JP6394243B2 - Optical waveguide device - Google Patents

Description

  The present invention relates to an optical waveguide device, and more particularly to an optical waveguide device including an optical waveguide formed on a substrate and a control electrode for controlling a light wave propagating in the optical waveguide.

  In recent years, in the fields of optical communication and optical measurement, a waveguide type including an optical waveguide formed on a substrate having an electro-optic effect and a control electrode for controlling a light wave propagating in the optical waveguide. Many optical waveguide elements such as optical modulators are used.

As one of such waveguide type optical elements, for example, a ferroelectric crystal lithium niobate (LiNbO ₃ ) (also referred to as “LN”) is used as a substrate having an electro-optic effect, and a Mach-Zehnder type optical waveguide and a control electrode are used. A Mach-Zehnder type optical modulator having a structure is widely used. The Mach-Zehnder type optical waveguide includes an incident waveguide for introducing light from the outside, two parallel waveguides for propagating the light introduced by the incident waveguide in two paths, and the two This is an optical waveguide pattern generally referred to as a Mach-Zehnder interferometer type, which is composed of an output waveguide for multiplexing and outputting the light propagated through the parallel waveguides to the outside. The Mach-Zehnder optical modulator includes a control electrode for controlling the phase of the light wave propagating in the parallel waveguide by changing the electro-optic effect.

This control electrode is generally an RF (high frequency) signal electrode (hereinafter referred to as “RF electrode”) formed on or near the parallel waveguide, and a ground that runs parallel to the RF electrode. It consists of electrodes. Especially in applications that perform high-speed optical modulation of several tens of Gb / s or more, the propagation speed of light propagating through the optical waveguide and the propagation of high-speed (high-frequency) modulation signals propagating through the control electrode formed along the optical waveguide Since it is necessary to match the speed and to match the electrical characteristics with the electric circuit such as the drive driver, it is one of the determinants of the appropriate adjustment of the propagation speed and the electrical characteristics. A buffer layer such as SiO ₂ is often provided between the LN substrate surface and the control electrode in order to achieve both an appropriate adjustment range of the characteristic impedance of a certain control electrode.

  In this case, the distance from the control electrode on the buffer layer to the optical waveguide in the LN substrate increases, and the electric field applied to the optical waveguide is weakened. As a result, the half-wave voltage (referred to as Vπ) of the Mach-Zehnder type optical modulator increases, and the drive voltage to be applied to the control electrode increases.

  One way to prevent such an increase in drive voltage is to reduce the substrate thickness (for example, 10 μm or less) instead of using a buffer layer. It is known to secure an adjustment range (for example, see Non-Patent Documents 1 to 4). However, when such a buffer layer is not used, the optical waveguide may be in direct contact with the metal of the wiring pattern for applying a high-speed (high-frequency) modulation signal to the control electrode from the side of the control electrode and the substrate. In that case, an absorption loss of light occurs at the contact portion.

  When an X-cut LN substrate is used, the control electrode (RF electrode and ground electrode) is generally formed in parallel with the waveguide on the side of the parallel optical waveguide. Even in the case where the material constituting the metal is, for example, a metal, absorption of light propagating through the parallel optical waveguide by the metal can be avoided if the distance between the electrode and the optical waveguide is sufficient. However, it is inevitable that the wiring pattern intersects the optical waveguide formed on the substrate, and it may be necessary to reduce the optical loss at the intersection between the wiring pattern and the optical waveguide. In particular, in a wiring pattern of a coplanar control electrode that runs parallel with a signal electrode sandwiched between ground electrodes, the area where the ground electrode of the wiring pattern and the optical waveguide overlap increases, which can reduce the light absorption loss that can occur. It is highly desired. In addition, for example, even in a more complicated optical waveguide structure including a plurality of Mach-Zehnder optical waveguides and control electrodes such as a nested modulator using an optical waveguide, there is a region where the ground electrode and the optical waveguide of the wiring pattern overlap each other. It is highly desirable to reduce the light absorption loss that can be increased.

  As a technique for reducing the optical loss at the intersection between the ground electrode and the optical waveguide, a configuration in which a portion where the ground electrode straddles the optical waveguide is a narrow metal pattern is known (Patent Document 1). . In addition, the ground electrode is formed so as to be divided into two parts in the immediate vicinity of the optical waveguide, and the two parts of the divided ground electrode are formed so as to straddle the optical waveguide so as not to cover the optical waveguide. A configuration in which a bridge connection is made by using bonding or a conductor is known (Patent Document 1).

  In the above configuration, the coupling between the bridge connection portion and the signal line is weak, and the characteristic impedance in the bridge connection section is higher than the characteristic impedance in the other sections. If the length of the bridge connection section (gap width of the divided portion of the ground electrode) is larger than the mode field diameter (about 10 μm) of light propagating through the optical waveguide passing through the gap section, the light loss in the section is large. When it is reduced to about twice the mode field diameter of light (that is, about 20 μm), the optical loss is reduced to almost negligible level. On the other hand, even if the length of the bridge connection section is about 20 μm, this length is smaller than the wavelength of the high-frequency signal propagating through the signal line (for example, about 1.3 mm at 110 GHz), so that the broadband characteristics as a signal transmission path The flatness of the reflection characteristic S11 and the transmission characteristic S21 does not deteriorate greatly even if there is a discontinuity in characteristic impedance in the bridge connection section.

  However, in the high-frequency signal line, the discontinuous part of the characteristic impedance becomes a signal scattering part, and in particular, radiation loss of the signal can occur. Further, as the size of the bridge connection portion (the height of the connection wire or ribbon, the height of the loop or the connection distance) is increased, this radiation loss increases, and there may be a problem that crosstalk with other adjacent electrodes increases. . For example, in polarization multiplexing QPSK (Quadrature Phase Shift Keying) modulation, which is currently mainstream, a loss of about 0.1 dB of the electrical signal of each tributary is an acceptable value, but 0.1 dB between the tributary signals. Crosstalk is a very serious problem.

  Furthermore, for example, in a Mach-Zehnder type optical modulator, control is performed on the waveguide due to the positional relationship between the control electrode, the wiring pattern, and the electrode pad provided on the substrate, and the waveguide pattern. A bent portion of the electrode or the wiring pattern may be formed. In such a bent portion of the control electrode or the wiring pattern, a radiation loss of an electric signal can be caused due to the bent shape itself of the portion.

  That is, when the bent portion of the electrode pattern is formed on the optical waveguide, the bridge connection is performed so as to straddle the optical waveguide at the bent portion, and the characteristic impedance discontinuity due to the bridge connection and the bending Due to the shape, the radiation loss of the signal in the part can be further increased. For this reason, the crosstalk is further increased, and the drive voltage must be increased to compensate for the signal amplitude lost due to the radiation loss.

  As a technique for reducing radiation loss at a bent portion in a wiring pattern having a coplanar structure, conventionally, a configuration in which the radius of curvature of a bent portion is set to a predetermined value or more (Patent Document 2), and a gap between a signal line and a ground electrode are adjusted. Various techniques such as a configuration (Patent Document 3) are known. However, these configurations are exclusively for preventing radiation loss, and when the bent portion of the electrode pattern intersects the optical waveguide, in addition to the means for preventing radiation loss, the bent shape is used. Therefore, it is necessary to provide a means for preventing signal radiation loss associated with (for example, using the configuration described in Patent Document 1), and the manufacturing process becomes complicated.

JP2013-242592A Japanese Patent No. 3548042 JP 2010-72129 A

E. Yamashita, and K. Atsuki, "Distributed Capacitance of a Thin-Film Electrooptic Light Modulator," Microwave Theory and Techniques, IEEE Transactions on, vo.no. 23 (1), pp. 177-178. (1975). T. Sueta, and M. Izutsu, "High speed guided-wave optical modulators," J. Opt. Commun., Vol. 3, pp. 52-58 (1982). K. Atsuki, and E. Yamashita, "Transmission Line Aspects of the Design of Broad-Band Electrooptic Traveling-Wave Modulators," Journal of Ligihtwave Technology, vol. LT-5, pp. 316-319 (1987). J. Kondo, K. Aoki, A. Kondo, T. Ejiri, Y. Iwata, A. Hamajima, T. Mori, Y. Mizuno, M. Imaeda, Y. Kozuka, O. Mitomi, and M. Minakata, " High-Speed and Low-Driving-Voltage Thin- Sheet X Cut LiNbO3 Modulator With Laminated Low-Dielectric-Constant Adhesive, "IEEE Photon. Tech. Lett., Vol. 17, pp. 2077-79 (2005).

  From the above background, in the optical waveguide element, it is possible to prevent radiation loss at the bent portion of the electrode pattern and prevent light absorption loss at the optical waveguide that passes through the lower portion of the electrode of the bent portion without complicating the manufacturing process. Therefore, realization of an element structure is required.

One aspect of the present invention includes an optical waveguide formed on a substrate, a control electrode 120 formed on the substrate for controlling a light wave propagating through the optical waveguide, and a coplanar having the control electrode 120 as a central conductor. An optical waveguide device comprising two ground electrodes 130 and 132 constituting a type line, wherein one ground electrode 132 located inside the bent part and the control electrode 120 at a bent part of the coplanar type line. The speed of the slot mode propagating between the first and second electrodes is lower than the speed of the slot mode propagating between the one ground electrode 130 located outside the bent portion and the control electrode.
Another aspect of the present invention is an optical waveguide formed on a substrate, a control electrode formed on the substrate for controlling a light wave propagating through the optical waveguide, and a coplanar line having the control electrode as a central conductor. An optical waveguide device comprising two ground electrodes, wherein, in a bent portion of the coplanar line, a side wall of the one ground electrode located inside the bent portion is opposed to the control electrode. A predetermined number of bridges having predetermined size openings and predetermined depths are formed.
According to another aspect of the present invention, in the bent portion, a bridge having a predetermined size opening and a predetermined depth on a side wall facing the control electrode of another ground electrode located outside the bent portion. Are formed in a predetermined number.
According to another aspect of the present invention, at least one of the bridge provided on the one ground electrode and the bridge provided on the other ground electrode includes all or one of the optical waveguides formed on the substrate. It is formed so as to straddle the part.
According to another aspect of the present invention, the substrate is made of a material having an electro-optic effect, and the optical waveguide is a Mach-Zehnder type optical waveguide.
According to another aspect of the invention, the thickness of the substrate is 50 μm or less.
According to another aspect of the present invention, the control electrode is formed using a plating method, and the thickness of the control electrode is 30 μm or more.
According to another aspect of the present invention, the optical waveguide includes an optical waveguide constituting a directional coupler, an optical waveguide intersecting with each other, or an optical waveguide having a Y branch portion.

It is a figure which shows the structure of the optical waveguide element which concerns on embodiment of this invention. It is an AA cross-sectional arrow view of the optical waveguide element shown in FIG. FIG. 2 is a cross-sectional view of the optical waveguide element shown in FIG.

Embodiments of the present invention will be described below with reference to the drawings.
FIG. 1 is a diagram showing the structure of an optical waveguide device according to an embodiment of the present invention.
The present optical waveguide device 100 is a Mach-Zehnder type optical modulator formed on a substrate 102, and includes an incident waveguide 110 for introducing light from the outside and two paths for the light introduced by the incident waveguide 110. A Mach-Zehnder type optical waveguide composed of two parallel waveguides 112 and 114 to be propagated separately and an output waveguide 116 for combining and outputting the light propagated through the parallel waveguides 112 and 114 (In FIG. 1, each waveguide is indicated by a thick broken line).

  In addition, the optical modulator 100 includes a high-frequency signal electrode (RF electrode) 120 formed between the parallel waveguides 112 and 114 at a predetermined distance from the parallel waveguides 112 and 114. The RF electrode 120 is provided in parallel with the parallel waveguides 112 and 114 over a section of a predetermined distance, and extends in a downward direction in the figure at the left and right end portions of the section, and externally at the lower end of the board 102 in the figure. Connected to the control circuit.

The substrate 102 is, for example, an LN (lithium niobate, LiNbO ₃ ) substrate, and the waveguides 110 to 116 are deposited on the substrate 102 so as to form a desired waveguide pattern by a technique such as photolithography and sputtering. The formed metal titanium (Ti) can be formed by thermally diffusing into the substrate 102. The formation of the waveguide on the substrate 102 using LN is not limited to the Ti thermal diffusion method described above, and can be performed by other known methods such as a proton exchange method. In addition to the above LN, the substrate 102 has an electro-optic constant capable of inducing a required refractive index change by applying an electric field, such as lithium tantalate, PLZT (lead lanthanum zirconate titanate), and an electro-optic polymer. It can be used as a material.

  The RF electrode 120 may be a metal electrode formed of a metal such as Au or Al with Ti or Cr as a base, for example.

  In this embodiment, the LN substrate used as the substrate 102 is an X-cut substrate, and it is necessary to apply an electric field in a direction parallel to the substrate surface with respect to the waveguide. For this reason, the RF electrode 120 is not directly above the parallel waveguides 112 and 114 (that is, not formed so as to overlap), but is located at a predetermined distance from the parallel waveguides 112 and 114 as described above. Is formed.

  On the substrate 102, ground electrodes 130 and 132 are formed on both sides of the RF electrode 120 in the width direction so as to be spaced apart from the RF electrode 120 by a predetermined distance so as to sandwich the parallel waveguides 112 and 114, respectively. As a result, the RF electrode 120 forms a coplanar type line with the ground electrodes 130 and 132, and an electric field is applied to the parallel waveguides 112 and 114 sandwiched between the RF electrode 120, the ground electrode 130, and the ground electrode 132. The phase of the light wave propagating through the parallel waveguides 112 and 114 is controlled.

  In addition, bridges 160 and 162 are provided at portions of the RF electrode 120 that intersect the parallel waveguide 114 so as to straddle the parallel waveguide 114. However, since these intersections have a short distance (usually 4 to 30 μm) and the amount of light absorption loss at the intersections is small, the formation of the bridges 160 and 162 can be omitted.

  Further, at the corners on the left and right sides of the ground electrode 132 located inside the bent portions on the left and right sides of the RF electrode 120 in the drawing (that is, on the side close to the center of curvature), openings of predetermined sizes are respectively provided. Five bridges (cavities) 140a, 140b, 140c, 140d, 140e (hereinafter abbreviated as 140a-e) having a predetermined depth, and bridges 142a, 142b, 142c, 142d, 142e (hereinafter, 142a-e). (Abbreviated as “dotted line”).

  Further, the left and right curved portions of the ground electrode 130 located on the outside of the left and right curved portions of the RF electrode 120 in the drawing (that is, the side far from the center of curvature) have three bridges 150a and 150b, respectively. , 150c (hereinafter abbreviated as 150a to c) and bridges 152a, 152b and 152c (hereinafter abbreviated as 152a to c) (respectively indicated by dotted lines).

  2A is an AA cross-sectional view taken along the dashed line in the figure, and shows details of a portion where the bridges 140a to 140e are formed at the left corner of the ground electrode 132 shown in FIG. As illustrated, the bridges 140 a to 140 e are formed on the boundary surface with the substrate 102 in the ground electrode 132 formed on the substrate 102. The bridges 142a to 142e provided at the right corner of the ground electrode 132 shown in FIG. 1 have the same configuration as that in FIG. 2A.

  2B is a BB cross-sectional view taken along the two-dot chain line in the drawing, and shows details of a portion where the bridges 150a to 150c are formed in the bent portion on the left side of the ground electrode 130 shown in FIG. As illustrated, the bridges 150 a to 150 c are formed on the boundary surface with the substrate 102 in the ground electrode 130 formed on the substrate 102. The bridge 150b is formed to have a predetermined size opening and a predetermined depth, and the bridges 150a and 150c are formed so as to straddle the parallel waveguides 112 and 114, respectively. By these bridges 150a and 150c, the light in the parallel waveguides 112 and 114 can propagate without receiving absorption loss due to the metal material of the ground electrode 130.

  In particular, the portion where the optical waveguide draws a curve, such as the parallel waveguides 112 and 114 covered by the ground electrode 130, is a portion that easily loses part of the propagation light as leakage light compared to the straight portion. It is preferable to avoid applying strain and stress to the substrate 102 in the portion as much as possible. In the optical waveguide device 100 of the present embodiment, the bridges 150a, 150c, 150c, 150c, 150c, 150c, 150c, and 150c are not used for the ground electrode 1320 that covers the curved portions of the parallel waveguides 112 and 114, without using means such as wire bonding. Since the contact between the ground electrode 130 and the parallel waveguides 112 and 114 is avoided by the 152a and 150c, the light propagation state in the parallel waveguides 112 and 114 can be kept stable.

  Further, these bridges provided on the ground electrodes 130 and 132 also have an action of reducing the radiation loss of the high frequency signal at the bent portion of the RF electrode 120. The principle will be described below.

  As shown in FIG. 1, the coplanar line constituted by the RF electrode 120 and the ground electrodes 130 and 132 includes a first slot line constituted by the RF electrode 120 and the ground electrode 130 serving as a central conductor, The second slot line composed of the ground electrode 132 can be regarded as a state in which two slot lines are coupled (for example, Knorr J. B and Kucher K. "Analysis of coupled slots and coplanar strips on dielectric substrate "See IEEE Trans. Microwave Theory & Tech. Vol. MTT-23, no.7 pp.541-548 (1975)).

  In the bent part (FIG. 1) of the coplanar line, the second slot line is located inside the bent part and the first slot line is located outside the bent part of the two slot lines. Therefore, the first slot line located outside (hereinafter also referred to as “outside slot line”) has a larger distance than the second slot line located inside (hereinafter also referred to as “inside slot line”). Therefore, the signal propagating through the outer slot line is delayed from the signal propagating through the inner slot line.

  As a result, the symmetry of the signal propagation mode (slot mode) between these two slot lines sandwiching the RF electrode 120 is lost, and signal radiation loss occurs at the bent portion.

  Therefore, in order to prevent radiation loss in the bent portion of the coplanar line formed by the RF electrode 120, the signal propagating through the outer slot line is reduced by reducing the speed of the signal propagating through the inner slot line of the bent portion. What is necessary is just not to be delayed with respect to the signal propagating through the inner slot line. In other words, at the bent portion, the speed of the slot mode in the inner slot line is made slower than the speed of the slot mode in the outer slot line, and the signal delay between the two slot modes due to the bent portion is reduced. Can be prevented.

  In the optical waveguide device 100 of this embodiment, bridges 140 a to 140 e and 142 a to 142 e are formed at each corner of the ground electrode 132 located inside each bent portion of the RF electrode 120. Therefore, in the bent portion, the area of the side wall portion of the ground electrode 132 that faces the side wall portion inside the bent portion of the RF electrode 120 is reduced by the sum of the opening area of the bridges 140a to 140e. Therefore, the inner slot line composed of the RF electrode 120 and the ground electrode 132 has a reduced distributed capacitance in the bent portion (therefore, the characteristic impedance of the portion increases, and the electric power passing through the substrate 102 having a high dielectric constant passes through. As the line of force increases, the propagation speed of the slot mode of the inner slot line (hereinafter also referred to as “inner slot mode”) decreases in the bent portion. Thereby, in the bent portion, the propagation time of the inner slot mode becomes substantially equal to the propagation time of the slot mode (hereinafter also referred to as “outer slot mode”) in the outer slot line. As a result, the outer slot mode and the inner slot mode balance with each other without causing a signal delay, so that radiation loss and radiated electromagnetic noise are less likely to occur.

  The propagation speed of the slot mode becomes slower as the distributed capacity between the central conductor and the ground electrode is smaller. Therefore, the larger the size of the bridge (opening area) provided in the ground electrode, and the larger the number of bridges. Become slow.

  In addition, the effect of delaying the propagation of the signal in the slot mode greatly depends on the depth of the bridge, and if the depth is shallow, the effect of the signal delay is small. The depth of each bridge needs to be determined according to the material of the ground electrodes 130 and 132, the frequency of the signal propagating through the RF electrode 120, and the size of the opening of the bridge. It is necessary to make the depth more than three times the depth at which the skin effect occurs (so-called skin depth), or preferably through holes. Further, even if the bridge is not formed so that the depth direction thereof coincides with the perpendicular direction of the side surface of the ground electrode, the effect of signal delay can be obtained. Therefore, when the bridge is formed above the optical waveguide, the bridge is formed as a through hole extending in a direction not orthogonal to the wall surface of the ground electrode, such as the bridges 150a, 150c, and 160 shown in FIG. be able to.

  In the present embodiment, the bridges 140a to 140e and 142a to 142e provided on the ground electrode 132 are formed as cavities with a depth of 20 μm having a semicircular opening with a diameter of 7.5 μm. The bridges 150a, 150c, 152a, and 152c provided on the ground electrode 130 so as to straddle the parallel waveguides 112 and 114 have a mode field diameter of light propagating through these waveguides (effective diameter of the optical power field distribution). Since it is about 10 μm, it is formed to have a semicircular opening having a diameter of 15 μm. In order to reduce the optical loss to an almost negligible level, it is necessary to make the opening 20 μm or more in diameter. However, since the length of the ground electrode 130 straddling the parallel waveguides 112 and 114 is as short as about 100 to 500 μm, the opening Even with a bridge of about 15 μm, the optical loss can be suppressed to 0.05 dB or less. The bridge 150b provided on the ground electrode 130 is for reducing the signal propagation speed of the outer slot mode and finely adjusting the propagation speed of the inner slot mode and the propagation speed of the outer slot mode. is there. Note that the number of bridges such as 150b for speed adjustment is not limited to one, and a plurality of bridges may be provided as necessary.

  Thus, by providing a speed adjusting bridge on the ground electrode 130 on the outer side of the bent portion, for example, each bridge (eg, 140a) provided on the ground electrode 132 on the inner side of the bent portion has the same shape with a large opening. The bridge (for example, 150b) provided on the ground electrode 130 outside the bent portion is formed in a shape having a smaller opening while adjusting the speed of the inner slot mode according to the number of the outer slot mode. The speed can be finely adjusted, and the degree of freedom in designing the bridge formation in the ground electrodes 130 and 132 is improved.

  The number of bridges provided in the ground electrode 130 outside the bent portion of the RF electrode 120, the shape and size of the opening, the number of bridges provided in the ground electrode 132 inside the bent portion, and the shape and size of the opening are as described above. Not limited to this, any number, any shape, and any size can be used as long as the signal delay between the inner slot mode and the outer slot mode can be eliminated. For example, the shape of the opening of the bridge can be an arbitrary curved shape such as a long semi-ellipse, or an arbitrary polygonal shape.

Here, a trial calculation result about the size of the bridge opening required from the viewpoint of eliminating the propagation time difference between the inner slot mode and the outer slot mode is shown.
When the bending radius (the bending radius of the arc formed by the center line of the RF electrode 120) at the bending portion of the quarter arc (90-degree bending) of the RF electrode 120 is r, the RF electrode 120 and the ground electrodes 130 and 132 are configured. The propagation distance of the microwave propagating through the coplanar electrode (the length of the arc of the center line of the RF electrode 120) is (πr / 2), and the propagation time t of the distance is
t = (πr / 2) / V _cpw = (πr / 2) (N _cpw / c ₀ ) (1)
Given in. Here, V _cpw , N _cpw , and c ₀ are the microwave group velocity, refractive index, and speed of light, respectively.

Further, the width of the RF electrode 120 is S, the distance between the RF electrode 120 and the ground electrodes 130 and 132 is g, and the distance of the arc formed by the center line of the gap between the RF electrode 120 and the ground electrode 130 is the outer slot mode. When the propagation distance, the distance of the arc formed by the center line of the gap between the RF electrode 120 and the ground electrode 132, is the propagation distance of the inner slot mode, the propagation distance difference ΔL between the two slot modes is
ΔL = π (S + g) / 2 (2)
Given in.

Therefore, if the difference between the propagation distance of the outer slot mode and the propagation distance of the inner slot mode generated during the time t given by Expression (1) is equal to the propagation distance difference ΔL given by Expression (2), that is, ,
ΔL = c ₀ (1 / N _out −1 / N _in ) t (3)
Is satisfied, no propagation delay occurs between both slot modes. Here, N _out and N _in are the refractive index for the outer slot mode of the transmission line composed of the RF electrode 120 and the ground electrode 130, and the inner slot of the transmission line composed of the RF electrode 120 and the ground electrode 132, respectively. Refractive index for the mode.

According to the above formula (3) and a general calculation formula for the refractive index in the transmission line, for example, when S = 20 μm, g = 30 μm, r = 200 μm, N _in should be set to be about 3% larger than N _out. Thus, the propagation delay between both slot modes can be eliminated.

As described above, the refractive index of the outer slot mode and the inner slot mode can be adjusted by changing the capacitance distribution of the slot line constituted by the RF electrode 120 and the ground electrodes 130 and 132. The capacitance of these slot lines includes the capacitance of a capacitor configured to sandwich the material of the substrate 102 along the lines of electric force passing through the inside of the substrate 102, the side wall of the RF electrode 120, and the ground electrode 130. It is given by the sum of the capacitance of each capacitor formed by facing the side wall of 132 on the substrate. In the case of a broadband optical modulator, normally, the capacitance of the capacitor composed of the opposing side walls is dominant, and _in order to make N _in about 3% larger than N _out , the side wall of the RF electrode 120 and the ground are grounded. What is necessary is just to make the electrostatic capacitance between the side wall of the RF electrode 120 and the side wall of the ground electrode 132 a little less than 3% with respect to the electrostatic capacitance between the side wall with the electrode 130. In order to realize this capacitance difference, the shape of the bridge opening and the size of the opening also depend, but approximately, the total area of the opening of the bridge provided in the ground electrode 132 is approximately equal to the side of the bridge 132. What is necessary is just to make it become about 2%-4% with respect to an area.

  For example, when the height of the ground electrode 132 is 30 μm, if the openings of the bridges 140a to 140e are semicircular each having a diameter of about 15 μm and these bridges are formed at a pitch of about 100 μm, the speed of the inner slot mode can be increased. It can be about 3% slower. The speed reduction may be adjusted by the size of the opening of each bridge, or may be adjusted by the pitch (interval) between adjacent bridges (if the pitch is increased, the electrostatic force per propagation distance is increased). Since the capacity decreases, the propagation speed also decreases).

  The formation of the ground electrode 132 including the bridges 140a to e and 142a to e and the ground electrode 130 including the bridges 150a to c and 152a to c is performed by, for example, air connection wiring between circuit elements in a manufacturing process of MMIC (Monolithic Microwave Integrated Circuits). It can be performed by the same process as that used for forming the (air bridge). In this case, as a material for the sacrificial layer (portion that is removed after metal deposition to form the bridge) for forming the bridge portion, a photoresist or a metal such as nickel, chromium, or nichrome can be used.

  The material deposition of the ground electrodes 130 and 132 can be performed using the same plating process (plating method) as that used for the material deposition of the RF electrode 120, for example, in the same process as the process of depositing the RF electrode material. For example, after depositing the above-described sacrificial layer (photoresist or the like) serving as a bridge portion on the substrate 102, when depositing the material of the RF electrode 120, the ground electrodes 130 and 132 are also deposited using the same material, and thereafter The bridges 140a and the like can be easily formed by removing the sacrificial layer by etching. In this case, although there are limitations on the material of the ground electrodes 130 and 132 and the material used for the sacrificial layer, the ground electrodes 130 and 132 can be formed with fewer steps, which is advantageous in terms of manufacturing cost.

  The ground electrodes 130 and 132 may be formed by stacking a plurality of metal films on the substrate 102 including at least one metal layer having an opening with a predetermined width.

  Further, the ground electrodes 130 and 132 do not necessarily have to be formed entirely by material deposition on the substrate 102. Therefore, for example, a substantially plate-like conductive member having grooves provided so as to constitute a bridge of a predetermined size is bonded to a metal film formed on the substrate 102 in advance by, for example, flip chip bonding. The ground electrodes 130 and 132 can also be configured.

  As described above, in the optical waveguide device of the present embodiment, the RF electrode as the control electrode constitutes a coplanar line, and the ground electrode (inner ground electrode) located inside the bent portion at the bent portion. One or more bridges having a predetermined size opening are provided. As a result, the propagation speed of the signal propagation mode (inner slot mode) of the inner slot line constituted by the RF electrode and the inner ground electrode is set to the outer slot line constituted by the RF electrode and the ground electrode outside the bent portion. The signal delay between the inner slot mode and the outer slot mode due to the bent portion is eliminated by lowering the propagation speed of the signal propagation mode (outer slot mode), and between the two slot modes due to the signal delay. Thus, signal radiation loss can be effectively prevented.

  Further, in the optical waveguide element of the present embodiment, a bridge similar to the above is formed on the ground electrode covering the optical waveguide formed in the vicinity of the bent portion so as to straddle the optical waveguide, and the material metal of the ground electrode Occurrence of light absorption loss in the optical waveguide due to is prevented. That is, when there is an optical waveguide that passes through a bent portion of a coplanar line formed by an RF electrode, the light absorption loss is reduced by using a bridge structure similar to a bridge that is a structure for preventing the radiation loss. Can be prevented.

  In this embodiment, the bridge 150a and the like are provided so as to straddle the parallel waveguides 112 and 114 over the entire portion of the ground electrode 130 overlapping the parallel waveguides 112 and 114. The bridge 150a or the like may be provided so as to straddle the parallel waveguides 112 and 114 only for a part of the overlapping portion according to the tolerance for the loss of light propagating in the optical waveguide. .

  The optical waveguide device 100 shown in the present embodiment has a Mach-Zehnder type optical waveguide, and no optical waveguide is formed below the ground electrode 132 inside the bent portion of the RF electrode 120. However, the present invention is not limited to this, and an optical waveguide having another configuration may be used in any positional relationship with respect to the RF electrode 120 and the ground electrodes 130 and 132. For example, the optical waveguide may be configured by two parallel waveguides separated by a predetermined distance, such as a directional coupler, and is configured by a waveguide that intersects or branches in an X shape or a Y shape. It may be. Further, a part of these optical waveguides may pass through the lower part of the inner ground electrode 132.

  When a part of the optical waveguide is formed below the inner ground electrode 132, at least one of the bridges formed on the ground electrode 130 is configured to straddle part or all of the optical waveguide. Can be. Thereby, the bridge provided on the inner ground electrode 132 can simultaneously prevent the radiation loss of the high-frequency signal in the bent portion and the absorption loss due to the ground electrode 132 in the optical waveguide.

  In the optical waveguide device shown in this embodiment, the buffer layer is not provided on the upper surface of the substrate 102. However, the present invention is not limited to this, and the above-described bridge 140a and the like are provided on the buffer layer formed on the substrate 102. It may be a thing. Also in this case, radiation loss of the high frequency signal can be prevented by forming a bridge to the ground electrode inside the bent portion of the RF electrode. At the same time, for example, if the buffer layer is thin and light absorption loss due to the ground electrode cannot be completely avoided with the buffer layer alone, a bridge is formed over the optical waveguide on the ground electrode above the optical waveguide. Thus, the occurrence of the absorption loss can be completely avoided.

  In the above-described embodiment, the X-cut LN substrate is used as the substrate 102. However, the present invention is not limited to this, and a Z-cut LN substrate may be used. When a Z-cut LN substrate is used, a buffer layer is generally formed on the substrate surface in order to provide the RF electrode directly over the parallel waveguide. However, as described above, such a buffer layer is formed. Even when the layers are formed, radiation loss of high-frequency signals can be effectively prevented by using the bridge structure shown in this embodiment.

  In addition to the formation of the bridge to the inner ground electrode at the bent portion of the RF electrode shown in the present embodiment, the distance between the RF electrode and the inner ground electrode is made larger than the distance between the RF electrode and the outer ground electrode. Configurations can be used together. Accordingly, for example, the distributed capacity of the inner slot line can be supplementarily reduced by the configuration, and the adjustment range by the bridge formation can be narrowed to enable finer adjustment.

  Furthermore, in the above-described embodiment, the space between the side wall of the RF electrode 120 and the side wall of the inner ground electrode 132 is air, but an insulating material (eg, resin) having a low dielectric loss is applied to the space. It can also be arranged to supplement the speed reduction effect of the inner slot mode.

  In addition, an insulating material (for example, resin) having a low dielectric loss is disposed in a space between the sidewall of the RF electrode 120 and the sidewall of the outer ground electrode 130 to supplementarily enhance the effect of reducing the propagation speed of the outer slot mode. Thus, signal delay between the inner slot mode and the outer slot mode can be prevented, and radiation loss can be reduced.

  DESCRIPTION OF SYMBOLS 100 ... Optical waveguide element, 102 ... Board | substrate, 110 ... Incident waveguide, 112, 114 ... Parallel waveguide, 116 ... Outgoing waveguide, 120 ... RF electrode, 130, 132 ... ground electrodes, 140a-e, 142a-e, 150a-c, 152a-c, 160, 162 ... bridges.

Claims (7)

An optical waveguide formed on the substrate; a control electrode formed on the substrate for controlling light waves propagating through the optical waveguide; and two ground electrodes constituting a coplanar line having the control electrode as a central conductor; An optical waveguide device comprising:
In the bent portion of the coplanar type line, the speed of the slot mode propagating between the one ground electrode located inside the bent portion and the control electrode is reduced to one ground electrode located outside the bent portion. It is slower than the speed of the slot mode propagating between the control electrodes,
Optical waveguide element. An optical waveguide formed on the substrate; a control electrode formed on the substrate for controlling light waves propagating through the optical waveguide; and two ground electrodes constituting a coplanar line having the control electrode as a central conductor; An optical waveguide device comprising:
The bent portion of the coplanar line, one of the ground electrode located inside of the bent portion, on the side wall facing the control electrode, the bridge having an opening with a predetermined depth of a predetermined size, a predetermined number In the bent portion of the coplanar line, the speed of the inner slot mode propagating between the one ground electrode located inside the bent portion and the control electrode is positioned outside the bent portion. Set to be slower than the speed of the outer slot mode propagating between the other ground electrode and the control electrode,
At least one of the bridges provided on the one ground electrode is formed so as to straddle all or part of the optical waveguide formed on the substrate, or another bridge located inside the bent portion A portion of the ground electrode formed on the side wall facing the control electrode so as to straddle all or part of the optical waveguide;
Optical waveguide element. In the bend portion, the other of the ground electrode located outside of the bend portion, the side wall facing the control electrode, the bridge having an opening with a predetermined depth of a predetermined size is formed by a predetermined number The speed of the outer slot mode is adjusted relative to the speed of the inner slot mode,
At least one of the bridge provided on the one ground electrode and the bridge provided on the other ground electrode is formed so as to straddle all or part of the optical waveguide formed on the substrate. ,
The optical waveguide device according to claim 2. The substrate is made of a material having an electro-optic effect,
The optical waveguide is a Mach-Zehnder optical waveguide,
The optical waveguide element as described in any one of Claims 1 thru | or 3 . The optical waveguide device according to claim 4 , wherein the thickness of the substrate is 50 μm or less. The control electrode is formed using a plating method, and
The control electrode has a thickness of 30 μm or more.
The optical waveguide device according to any one of claims 1 to 5 . The optical waveguide element according to any one of claims 1 to 6 , wherein the optical waveguide includes an optical waveguide constituting a directional coupler, an optical waveguide that intersects each other, or an optical waveguide having a Y branching portion.

Images