JP4991909B2 - Light control element - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an optical control element that enables stable operation at a low driving voltage, in particular an optical control element using two resonance type electrodes that enables stable operation even if a cross talk (coupling) between both the electrodes is produced. <P>SOLUTION: In an optical control element having a substrate 1 having an electro-optic effect, plural optical waveguides 2 formed on the substrate, and a control electrode 3 formed on the substrate for controlling the phase of light which propagates through the optical waveguide, the control electrode comprises at least resonance type electrodes 31 and 32 having the same resonance frequency; and power supply electrodes 41 and 42 that supply a control signal to each resonance type electrode. A shape and a formation position of each resonance type electrode, and a power supply position of the power supply electrode to each resonance type electrode are provided such that odd mode coupling with each other is enabled. Each resonance type electrode is supplied with an in-phase control signal by the power supply electrode. <P>COPYRIGHT: (C)2012,JPO&amp;INPIT

Description

  The present invention relates to a light control element, and more particularly to a light control element provided with a resonant electrode that modulates a light wave propagating through an optical waveguide.

  Resonates with transmitters of optical communication systems such as optical modulators for optical transmission of high-frequency signals of several GHz or more used for radio and pulsar modulators for optical clock generation used with data modulation in long-distance transmission. Light control elements such as type optical modulators are used. A resonant optical modulator uses a substrate material having an electro-optic effect such as lithium niobate, and uses a control electrode having a resonant electrode to change the refractive index of the optical waveguide formed on the substrate, It is configured to modulate the intensity and phase of light propagating through the optical waveguide.

  In the resonance type electrode, when an electric signal having a specific frequency is input from the feeding point, a standing wave of the electric signal is generated at the electrode. As described above, since the resonance type optical modulator uses the resonance phenomenon of the electric signal, it operates particularly efficiently when a specific frequency is inputted, and generally has a unit length longer than that of the traveling wave type optical modulator. Modulation efficiency per unit is good.

  Because of such characteristics, there are many examples in which the length of the electrode of the conventional resonant optical modulator is designed to be shorter than one wavelength of the electric signal. However, in the condition where the speed of light propagating through the optical waveguide and the speed of the control signal propagating through the electrode of the action part are approximately the same, the electrode can be lengthened and affected by the attenuation of the control signal. The driving voltage can be improved according to the length of the electrode.

  Non-Patent Document 1 discloses that the combined use of speed matching and a resonant electrode is effective for improving the efficiency, and Non-Patent Document 2 discloses a resonant electrode type optical modulator using lithium niobate as a substrate. It describes the case where good characteristics were obtained by setting the refractive index (nm) of the electrical signal to approximately 2.2 (the refractive index of lithium niobate with respect to light is approximately 2.2).

  On the other hand, under conditions where the speeds of light and electrical signals do not match, the length of the electrode cannot be made sufficiently long, and even if the modulation efficiency per unit length is high, the overall modulation efficiency can be improved as a result. Can not. For this reason, the half-wave voltage Vπ, which is a parameter representing the efficiency of the modulator, is approximately 10 V or higher, and sufficient operation cannot be obtained unless a very high voltage is applied.

  Further, if a part of the optical waveguide is branched into two paths to form a Mach-Zehnder (MZ) interferometer structure, if the two branched lights are made to interfere with each other, the optical waveguide operates as a light intensity modulator. In a long-distance transmission, a pulsar modulator for generating an optical clock used with data modulation, etc. can be obtained by operating the amount of phase change of the two split lights with the same magnitude and opposite phase change. It is desirable that the wavelength chirp is zero, and a configuration is adopted in which signals of the same magnitude and opposite signs are applied to the respective branch waveguides of the MZ interferometer. In addition, when signals having opposite signs are applied to the two electrodes corresponding to each of the branched waveguides (referred to as “two-electrode type”), there is an effect of lowering the driving voltage.

  From these facts, it is expected that a drastic reduction in driving voltage can be realized if a two-electrode type MZ modulator is configured using a long resonant electrode in which the speeds of light and control signals are matched. . However, since the resonance electrode has high efficiency and a very strong electric field for a signal having a resonance frequency, coupling with a surrounding conductive material (signal crosstalk) is remarkable. In a situation where the signals of both control electrodes are cross-talked, the phase of the control signal is disturbed and the desired light modulation cannot be obtained.

  In particular, signals are likely to be coupled to elements (components) that satisfy the resonance condition for the same frequency. For the two-electrode type MZ optical modulator, an electrode having the same basic structure is adopted as a control electrode corresponding to each branching waveguide. For this reason, when a resonant electrode is configured, each control electrode (resonant electrode) has the same resonant frequency, and coupling (signal crosstalk) becomes significant. In addition, when a long electrode with a speed matching is used, the influence of the coupling between the two electrodes becomes more remarkable because the electrode is long.

  Patent Document 1 discloses a countermeasure for preventing crosstalk of a control signal between a plurality of electrodes in the case of an optical modulator manufactured by forming a coplanar electrode on a substrate having an electro-optic effect such as lithium niobate. There is an example of forming a groove. In general, the strength of the control electric field decreases as the distance from the control electrode increases. This is a phenomenon that depends on the structure of the substrate and the electrode. As shown in FIG. 1, when the waveguide interval is about 150 μm, the electric field strength exerted on the other optical waveguide is about 1%. It is also disclosed that it is about 0.2% for about 300 μm and about 0.1% for about 400 μm.

  Thus, it is disadvantageous in terms of device size and cost to increase the distance between the two branch optical waveguides of the MZ interferometer. In addition, although the method of forming a groove shown in Patent Document 1 can be expected to improve, it is disadvantageous in terms of device manufacturing costs such as processing of an additional structure for prevention.

JP 2009-53444 A

Mark Yu and Anand Gopinath, "Velocity Matched ResonantSlow-Wave Structure for Optical Modulator", Proceedings of Integrated Photonics Research (IPR), ITuH7-1, pp.365-369, Palm Springs, California, March22, 1993 Roger Krahenbuhl and M. M. Howerton, "Investigations on Short-Path-Length High-Speed Optical Modulators in LiNbO3 With Resonant-Type Electrodes", JOURNAL OF LIGHTWAVE TECHNOLOGY, VOL.19, No. 9, pp.1287-1297, SEPTEMBER 2001

  The problem to be solved by the present invention is to solve the above-mentioned problems and to provide a light control element capable of stable operation with a low driving voltage, and in particular, using two resonant electrodes. An object of the present invention is to provide a light control element capable of stable operation even when crosstalk (coupling) between both electrodes occurs.

In order to solve the above problems, an invention according to claim 1 is directed to a substrate having an electro-optic effect, a plurality of optical waveguides formed on the substrate, and a phase of light that is provided on the substrate and propagates through the optical waveguide. The control electrode includes a control electrode for controlling the at least two resonance-type electrodes having the same resonance frequency, and a power supply electrode for supplying a control signal to each of the resonance-type electrodes. The optical waveguide constitutes one or a plurality of Mach-Zehnder interferometers, and the two resonant electrodes are arranged corresponding to the two branch waveguides constituting the Mach-Zehnder interferometer, and each resonant type The shape and formation position of the electrodes, and the power feeding position by the power feeding electrode to each resonance type electrode are set so that the odd mode coupling is possible, and a control signal having the same phase is fed to each resonance type electrode by the power feeding electrode. It is In a state where the control signal is fed, the resonant electrode is characterized that you bond odd mode together.

The invention according to claim 2 is the optical control device according to claim 1, the shape and formation position of the two resonant-type electrode, and the feeding position by the feeding electrode to the resonant electrode is a point-symmetric with It is set as follows.

According to a third aspect of the present invention, in the light control element according to the first or second aspect , the resonant electrode includes one signal electrode and a ground electrode surrounding the signal electrode, and the two end portions of the signal electrode Are either open to the ground electrode, shorted together, or one shorted and the other open.

According to a fourth aspect of the present invention, in the light control element according to any one of the first to third aspects, the resonant electrode has one signal electrode, and the length of the signal electrode has a predetermined frequency. The control signal is longer than the wavelength formed on the signal electrode.

According to a fifth aspect of the present invention, in the light control element according to any one of the first to fourth aspects, the resonant electrode has one signal electrode, and the power supply position to the signal electrode is a resonant type. It is a position where the impedance of the electrode is 50Ω.

The invention according to claim 6 is the light control element according to claim 5 , wherein the power supply position to the signal electrode is a position closest to the center of the signal electrode.

According to the first aspect of the present invention, a substrate having an electrooptic effect, a plurality of optical waveguides formed on the substrate, and a control electrode for controlling the phase of light provided on the substrate and propagating through the optical waveguide in the light control element having bets, the control electrode includes at least two resonant electrode having the same resonant frequency, and a feeding electrode for feeding a control signal to each of the resonant electrodes, the optical waveguide is a single Constituting one or a plurality of Mach-Zehnder interferometers, and the two resonance-type electrodes are arranged corresponding to two branching waveguides constituting the Mach-Zehnder interferometer, and the shape and formation position of each resonance-type electrode, In addition, the feeding positions of the resonance electrodes by the feeding electrodes are set so that odd-mode coupling is possible. The control signals of the same phase are fed to the resonance electrodes by the feeding electrodes, and the control signals are fed. The In state, because the resonant electrodes to bond odd modes together, even if generated crosstalk (bonds) between the resonant electrodes, the same amount of electric field energy and minute attached to the other electrode in the same phase Since it receives, it works in the same way as when there is no coupling, and a stable light modulation operation is possible. In addition, this is the same regardless of the magnitude of the coupling between both electrodes. Therefore, it is possible to provide a light control element that can stably operate with a low driving voltage.

Further , the optical waveguide constitutes one or a plurality of Mach-Zehnder interferometers, and the two resonant electrodes are arranged corresponding to the two branch waveguides constituting the Mach-Zehnder interferometer. It is possible to provide a light control element with a lower driving voltage, such as a two-electrode optical modulator using a mold electrode.

According to the invention of claim 2 , the shape and formation position of the two resonance electrodes, and the power supply position by the power supply electrode to each resonance electrode are set to be point-symmetric with each other. Even when crosstalk (coupling) occurs, electric field energy is always transmitted and received in the in-phase state, so that stable operation can be performed in the same manner as when there is no coupling.

According to the invention of claim 3 , the resonance electrode is composed of one signal electrode and a ground electrode surrounding the signal electrode, and are two ends of the signal electrode open to the ground electrode? Or both are short-circuited, or one is short-circuited and the other is open, so it is possible to form resonant electrodes of various wavelengths using the same length of signal electrode It becomes. In addition, when the two ends of the signal electrode are both opened or short-circuited together, the signal electrodes of each resonance electrode are arranged in parallel with the ends aligned. For this reason, the entire length when the two resonance-type electrodes are arranged side by side can be minimized. In addition, when used in a Mach-Zehnder interferometer, the position of the action portion where the signal electrode applies an electric field to each branching waveguide is symmetrical with respect to the optical axis that is the propagation direction of the light wave of the Mach-Zehnder interferometer. Therefore, it is possible to realize a light control element that makes the wavelength chirp zero.

According to the invention of claim 4 , the resonant electrode has one signal electrode, and the length of the signal electrode is longer than the wavelength that the control signal having a predetermined frequency forms on the signal electrode. It is possible to provide a light control element that can be driven at a low voltage.

According to the invention of claim 5 , the resonance electrode has one signal electrode, and the power supply position to the signal electrode is a position where the impedance of the resonance electrode is 50Ω. When a signal is input, reflection due to impedance mismatching or the like is suppressed, and a light control element with a low driving voltage can be provided.

According to the invention of claim 6 , in the light control element of claim 5 , the power feeding position to the signal electrode is a position closest to the center of the signal electrode. Furthermore, since the electric field intensity distribution exerted on the optical waveguide by each resonant electrode can be made substantially the same, wavelength chirp can be suppressed.

10 is a graph showing the relationship between electric field strength and distance (interval of optical waveguide) disclosed in Patent Document 1. It is a figure explaining the 1st Example (what open | released both ends of the signal electrode from the ground electrode) concerning the light control element of this invention. It is a figure which shows the general view of the electrode of the light control element of FIG. It is a figure which shows the cross-section of the light control element of FIG. It is a figure explaining the 2nd Example (what shorted both ends of the signal electrode to the ground electrode) concerning the light control element of the present invention. When the length of the signal electrode in the light control element of FIG. 2 is a half-wavelength λ / 2 (λ: signal wavelength), it is a diagram for explaining the relationship between the feeding position and the impedance and the state of the electric field vector at a specific timing. FIG. 7 is a diagram for explaining how a crosstalk phenomenon occurs in a lower resonance electrode that is slightly apart when a control signal is input only to the upper resonance electrode in the same case as in FIG. 6. FIG. 8 is a diagram for explaining how the crosstalk phenomenon occurs when the upper resonance electrode and the lower resonance electrode are close to each other in the same case as in FIG. 7. FIG. 7 is a diagram for explaining a state in which a crosstalk phenomenon occurs in the other resonance electrode that is slightly apart when a control signal is input to each resonance electrode in the same case as in FIG. 6. FIG. 10 is a diagram for explaining how the crosstalk phenomenon occurs when the upper resonance electrode and the lower resonance electrode are close to each other in the same case as in FIG. 9. When the length of the resonance type signal electrode (with both ends of the signal electrode open from the ground electrode) is 3λ / 2 (λ: signal wavelength), the relationship between the feeding position and the impedance, and the electric field vector at a specific timing It is a figure explaining a mode. FIG. 6 is a diagram for explaining the relationship between a feeding position and impedance and the state of an electric field vector at a specific timing when the length of a signal electrode in the light control element of FIG. 5 is λ (λ: signal wavelength). When the length of the resonant electrode signal electrode (with the left end of the signal electrode shorted and the right end open with respect to the ground electrode) is 3λ / 4 (λ: signal wavelength), the relationship between the feeding position and the impedance It is a figure explaining the mode of the electric field vector in specific timing. When the length of the signal electrode of the resonance electrode (with the left end of the signal electrode open and the right end short-circuited with respect to the ground electrode) is 3λ / 4 (λ: signal wavelength), the relationship between the feeding position and the impedance and It is a figure explaining the mode of the electric field vector in specific timing. It is a figure explaining an example which combined the resonance type electrode of FIG. 13 and FIG. It is a figure explaining the structure which made resonance type electrode unevenly distributed in a part of MZ interferometer. It is a figure explaining the structure which uses only a part of resonance type electrode as an action part which acts on an optical waveguide. It is a figure explaining a mode that the drive circuit which inputs a control signal to the light control element of the present invention has been arranged. It is another figure explaining a mode that the drive circuit which inputs a control signal is arrange | positioned to the light control element of this invention.

Hereinafter, the light control element of the present invention will be described in detail.
As shown in FIG. 2 or 5, the light control element of the present invention includes a substrate 1 having an electro-optic effect, a plurality of optical waveguides 2 formed on the substrate, and the optical waveguide provided on the substrate. In the light control element having the control electrode 3 for controlling the phase of the light propagating light, the control electrode includes at least the resonance electrodes 31 and 32 having the same resonance frequency, and a control signal to each of the resonance electrodes. The optical waveguide 2 constitutes a Mach-Zehnder interferometer, and the two resonance-type electrodes 31 and 32 are connected to two branch waveguides constituting the Mach-Zehnder interferometer. arranged corresponding shape and the formation position of each resonance electrode, and the feeding position by the feeding electrode to the resonant electrode is set to allow odd-mode coupled together, each resonant electrode, said In-phase with feed electrode Control signal is powered, in a state where the control signal is fed, the resonant electrode is characterized that you bond odd mode together.

  As the substrate 1 having an electro-optic effect, for example, lithium niobate, lithium tantalate, PLZT (lead lanthanum zirconate titanate), quartz-based materials, and combinations thereof can be used. In particular, lithium niobate (LN) or lithium tantalate (LT) crystals having a high electro-optic effect are preferably used.

  The optical waveguide can be formed by a method of forming a ridge on the substrate, a method of adjusting the refractive index of a part of the substrate, or a method of combining both. In the ridge type waveguide, other portions are removed by mechanical cutting or chemical etching so as to leave a substrate portion to be an optical waveguide. It is also possible to form grooves on both sides of the optical waveguide. In the method of adjusting the refractive index, the refractive index of a part of the substrate surface corresponding to the optical waveguide becomes higher than the refractive index of the substrate itself by using a thermal diffusion method such as Ti or a proton exchange method. Configure as follows.

Control electrodes such as signal electrodes and ground electrodes can be formed by forming a Ti / Au electrode pattern, a gold plating method, or the like. Also, each electrode is optionally arranged via a buffer layer such as SiO ₂ film between a substrate. The buffer layer has an effect of preventing light waves propagating through the optical waveguide from being absorbed or scattered by the control electrode. Moreover, as a structure of a buffer layer, in order to relieve the pyroelectric effect of a thin plate, it is possible to incorporate a Si film or the like as necessary.

The light control element of the present invention employs a configuration in which at least two resonance electrodes are formed on the control electrode, and the modulation efficiency of the control signal (modulation signal) is not affected even if both resonance electrodes cross-talk. is doing. For this purpose, the following two requirements are required.
(1) Both resonance-type electrodes have basically the same shape and are in a condition of being coupled with odd (symmetric) modes.
(2) An in-phase signal is supplied to both resonant electrodes.

  The resonant electrode is mainly composed of one signal electrode and a ground electrode surrounding it. As a combination of the two resonance-type electrodes, as described later, both ends of the signal electrode are both open from the ground electrode, or both ends of the signal electrode are short-circuited to the ground electrode. The combination of “short-circuited at both ends—short-circuited at both ends” is one of the preferred embodiments. It goes without saying that various combinations such as “short-circuited at both ends—one short-circuited and other open” are possible.

  Since the two resonance-type electrodes are basically the same shape electrodes, the resonance frequency is the same, and it is easy to combine them. Normally, when this electrode is arranged so as to be symmetric with respect to the central axis (light propagation direction) of the MZ interferometer, it is supplied to the signal electrode by the position on one signal electrode constituting the resonance electrode. Since the state (the direction of the electric field vector) between the electric field generated by the control signal and the electric field generated by the crosstalk between the electric field formed by the other resonant electrode is different, the control signal interferes in a complicated manner, and the resonant electrode (particularly, The normal propagation of the control signal on one signal electrode) is hindered.

  On the other hand, if both resonant electrodes are placed under the condition that odd-mode coupling occurs even when coupled, and a large signal with the same phase is fed to each other, the same amount of electric field energy as that coupled to the other is fed to the same phase. It works in the same way as when there is no connection. This is the same regardless of the strength of the coupling between both electrodes.

  In order to arrange so that the odd mode coupling is achieved even if both resonant electrodes are coupled, the configuration is as shown in FIG. In other words, each resonance electrode has the same shape, and rotates any point located on a plane equidistant from the action part of the MZ interferometer (the part where the electric field formed by each resonance electrode acts on the optical waveguide). As a center, it is arranged at a position that is 180 ° rotationally symmetric (point symmetry, see fixed point O in FIG. 2). A control signal having the same phase and the same magnitude is fed to the feeding point (feeding position) of one signal electrode (31, 32) of each resonance type electrode using the feeding electrodes 41,.

  FIG. 6 shows the relationship between the feeding position and the impedance and the electric field at a specific timing when the length L of the resonant electrode (signal electrode) in the light control element of FIG. 2 is half wavelength λ / 2 (λ: signal wavelength). It is a figure explaining the mode of a vector (arrow). As is apparent from the graph showing the relationship between the power feeding position and impedance in FIG. 6, when both ends of the signal electrodes (resonance type electrodes) 31 and 32 are open from the ground electrode 33, the length of the signal electrode However, when the signal wavelength is a half wavelength, there are two feeding positions where the impedance is 50Ω.

  In addition, if the power supply position is different even if an in-phase control signal is input, the electric field vector at a specific timing, such as the electric field vectors shown in the upper resonance electrode and the lower resonance electrode in FIG. Are opposite to each other.

  FIGS. 7 and 8 are diagrams for explaining the formation state of the electric field vector when the control signal is fed only to the upper resonance electrode. The electric field vector formed on the lower resonance electrode is the upper resonance electrode. This is due to crosstalk (coupling) from the mold electrode. Thus, in crosstalk, a signal (electric field vector) whose phase is shifted by π is excited. The magnitude of the excited electric field vector is also larger when the distance between the two resonant electrodes is narrow (FIG. 8).

  Here, a control signal is supplied as shown in FIG. 6, and an electric field vector generated at each resonance electrode and an electric field vector (FIG. 7 or 8) where one resonance electrode excites the other resonance electrode. As shown in FIG. 9 or FIG. 10, even if crosstalk (coupling) occurs between both resonance-type electrodes, an odd (symmetric) mode is obtained. The control signal (electric field vector by feeding) is not disturbed, and desired light modulation can be performed. Naturally, the effect of crosstalk becomes significant when the distance between the resonant electrodes is narrow as shown in FIG. 10, but the electric field generated by feeding the signal electrode and the electric field excited by the crosstalk are in the same direction. Therefore, there is no disturbance of modulation due to crosstalk.

  As shown in FIG. 11, the resonance electrode whose signal electrode (resonance type electrode) 31, 32 is longer than the wavelength of the control signal has a specific impedance value and performs the same resonance operation. There are a plurality of excitation points (feeding positions). Therefore, as for the power feeding position, any excitation point may be used as long as resonance of the same condition is excited. However, when the feeding positions of the two resonance electrodes are close to each other, the intensity distribution of the electric field formed by each resonance electrode in the propagation direction of the optical waveguide is almost the same. In this case, the wavelength chirp can be made zero, which is more preferable.

  FIG. 3 simply shows the configuration of an actual electrode used in the light control element of the present invention. Since it has a two-electrode structure, a Z-cut LN substrate is most suitable as the substrate. The optical waveguide has an MZ interferometer type shape. In the prior art, in order to reduce the influence of control signal interference, it is necessary to increase the distance between the branch optical waveguides so that the distance between the signal electrodes (hot electrodes) is 400 μm or more. This is not necessary for the control element. Rather, since the light control element itself can be reduced in size by narrowing the waveguide interval, it may be 100 μm or less where crosstalk (coupling) becomes significant in the conventional resonance type modulation.

  The shape of the resonant electrode is preferably a coplanar (CPW) structure (a configuration in which the signal electrode is sandwiched by the ground electrode) so that the speed of the optical signal propagating through the optical waveguide and the speed of the control signal propagating through the electrode are approximately equal. Constitute. When the propagation speeds of the two are substantially equal (when the speed matching condition is substantially satisfied), the length of the signal electrode (resonance type electrode) can be made longer than the wavelength of the resonance frequency of the control signal, thereby reducing the drive voltage. Is advantageous.

  The resonant electrode is not limited to the above-described CPW structure, but CPS (a configuration in which a ground electrode is provided on one side of the signal electrode), G-CPW (a CPW is formed on the surface of the substrate, and a ground electrode on the back surface of the substrate) Various configurations can be employed such as a configuration of Various configurations can be adopted for the power supply electrode as well as the resonance type electrode. However, in order to facilitate electrical connection between the power supply electrode and the resonance type electrode, it is preferable to employ the same type of configuration. Further, if necessary, the power supply electrode may be provided with a capacitor, a resistor, or the like in the middle of the power supply electrode, and a filter circuit or the like may be provided.

  In the light control element of the present invention, as shown in FIG. 4 (cross-sectional view of FIG. 3), a ridge type optical waveguide in which a control signal is effectively applied to the optical waveguide portion is used for the purpose of further reducing the drive voltage. . Of course, the optical waveguide can be used not only to form a substrate in a ridge shape but also to adjust the refractive index by subjecting the ridge portion to thermal diffusion of Ti, if necessary. Further, any electrode type or optical waveguide may be used as long as the velocity matching is substantially satisfied, regardless of whether the electrode has a non-coplanar (CPW) structure or a non-ridge waveguide. .

  2 or 3, the signal electrodes (resonance type electrodes) 31 and 32 are of the type in which both ends are open from the ground electrode. The feeding position is not at the center of the signal electrode (resonance type electrode) but at an asymmetric position. Here, the position closest to the center of the electrode among the positions where the impedance on the signal electrode (resonance type electrode) is 50Ω (usually 50Ω is generally not limited to this impedance value) is fed. As a point, the drive circuit directly feeds the control signal without using the impedance matching circuit. This is because, compared with the case where power is supplied to the end of the electrode, there is a problem of reproducibility of the manufacturing process, and there is less change in characteristics when the shape of the end of the electrode varies.

  Next, FIG. 5 shows a second embodiment of the light control element of the present invention. The difference from the embodiment of FIG. 2 is that in the embodiment shown in FIG. 2, both ends of the signal electrode (resonance type electrode) are opened from the ground electrode, but in FIG. 5, the both ends are short-circuited to the ground electrode. ing.

  In the embodiment of FIG. 5 as well, as in FIG. 2, both resonant electrodes have the same shape, and are arranged at 180 ° rotational symmetry (point symmetry) with the center of the MZ interferometer as the center of rotation. However, control signals having the same phase and the same magnitude are fed. Under this condition, the electric field at the position of each signal electrode (resonance type electrode) 31, 32 is an electric field having the same magnitude and opposite sign. For this reason, if the electrode arrangement conditions and the control signal power supply conditions as shown in FIG. 5 are satisfied, even if coupling occurs between both resonance-type electrodes, an odd (symmetric) mode is obtained. The control signal on the (resonance type electrode) is not disturbed.

  FIG. 12 is a diagram for explaining the relationship between the feeding position and impedance and the state of the electric field vector at a specific timing when the length L of the signal electrode in the light control element of FIG. 5 is λ (λ: signal wavelength). . As is apparent from the graph showing the relationship between the power feeding position and impedance in FIG. 12, the length of the signal electrode in the case where both ends of the signal electrodes (resonance type electrodes) 31 and 32 are short-circuited to the ground electrode 33 is shown. When L is one wavelength of the signal wavelength, there are four feeding positions where the impedance is 50Ω.

  In addition, even if an in-phase control signal is input, if this feeding position is in the relationship (point symmetry) as shown in FIG. 12, the upper resonance type electrode and the lower resonance type electrode shown in FIG. Like the electric field vector, the electric field vectors at the specific timing are opposite to each other. It is also possible to select a power feeding position that is close to the end of the signal electrode (resonance type electrode) as the power feeding position, but when the shape of the end of the electrode varies due to the reproducibility of the manufacturing process. In addition, the change in the characteristics of the two becomes large, and the electric field intensity distribution formed by the two resonant electrodes in the light propagation direction is likely to be different, which is not preferable from the viewpoint of reducing the wavelength chirp to zero.

  Further, some of the signal electrodes constituting the resonance electrode open one end and short the other end with respect to the ground electrode. As shown in FIGS. 13 and 14, when the length L of the signal electrode of the resonance electrode is 3λ / 4 (λ: signal wavelength), the relationship between the feeding position and the impedance and the electric field vector at a specific timing are shown. FIG. 13 shows a state in which the left end of the signal electrode is short-circuited and the right end is opened with respect to the ground electrode. FIG. 14 shows an open state of the left end of the signal electrode with respect to the ground electrode. The right end is short-circuited.

  As shown in FIGS. 13 and 14, there are three feeding positions where the impedance is 50Ω even with the resonant electrode having such a shape. In addition, the electric field vector at a specific timing generated in the resonant electrode can be selected in different directions depending on the feeding position.

  However, even if one of the resonance electrodes illustrated in FIG. 13 and one of the resonance electrodes illustrated in FIG. As shown in FIG. 15, it is necessary to dispose both of them, the light control element itself becomes large, and light modulation is performed at different positions with respect to the propagation direction of the light wave, so that the wavelength chirp is made zero. It is difficult.

  When the arrangement of the resonance type electrodes is not point-symmetric, it is possible to perform odd mode coupling by combining the cases 1 and 5 in FIG. 13, for example.

  In the above description, the linear signal electrode (resonance type electrode) has been mainly described. However, the present invention is not limited to this, and when the branched waveguide is curved or bent, it resonates according to the waveguide. The electrode may be bent or bent. In addition, the signal electrodes (resonant electrodes) are not limited to a linear type but a ring type as long as the operation efficiency of the branched optical waveguides is the same and the requirements of being arranged at positions where they are odd-mode coupled to each other are satisfied. There is no problem.

  Further, as described above, the position of the resonance electrode with respect to the optical waveguide constituting the MZ interferometer is not limited to the arrangement of the two resonance electrodes so as to coincide with the center of the MZ interferometer. 16, the resonance type electrodes 31 and 32 are configured to be unevenly distributed in a part of the branch waveguide of the MZ interferometer, or only a part of the resonance type electrode is branched as shown in FIG. It is also possible to arrange so that it overlaps with the waveguide, and the resonance type electrode is constituted by an action part (range S) acting on the optical waveguide and a non-action part. In the case as shown in FIG. 17, the shape and arrangement of the resonance electrode of the non-acting part increases the degree of design freedom compared to the case where the entire resonance electrode described above becomes the action part.

  In order to drive the light control element of the present invention, control signals having the same phase and the same frequency are applied to the resonance electrodes. As shown in FIG. 18, a control signal (arrow) is input to the power supply electrodes 41 and 42 of the resonance electrodes 31 and 32 using two systems of drive circuits. As an example of the drive circuit, a signal having a predetermined frequency from a signal source is input to a driver 1 (driver 2), amplified to a predetermined signal voltage, and then passed through a band filter 1 (band filter 2) that removes noise. Input to the power supply electrodes 41 and 42 of the light control element. In addition, as shown in FIG. 19, it is also possible to divide the control signal into two by using one drive circuit and to supply the control signals to the resonance electrodes 31 and 32. However, in this case, in order to adjust the phase of the control signal to be fed, it is desirable to interpose the phase adjuster on at least one of the feed lines. Of course, it is possible to adjust the length of the feeder line in advance to be in phase, and in that case, the phase adjuster can be omitted.

  When the light control element of the present invention employs an MZ interferometer for the optical waveguide, the amount of phase change of the light in each branch optical waveguide has the same reverse sign, and the wavelength at the optical output unit of the MZ interferometer An ON / OFF pulsed optical signal without chirp is generated. This ON / OFF pulse optical signal without wavelength chirp is the most desirable characteristic as an optical clock.

  Furthermore, in the light control element of the present invention, it has become possible not only to use a resonance-type long electrode and a two-electrode arrangement structure, which has been impossible to be used at the same time, but also to perform crosstalk (coupling). By using this, it is possible to obtain effects such as dramatic improvement in modulation efficiency and miniaturization by reducing the gap between electrodes.

When the light control element of the present invention is applied to a pulser, the following effects can be expected.
・ Realization of high-speed, ultra-low voltage, small pulsar ・ Innovative reduction in power consumption ・ Use of low-cost drive system reduces user costs Cost reduction by

  As described above, according to the present invention, an optical control element capable of stable operation with a low driving voltage is provided, and in particular, crosstalk between both electrodes using two resonance electrodes. It is possible to provide a light control element capable of stable operation even when (coupling) occurs.

DESCRIPTION OF SYMBOLS 1 Substrate having electro-optic effect 2 Optical waveguide 21, 22 Branched waveguide 3 Control electrode 31, 32 Signal electrode (resonance type electrode)
33 Ground electrode 41, 42 Feed electrode

Claims (6)

In a light control element having a substrate having an electro-optic effect, a plurality of optical waveguides formed on the substrate, and a control electrode provided on the substrate for controlling the phase of light propagating through the optical waveguide,
The control electrode includes at least two resonance electrodes having the same resonance frequency, and a feeding electrode that supplies a control signal to each of the resonance electrodes,
The optical waveguide constitutes one or a plurality of Mach-Zehnder interferometers, and the two resonant electrodes are arranged corresponding to two branching waveguides constituting the Mach-Zehnder interferometer,
The shape and formation position of each resonance electrode, and the power supply position by the power supply electrode to each resonance electrode are set so that odd mode coupling is possible with each other,
A control signal having the same phase is fed to each resonance electrode by the feeding electrode ,
In a state where the control signal is powered, the light control element characterized Rukoto the resonant electrodes to bond odd mode together. 2. The light control element according to claim 1 , wherein the shape and formation position of the two resonance electrodes and the power supply position by the power supply electrode to each resonance electrode are set to be point-symmetric with each other. Characteristic light control element. 3. The light control element according to claim 1, wherein the resonance-type electrode includes one signal electrode and a ground electrode surrounding the signal electrode, and two ends of the signal electrode are connected to the ground electrode. A light control element characterized in that it is either open or shorted together, or one is shorted and the other is open. In the optical control element according to any one of claims 1 to 3, the resonant electrode has one signal electrode, the length of the signal electrode, the control signal is the signal electrode on having a predetermined frequency A light control element characterized in that the light control element is longer than the wavelength to be formed. In the optical control element according to any one of claims 1 to 4, the resonant electrode has one signal electrode, the feeding position to the signal electrode, the position where the impedance of the resonant electrode is 50Ω A light control element characterized by the above. 6. The light control element according to claim 5 , wherein a power feeding position to the signal electrode is a position closest to a center of the signal electrode.

Images