Patent Application
20150293384
1. Field of the Invention
The present invention relates to an optical waveguide element having high speed and low optical loss, and an optical modulator including the optical waveguide element.
2. Description of the Related Art
In recent years, the use of an optical integrated circuit, such as a planar optical waveguide, in a device for optical fiber communication, especially, long-distance wavelength multiplex optical fiber communication, has been considered.
PCT International Publication No. WO95/08787 discloses a single-mode silicon rib waveguide which has a PIN junction in a lateral direction and changes the earner density in the PIN junction to control refractive index.
U.S. Pat. No. 7,085,443 describes an optical modulator which is a silicon rib waveguide having a PN diode in a lateral direction, and in which the dopant density of each of P and N regions is uneven and is changed nonlinearly, the dopant density has a maximum value in a portion which is continuous to the electrode of each of the P and N regions, and the dopant density has a minimum value at the center of a waveguide mode.
Kensuke Ogawa, Kazuhiro Goi, Yong Tsong Tan, Tsung-Yang Liow, Xiaoguang Tu, Qing Fang, Guo-Qiang Lo, and Dim-Lee Kwong, “Silicon Mach-Zehnder modulator of extinction ratio beyond 10 dB at 10.0-12.5 Gbps”, Optics Express, 2011, Vol 19, No. 26, pp. B26-B31 discloses a high-speed optical modulator of 10 Gbps (gigabits/second) which uses a Mach-Zehnder waveguide having a silicon rib waveguide with a PN diode in a lateral direction as a phase modulation unit, and in which it is reported that an extinction ratio is equal to or greater than 10 dB, and optical loss is equal to or less than 10.5 dB.
An electro-optical element of PCT International Publication No. WO95/08787 uses refractive index control in a single-mode silicon waveguide. However, if a silicon waveguide with the width and thickness of a core of submicron scale is used, optical loss is increased due to optical scattering caused by lateral roughness of the waveguide core, and it is not possible to realize low optical loss, which is required for an optical integrated circuit for long-distance wavelength multiplex optical fiber communication.
Based on the distribution of the dopant density of the PN diode disclosed in U.S. Pat. No. 7,085,443, when expanding the core width in order to reduce optical loss by lateral roughness of the waveguide core, a higher order mode is generated. As a result, if a higher order mode component is superimposed on an intensity modulation signal or a phase modulation signal, an extinction ratio in intensity modulation or a Q value (Q-factor) in phase modulation is decreased, a bit error rate is increased, and an excellent optical communication system is not constructed.
In the silicon Mach-Zehnder waveguide having the PN diode disclosed in Kensuke Ogawa, Kazuhiro Goi, Yong Tsong Tan, Tsung-Yang Liow, Xiaoguang Tu, Qing Fang, Guo-Qiang Lo, and Dim-Lee Kwong, “Silicon Mach-Zehnder modulator of extinction ratio beyond 10 dB at 10.0-12.5 Gbps”, Optics Express, 2011, Vol. 19, No. 26, pp. B26-B31, it is difficult to reduce optical loss in a wavelength region including a C band (1530 to 1565 nm) and an L band (1565 to 1625 nm) more than 10 dB. Therefore, it is not possible to realize low optical loss which is required for optical components for long-distance wavelength multiplex optical fiber communication.
From the above, the present invention provides an optical waveguide element which is suitable for an optical integrated circuit, such as an optical modulator having high-speed and low optical loss, and which has a high extinction ratio in intensity modulation or a high Q value in phase modulation in an entire wavelength region for use in long-distance wavelength multiplex optical fiber communication.
According to a first aspect of the present invention, an optical waveguide element is provided, including: a rib waveguide which includes a core including a rib portion and a pair of slab portions including a first slab portion and a second slab portion connected to both sides of the rib portion so as to sandwich the rib portion, wherein the rib portion includes a cross-sectional dimension which allows the propagation of a fundamental mode and a higher order mode in a predetermined single polarization state, and includes a first P-type semiconductor and a first N-type semiconductor forming a PN junction, the first slab portion includes a second P-type semiconductor and a P-type conductor connected to each other, and the second P-type semiconductor is connected to the first P-type semiconductor of the rib portion, the second slab portion includes a second N-type semiconductor and an N-type conductor, and the second N-type semiconductor is connected to the first N-type semiconductor of the rib portion, the first and second P-type semiconductors are made of a semiconductor material including a P-type dopant, the first and second N-type semiconductors are made of a semiconductor material including an N-type dopant, the P-type conductor is made of a semiconductor material including a P-type dopant in a higher concentration than the first and second P-type semiconductors, is electrically connected to a first electrode, and is arranged in a region where the higher order mode propagates, and the N-type conductor is made of a semiconductor material including an N-type dopant in a higher concentration than the first and second N-type semiconductor, is electrically connected to the second electrode, and is arranged in a region where the higher order mode propagates.
According to a second aspect of the present invention, in the optical waveguide element according to the first aspect, it is preferable that the semiconductor material be silicon.
According to a third aspect of the present invention, it is preferable that the optical waveguide element according to the first or second aspect further include: an upper cladding which is arranged in a surrounding of the core, the first and second electrodes being formed on the upper surface of the upper cladding; and connection conductors which electrically connect the first electrode and the P-type conductor, and the second electrode and the N-type conductor, and pass through the upper cladding.
According to a fourth aspect of the present invention, in the optical waveguide element according to any one of the first to third aspects, it is preferable that the width of the rib portion of the rib waveguide be 500 to 600 nm.
According to a fifth aspect of the present invention, it is preferable that the optical waveguide element according to any one of the first to fourth aspects further include: a rectangular waveguide which includes a rectangular core portion having the same width and thickness as the rib portion; and a connection portion which connects at least one end of the rib waveguide and the rectangular waveguide, and comprises a transition region where the width of each of the pair of slab portions changes from zero in succession.
According to a sixth aspect of the present invention, in the optical waveguide element according to the fifth aspect, it is preferable that the width of each of the pair of slab portions in the transition region change linearly in a propagation direction of light.
According to a seventh aspect of the present invention, in the optical waveguide element according to the fifth aspect, it is preferable that the width of each of the pair of slab portions in the transition region change quadratically in a propagation direction of light.
According to an eighth aspect of the present invention, it is preferable that the optical waveguide element according to any one of the fifth to seventh aspects further include: an inverted tapered waveguide which is connected to an end portion on an opposite side to the connection portion of the rectangular waveguide, and has a narrower width toward the tip thereof.
According to a ninth aspect of the present invention, it is preferable that the optical waveguide element according to any one of the fifth to seventh aspects further include: a Mach-Zehnder waveguide which includes two arm portions between two optical branching parts, and each of the 1×2 optical branching parts be constituted by a multimode interferometer, each of the two arm portions includes the rib waveguide, and the rib waveguides and the 1×2 optical branching parts be connected by the rectangular waveguides.
According to a tenth aspect of the present invention, in the optical waveguide element according to any one of the first to ninth aspects, it is preferable that an inverted tapered waveguide with a narrower width toward the tip be connected to an end portion of a rectangular waveguide in an end portion in which light is incident on the optical waveguide element or light is emitted from the optical waveguide element.
According to an eleventh aspect of the present invention, an optical modulator is provided, including: the optical waveguide element according to any one a the first to tenth aspects, wherein a constant reverse bias voltage is applied to a first electrode connected to the P-type conductor and a second electrode connected to the N-type conductor and an electrical signal only including an AC component is transmitted to either the first electrode or the second electrode to perform optical modulation.
According to the aspects of the present invention, it is possible o provide an optical waveguide element which is suitable for an optical integrated circuit, such as an optical modulator having high-speed and low optical loss, and which has a high extinction ratio in intensity modulation or a high Q value in phase modulation.
Hereinafter, the present invention will be described based on a preferred exemplary embodiment referring to the drawings.
A core 12 of an optical waveguide has a rib portion 22, and a pair of slab portions 23 and 24 respectively connected to both sides of the rib portion 22. The thickness trib of the rib portion 22 is greater than the thickness tslab of each of the slab portions 23 and 24. The slab portions 23 and 24 are connected to the sides (both sides) in the lower portion of the rib portion 22, and the lower surface of the rib portion 22 facing a lower cladding 2 is flush with the lower surface of each of the slab portions 23 and 24 facing the lower cladding 2.
In the surrounding of the core 12, an upper cladding 7 and a lower cladding 2 are disposed. The claddings are made of a material having a refractive index lower than that of the core 12 When the core 12 is made of a semiconductor material having a high refractive index, such as silicon (Si), the claddings can be made of silica (SiO2) silicon nitride, or the like. The claddings may be made of an electrical insulation material.
A substrate 1 may be a semiconductor substrate or an insulator substrate, and when an SOI (Silicon-on-insulator) wafer is used, the substrate 1, the lower cladding 2, and the core 12 can be respectively made of a silicon substrate, a buried oxide (BOX) layer, and an SOI layer. In this case, the substrate 1 may be undoped silicon, and silicon of the core 12 is doped as described below.
The lower cladding 2 may have, for example, about 2 μm or a sufficient thickness. The upper cladding 7 can be formed by synthesizing a compound, such as silica (SiO2), and depositing the compound on the rib portion 22 and the slab portions 23 and 24 at a thickness of 1 to 2 μm using a manufacturing method, such as CVD (Chemical Vapor Deposition).
In the core 12, a P-type semiconductor 3 and an N-type semiconductor 4 constituting a PN diode exist. In the boundary between the P-type semiconductor 3 and the N-type semiconductor 4, a PN junction 13 of capacitance C (see an equivalent circuit of
A P-type conductor 5 is connected to the end portion of the P-type semiconductor 3 opposite to the PN junction 13. Similarly, an N-type conductor 6 is connected to the end portion of the N-type semiconductor 4 opposite to the PN junction 13. The P-type conductor 5 is made of a semiconductor material containing a P-type dopant in a higher concentration than the P-type semiconductor 3, and the N-type conductor 6 is made of a semiconductor material containing an N-type dopant in a higher concentration than the N-type semiconductor 4.
As the semiconductor materials constituting the P-type conductor 5 and the N-type conductor 6, the same semiconductor materials as the P-type semiconductor 3 and the N-type semiconductor 4 are preferably used. For the P-type dopant and the N-type dopant, the same materials are preferably used in the semiconductors and the conductors
Examples of the semiconductor material include group IV semiconductors, such as silicon (Si), germanium(Ge), and silicon germanium (SiGe), group III-V semiconductors, such as AlN, GaP, GaAs, and InP, and group II-VI semiconductors, such as ZnO, ZnS, ZnSe, and CdTe. The group IV semiconductors are preferably used, and for example, silicon (Si) can be used.
In the case of the group IV semiconductors, as the P-type dopant, group III elements, such as boron (B), are used, As a method of forming, the P-type semiconductor 3 and the P-type conductor 5, a method of adding the P-type dopant to the semiconductor material layer constituting the core 12 by ion implantation or the like is used.
As the N-type dopant of the group IV semiconductors, group V elements, such as phosphorus (P) and arsenic (As), are used. As a method of forming the N-type semiconductor 4 and the N-type conductor 6, a method of adding the N-type dopant to the semiconductor material layer constituting the core 12 by ion implantation or the like is used.
The rib portion 22 has a P-type semiconductor 3a and an N-type semiconductor 4a, and the P-type semiconductor 3a and the N-type semiconductor 4a in the rib portion 22 form the PN junction 13.
A first slab portion 23 which is one of a pair of slab portions has a P-type semiconductor 3b and a P-type conductor 5 connected to each other, and the P-type semiconductor 3b of the first slab portion 23 and the P-type semiconductor 3a of the rib portion 22 are connected to each other to form an integrated P-type semiconductor 3.
A second slab portion 24 which is the other slab portion of a pair of slab portions has an N-type semiconductor 4b and an N-type conductor 6 connected to each other, and the N-type semiconductor 4b of the second slab portion 24 and the N-type semiconductor 4a of the rib portion 22 are connected to each other to form an integrated N-type semiconductor 4.
The P-type conductor 5 and the N-type conductor 6 are respectively electrically connected to electrodes 10 and 11 in order to apply a voltage to the P-type semiconductor 3 and the N-type semiconductor 4 constituting the PN diode.
In the example shown in
The connection conductors and 9 and the electrodes 10 and 11 can be formed by depositing a conductor, such as a metal, using plating, vapor deposition, or the like. For example, the connection conductors 8 and 9 are formed by tilling openings formed in the upper cladding 7 using photolithography or the like with a conductor, and the electrodes 10 and 11 are formed by laminating a conductor on the upper surface of the upper cladding 7.
In the example shown in
A constant reverse bias voltage is applied between the electrode 10 connected to the P-type conductor 5 and the electrode 11 connected to the N-type conductor 6, and an electrical signal including only an AC component is transmitted to either electrode, thereby performing optical modulation. An electrical signal for use in modulation preferably includes only an AC component oscillating to positive and negative, without including a DC component. As the electrical signal, a high-frequency signal of for example, 1 GHz to 20 GHz is preferably used for a high-speed operation.
The P-type conductor 5 and the N-type conductor 6 are doped in a high concentration in order to reduce electrical resistance which adversely affects the high-speed operation. For example, while the dopant concentration in the P-type semiconductor 3 and the N-type semiconductor 4 is about 1018 cm−3, the dopant concentration in the P-type conductor 5 and the N-type conductor 6 is about 1020 cm−3. The P-type conductor 5 and the N-type conductor 6 have electrical resistance R and inductance L of the substantially same values (see the equivalent circuit of
As described above, in the waveguide using the semiconductor material, such as silicon, for example, light scattering due to lateral roughness of the waveguide core when forming a waveguide pattern using photolithography significantly affects optical loss of the waveguide. In order to reduce optical loss due to lateral roughness, it is desirable that the position of the core lateral is separated from the center of a fundamental mode of the waveguide by widening the core width.
In the case of a rib waveguide, the rib portion 22 is a principal portion of the core 12 which allows the propagation of the fundamental mode. For this reason, if the width wrib of the rib portion 22 is wide, optical loss of the waveguide can be reduced. For example, it is preferable that the rib width wrib be 500 to 600 nm.
Table 1 shows optical loss with respect to the rib width of the rib waveguide having the silicon core.
However, if the rib width wrib is widened, a higher order mode more strongly localizes in the rib portion 22 of the waveguide core. That is, the rib portion 22 has a cross-sectional dimension such that a fundamental mode and a higher order mode in a specific polarization state propagate therethrough. In a planar optical waveguide, there are a TE component horizontal to the substrate 1, the slab portions 23 and 24, and the like, and a TM component vertical to the substrate 1, the slab portions 23 and 24, and the like. In general, the TE component contributes to waveguide, and the waveguide mode of the TM component does not contribute to waveguide. This is because an effective refractive index of the fundamental mode of the TM component is lower than an effective refractive index of the higher order mode (for example, N=2) of the TE component.
As an example,
At the rib width of 400 nm, the higher order mode of N=2 localizes on the slab side. Therefore, although the rib waveguide is a single-mode waveguide, it is understood from Table 1 that optical loss is high. When the rib width is greater, optical loss is smaller; however, a higher order mode tends to localize at the rib center. When the rib width is increased to a maximum of 700 nm and the same mode calculation is executed, it has been confirmed that a mode of a. TE component of a higher order than N=2 does not propagate through the rib waveguide.
Hereinafter, the “fundamental mode” indicates an N=1 mode of a TE component, and the term “higher order mode” indicates an N=2 mode of a TE component.
Accordingly, in the optical waveguide element of the present exemplary embodiment, in order to avoid degradation of an extinction ratio, a Q value, and the like by propagation of a higher order mode, a member which attenuates a higher order mode without attenuating a fundamental mode is disposed. With this, it is possible to constitute an optical waveguide which is a multimode waveguide allowing the propagation of a higher order mode and substantially functions as a single-mode waveguide with the dimension of the core cross-section.
Specifically, the P-type conductor 5 and the N-type conductor 6 are disposed in the slab portions 23 and 24. In the related art, in order to enhance electrical connection of the slab portions and the electrodes, there is a case where a region having a high dopant concentration is disposed around the connection portions of the slab portions and the electrodes for example, see U.S. Pat. No. 7,085,443). In the present invention, the forming range of the P-type conductor 5 and the N-type conductor 6 extends near the rib portion 22, and is disposed in a region where a higher order mode propagates.
In the P-type conductor 5 and the N-type conductor 6, holes and electrodes are distributed as carriers with a high density. Therefore, light is attenuated by free carrier absorption. It is known that an optical absorption coefficient by free carrier absorption is increased in proportion to the carrier density.
As described above, since the higher order mode spreads to the slab portions 23 and 24, attenuation in the P-type conductor 5 and the N-type conductor 6 occurs strongly with respect to the higher order mode. With this, it is possible to selectively attenuate the higher order mode. Accordingly, the rib waveguide according to the present invention can reduce the influence of optical loss by lateral roughness by widening the rib width and transmission loss as a single-mode waveguide, which allows the propagation of only the fundamental mode, can be realized. Therefore, it is possible to avoid deterioration of characteristics in modulation (for example, degradation of an extinction ratio in intensity modulation, degradation of a Q value in phase modulation, or the like).
Free carrier absorption causes a change in a refractive index. This is generally a phenomenon which is called plasma dispersion or the like. In an optical modulator using a silicon waveguide, the carrier density can be changed by a high-frequency electrical signal using plasma dispersion, and an effective refractive index of the waveguide can be modulated at high speed.
In the optical waveguide element of the present exemplary embodiment, if a high-frequency signal is applied, the carrier density is changed around the PN junction 13 located in the rib portion 22 of the waveguide core 12. With this, it is possible to modulate an effective refractive index of the waveguide at high speed. Since the electrical resistance R of the P-type conductor 5 and the N-type conductor 6 is small, and a voltage drop in the P-type conductor 5 and the N-type conductor 6 is small, a change in the carrier density in the P-type conductor 5 and the N-type conductor 6 is negligible. Accordingly, it is possible to attenuate the higher order mode without temporal fluctuations even when a high-frequency electrical signal is applied.
It is possible to optimize the interval wPN between the P-type conductor 5 and the N-type conductor 6 according to the rib width wrib. For example, the interval wPN is adjusted such that optical loss becomes as small as possible, and characteristics in modulation (for example, an extinction ratio in intensity modulation, a Q value in phase modulation, or the like) are improved as much as possible One procedure is illustrated, in which an intensity modulator or a phase modulator is constituted to adjust the interval wPN such that factors IER/Iloss, and IER/IQ are maximized thereby optimizing the interval wPN. The IER, Iloss, and respectively represent an extinction ratio, a Q value, and optical loss in terms of dB.
As a specific example, the interval wPN between the P-type conductor 5 and the N-type conductor 6 can be set to 2 μm (2000 nm) with respect to the rib width wrib of 500 to 600 nm. For example, the interval wPN may be set to about two or five times the rib width wrib. The middle point of the interval wPN in the rib width direction (lateral direction) exists substantially at the center of the rib portion 22. In order to suppress the pumping of the higher order mode, preferably, the PN junction 13 exists substantially at the center of the rib portion 22. This is because the electric field profile of the higher order mode is asymmetric in the lateral direction, and thus, when the position of the PN junction 13 is deviated from the center of the rib portion 22 and is asymmetric in the lateral direction, the higher order mode is easily pumped.
In the example of
In order to suppress the pumping, of the higher order mode, preferably, the distribution of the doping density (dopant concentration) of the P-type conductor 5 and the N-type conductor 6 is substantially even. This is because, if the doping density is uneven, when a high-speed electrical signal is applied, the higher order mode is pumped by intensity modulation in the slab portions 23 and 24, resulting in deterioration of characteristics, such as an extinction ratio, a Q value, and the like.
When manufacturing a waveguide, since there is misalignment due to restriction of manufacturing accuracy, there is an error of ±10 nm order. In the thickness control of each layer, there is an error of ±tens of nm order. In order to evaluate the characteristics of the waveguide, it is desirable that a plurality of waveguides having the same specification is prepared, and an average value and a standard deviation are obtained. In the description of this specification, although description of errors as regards the dimension and characteristics of the waveguide is omitted, it is desirable to consider errors, such as a manufacturing error.
For example, when “the PN junction exists at the center of the rib portion in the lateral direction”, this means that there is a manufacturing error in the position of the PN junction or each of both laterals of the rib portion, and there is also an error in the addition range of the dopant of the P-type semiconductor and the N-type semiconductor. Since there is also roughness on both laterals of the rib portion, it is desirable that a condition is satisfied to the same degree as an error, for example, within a range of about one time to three times the error in an average position based on a reference, such as a predetermined section in a longitudinal direction. Regarding “the middle point of wPN exists at the center of the rib portion in the lateral direction”, there is an error in each of the positions of the boundary between either of the P-type and N-type conductor, which become both ends of wPN, and the semiconductor, and the positions of both laterals of the rib portion, and it is desirable to take the error into consideration similarly.
In the case of the silicon waveguide, the core is Si, and the cladding is SiO2. The refractive indexes of the materials of the core and the cladding are respectively 3.5 and 1.45. In this case, the refractive index difference between the core and the cladding is great, and contrast is high.
In the fundamental mode (
For comparison, the same calculation has been performed in the case of a compound semiconductor waveguide. In an InGaAsP/InP-based compound semiconductor material, a combination in which a core is In1-xGaxAsyP1-y (where x=0.33, y=0.72), and a cladding is InP is possible. In this combination, since there is no lattice mismatching between the core and the cladding and the influence of distortion can be eliminated, if the cladding thickness is set to be equal to or greater than 500 nm, the confinement of guided light to the core can be secured. The refractive indexes of the materials of the core and the cladding are respectively 3.45 and 3.18. A refractive index contrast is low compared to an Si-based or SiO2-based compound semiconductor. Since the refractive index contrast is low compared to a silicon-based compound semiconductor, the intensity center of the higher order mode tends to localize on the slab side to about wrib=1100 nm. Accordingly, the higher order mode does not become a propagation mode and it is not necessary to attenuate the higher order mode. If wrib is wider than 1100 nm, the intensity center of the higher order mode localizes in the rib, and the higher order mode becomes a propagation mode. However, since the intensity center of the higher order mode localizes strongly in the rib, the influence of intensity attenuation in a highly doped region of the lateral slab is very small and is not effective. The dimension other than wrib is the same as in the silicon waveguide. The same result is obtained even in a GaAs-based or GaAlAs-based compound semiconductor.
Accordingly, in the optical waveguide element of the present exemplary embodiment, the effect of performance improvement by higher order mode attenuation is remarkable for a high refractive index contrast waveguide. In particular, it is effective for a silicon-based waveguide (Si core and SiO2 cladding). Similarly, it is considered that this embodiment is effective for a waveguide in which the refractive index difference (or refractive index ratio) between a core and a cladding is great.
An optical waveguide which is usable in a high-speed intensity modulator or phase modulator of 10 Gbps or greater will be described in detail. Like the equivalent circuit of
When the waveguide length is equal to or greater than 3 mm, if a high-speed electrical signal (output power 500 mW to 750 mW) including only an AC component is transmitted along one of the electrode connected to the P-type conductor and the electrode connected to the N-type conductor from an electrical signal generator (the other electrode is grounded) having output impedance of 50 Ω, phase modulation amplitude can be set to be equal to or greater than π, and intensity modulation of a high extinction ratio or phase modulation of a high Q value is possible. The output power of the high-speed electrical signal is adjusted according to the wavelength. In order to avoid an increase in optical loss or deterioration of the waveform of the high-speed electrical signal, it is preferable that the waveguide length be about 6 mm or less.
As described above, in order to reduce optical loss, the rib width wrib needs to be set to 500 nm or greater. If wrib, exceeds 600 nm, since the higher order mode localizes more strongly in the rib portion, the attenuation of the higher order mode in the P-type conductor and the N-type conductor is insufficient, and the extinction ratio or the Q value is decreased Therefore, it is preferable that the rib width wrib be substantially to 500 to 600 nm or greater.
In order to constitute an optical integrated circuit, such as an optical modulator, using the rib waveguide shown in
It is preferable that the length of the transition region B be sufficiently longer than the wavelength of light. The length of the transition region B is not particularly limited, and is, for example, about 20 μm.
Similarly to the rib waveguide, in order to reduce optical loss due to lateral roughness, it is preferable that the rectangular waveguide has a cross-sectional dimension capable of allowing the propagation of the higher order mode. Accordingly, the width of the rectangular waveguide is set to be the same as the rib width. Therefore, the higher order mode propagates even through the rectangular waveguide; however, the higher order mode propagating the rectangular waveguide can be attenuated in the rib waveguide by the above-described principle. With this, it is possible to constitute an optical integrated circuit which has lower loss than an optical integrated circuit constituted by a single-mode waveguide and has performance equivalent to an optical integrated circuit constituted by a single-mode waveguide in the functions, such as the extinction ratio.
The rectangular waveguide can be connected to an inverted tapered waveguide whose width is gradually narrowed toward the end surface (or the tip). In this case, the wide end portion of the inverted tapered waveguide is connected to the rectangular waveguide, and the width of the inverted tapered waveguide is narrowed from the side connected to the rectangular waveguide toward the tip. In the end portion where the inverted tapered waveguide is connected to the rectangular waveguide, the widths of both waveguides may match each other. When optically coupling a waveguide on a substrate to an optical fiber, the narrow tip of the inverted tapered waveguide is arranged toward the optical fiber. The counterpart of optical coupling is not limited to the optical fiber, and when optically coupling a waveguide of a silicon core having a small mode field diameter (MFD) to another waveguide, such as an optical fiber having a large mode field diameter, it is preferable that an inverted tapered waveguide be disposed at the tip of the silicon waveguide. The rectangular waveguide or the inverted tapered waveguide can be constituted by a semiconductor material, such as silicon, like the rib waveguide. In the rectangular waveguide or the inverted tapered waveguide, P-type or N-type dope may not be applied.
If the inverted tapered waveguide is used, when optically coupling the optical fiber and the rectangular waveguide, it is possible to reduce mode mismatching between the end surfaces of both of the optical fiber and the rectangular waveguide and to reduce loss in optical coupling. If mode mismatching is reduced, the generation of the higher order mode can be reduced, but cannot be completely removed. When constituting an optical integrated circuit, mode mismatching mostly causes the generation of the higher order mode. The rib waveguide of the present exemplary embodiment is effective for attenuating the higher order mode.
The inverted tapered waveguide is an example of a structure of spot size conversion in which the core size is gradually narrowed to expand the mode field. Since the direction of expanding the size of the core is reverse to the direction of expanding the size of the mode field, this type is called an “inverted tapered type”.
Since the thickness of the rib waveguide and the rectangular waveguide is a submicron scale, the tip portion of the inverted tapered waveguide has minute width and end surface. Accordingly, near the end swine of the inverted tapered waveguide, it is possible to suppress reflectance of light incident on the cross-sectional area of a high refractive index core on the end surface of the inverted tapered waveguide to about −30 dB. As a result, an optical integrated circuit having high return loss (small reflection loss) as well as low optical loss and a high extinction ratio can be constituted.
The optical waveguide element of the present exemplary embodiment may constitute a Mach-Zehnder waveguide using a 1×2 multimode interferometer (MMI) as an optical branching and coupler section.
The 1×2 MMI has an advantage in that optical loss can be reduced compared to a Y-shaped optical branching and coupler section; while the extinction ratio is decreased since the higher order mode becomes the propagation mode, not a radiation mode. In the optical waveguide element of the present exemplary embodiment, the higher order mode incident, on the 1×2 MMI can be removed by the rib optical waveguide, the decrease in the extinction ratio is eliminated.
Light incident on the MMI 30A from the left waveguide 31A is branched into the right two waveguides 32A and 33A, and propagates through the arm portions 35 and 36 of the Mach-Zehnder waveguide 34. A higher order mode generated by the inverted tapered waveguide, the MMI, or the like is attenuated while propagating through the arm portions 35 and 36 and disappears. Only a fundamental mode can be incident on the MMI 30B on the combining side from both arm portions 35 and 36 from the left. Accordingly, if a high-speed electrical signal is input to the rib waveguide, signal light subjected to optical modulation with a high extinction ratio is output from the right waveguide 31B of the MMI 30B on the multiplexing side. With this, it is possible to constitute a Mach-Zehnder waveguide suitable for optical modulator with low optical loss and a high extinction ratio. If the Mach-Zehnder waveguide is used, it is possible to control the chirp of an optical signal, and to constitute a high-speed optical modulator suitable for long-haul transmission.
Although the present invention has been described based on the preferred exemplary embodiment, the present invention is not limited to the above-described embodiment, and various alteration or modifications can be made without departing from the spirit and scope of the present invention.
For example, an optical modulator using the Mach-Zehnder waveguide is an example of an optical integrated circuit using the optical waveguide element of the present invention. It is possible to constitute an optical integrated circuit, such as a multiport optical switch, by connecting Mach-Zehnder waveguides in multiple stages, by disposing a plurality of Mach-Zehnder waveguides in parallel, or by using a 1×N (where N is three or more) optical branching and coupler section (for example, an MMI).
In the Mach-Zehnder waveguide, the directions (polarities) of disposing the P-type conductor and the N-type conductor in the rib waveguides of both arm portions are not particularly limited, and any of the following may be applied.
(1) P type-rib portion—N type—space between arms—P type-rib portion—N type
(2) P type-rib portion—N type—space between arms—N type-rib portion—P type
(3) N type-rib portion P type—space between arms—P type-rib portion—N type
(4) N type-rib portion—P type—space between arms—N type-rib portion—P type
When connecting the P-type and N-type conductors and the electrodes, the lateral surfaces or lower surfaces of the P-type conductor and the N-type conductor may be connected to the electrodes.
A light source or a light receiver may be disposed in the plane of the substrate of the optical integrated circuit, and may be connected to a waveguide having the rib waveguide of the present invention.
It is preferable that, in addition to the end portion of the rectangular waveguide 21 of
Hereinafter, the present invention will be specifically described based on examples.
In the configuration shown in
A reverse bias voltage of about 1 V to 5 V was applied to the electrodes 10 and 11, a high-frequency signal including only an AC component was transmitted to any electrode (signal electrode), and the other electrode (GND) was grounded, whereby an optical modulator suitably usable for optical fiber transmission of 10 Gbps to 100 Gbps was constituted.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2012-283964 | Dec 2014 | JP | national |
This application is a continuation application based on a PCT Patent Application No. PCT/JP2013/073828, filed Sep. 4, 2013, whose priority is claimed on Japanese Patent Application No. 2012-283964 filed Dec. 27, 2012, the entire contents of which are hereby incorporated by reference.
| Number | Date | Country | |
|---|---|---|---|
| Parent | PCT/JP2013/073828 | Sep 2013 | US |
| Child | 14748331 | US |
