For this reason, in the semiconductor optical modulator according to the present embodiment, as shown in FIG. 2, the length of the upper delay portion 14a is made longer than that of the lower delay portion 14b, whereby the inner folded portion 15a (FIG. In this case, a delay difference is provided so as to compensate for the shortening of the inner folded portion 15a by reducing the width asymmetrically after passing through the modulation electrodes 13a and 13b. Thereby, the phase difference generated between the two arm waveguides can be eliminated.

Next, a method for mounting the semiconductor optical modulator according to the present embodiment on a package will be described.
Polishing is performed on the back surface of the semi-insulating-InP substrate 50 to the extent that cleavage can be performed after manufacturing the semiconductor optical modulator 1 described above. After depositing a metal film so that the fixed solder adheres to the back surface, and then cleaving various elements, there is no reflection on the end face of the semiconductor optical modulator 1 in which the input waveguide 10 and the output waveguides 18a and 18b are formed. A coat was applied. In the semiconductor optical modulator 1 according to the present embodiment, since the input waveguide 10 and the output waveguides 18a and 18b are formed only on one side, the number of times of applying the antireflection coating is reduced from two times to one time. And the number of processes can be reduced.

FIG. 3 is a schematic diagram illustrating a method of mounting the semiconductor optical modulator according to the present embodiment on a package.
As shown in FIG. 3, the semiconductor optical modulator 1 according to the present embodiment is mounted on a submount 19 made of aluminum nitride with a standard chip bonder and then fixed by heating, and subsequently, a termination resistor, a capacitor, a thermistor (resistance) A temperature sensor that detects the temperature as a change) and various elements 20 such as a wiring board for transmitting a high-speed signal were similarly mounted by a chip bonder and fixed by reflow. After the semiconductor optical modulator 1 and various elements and wiring electrodes on the fixed submount 19 were connected by wire bonding, the submount 19 was reflow-fixed to the mount 21 made of CuW again.

The mount 21 was fixed with solder in an airtight package in which the end fiber 22 can be fixed, and a 127 μm pitch two-core metallized fiber was fixed with solder.
The lens 23 was inserted between the fiber at the end (hereinafter referred to as the direction end fiber 22) and the semiconductor optical modulator 1 for alignment, and then the lens 23 was fixed using a YAG laser. Conventionally, two lenses 23 are necessary for input and output, but by using the semiconductor optical modulator 1 according to this embodiment, only one lens 23 is required, and the number of members can be reduced to half. Further, since the work for fixing the lens 23 can be halved by using the semiconductor optical modulator 1 according to the present embodiment, the number of mounting steps can be reduced.

  In this embodiment, a 1-input 2-output MMI (multi-mode-interferometer) is used as the input-side coupler 11, and a 2-input 2-output MMI is used as the output-side coupler 17. The side coupler 17 may have one input and two outputs, or vice versa. Also, other couplers such as other directional couplers may be used.

Hereinafter, a second embodiment of the semiconductor optical modulator according to the present invention will be described with reference to the drawings.
The semiconductor optical modulator according to the present invention has the same manufacturing method as the semiconductor optical modulator according to the first embodiment. However, since the semiconductor optical modulator according to the present invention is applied to a fiber that is not the end, a non-reflective coating is applied to both sides of the semiconductor optical modulator, and both sides of the semiconductor optical modulator are also packaged. Position and position the lens. The other configuration of the semiconductor optical modulator according to the present invention is basically the same as that of the semiconductor optical modulator 1 according to the first embodiment.

First, the configuration of the semiconductor optical modulator according to the present embodiment will be described.
FIG. 4 is a schematic diagram showing the configuration of the semiconductor optical modulator according to the present embodiment. FIG. 4 schematically shows the semiconductor optical modulator according to the present embodiment as viewed from above.
As shown in FIG. 4, the semiconductor optical modulator 2 according to the present embodiment includes an input-side coupler 31 that branches light from the input waveguide 30, and an interval between two arm waveguides that are connected after the input-side coupler 31. For spreading the light, modulation electrodes 33a, 33b connected after the expansion portions 32a, 32b for modulating light, and connected between the modulation electrodes 33a, 33b, and the distance between the two arm waveguides being narrowed. In addition, the first return portion 35a is formed of delay portions 34a and 34b that make the propagation distances of the two arm waveguides equal, and a curved waveguide that is connected after the delay portions 34a and 34b and changes the propagation direction of light by 180 degrees. , 35b, after the first phase adjustment electrodes 36a, 36b, connected after the first folding portions 35a, 35b, and for adjusting the phase of the light, and after the first phase adjustment electrodes 36a, 36b. Second folded portions 37a and 37b made of curved waveguides that are connected and convert the propagation direction of light again by 180 degrees, and second phase adjustment electrodes 36c and 36d that are connected after the phase adjustment portion 36 and perform phase adjustment of light. Then, an output side coupler 38 connected after the second phase adjustment electrodes 36c and 36d and causing interference of the respective lights, and an input waveguide 30 connected after the output side coupler 38 and outputting the interfered light are formed. The output waveguides 39a and 39b are formed on different surfaces.

Next, features of the semiconductor optical modulator according to the present embodiment will be described.
The high frequency modulation unit 33 sets the distance D1 between the two arm waveguides to a certain extent in consideration of electrical crosstalk, and sets “D1 = 100 μm”. When the GSSG (GND-SIGNAL-SIGNAL-GND) electrode design is used for the two modulation electrodes 33a and 33b through which the high frequency passes, it is not a problem if the distance D1 is reduced to some extent. Manufacturing becomes difficult.

  In addition, when designing with general GSG (GND = SIGNAL = GND), a certain distance D1 is required. Furthermore, since different signals may be input depending on the application, it is necessary to increase the distance D1 in order to reduce electrical crosstalk. That is, in the modulation unit 33, electrical crosstalk can be reduced by adopting a configuration in which the interval D1 is separated to some extent.

  Therefore, after passing through the modulation section 33, the distance D2 between the two arm waveguides is once narrowed to “D2 = 10 μm”. In addition, when 180 degree | times folding is performed without narrowing the space | interval D2, the width | variety of the 1st folding | turning part 35a, 35b will become wide on design, and an occupation area will become large. That is, the width of the first folded portions 35a and 35b can be reduced by reducing the distance D2.

  Further, since the first folded portions 35a and 35b, which are the folded portions, are only the phase adjusting unit 36 that is driven by applying a DC voltage, there is a problem in operation even if the interval D2 between the two arm waveguides is narrowed. Absent. Note that the range in which the distance D2 can be narrowed can be set as long as it is within a range that can be manufactured and is a distance that does not couple light propagating through the two arm waveguides.

  Thereafter, the first phase adjustment electrodes 36a and 36b are formed while maintaining the interval D2, and the second return portions 37a and 37b for performing the 180-degree return again after the first phase adjustment electrodes 36a and 36b are provided. Form. Further, the second phase adjusting electrodes 36c and 36d are formed after the second folded portions 37a and 37b with the arm waveguide interval D3 set to “D3 = D2,” and output after the second phase adjusting electrodes 36c and 36d. A side coupler 38 is installed.

  In the semiconductor optical modulator 2 according to this embodiment, the electrodes on the same arm waveguide of the first phase adjustment electrodes 36a and 36b and the second phase adjustment electrodes 36c and 36d are connected in parallel. . Thereby, the entire length of the phase adjustment unit 36 can be shortened. In other words, it can also be said that the phase adjustment unit 36 includes the second folding portions 37a and 37b.

  In this case, the length of the semiconductor optical modulator 2 is more precisely determined by the modulation electrodes 33a and 33b that mainly perform the main function, more precisely “length of the input waveguide 30 + length of the modulation electrodes 33a and 33b + first length”. Therefore, the length in the length direction is determined by the phase adjustment electrodes 104a and 104b (see FIG. 7) as in the conventional linear MZ modulator. Can be avoided. Therefore, according to the semiconductor optical modulator 2 according to the present embodiment, the length in the longitudinal direction can be made very short, and the first phase adjustment electrodes 36a and 36b and the second phase adjustment electrodes having the sufficient length can be obtained. Phase adjustment electrodes 36c and 36d can be obtained.

  As described above, in the semiconductor optical modulator 2 according to the present embodiment, by arranging the modulation electrodes 33a and 33b and the phase adjustment electrodes 36a and 36b in a relationship of “D1> D2” at least, Since the first phase adjustment electrodes 36a and 36b and the second phase adjustment electrodes 36c and 36d can be made sufficiently long while the length of the phase modulation unit 36 can be shortened, the semiconductor optical modulator 2 The length in the length direction can be shortened, the increase in the width of the semiconductor optical modulator 2 can be suppressed, and a folded structure can be realized. Further, electrical crosstalk in the modulator 33 can be suppressed. You can also.

  In the present embodiment, the second phase adjusting electrodes 36c and 36d are also disposed after the second folded portions 37a and 37b. However, the second phase adjusting electrodes 36c and 36d are not necessarily disposed unless particularly necessary. However, since it is easier to obtain long-term reliability in the phase adjustment unit 36 by applying a lower DC voltage, it is preferable to form the second phase adjustment electrodes 36c and 36d if the occupied area does not change. In the present embodiment, the distances D1 and D2 are set as “D1> D2,” but other settings may be made.

A third embodiment of the semiconductor optical modulator according to the present invention will be described below with reference to the drawings.
The semiconductor optical modulator according to the present invention is the same in manufacturing method as the semiconductor optical modulator 1 according to the first embodiment. However, since the fiber to which the semiconductor optical modulator according to the present invention is applied is not the opposite end, a non-reflective coating is applied on both sides of the semiconductor optical modulator, and the lens is positioned on both sides of the semiconductor optical modulator even during packaging. And place it. The other configuration of the semiconductor optical modulator according to the present invention is basically the same as that of the semiconductor optical modulator 1 according to the first embodiment.

First, the configuration of the semiconductor optical modulator according to the present embodiment will be described.
FIG. 5 is a schematic diagram showing the configuration of the semiconductor optical modulator according to the present embodiment. FIG. 5 schematically shows the semiconductor optical modulator according to the present embodiment as viewed from above.
As shown in FIG. 5, the semiconductor optical modulator 3 according to this example includes the second optical modulator 3 according to the second example after the phase adjustment electrodes 16 a and 16 b in the semiconductor optical modulator 1 according to the first example. The DQPSK is provided with a phase difference adjustment unit that uses two semiconductor optical modulators having the folded portions 37a and 37b and shifts the phase of the output light of each semiconductor optical modulator by 90 degrees.

  The semiconductor optical modulator 3 according to the present embodiment includes an input-side coupler 41 connected after the input waveguide 40, and a deploying portion 42a for increasing the interval between the two arm waveguides connected after the input-side coupler 41. 42b, an input-side coupler 11 that is connected after each of the development part 42a and the development part 42b and branches light, and development parts 12a and 12b that are connected after the input-side coupler 11 and widen the distance between the two arm waveguides. The modulation electrodes 13a and 13b that are connected after the expanding portions 12a and 12b and modulate the light, and the spaces between the two arm waveguides that are connected after the modulation electrodes 13a and 13b are narrowed and propagated through the two arm waveguides. Folding consisting of delay sections 14a and 14b having the same distance and a curved waveguide connected after the delay sections 14a and 14b and converting the propagation direction of light by 180 degrees. The phase adjustment electrodes 16a and 16b connected after the portions 15a and 15b, the turn-back portions 15a and 15b, and the phase adjustment electrodes 16a and 16b, and the curves connected after the phase adjustment electrodes 16a and 16b and converting the light propagation direction again by 180 degrees. Second folded portions 37a and 37b made of waveguides, an output side coupler 38 connected after the second folded portions 37a and 37b and interfering with each light, and an output side coupler 38 connected after the output side coupler 38. Of phase adjusting electrodes 43a and 43b that give a phase difference of 90 degrees between the respective output lights from the output side, an output side coupler 44 that is connected after the phase adjustment electrodes 43a and 43b, and interferes with the respective lights, and an output side coupler 44 An output waveguide 45 formed on a surface different from the surface on which the input waveguide 40 that outputs light that is connected and interfered later is formed. And it is made of.

Next, features of the semiconductor optical modulator according to the present embodiment will be described.
The DPSK is configured to include two single MZ modulators, and also includes a phase adjustment electrode for the phase difference adjustment unit 43 that gives a phase difference of 90 degrees between the output lights from the two single MZ modulators. 43a and 43b need to be formed. For this reason, there is a problem that the length in the length direction becomes much longer because it is not necessary to simply arrange two single MZ modulators side by side.

  Therefore, in the semiconductor optical modulator 3 according to the present embodiment, the second folded portions 37a and 37b are provided after the phase adjusting electrodes 16a and 16b, and the output side coupler 38 is provided after the second folded portions 37a and 37b. Thereafter, phase adjustment electrodes 43a and 43b are provided that give a phase difference of 90 degrees between the respective output lights from the two output side couplers 38. In the present embodiment, the length of the phase adjustment electrodes 43a and 43b of the phase difference adjustment unit 43 is 1.1 mm, which is the same as that of the phase adjustment electrodes 16a and 16b.

  According to the configuration of the semiconductor optical modulator 3 according to the present embodiment, the length in the length direction only needs to be increased by the arrangement of the input side coupler 41 and the development portions 42a and 42b. The total length of the modulator 3 can be kept short. Note that. In actual design, it was possible to design with a length of 2.8 μm. However, when two single MZ modulators are simply arranged side by side and linearly laid out, a length of 6.1 mm is required, which is more than double that of the semiconductor optical modulator 3 according to the present embodiment. there were.

  The width of the semiconductor optical modulator 3 according to this embodiment can be made narrower than the conventional DQPSK in which two single MZ modulators are simply arranged side by side, and a layout of 1.8 mm is possible. When assuming TOSA (Transmitter Optical Sub-Assembly), the width inside the container can be about 5.5 mm, and the semiconductor optical modulator 3 according to this embodiment has a width of 1.8 mm. And can be stored in the TOSA. Further, since the length of the semiconductor optical modulator 3 according to the present embodiment is shorter than the conventional DQPSK in which two single MZ modulators are arranged side by side, the wavelength tunable laser can be integrated in the TOSA. Can fit in length.

  In this embodiment, two semiconductor optical modulators are arranged vertically symmetrically, and the phase adjustment electrodes 16a and 16b to which a DC voltage is applied are located inside. For this reason, since the distance between the modulation electrodes 13a and 13b to which the high frequency is applied in the two semiconductor optical modulators can be increased, electrical crosstalk can be suppressed.

  In addition, the semiconductor optical modulator 3 according to the present embodiment has been described as an example in which the semiconductor optical modulator 3 is configured as DQPSK. However, two semiconductor optical modulators 3 according to the present embodiment are arranged side by side to form a DP-QPSK (Dual Polarization Differential Quadrature Shift). Even if it is a case where (Keying) etc. are comprised, the same effect | action and effect are acquired, and the length of the length direction of DP-QPSK etc. can be shortened.

  The present invention can be used for, for example, a high-speed modulator of 10 GHz or more, particularly a high-speed modulator housed in a small package.

1, 2, 3 Semiconductor optical modulator 10 Input waveguide 11 Input coupler 12a, 12b Expanding unit 13 Modulating unit 13a, 13b Modulating electrode 14a, 14b Delay unit 15a, 15b Folding unit 16 Phase adjusting unit 16a, 16b Phase adjusting electrode 17 Output side couplers 18a and 18b Output waveguide 19 Submount 20 Various elements 21 Mount 22 End fiber 23 Lens 30 Input waveguide 31 Input side couplers 32a and 32b Expanding part 33 Modulating parts 33a and 33b Modulating electrodes 34a and 34b Delay part 35a, 35b First folding unit 36 Phase adjusting units 36a, 36b First phase adjusting electrodes 36c, 36d Second phase adjusting electrodes 37a, 37b Second folding unit 38 Output side couplers 39a, 39b Output waveguide 40 Input Waveguide 41 Input side coupler 42a, 42b Exhibition Unit 43 phase difference adjustment unit 43a, 43 phase adjustment electrode 44 output side coupler 45 output waveguide 50 semi-insulating-InP substrate 51 first n-type electrode layer 52 first n-type cladding layer 53 first intermediate layer 54 Multiple quantum well core layer 55 second intermediate layer 56 first low-concentration cladding 57 p-type cladding layer 58 second n-type cladding layer 59 second n-type electrode layer 60 first n-type electrode 61 second n-type electrode 100 input waveguide 101 entrance couplers 102a and 102b arm waveguide 103 modulation sections 103a and 103b modulation electrodes 104 phase adjustment sections 104a and 104b phase adjustment electrode 105 exit side couplers 106a and 106b output waveguides

Claims (5)

Two couplers and two arm waveguides connecting them are provided, and a modulation electrode that modulates by applying a high frequency on the same arm waveguide and a phase adjustment electrode that adjusts the operating point by applying a DC voltage are provided A Mach-Zehnder type semiconductor optical modulator made of a semiconductor,
An input side coupler for branching light from the input waveguide;
A development part connected after the input side coupler for widening the distance between the two arm waveguides;
A modulation electrode that is connected after the expanding portion and modulates light;
A delay unit connected after the modulation electrode to reduce the interval between the two arm waveguides and equalize the propagation distance of the two arm waveguides;
A folded portion made of a curved waveguide connected after the delay portion and converting the propagation direction of light by 180 degrees;
A phase adjusting electrode for adjusting the phase of light connected after the folded portion;
An output-side coupler connected after the phase adjustment electrode to interfere with each light;
An output waveguide formed on the same surface as the surface on which an input waveguide that outputs light that is connected and interfered after the output-side coupler is formed ;
The delay unit is by reducing the distance between the two arm waveguides in the asymmetric semiconductor optical modulator according to claim equal to Rukoto the propagation distance of the two arm waveguides. Two couplers and two arm waveguides connecting them are provided, and a modulation electrode that modulates by applying a high frequency on the same arm waveguide and a phase adjustment electrode that adjusts the operating point by applying a DC voltage are provided A Mach-Zehnder type semiconductor optical modulator made of a semiconductor,
An input side coupler for branching light from the input waveguide;
A development part connected after the input side coupler for widening the distance between the two arm waveguides;
A modulation electrode that is connected after the expanding portion and modulates light;
A delay unit connected after the modulation electrode to reduce the interval between the two arm waveguides and equalize the propagation distance of the two arm waveguides;
A first folded part comprising a curved waveguide connected after the delay part and converting the propagation direction of light by 180 degrees;
A first phase adjustment electrode connected after the first folded portion and performing phase adjustment of light;
A second folded portion comprising a curved waveguide connected after the first phase adjusting electrode and converting the propagation direction of light again by 180 degrees;
A second phase adjustment electrode connected after the second folded portion and performing phase adjustment of light;
An output-side coupler connected after the second phase adjustment electrode to interfere with each light;
An output waveguide formed on a surface different from a surface on which an input waveguide that outputs light that is connected and interfered after the output coupler is formed ; and
The delay unit is by reducing the distance between the two arm waveguides in the asymmetric semiconductor optical modulator according to claim equal to Rukoto the propagation distance of the two arm waveguides. Two couplers and two arm waveguides connecting them are provided, and a modulation electrode that modulates by applying a high frequency on the same arm waveguide and a phase adjustment electrode that adjusts the operating point by applying a DC voltage are provided A Mach-Zehnder type semiconductor optical modulator made of a semiconductor,
An input side coupler for branching light from the input waveguide;
A development part connected after the input side coupler for widening the distance between the two arm waveguides;
A modulation electrode that is connected after the expanding portion and modulates light;
A delay unit connected after the modulation electrode to reduce the interval between the two arm waveguides and equalize the propagation distance of the two arm waveguides;
A folded portion made of a curved waveguide connected after the delay portion and converting the propagation direction of light by 180 degrees;
A phase adjusting electrode for adjusting the phase of light connected after the folded portion;
A second folded portion comprising a curved waveguide connected after the phase adjustment electrode and converting the propagation direction of light again by 180 degrees;
An output-side coupler that interferes with each light from the second folded portion;
An output waveguide connected after the output coupler and outputting the interfered light ,
The delay unit is by reducing the distance between the two arm waveguides in the asymmetric semiconductor optical modulator according to claim equal to Rukoto propagation distance between the two arm waveguides. Two semiconductor optical modulators according to claim 3;
An input-side coupler that splits light into the two semiconductor optical modulators;
A phase difference adjusting unit that is connected to the output waveguides of the two semiconductor optical modulators and gives a phase difference of 90 degrees between the output lights;
A semiconductor optical modulator comprising: an output-side coupler connected after the phase difference adjusting unit and causing interference of each light. The distance between the two arm waveguides having the modulation electrode is D1, and the distance between the two arm waveguides having the phase adjustment electrode is D2, and "D1>D2" is set. The semiconductor optical modulator according to claim 1.

Images