JP5081777B2 - Mach-Zehnder interferometer type optical functional element - Google Patents

Description

  The present invention is a Mach-Zehnder interferometer type optical functional element composed of a substrate-type optical waveguide component used in optical fiber communication or the like. In principle, transmitted light is attenuated as a phase controller, and excessive transmission loss occurs in the part. The present invention relates to a technique for improving the output light intensity and improving the extinction ratio when a mechanism such as this is applied.

Conventionally, an optical functional element such as an optical modulator or an optical switch including a Mach-Zehnder interferometer is known and widely used in an optical fiber communication system. For example, Patent Document 1 discloses an optical switch in which a thin film heater is formed on a quartz-based substrate type optical waveguide and a thermo-optic phase shifter is formed as an optical functional element utilizing the thermo-optic effect.
In addition, an optical modulator using lithium niobate (LiNbO ₃ ) is widely used as an optical functional element utilizing the electro-optic effect.
However, the technique described above uses a thermal optical effect for the technique using the thermo-optic effect, so the response speed is slow, and the switching speed when used as an optical switch is about several milliseconds. In addition, the optical modulator using lithium niobate has a large element size, and cannot be integrated at high density, and is expensive.

As a technique for overcoming these disadvantages, silicon photonics device technology for manufacturing optical waveguide components using a complementary metal oxide semiconductor (CMOS) manufacturing process has recently attracted attention. According to this technique, a small and high density integrated optical component can be manufactured at low cost. However, when the refractive index change due to carrier injection is used in silicon photonics device technology, loss due to free carrier absorption (FCA) occurs.
Therefore, a phase controller is prepared only in one of the two optical paths of the Mach-Zehnder interferometer (MZI), thereby configuring an optical modulator or optical switch configured to control the interference and turn on / off the output light. In this case, since the loss increases only in the optical path having the phase controller, there is a problem that the intensity balance of the interference light cannot be achieved and the extinction ratio is deteriorated.

An example of a countermeasure against the large loss of one optical path in the Mach-Zehnder interferometer is disclosed in Patent Document 2 below. In the technique described in Patent Document 2, an electrode is provided as a branching ratio adjusting unit that changes the branching ratio of at least one of the two branching units (optical coupler units) of the Mach-Zehnder interferometer. A different structure is disclosed.
However, this technique has a structure in which an electrode is provided in the optical coupler portion of the Mach-Zehnder interferometer and the branching ratio is controlled by this electrode, and there is a problem that the control becomes complicated as the number of electrodes increases.
JP-A-6-34924 Japanese Patent Laid-Open No. 7-9524

  The present invention has been made in view of the above circumstances, and in a Mach-Zehnder interferometer type optical functional element, a mechanism in which transmitted light is attenuated in principle as a phase controller, and excessive transmission loss occurs in the part. It is an object of the present invention to provide a technique capable of improving the output light intensity and improving the extinction ratio even when applied.

In order to solve the above problems, the present invention includes an input optical waveguide, a demultiplexing unit that is connected to the input optical waveguide and that divides and emits the signal light incident from the input optical waveguide, The phase of the signal light is adjusted according to the first optical path and the second optical path where the light emitted from the wave part is incident, and a phase control signal provided as a part of the second optical path. An optical modulation unit that couples the signal light from the first optical path and the signal light from the second optical path connected to the first optical path and the second optical path, A Mach-Zehnder interferometer-type optical functional element having an output waveguide from which the multiplexed light from the multiplexing unit is emitted, wherein the second optical path is lighter than the first optical path due to the optical modulation unit. Propagation loss is increased, the demultiplexing section has a fixed unequal branching ratio, and light incident on the second optical path Degree Ri is Na is stronger than the intensity of light incident on the first optical path, a branching ratio of the branching unit, a phase in the optical modulating unit so as to maximize the output light intensity from the output waveguide In the controlled state, the signal light intensity for the multiplexing part from the first optical path is determined to be equal to the signal light intensity for the multiplexing part from the second optical path, and the demultiplexing is performed. The part is a Y-branch type optical waveguide, and a tapered part whose width gradually increases from the incident side to the outgoing side of the branching part is formed, and the inclination angles of both side surfaces of the tapered part with respect to the extension axis at the center of the input optical waveguide Are different between the first optical path side and the second optical path side, and two S-bend type optical waveguides are connected to the output end of the branching unit to form the first optical path and the second optical path. In addition, in the demultiplexing part, the incident part before entering the tapered part The force optical waveguide portion is linear, and of the two side walls of the tapered portion, the side wall on the first optical path side is parallel to the optical axis of the linear portion in the vicinity of the tapered portion of the input optical waveguide. The two side walls on the optical path side are inclined with respect to the optical axis of the linear portion in the vicinity of the tapered portion of the input optical waveguide, and the width of the output end of the tapered portion is connected to the output end of the tapered portion. The sum of the widths of the S-bend type optical waveguides is equal to the sum of the gap widths at the start ends of the two S-bend type optical waveguides, and the distance between the two S-bend type optical waveguides is Narrowest at the start end side connected to the output end of the taper portion, and gradually separated from each other toward the end, and the end surface of the taper portion connecting the S bend type optical waveguide is the end of the input optical waveguide the end faces of the parallel, characterized in Rukoto such by.

In the Mach-Zehnder interferometer-type optical functional element of the present invention, the optical modulation unit includes either or both of a region that is a P-type semiconductor and a region that is an N-type semiconductor, and two optical modulation units are provided as phase control signals. In this structure, a refractive index change due to a change in carrier concentration is induced in the semiconductor region by applying a voltage to the electrode, and the second optical path has an excessive light propagation loss compared to the first optical path. The loss due to free carrier absorption is preferred .

In the Mach-Zehnder interferometer type optical functional element of the present invention, each part through which the signal light passes is made of a material having a high refractive index in the input optical waveguide, the demultiplexing unit, the first optical path, the second optical path, and the multiplexing unit. An optical waveguide structure in which a core layer is surrounded by a clad layer made of a material having a low refractive index, the core part is made of silicon, silicon nitride, or silicon oxynitride, and a semiconductor region of the light modulation part is It is preferably a substrate type optical waveguide component configured by injecting an element as a dopant into silicon.

The Mach-Zehnder interferometer-type optical functional element of the present invention includes the first output waveguide with respect to the incident optical power of the input optical waveguide, with the length of the tapered portion of the branching portion as the horizontal axis and the branching ratio as the vertical axis. The graph of the result of measuring the intensity ratio of the output light power from the second output waveguide is drawn, and the first section where the branching ratio increases as the length of the tapered portion increases is the first section, It is preferable that the length of the tapered portion is selected from the range of the second section when the section in which the branching ratio decreases as the length of the subsequent tapered section becomes the second section.
Mach-Zehnder interferometer type optical functional device of the present invention, the first even Ri by the optical path only a short said second optical path of the optical path length, it is preferable that the phase controller is provided.

As described above, according to the present invention, the branching ratio of the demultiplexing unit is set to be an unequal branching ratio, and a structure in which a large amount of light is incident on the second optical path having a large loss can be obtained. It is possible to provide a Mach-Zehnder interferometer type optical functional element in which the emission intensity and the emission intensity of the second optical path can be made uniform, the decrease in the light output intensity is suppressed, and the deterioration of the extinction ratio is suppressed. In addition, since the branching ratio of the branching unit is fixed, the above-described effects can be obtained without complicating the control.
If the optical modulator includes at least one of P-type and N-type semiconductor regions and phase control is performed with a structure in which a refractive index change caused by a change in carrier concentration of the semiconductor region is induced by a voltage from the electrodes, the response as an optical switch Speed can be increased.

  Moreover, if a substrate-type optical waveguide component configured by injecting an element serving as a dopant into silicon as a semiconductor region of the light modulation unit as a core structure including a core part and a clad layer is used, a substrate using silicon photonics technology is used. An optical waveguide can be formed thereon, and an optical component with a miniaturized element can be obtained.

By adopting a structure having a taper part with an expanded core part width as a demultiplexing part, the branching ratio can be easily fixed, and the outgoing light intensity of the first optical path can be obtained without complicating the control. The emitted light intensity of the second optical path can be made uniform, a decrease in the light output intensity can be suppressed, and the deterioration of the extinction ratio can be suppressed.
As a branching unit, a structure having a fixed branching ratio lower than 9 to 1 and an intermediate branching ratio such as unequal branching has low loss in a single-mode Y-branch substrate type waveguide, and is large This can be achieved with manufacturing tolerance.
Further, as the taper length of the demultiplexing part, a graph showing the result of measuring the intensity ratio of the output light power from the first output waveguide and the second output waveguide with respect to the incident light power of the input optical waveguide is used. By using the second section in which the branching ratio decreases as the length of the tapered portion increases, instead of the first first section in which the branching ratio increases as the length of the section increases. It is possible to provide a Mach-Zehnder interferometer type optical functional element that can achieve a low loss and a large manufacturing tolerance.
In addition, by adopting a structure in which the phase controller is provided in the shorter optical path length for the first optical path and the second optical path, the transmission loss difference is suppressed compared to the structure in which the phase controller is provided in the longer optical path length. be able to.

An embodiment of a Mach-Zehnder interferometer type optical functional element of the present invention will be described below with reference to the drawings. In addition, each embodiment described below is specifically described so that the gist of the present invention can be easily understood, and the present invention is not limited to the following embodiment.
1 to 4 schematically show a Mach-Zehnder interferometer-type optical functional element of this embodiment. A Mach-Zehnder interferometer-type optical functional element A of this embodiment includes a substrate 1 and a transmission line formed thereon. 2 and a phase controller 3 provided in the middle of the transmission path 2.
The transmission line 2 of the present embodiment is a so-called Mach-Zehnder type optical waveguide, for example, a core part such as silicon nitride (SiN) whose refractive index is adjusted, and a refractive index provided surrounding the core part is adjusted. An optical waveguide composed of a clad layer is provided.

In this example, the transmission path 2 branches an input optical waveguide 5 constituted by one optical waveguide into two optical paths 7 and 8 by a Y-type demultiplexing unit 6, and the first optical path 7 and the second optical path 8 are branched. By providing a desired optical path length difference, specifically, the second optical path 8 is formed to be shorter than the first optical path 7 so that a Mach-Zehnder interferometer is configured. Further, the first optical path 7 and the second optical path 8 are configured as a multiplexing unit 9 by arranging the core portions of the extension portions of the paths close to each other, and after the light is interfered in the multiplexing unit 9, It is supposed to be output light. Further, the portion extended to the tip of the combining portion 9 on the first optical path 7 side is the first output waveguide 10, and the portion extended to the tip of the combining portion 9 on the second optical path 8 side is The second output waveguide 11 is used. Further, a photonic crystal structure or the like is used for the middle part of the shorter second optical path 8, that is, for a part of the second optical path 8 located between the demultiplexing unit 6 and the multiplexing unit 9. A phase controller 3 is provided. Note that the previous multiplexing unit 9 may be a Y-shaped coupler using a Y-branch optical waveguide in the reverse direction, or may be configured by a directional coupler.
Next, the phase controller 3 includes a light modulation unit 12 provided in the middle part of the second optical path 8 and a controller 14 electrically connected to the light modulation unit 12 through the electrodes 3A and 3B via the wirings 13. The light passing through the second optical path 8 can be modulated via the light modulator 12 by a control signal from the controller 14.

As an example of the optical waveguide formed on the substrate 1, a core portion 17 is provided on the substrate 1 so as to be surrounded by a lower cladding layer 15 and an upper cladding layer 16 as shown in a cross-sectional structure in FIG. The structure can be exemplified. As the substrate 1, for example, a silicon substrate, an SOI (Silicon On Insulator) substrate, a quartz substrate, or the like can be applied. As a core material or a cladding material, silicon nitride (Si ₃ N ₄ ), silicon oxynitride (SiOxNy), silicon dioxide (SiO ₂ ), or the like can be used as appropriate.
The branching unit 6 of the present embodiment has unequal branching with a fixed branching ratio that branches the signal light incident from the input optical waveguide 5 into the first optical path 7 and the second optical path 8 at a desired branching ratio. It functions as a Y-branch optical waveguide. Note that the branching unit 6 may be a directional coupler having the same function.

Next, the first optical path 7 and the second optical path 8 and the output optical waveguides 10 and 11 constituting the input optical waveguide 5 and the Mach-Zehnder interferometer are configured to be single-mode optical waveguides for the wavelength of the signal light. ing. In the Mach-Zehnder interferometer of this example, when the optical path difference between the first optical path 7 and the second optical path 8 is ΔL, and the effective refractive index when an optical signal having a predetermined wavelength is propagated through the waveguide is Neff, When ΔL · Neff is m · λ, where m is an integer, the phase difference is m · 2π and is strengthened by interference, and when (m + 1/2) λ, the phase difference is (2m + 1) π. And function to counteract by interference.
In general, since the Mach-Zehnder interferometer multiplexes and interferes optical signals with different phases that are branched in a one-to-one relationship, the propagation losses of the two optical waveguides that give the phase difference must match. is there. Here, if the amount of propagation loss differs greatly between the first optical path 7 and the second optical path 8, the output light corresponding to the difference in the amount of propagation loss is not sufficiently canceled when interference is made to cancel each other. It will remain. Therefore, when used as a modulator or as an optical switch, the extinction ratio is small.

  The cause of the difference in propagation loss is that the substrate type optical waveguide component using silicon photonics technology, such as an optical splitter for FTTH (Fiber To The Home), such as a conventional quartz type substrate type light source waveguide or polymer having a small relative refractive index difference. Since the relative refractive index difference is large compared to a substrate-type optical waveguide, etc., the large scattering due to the roughness of the core side wall caused by the manufacturing process is the difference in propagation loss depending on the optical path length difference. It is done. Further, when the plasma effect due to the carrier concentration change is used as the optical characteristic variable mechanism in the phase controller, the free carrier absorption increases and the loss increases as the carrier concentration increases. Considering these two factors, the phase controller 3 should be provided on the side of the second optical path 8 having the shorter optical path length, out of the first optical path 7 and the second optical path 8. Compared with the case where the phase controller is provided on the first optical path 7 side, the difference in the propagation loss amount can be reduced.

For example, as an example of a simplified model, the demultiplexing unit 6 is assumed to be a 3 dB splitter, that is, an equal branch, the propagation loss of each of the first optical path 7 and the second optical path 8 is 1 dB, and the phase controller 3 Consider the case where the increase in loss due to free carrier absorption when a desired phase shift is performed is 3 dB.
When the phase controller 3 does not give a phase shift, an optical signal is output, and when the optical signal is canceled by giving a π phase shift, there is no 3 dB loss difference, and if the signals are sufficiently canceled, the optical signal is output. However, an extinction ratio of about 20 dB is generally obtained as the performance of the actually manufactured device. However, if there is a loss difference, the optical output of about 15% remains without being canceled out, and the extinction ratio Deteriorates to 7 dB.
On the other hand, in the present embodiment, the branching ratio of the branching unit 6 is set to be an unequal branch in advance so that the propagation loss difference between the first optical path 7 and the second optical path 8 becomes equal when a desired phase shift is given. Ratio. For example, in the case of the above-described example, when the demultiplexing unit 6 is configured so that more light is distributed to the second optical path 8 than the first optical path 7 in a 4: 6 ratio, the extinction ratio of 10 and several dB. Can be obtained. This corresponds to, for example, selecting a taper length L that gives a branching ratio of 1.5 dB shown in FIG. 9 or FIG. 12 to be described later, and this causes the extinction ratio to deteriorate to 7 dB. It means that it becomes possible to obtain.

 It is not appropriate to apply the technique of the present invention to a structure in which an optical signal is output from the phase controller 3 when a phase shift is applied. In this case, when the technology of the present invention is not applied and the branching is made equal, the output light intensity is only reduced to about 65% of the input light. 6, when more light is distributed to the second optical path 8 than to the first optical path 7, the output light intensity decreases to about 62% and the characteristics deteriorate. This is a result of distributing much energy to the second optical path 8 having a large propagation loss.

The specific structures of the phase controller and the unequal branching / demultiplexing unit applicable to the Mach-Zehnder interferometer type optical functional element of the present embodiment will be described in more detail below.
"Phase controller"
The phase controller applicable in the present embodiment varies the refractive index of a partial region of the optical waveguide by applying a voltage to the electrode pad formed in the element or by passing a current, and the phase of the signal light Is a functional element that controls There are those using the electro-optic effect and those using the thermo-optic effect. Among them, those using an N-type semiconductor material or a P-type semiconductor material and utilizing a refractive index change due to the free carrier plasma effect are high-speed responsiveness. In addition, a silicon photonics device that realizes this using a silicon material is useful because it can be manufactured at low cost in the CMOS LSI manufacturing process.

 Such an optical functional element can be realized by using a silicon material as the whole or a part of the core region of the optical waveguide and using an SOI (Silicon On Insulator) substrate. If the silicon of the SOI layer of the substrate is doped with impurities such as B (boron), Al (aluminum), Ga (gallium), In (indium), Tl (thallium), etc., the N-type semiconductor and N ( Nitrogen), P (phosphorus), As (arsenic), Sb (antimony), Bi (bismuth) or the like can be doped to form a P-type semiconductor.

FIG. 5 is a schematic diagram of an example of a cross-sectional structure of a phase controller including a silicon semiconductor as a part of an optical waveguide core.
The SOI substrate 20 is obtained by forming a SiO ₂ layer 22 and an SOI layer 23 on a Si substrate 21. This SOI layer 23 is silicon, and a photolithographic film is formed by periodically forming fine cylinders thereon. A nick crystal portion 25 is formed. Further, by doping with impurities, it is partially formed as an N-type semiconductor and a P-type semiconductor, and a PN junction is formed. The transmission line 26 is, for example, a silicon nitride core having a refractive index of 2 and a rectangular cross section, and the silicon nitride core and the SOI layer 23 positioned below the silicon nitride core constitute a single mode optical waveguide core. Electrodes 27 and 28 are arranged so as to sandwich the photonic crystal part 25 in the plane direction, and a SiON layer 29 is covered thereon.
In the above configuration, by applying a voltage between the electrodes 27 and 28 arranged so as to sandwich the photonic crystal part 25, the carrier concentration distribution of the SOI layer 23 is changed, and the refractive index is changed by the plasma effect. Thus, it is possible to control the phase of the propagated signal light. However, the increase in free carriers also increases propagation loss due to free carrier absorption.

  6 and 7 are configuration diagrams showing a first example and a second example in which the phase controller having the basic structure shown in FIG. 5 is applied to a Mach-Zehnder interferometer type optical functional element.

In the first example shown in FIG. 6, the Mach-Zehnder interferometer type optical functional element B includes an input optical waveguide (transmission path) 31, a branching unit 32, a first optical path 33, and a second optical path 34 on an SOI substrate. The optical modulation unit 39, the multiplexing unit (coupling unit) 40, the first output waveguide 41, and the second output waveguide 42 including the photonic crystal unit 36 and the electrodes 37 and 38 are formed. The input optical waveguide 31 can guide light incident from an optical fiber or the like (for example, having a wavelength in the 1.55 μm band, which is a communication wavelength band) to an optical modulation unit that applies predetermined phase modulation, so that phase modulation can be applied. It is configured.
In FIG. 6, configurations other than the transmission path 31, the photonic crystal portion 36, the electrodes 37 and 38, and the barrier region 43, such as a SiON (silicon oxynitride) layer that covers the transmission path 31, are omitted.

In the second example shown in FIG. 7, in the Mach-Zehnder interferometer type optical functional element C, the SiO ₂ layer 50 is formed on the Si substrate 49, and the input optical waveguide (transmission path) 51 and the branching unit are formed thereon. 52, a first optical path 53, a second optical path 54, a light modulation section 59 comprising a photonic crystal section 56 and electrodes 57, 58, a multiplexing section (coupling section) 60, a first output waveguide 61, a first Light that is produced by forming two output waveguides 62 and that imparts predetermined phase modulation to light (for example, having a wavelength in the 1.55 μm band that is a communication wavelength band) incident from an optical fiber or the like by the input optical waveguide 51 It is configured such that it can be guided to the modulation unit and phase modulation can be applied.
In FIG. 7, configurations other than the transmission line 51, the photonic crystal unit 56, the electrodes 57 and 58, and the barrier region 63 are omitted.

  In these Mach-Zehnder interferometer type optical functional elements B and C, for example, in the photonic crystal portions 36 and 56, the light incident end side has a cross-sectional area perpendicular to the light traveling direction, and the light traveling It is formed in a reverse taper shape that gradually increases gradually in the direction, and the exit end side is formed in a forward taper shape in which the cross-sectional area continuously decreases gradually in the light traveling direction. The For example, a dielectric cylinder is arranged in a triangular lattice pattern on the base material, and at this time, the incident end side is one, two, three,... A predetermined number in the direction orthogonal to the propagation of light on the transmission line. Are arranged so as to increase gradually one by one, and the emission end side gradually decreases from a predetermined number of three, two, one,... In the direction perpendicular to the propagation of light on the transmission line 11. Arrange so that. This is because the light propagating through the transmission line 11 is prevented from being reflected by the end face of the photonic crystal portion 12 and phase modulation can be applied with low loss and high efficiency.

  Then, a base material (for example, silicon (Si)) constituting the photonic crystal portions 36 and 56 is applied to the photonic crystal portions 36 and 56 by applying current or voltage (energy application) with an electrode as an energy applying portion. , And the phase modulation is applied to the light propagating through the transmission line in the branched light modulation sections 39 and 59. Similarly, a base material (for example, silicon (Si)) constituting the photonic crystal portions 36 and 56 is applied to the photonic crystal portions 36 and 56 by applying current or voltage (energy application) to the photonic crystal portions 36 and 56 with an electrode as an energy applying portion. ) Is changed, and phase modulation is applied to the light propagating through the second optical paths 34 and 54 after branching. One electrode may be grounded and a current or a voltage may be applied (energized) only to the other electrode.

In addition, barrier regions 43 and 63 are provided by removing the SOI layer in regions other than the vicinity of the second optical paths 34 and 54 by etching or the like. These barrier regions 43 and 63 are configured by being filled with a material having a refractive index smaller than that of the base material constituting the photonic crystal portions 34 and 54. Thereby, it is possible to prevent (suppress) the light propagating through the second optical paths 34 and 54 from diffusing in the direction perpendicular to the light traveling direction through the SOI layer. As a material having a refractive index smaller than that of the base material forming the photonic crystal portions 36 and 56, for example, when the base material forming the photonic crystal portions 36 and 56 is silicon (Si), silicon oxide is used. (SiO ₂ ) or silicon oxynitride (SiON) may be used.

Further, impurities are added to the SOI layer and carriers are distributed. Examples of such impurities include P-type elements B (boron), Al (aluminum), Ga (gallium), In (indium), Tl (thallium), and N-type elements N (nitrogen), which are also used in semiconductors. ), P (phosphorus), As (arsenic), Sb (antimony), Bi (bismuth), and the like. Among these, B (boron) is particularly suitable for the P-type element because it can be easily doped. Therefore, here, boron (B), which is a P-type impurity, is added and holes ( Hole) is distributed.
In addition, when an N-type impurity such as phosphorus (P) is added, electrons are distributed as carriers, but even when electrons are distributed, the action generated by voltage application described below is a change in refractive index due to free carrier absorption. Except for qualitatively.
Further, the impurities may be added uniformly in the plane, or P-type impurities and N-type impurities may be added so as to provide a PN junction at the lower part of the transmission line.

`` Action generated by voltage application ''
In the photonic crystal portions 36 and 56, if the carrier velocity of the SOI layer exceeds the saturation velocity by increasing the voltage applied from the electrodes, the space charge density is locally increased near the region where the electric field is concentrated. A dipole region called a dipole layer is generated. In such a dipole layer, the electric field is locally strengthened, and as a result, the energy band gap of the electron system has a spatial gradient. As a result, due to the Franz-Keldysh effect, the band gap is shifted by low energy, and the refractive index increases due to the Kramers-Kronig relationship.
Further, if the dipole layer is present, the carrier density locally increases or decreases. As the carrier density increases, the band gap shifts to a lower energy due to the many-body correlation effect based on electrostatic interactions. In this case as well, the refractive index increases due to the Kramers-Kronig relationship. The effect of low energy shift is stronger when the carrier density is higher and weaker when the carrier density is low. Therefore, the refractive index increases in the entire photonic crystal part.
From the principle described above, voltage application to the photonic crystal portions 36 and 56 acts on the photonic crystal portions 36 and 56 as described above, and as a result, the refractive index of the base material of the photonic crystal portions 36 and 56 is increased. It can change and give phase modulation.

`` Non-equal branching demultiplexer ''
The specific structure of the non-equally branched branching unit applicable to the Mach-Zehnder interferometer type optical functional element A of the previous embodiment will be described in detail below.
Conventionally, an optical branching device using a substrate-type optical waveguide has been widely used. In most cases, the structure is a Y-branch type or a directional coupler type. The mold is advantageous. Many of them are Y-branch type with one input and two outputs for optical power distribution, ie, one-to-one output, or 95 to 5 or 99 pairs for tap couplers. 1 or the like having a large branching ratio.
An optical branching element having an intermediate branching ratio, that is, a ratio of 55:45, 6: 4, 7: 3, 8: 2, 9: 1, and a branching ratio less than 9: 1 and unequal branching It is disclosed in Japanese Patent Laid-Open No. 6-235843 that it is difficult to realize an optical branching device having an intermediate branching ratio as described above with a single-mode Y-branch substrate type optical waveguide.
In this patent specification, in a single-mode asymmetric Y U-type optical branching path, most of the two branched optical waveguides flow into the optical waveguide side having a large propagation constant, and the optical waveguide having a small propagation constant. In the single-mode optical unequal branch circuit, it is difficult to control the branching ratio, which is a non-patent document WKBurns et al., IEEE J. Quantum Electron, QE-11, P.32 (1975).

However, in the present invention, in order to solve such problems of the prior art, an optical branch circuit having a new configuration described below is devised, its structure is examined, and a suitable structure is provided.
FIG. 2 is an overall view of a demultiplexing section (optical branching element) applied to the present invention having a display aspect ratio of 1: 4. FIG. 3 is an enlarged view of the vicinity of the branching section. For example, as shown in FIG. 4, it has a composite structure of a core part and a clad layer similar to the optical waveguide described above.

  As shown in FIG. 3, the branching unit 6 in this example includes a tapered portion 6 a having a trapezoidal shape in plan view, and is connected to the input optical waveguide 5 and has the same width as the input optical waveguide 5. The end face 6A and the input side end face 6A are arranged at right angles, and are extended flush with the side face of the linear input optical waveguide 5 (extend parallel to the extension axis 5A of the optical axis of the input optical waveguide 5). The first side surface 6B is continuously inclined with a slight inclination from a direction perpendicular to the input side end surface 6A, and the first side surface 6B is inclined in a direction that widens as the distance from the input side end surface 6A increases. The second side surface 6C is formed on the terminal side of the first side surface 6B and the second side surface 6C, and is surrounded by the output side end surface 6D parallel to the input side end surface 6A. Further, the second side surface 6C is inclined so as to have an inclination angle θ resulting from a difference in width between the input side end surface 6A and the output side end surface 6D.

The cross section of the branching portion 6 in this example has a cross-sectional structure in which the core portion 17 is surrounded by the cladding layers 15 and 16 as shown in FIG. 4, but the core portion 17 provided inside is from the input side end face 6A side. Also, the core structure is such that the width gradually increases on the output side end face 6D side. That is, the width of the core portion of the output side of the branching portion 6, S bend type waveguide 7A to be described later, and has a width + gap width G of 8A. The gap width G is the width of the cladding layer between the cores of the S-bend type waveguides 7A and 8A.

S-bend type waveguides 7A and 8A are optically connected to the output side end face 6D of the branching section 6 of this core structure. These waveguides 7A and 8A have the same internal structure as the previous input optical waveguide 5, and their width W is the same as the width W of the input optical waveguide 5, but is curved and formed in an S-shape in plan view. .
The start end 7a of one waveguide 7A is optically connected to the end on the first side face 6B side at the output side end face 6D of the branching section 6, and the start end 8a of the other waveguide 8A is connected to the branching section 6. The output side end face 6D is optically connected to the end part on the second side face 6C side. In addition, a gap G is provided between the waveguides 7A and 8A on the width direction center portion side of the output side end face 6D of the branching portion 6. That is, the width AW of the output-side end face 6D of the branching section 6 is formed to be a width obtained by adding the value of the gap G to the sum of the widths W of the waveguides 7A and 8A. (2W + G = width of output side end face 6D)
Further, the second side surface 6C of the branching portion 6 is formed to be a tapered surface that reaches from the end portion of the input side end surface 6A to the end portion of the output side end surface 6D, and in the branching portion 6, the input side end surface 6A. And the output side end face 6D are arranged in parallel.

  For example, the demultiplexing unit 6 forms a lower clad layer having a low refractive index on a wafer made of silicon or quartz having a thickness of 1 mm or 0.7 mm, a core layer having a high refractive index, and a photo layer. The core layer is formed into a desired shape by a lithography process and an etching process, and an upper cladding layer having a low refractive index is further formed to cover the core, which is cut to obtain an optical branching element. Various organic / inorganic materials that are transparent with respect to the wavelength of the optical signal used in the element can be applied to the core material and the clad material here.

In this embodiment, one input optical waveguide 5 may be a straight optical waveguide or a curved optical waveguide, but is connected to the vicinity of the second side surface 6C, that is, the tapered portion of the branching portion 6, that is, the tapered portion. In the vicinity of the end portion, a straight line is preferable.
The core width of the input optical waveguide 5 and the core width of each part of the demultiplexing unit 6 are determined by determining the condition range for a single mode optical waveguide by calculation from the refractive index of the core material and the refractive index of the cladding material.
From the viewpoint of reducing the propagation loss due to scattering due to the roughness of the core side wall and the connection loss in the element input / output section, it is preferable that the core diameter is large. It causes interference and multimode dispersion.
The width G of the gap is determined by determining a numerical range in which the gap G can be stably formed from the capability values of the photolithography process and the etching process.
The narrower one can reduce excess loss, but the wider one allows stable machining.

The S-bend type waveguides 7A and 8A are bent waveguides bent into an S-shape, and the surface direction of the connection portion of the output side end face 6D of the branching waveguide 6 is made parallel to the end face of the terminal end of the input optical waveguide 5. In addition, the surface direction of the output optical waveguide connecting portion is parallel to the surface of the output optical waveguide starting end.
A plurality of S-bend S-shaped bent shapes can be proposed depending on how the center line is formed. However, if the design is performed using a sufficiently gentle radius of curvature according to the relative refractive index difference between the core portion and the cladding layer. Any S U shape may be used.
In the following case study, an S-shaped waveguide based on an arc shape was used. The distance between the output ends of the S-bend type waveguides 7A and 8A depends on the arrangement of the next phase controller and the waveguide that provides the optical path length difference. In this embodiment, the two waveguides following the S-bend type waveguides 7A and 8A are either a straight optical waveguide or a bent optical waveguide, and the vicinity of the start end connected to the S-bend type waveguides 7A and 8A is a straight line. It is formed in a shape.

  In the structure of the present invention, the core part of the optical waveguide of each part excluding the phase controller may have a structure similar to that of the phase controller, and may be a waveguide core containing a silicon material into which no dopant is implanted. A transition portion using a tapered structure may be provided in front of and behind the phase controller as an embedded optical waveguide that does not include the phase controller.

An example in the case where the structure of the present invention is specifically designed will be described below.
As a concrete example of the first design, an embedded optical waveguide having a rectangular cross section, SiON (silicon oxynitride) whose refractive index is adjusted to 1.5 is used as a core material, and the refractive index is 1.465. A three-dimensional beam propagation method (BPM) simulation of a high relative refractive index difference optical waveguide with a relative refractive index difference Δ = 2.3% using a certain SiO ₂ (silica) in the lower cladding layer and the upper cladding layer The result of examination is shown. In this type of waveguide design, simulation by the three-dimensional beam propagation method is a simulation means that is commonly used by those skilled in the art at the time of design.

The simulation was performed assuming that the thickness of the lower cladding layer was 5 μm, the width W of the core portion was 3.2 μm, the height was 3.2 μm, and the thickness of the upper cladding layer was 8.2 μm. Since the width G of the branch gap is 0.4 μm, the width of the output end of the tapered portion is 6.8 μm. The radius of curvature of the S-bend type waveguide was 1000 μm, and the separation distance of the output waveguide subsequent thereto was 127 μm.
FIG. 8 shows the intensity ratio of the output light power from the first and second output waveguides to the input light power of the input waveguide. In addition, the taper part length of the horizontal axis in FIG. 8 shows the value of the X coordinate length on the simulation software used in this example.
Here, the output waveguide connected to the upper side of FIG. 2 and FIG. 3, that is, the side of the taper portion whose side wall is parallel to the extension axis of the optical axis of the input waveguide linear portion is the first output waveguide. The other is the second output waveguide. The output light intensity of each output waveguide is obtained by calculating the overlap integral between the light intensity distribution of the output light from the output waveguide and the light intensity distribution of the basic propagation mode of the output waveguide, and calculating the relative value with respect to the incident light intensity. I asked for it.

FIG. 9 is a branching ratio obtained by dividing the output light intensity of the waveguide having the larger output by the output light intensity of the waveguide having the smaller output. In addition, the scale of the vertical axis | shaft in the intersection of the vertical axis | shaft of FIG. The taper length of the horizontal axis in FIG. 9 indicates the value of the X coordinate length on the simulation software used in this example. In the simulation result shown in FIG. 8, the output of the first waveguide is larger by one data point with the taper length L = 20 μm. The branching ratio increases as L increases, and increases to 3.4 when L = 60 μm.
During this time, the excess loss shown in FIG. 10 is 0.3 to 0.2 dB. This section having a taper length of 20 μm to 60 μm is referred to as a first section.
The branching ratio starts to decrease from the position of the taper length L = 60 μm, and becomes 1.1 at L = 140 μm. In the range of the taper length L = 60 to 140 μm, the excess loss is reduced to 0.02 dB to 0.01 dB. This range will be referred to as the second section.
As shown in FIG. 8, after that, the branching ratio repeatedly increases and decreases, but the maximum branching ratio becomes smaller. In order to obtain a desired branching ratio, it is necessary to select the taper length L from the first section or the second section, and the second section has less excess loss than the first section. In addition, since the change in the branching ratio with respect to the change in the taper length is gradual, the tolerance for the processing error during manufacturing can be increased.

 From the above examination results, in the design of this high relative refractive index difference optical waveguide, by selecting a numerical value capable of realizing a desired branching ratio from the range of L = 60 to 140 μm as the taper length, the excess loss is reduced and the manufacturing is performed. It is possible to obtain an optical branching element having a branching ratio of 1.1 to 3.4 with a large tolerance.

As a specific example of the second design, an embedded optical waveguide having a rectangular cross section, a core material is fluorinated polyimide whose refractive index is adjusted to 1.534, and the refractive index is adjusted to 1.529. The result of having examined by the two-dimensional BPM (Beam Propagation Method) simulation about the low relative refractive index difference optical waveguide with relative refractive index difference (DELTA) = 0.3% which used the fluorinated polyimide for the lower clad layer and the upper clad layer is shown. In this type of waveguide design, simulation by two-dimensional BPM is a simulation means commonly used by those skilled in the art at the time of design.
The calculation was performed assuming that the lower clad layer had a thickness of 15 μm, the core had a width: W : 7 μm × height: 7 μm, and the upper clad layer had a thickness of 22 μm. Since the width G of the branch gap is 2 μm, the width of the output end of the tapered portion is 16 μm. The radius of curvature of the S bend portion was 40 mm, and the separation distance of the subsequent waveguide was 127 μm.

11 shows the intensity ratio of the output light power from the first and second output waveguides with respect to the incident light power of the input waveguide, and FIG. 12 shows the branching ratio. In this simulation result, the output of the first waveguide is larger by two points of data of the taper length L = 50 μm and 100 μm. Except for these two points, the branching ratio increases as L increases and increases to 5.2 when L = 400 μm. During this time, the excess loss shown in FIG. 13 is 0.11 to 0.16 dB. This will be referred to as the first section. From here, the branching ratio starts to drop, and becomes 1.1 at L = 900 μm. The excess loss is reduced to 0.15 to 0.04 dB. This will be referred to as a second section. Excluding L = 400 μm, which is the boundary between the first section and the second section, and looking at L = 450 to 900 μm, the excess loss is as small as 0.12 to 0.04 dB, indicating an excellent value. ing.

Thereafter, also in this simulation example, the branching ratio increases and decreases, and the maximum branching ratio decreases. In order to obtain a desired branching ratio, it is necessary to select a taper length from the first section or the second section, and the second section has less excess loss than the first section, and Since the change in the branching ratio with respect to the change in the taper length is gradual, the tolerance for processing errors during manufacturing is large.
From the above simulation results, even in the design of this low relative refractive index difference optical waveguide, by selecting a numerical value capable of realizing a desired branching ratio from the range of L = 400 to 900 μm in the second section as the taper length, It is possible to obtain an excellent optical branching device having a branching ratio of 1.1 to 5.2 with low loss and high manufacturing tolerance.

  As described above, according to the structure of the above-described Mach-Zehnder interferometer type optical functional elements A, B, and C, the branching ratio of the demultiplexing units 6, 32, and 52 is the unequal branching ratio, and the second optical path having a large loss By adopting a structure in which a large amount of light is incident on 8, 34, 54, it is possible to make the emitted light intensity of the first optical paths 7, 33, 53 and the emitted light intensity of the second optical paths 8, 34, 54 uniform. Mach-Zehnder interferometer-type optical functional elements A, B, and C that suppress the decrease in the light output intensity and suppress the deterioration in the extinction ratio can be provided. In addition, since the branching ratios of the demultiplexing units 6, 32, and 52 are fixed, the above-described effects can be obtained without complicating the control.

  Next, if phase modulation is performed with a structure in which a P-type or N-type semiconductor region is included in the light modulators 12, 39, 59 and a refractive index change due to a change in carrier concentration of the semiconductor region is induced by a voltage due to an electrode, In the Mach-Zehnder interferometer type optical functional elements A, B, and C, the response speed as an optical switch can be increased.

  In each embodiment, a substrate-type optical waveguide having a core structure composed of the core portion 17 and the cladding layers 15 and 16 and configured by injecting an element serving as a dopant into silicon as a semiconductor region of the light modulation portions 12, 39 and 59. In the case of a component, an optical waveguide can be formed on the substrates 1, 21, and 49 using silicon photonics technology, so that an optical component with a miniaturized element can be obtained and can be provided at low cost.

The branching ratio can be easily fixed and controlled by adopting a structure including a taper portion 6a in which the core width is widened as the branching portion 6 of the Mach-Zehnder interferometer type optical functional element A shown in FIG. The output light intensity of the first optical path 7 and the output light intensity of the second optical path 8 can be made uniform without complicating the light intensity, and the decrease in the light output intensity can be suppressed and the deterioration of the extinction ratio can be suppressed. .
The branching unit 6 has a fixed branching ratio lower than 9 to 1, and a structure having an intermediate branching ratio that is non-equal branching has low loss in a single-mode Y-branch substrate type waveguide, It can be realized with a large manufacturing tolerance.

Further, as a taper length L of the demultiplexing unit 6, a graph of the result of measuring the intensity ratio of the output light power from the first output waveguide 7 and the second output waveguide 8 with respect to the incident light power of the input optical waveguide 5 , The first section in which the branching ratio increases as the length of the tapered portion 6a increases, and the second section in which the branching ratio decreases as the length of the subsequent tapered section increases. By utilizing as described above, it is possible to provide the Mach-Zehnder interferometer type optical functional element A that can achieve a further low loss and a large manufacturing tolerance.
In addition, since the Mach-Zehnder interferometer type optical functional elements B and C are also provided with the demultiplexing sections 32 and 52 having the same configuration as the demultiplexing section 6, the same operational effects as the Mach-Zehnder interferometer type optical functional element A can be obtained. Can do.
Further, in the Mach-Zehnder interferometer type optical functional elements A, B, and C, the second optical paths 8, 34, and the second optical paths 8, 34, and 54 having the shorter optical path lengths of the first optical paths 7, 33, and 53 and the second optical paths 8, 34, 54 is provided with the phase controllers 12, 39, 59, so that the transmission loss is smaller than the structure in which the phase controllers 12, 39, 59 are provided on the sides of the first optical paths 7, 33, 53 having a long optical path length. The difference can be suppressed.

FIG. 1 is a plan view showing a schematic structure of a first embodiment of a Mach-Zehnder interferometer type optical functional element according to the present invention. FIG. 2 is a plan view showing an input optical waveguide, a demultiplexing unit, a first optical path, and a second optical path applied to the first embodiment of the Mach-Zehnder interferometer type optical functional element. FIG. 3 is an enlarged view of a demultiplexing unit applied to the first embodiment of the Mach-Zehnder interferometer type optical functional element. FIG. 4 is a sectional view showing an internal structure of a demultiplexing unit applied to the first embodiment of the Mach-Zehnder interferometer type optical functional element. FIG. 5 is a partial cross-sectional view in the case where the phase controller applied to the first embodiment of the Mach-Zehnder interferometer type optical functional element is a silicon photonics device. FIG. 6 is a block diagram showing a second embodiment of the Mach-Zehnder interferometer type optical functional element according to the present invention. FIG. 7 is a block diagram showing a third embodiment of a Mach-Zehnder interferometer type optical functional element according to the present invention. FIG. 8 shows the result of a simulation analysis by a three-dimensional beam propagation method based on the material and specific dimensions of each structural part and the length of the tapered part of the demultiplexing part in the structure of the Mach-Zehnder interferometer type optical functional element according to the present invention. The figure which shows the relationship between the output optical power of each output waveguide, and the length of a taper part. FIG. 9 is a diagram showing the relationship between the branching ratio of the branching portion and the length of the tapered portion obtained as a result of the same analysis. FIG. 10 is a diagram showing a relationship between excess loss and the length of the tapered portion obtained as a result of the analysis. FIG. 11 shows each output obtained as a result of a two-dimensional BPM simulation analysis based on the material and specific dimensions of each structural part and the length of the tapered part of the demultiplexing part in the structure of the Mach-Zehnder interferometer type optical functional element according to the present invention. The figure which shows the relationship between the output optical power of a waveguide, and the length of a taper part. FIG. 12 is a diagram showing the relationship between the branching ratio of the branching portion and the length of the tapered portion obtained as a result of the same analysis. FIG. 13 is a diagram showing the relationship between excess loss and the length of the tapered portion obtained as a result of the same analysis.

Explanation of symbols

  A ... Mach-Zehnder interferometer type optical functional element, G ... Gap, 1 ... Substrate, 2 ... Transmission path, 3 ... Phase controller, 5 ... Input optical waveguide, 6 ... Demultiplexing part, 6A ... Input side end face, 6B ... First 1 side surface, 6C ... 2nd side surface, 6D ... output side end surface, 7 ... 1st optical path, 7A ... S bend type waveguide, 8 ... 2nd optical path, 8A ... S bend type waveguide, 9 DESCRIPTION OF SYMBOLS multiplex part, 10 ... 1st output waveguide, 11 ... 2nd output waveguide, 12 ... light modulation part, 13 ... wiring, 14 ... control part, 15 ... lower clad layer, 16 ... upper clad layer, 17 ... Core part, 31, 51 ... Input optical waveguide, 32, 52 ... Demultiplexing part, 33, 53 ... First optical path, 34, 54 ... Second optical path, 36, 56 ... Photonic crystal part, 37, 38, 57, 58... Electrode, 39, 59... Light modulator, 40, 60... Multiplexer, 41, 61. , 62 ... second output waveguides,

Claims (5)

An input optical waveguide, a demultiplexing unit that is connected to the input optical waveguide and that divides and emits signal light incident from the input optical waveguide, and a light that is output from the demultiplexing unit A first optical path, a second optical path, a light modulation unit that is provided as a part of the second optical path and adjusts the phase of the signal light according to a phase control signal input from the outside, and the first optical path; A combining unit that is connected to the second optical path and combines and emits the signal light from the first optical path and the signal light from the second optical path, and the combined light from the combining unit is output A Mach-Zehnder interferometer type optical functional element having an output waveguide,
The second optical path has a light propagation loss larger than that of the first optical path due to the optical modulation unit, and the branching unit has a fixed unequal branching ratio, and is incident on the second optical path. Ri Na is stronger than the intensity of light intensity incident on the first optical path,
In the state where the phase is controlled in the optical modulation unit so that the branching ratio of the demultiplexing unit maximizes the output light intensity from the output waveguide, the signal light intensity to the multiplexing unit from the first optical path and , The signal light intensity from the second optical path to the combined portion is determined to be equal,
The branching portion is a Y-branch type optical waveguide, and tapered portions having a gradually widening core portion are formed from the incident side to the outgoing side of the branching portion, and both side surfaces of the tapered portion with respect to the extension axis at the center of the input optical waveguide Is different between the first optical path side and the second optical path side,
Two S-bend type optical waveguides are connected to the output end of the demultiplexing unit to form the first optical path and the second optical path, and in the demultiplexing unit, before entering the tapered unit The input optical waveguide portion is straight, and of the two side walls of the tapered portion, the side wall on the first optical path side is parallel to the optical axis of the linear portion near the tapered portion of the input optical waveguide, The two side walls on the second optical path side are inclined with respect to the optical axis of the linear portion near the tapered portion of the input optical waveguide, and the width of the output end of the tapered portion is connected to the output end of the tapered portion. The sum of the widths of the S-bend optical waveguides is equal to the sum of the gap widths at the beginnings of the two S-bend optical waveguides, and the distance between the two S-bend optical waveguides is Narrowest at the start end side connected to the output end of the taper portion, and toward each other Gradually separated becomes in the S bend type the tapered portion of the end face the input optical waveguide termination end face and parallel to the has been a Rukoto Mach-Zehnder interferometer type optical functional device according to claim of connecting the optical waveguide of the . The light modulation unit includes a region that is a P-type semiconductor and / or a region that is an N-type semiconductor, and the light modulation unit applies a voltage to the two electrodes as a phase control signal, so that A structure in which a refractive index change due to a carrier concentration change is induced in a region, and an optical propagation loss that is excessive in the second optical path as compared with the first optical path is a loss due to free carrier absorption. The Mach-Zehnder interferometer-type optical functional element according to claim 1 , wherein In each of the input optical waveguide, the demultiplexing unit, the first optical path, the second optical path, and the multiplexing unit, each portion through which signal light passes is a cladding made of a material having a high refractive index and a cladding made of a material having a low refractive index. An optical waveguide structure formed by surrounding layers, wherein the core portion is made of silicon, silicon nitride, or silicon oxynitride, and a semiconductor region of the light modulation portion is configured by injecting an element serving as a dopant into silicon. , Mach-Zehnder interferometer type optical functional device according to claim 1 or 2, characterized in that a planar optical waveguide component. With the length of the tapered portion of the branching portion on the horizontal axis and the branching ratio on the vertical axis, the output light power from the first output waveguide and the second output waveguide with respect to the incident light power of the input optical waveguide The graph of the result of measuring the intensity ratio is drawn, and the first section where the branching ratio increases as the length of the tapered portion becomes the first section, and the length of the subsequent tapered section becomes longer when section branching ratio decreases is the second section, according to any one of claims 1 to 3 the length of the tapered portion is characterized in that it is selected from the range of the second section Mach-Zehnder interference type optical functional element. The first even Ri by the optical path only a short said second optical path of the optical path length, the Mach-Zehnder interferometer according to any one of claims 1 to 4, characterized in that the phase controller is provided Type optical functional element.

Images