Optical modulator element, optical transmitter, and optical transceiver

Abstract

To provide an optical modulator element or the like capable of widening the band of a modulator by ensuring speed matching between an electric signal and signal light.SOLUTION: An optical modulator element includes, on a board, an optical branching section and an optical multiplexing section including a first material, two optical waveguide arms for connecting the optical branching section and the optical multiplexing section, and an electrode for feeding an electric signal to the two optical waveguide arm waveguides. Each optical waveguide of the two optical waveguide arms includes: a first optical waveguide containing the first material; a second optical waveguide including a second material having a higher electrooptical effect than the first material; and a transition section for performing light transition between the first optical waveguide and the second optical waveguide. The board has a hollow section in which the entire part or one part of the board is removed while being located below the second optical waveguide in a plan view.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to an optical modulator element, an optical transmitter, and an optical transceiver. [Background technology]

[0002] For example, the rapid increase in the amount of Internet Protocol (IP) data traffic has created a demand for higher capacity optical networks. In addition, further miniaturization and integration of optical transmitters and receivers is required to increase the spatial efficiency of optical transceivers. Silicon (Si) waveguides used in optical transmitters and receivers have strong optical confinement and can reduce the bending radius to about 10 μm, so SiPh (photonics) elements are beginning to be applied to 64G baud rate optical transmitters and receivers.

[0003] Optical devices are also called SiPhotonics elements (hereafter abbreviated as SiPh elements) because the optical circuits are composed of Si waveguides fabricated using SOI (Silicon-On-Insulator) wafers. The optical modulator element and the optical receiver element in the optical device are connected by an optical waveguide. SiPh elements utilize the process technology and process equipment of Si electric semiconductor elements, and many elements can be fabricated at once using large-area Si wafers, for example, 8 to 12 inches. In addition, SiPh elements have a large refractive index of about 3.4 and strong light confinement, so the bending radius of the Si optical waveguide can be about 10um, making it possible to miniaturize the elements. Therefore, there are advantages of high economies of scale and low cost. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2003-270599 A [Patent Document 2] JP 2015-191031 A [Patent Document 3] US Patent Application Publication No. 2020 / 0152574 Summary of the Invention [Problem to be solved by the invention]

[0005] However, in Si modulators such as the X-polarized wave modulation section and Y-polarized wave modulation section in the optical modulator element of a conventional optical device, voltage is not applied directly to the waveguide with metal electrodes, but is applied via a Si layer doped with impurities, resulting in higher electrical resistance than when only metal electrodes are used. Moreover, the optical waveguide used in the Si modulator has a large capacity due to its P / N junction structure, resulting in high high-frequency loss. Therefore, assuming operation with a practical driver drive voltage (±2V or 4V or less), it is difficult to achieve a wide bandwidth of 50GHz or more and a high speed of 96G baud rate or more.

[0006] Therefore, many attempts have been reported at academic conferences to integrate electro-optical materials with electro-optical effects, such as LiNbO3 (hereinafter simply referred to as LN), which is said to enable speeds of 96G baud rates or higher, onto SiPh elements. (For example, Mingbo He1 et. al. “High-performance hybrid silicon and lithium niobate Mach-Zehnder modulators for 100 Gbit s-1 and beyond”, Nat. Photon. 13, 359-364 (2019))

[0007] However, the dielectric constant of the Si substrate of the SiPh element is relatively large at about 12, so the refractive index of the high-frequency electrical signal traveling through the electrode tends to be large. In particular, when using a capacitance-loaded electrode, which is advantageous for reducing high-frequency loss because the effective electrode size increases due to the wide distribution of current as the electrode structure of the modulator, the refractive index of the electrical signal increases due to the influence of the capacitance-loaded electrode in addition to the influence of the refractive index of the substrate. As a result, the speed of the electrical signal becomes somewhat slower than the speed of the signal light propagating through the LN waveguide, and band limitations occur due to mismatch in speed. Therefore, it is difficult to achieve a bandwidth of 100 GHz or more, which is required for a 200 G baud rate.

[0008] In one aspect, an object of the present invention is to provide an optical modulator element or the like that realizes a broadband modulator by ensuring speed matching between an electrical signal and signal light. [Means for solving the problem]

[0009] An optical modulator element according to one embodiment includes, on a substrate, an optical branching section and an optical multiplexing section each including a first material, two optical waveguide arms connecting the optical branching section and the optical multiplexing section, and electrodes for applying an electric signal to the two optical waveguide arms. Each optical waveguide of the two optical waveguide arms includes a first optical waveguide including the first material, a second optical waveguide including a second material having a higher electro-optic effect than the first material, and a transition section for optical transition between the first optical waveguide and the second optical waveguide. The substrate includes a recessed portion in which all or a part of the substrate below the second optical waveguide in a plan view is removed. Effect of the Invention

[0010] According to one aspect, a broadband modulator is realized by ensuring speed matching between an electrical signal and an optical signal. [Brief description of the drawings]

[0011] [Figure 1] FIG. 1 is a schematic plan view showing an example of an optical transceiver according to this embodiment. [Diagram 2] FIG. 2 is a schematic plan view showing an example of the slave MZM of the first embodiment. [Diagram 3] FIG. 3 is an explanatory diagram showing an example of a first transition section in a child MZM. [Figure 4] FIG. 4 is a schematic cross-sectional view showing an example of a cross section taken along line AA shown in FIG. [Diagram 5] FIG. 5 is a schematic cross-sectional view showing an example of a cross section taken along line BB shown in FIG. [Figure 6] FIG. 6 is a schematic cross-sectional view showing an example of a cross section taken along line CC shown in FIG. [Figure 7] FIG. 7 is a schematic plan view showing an example of a capacitively loaded electrode of a daughter MZM. [Figure 8] FIG. 8 is a schematic cross-sectional view showing an example of a cross section taken along line AA shown in FIG. [Figure 9] FIG. 9 is a schematic cross-sectional view showing an example of a cross section taken along line BB shown in FIG. [Figure 10] FIG. 10 is a schematic cross-sectional view showing an example of a cross section taken along line CC shown in FIG. [Figure 11] FIG. 11 is an explanatory diagram showing an example of the relationship between the substrate removal width and the high-frequency refractive index in the slave MZM. [Figure 12] FIG. 12 is an explanatory diagram showing an example of the frequency dependence of the EO characteristics in a slave MZM. [Figure 13] FIG. 13 is a schematic cross-sectional view showing an example of a slave MZM of the second embodiment. [Figure 14] FIG. 14 is a schematic cross-sectional view showing an example of a slave MZM of the second embodiment. [Figure 15] FIG. 15 is an explanatory diagram showing an example of the relationship between the substrate removal width and the high-frequency refractive index in the slave MZM. [Figure 16] FIG. 16 is an explanatory diagram showing an example of the relationship between the substrate removal rate and the high-frequency refractive index for each substrate removal width. [Figure 17] FIG. 17 is an explanatory diagram showing an example of the frequency dependence of the EO characteristics for each substrate removal rate of the slave MZM. [Figure 18] FIG. 18 is a schematic plan view showing an example of a slave MZM according to the third embodiment. [Figure 19] FIG. 19 is a schematic cross-sectional view showing an example of a cross section taken along line AA shown in FIG. [Figure 20] FIG. 20 is a schematic cross-sectional view showing an example of a cross section taken along line BB shown in FIG. [Figure 21] FIG. 21 is a schematic cross-sectional view showing an example of a cross section taken along line CC shown in FIG. [Figure 22] FIG. 22 is a schematic cross-sectional view showing an example of a slave MZM of the fourth embodiment. [Diagram 23] FIG. 23 is a schematic cross-sectional view showing an example of a slave MZM of the fourth embodiment. [Figure 24] FIG. 24 is a schematic cross-sectional view showing an example of a slave MZM of the fifth embodiment. [Diagram 25] FIG. 25 is a schematic cross-sectional view showing an example of a slave MZM of the fifth embodiment. [Figure 26] FIG. 26 is a schematic plan view illustrating an example of the modulator according to the sixth embodiment. [Figure 27] FIG. 27 is a cross-sectional schematic diagram showing an example of a slave MZM of the seventh embodiment. [Figure 28] FIG. 28 is a schematic cross-sectional view showing an example of a cross section taken along line AA shown in FIG. [Figure 29] FIG. 29 is a schematic cross-sectional view showing an example of a cross section taken along line BB shown in FIG. [Diagram 30] FIG. 30 is a schematic plan view showing an example of the modulator according to the eighth embodiment. [Diagram 31] FIG. 31 is a block diagram showing an example of an optical transceiver according to this embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] Hereinafter, the embodiments of the optical device and the like disclosed in the present application will be described in detail with reference to the drawings. Note that the present invention is not limited to these embodiments. In addition, the following examples may be appropriately combined as long as no contradiction occurs. EXAMPLES

[0013] Fig. 1 is a schematic plan view showing an example of an optical transceiver 1 according to the present embodiment. The optical transceiver 1 shown in Fig. 1 is, for example, a 200G baud rate DP-QPSK (Dual Polarization-Quadrature Phase Shift Keying) type optical device. The optical transceiver 1 has an optical modulator element 2 and an optical receiver element 3. The optical modulator element 2 and the optical receiver element 3 are, for example, SiPh elements. The optical transceiver 1 has a local oscillator optical waveguide 4, a transmitting optical waveguide 5, a receiving optical waveguide 6, a glass block 7, and a third branching section 8.

[0014] The local oscillation optical waveguide 4 is optically connected to the local oscillation optical fiber F1 via a glass block 7 by butt joint junction, and is, for example, a Si waveguide, which propagates the local oscillation light. The transmission optical waveguide 5 is optically connected to the output side optical fiber F2 via the glass block 7 by butt joint junction, and is, for example, a Si waveguide, which propagates the transmission light. The reception optical waveguide 6 is optically connected to the input side optical fiber F3 via the glass block 7 by butt joint junction, and is, for example, a Si waveguide, which propagates the reception light. The local oscillation light incident from the local oscillation optical waveguide 4 is branched into two at the third branching section 8, one of which is used as a light source for the optical modulator element 2, and the other is used as the local oscillation light for the optical receiver element 3. The branching ratio of the third branching section 8 is optimally adjusted according to the application.

[0015] The optical receiver element 3 has a PBS (Polarization Beam Splitter) 11 and a first PR (Polarization Rotator) 12. The optical receiver element 3 has a first optical hybrid circuit 13A, a second optical hybrid circuit 13B, and first to fourth PD (Photo Diode) pairs 14A to 14D (14).

[0016] The PBS 11 separates the received light input from the receiving optical waveguide 6 into two orthogonal polarization states, for example, an X-polarized component and a Y-polarized component. The X-polarized component is a horizontally polarized component, and the Y-polarized component is a vertically polarized component. The PBS 11 outputs the separated X-polarized component to the first optical hybrid circuit 13A. Furthermore, the first PR 12 rotates the Y-polarized component from the PBS 11 by 90 degrees, and outputs the Y-polarized component after the polarization rotation to the second optical hybrid circuit 13B.

[0017] The first optical hybrid circuit 13A causes the X-polarized component of the received light to interfere with the local light, thereby acquiring I-component and Q-component optical signals. The I-component is an in-phase axis component, and the Q-component is an orthogonal axis component. The first optical hybrid circuit 13A outputs the I-component signal light of the X-polarized component to the first PD pair 14A. The first optical hybrid circuit 13A outputs the Q-component signal light of the X-polarized component to the second PD pair 14B.

[0018] The second optical hybrid circuit 13B causes the Y polarization component after polarization rotation to interfere with the local light to obtain I and Q component signal light. The second optical hybrid circuit 13B outputs the I component signal light of the Y polarization component to the third PD pair 14C. The second optical hybrid circuit 13B outputs the Q component signal light of the Y polarization component to the fourth PD pair 14D.

[0019] The first PD pair 14A converts the I-component signal light of the X-polarized component from the first optical hybrid circuit 13A into an electric signal, and outputs an electric signal. The second PD pair 14B converts the Q-component signal light of the X-polarized component from the first optical hybrid circuit 13A into an electric signal.

[0020] The third PD pair 14C electrically converts the I-component signal light of the Y-polarized component from the second optical hybrid circuit 13B and outputs an electrical signal. The fourth PD pair 14D electrically converts the Q-component signal light of the Y-polarized component from the second optical hybrid circuit 13B and outputs an electrical signal.

[0021] The optical modulator element 2 has a first branching unit 21, an X-polarization modulation unit 22, a Y-polarization modulation unit 23, a second PR 24, and a PBC (Polarization Beam Combiner) 25. The first branching unit 21 branches and outputs local light to the X-polarization modulation unit 22 and the Y-polarization modulation unit 23.

[0022] The X-polarized modulation section 22, i.e., the parent MZM on the X-polarized side, has a second branching section 22A, two child MZMs (Mach-Zehnder Modulators) 22B, two parent DC phase shifters 22C, and a first multiplexing section 22D. The second branching section 22A branches and outputs the signal light from the first branching section 21 to each child MZM 22B. Each child MZM 22B has a branching section 31, two optical waveguide arms 32, a multiplexing section 33, and an RF electrode 34.

[0023] The branching unit 31 in the X-polarization modulation unit 22 outputs the signal light from the second branching unit 22A to the two optical waveguide arms 32. The multiplexing unit 33 multiplexes the signal light propagating through the two optical waveguide arms 32, and outputs the multiplexed signal light to the parent DC phase shifter 22C. One child MZM 22B in the X-polarization modulation unit 22 is, for example, a modulation unit that modulates the I-component signal light of the X-polarization propagating through the two optical waveguide arms 32 in response to a high-frequency signal from the RF electrode 34, and outputs the modulated I-component signal light to the parent DC phase shifter 22C. Furthermore, the other child MZM 22B in the X-polarized modulation unit 22 is, for example, a modulation unit that modulates the Q-component signal light of the X-polarized wave propagating through the two optical waveguide arms 32 in response to the high-frequency signal of the RF electrode 34, and outputs the modulated Q-component signal light to the parent DC phase shifter 22C.

[0024] The parent DC phase shifter 22C in the X-polarized modulation unit 22 is a phase adjustment unit that adjusts the phase of the modulated X-polarized I-component signal light from one child MZM 22B and adjusts the phase of the modulated X-polarized Q-component signal light from the other child MZM 22B according to the drive voltage signal. The phase adjustment in the parent DC phase shifter 22C can make the modulated X-polarized I-component signal light and the modulated X-polarized Q-component signal light orthogonal to each other. The X-polarized I-component signal light and the X-polarized Q-component signal light after passing through the parent DC phase shifter 22C are multiplexed in the first multiplexing unit 22D, and the multiplexed X-polarized IQ mixed signal is output to the PBC 25.

[0025] The Y-polarized modulation unit 23, i.e., the parent MZM on the Y-polarized side, has a second branching unit 23A, two child MZMs 23B, two parent DC phase shifters 23C, and a first multiplexing unit 23D. The second branching unit 23A branches and outputs the signal light from the first branching unit 21 to each child MZM 23B. Each child MZM 23B has a branching unit 31, two optical waveguide arms 32, a multiplexing unit 33, and an RF electrode 34.

[0026] The branching unit 31 in the Y-polarized wave modulation unit 23 outputs the signal light from the second branching unit 23A to the two optical waveguide arms 32. The multiplexing unit 33 multiplexes the signal light propagating through the two optical waveguide arms 32, and outputs the multiplexed signal light to the parent DC phase shifter 23C. One child MZM 23B in the Y-polarized wave modulation unit 23 is, for example, a phase modulation unit that modulates the I-component signal light of the Y-polarized wave propagating through the two optical waveguide arms 32 in response to a high-frequency signal from the RF electrode 34, and outputs the modulated I-component signal light to the parent DC phase shifter 23C. Furthermore, the other child MZM 23B in the Y polarization modulation unit 23 is, for example, a phase modulation unit that modulates the Q component signal light of the Y polarization propagating through the two optical waveguide arms 32 in response to the high-frequency signal of the RF electrode 34, and outputs the modulated Q component signal light to the parent DC phase shifter 23C.

[0027] The parent DC phase shifter 23C in the Y-polarized modulation unit 23 is a phase adjustment unit that adjusts the phase of the modulated Y-polarized I-component signal light from one child MZM 23B and adjusts the phase of the modulated Y-polarized Q-component signal light from the other child MZM 23B according to the drive voltage signal. The phase adjustment in the parent DC phase shifter 23C makes it possible to orthogonalize the modulated Y-polarized I-component signal light and the modulated Y-polarized Q-component signal light. After passing through the parent DC phase shifter 23C, the Y-polarized I-component signal light and the Y-polarized Q-component signal light are multiplexed in the first multiplexing unit 23D, and the multiplexed Y-polarized IQ mixed signal is output to the second PR 24.

[0028] The second PR 24 rotates the Y-polarized IQ mixed signal by 90 degrees in polarization, and outputs the Y-polarized IQ mixed signal after polarization rotation to the PBC 25. Then, the PBC 25 combines the X-polarized IQ mixed signal from the X-polarization modulation unit 22 and the Y-polarized IQ mixed signal after polarization rotation from the second PR 24, and outputs the combined IQ mixed signal from the transmission optical waveguide 5.

[0029] 2 is a plan view schematic diagram showing an example of the daughter MZM 22B (23B) of the first embodiment. The daughter MZM 22B in the X-polarized wave modulation unit 22 will be described. Note that the daughter MZM 23B in the Y-polarized wave modulation unit 23 has the same configuration as the daughter MZM 22B in the X-polarized wave modulation unit 22, so the same reference numerals are used to denote the same configurations and operations, and description of the overlapping configurations and operations will be omitted.

[0030] As described above, the child MZM 22B (23B) has a branching section 31, two optical waveguide arms 32, a multiplexing section 33, and an RF electrode 34. The branching section 31 is, for example, a Si branching section that branches and outputs the signal light from the second branching section 22A (23A) to each of the first Si waveguides 32A in the two optical waveguide arms 32. The two optical waveguide arms 32 are two arms having two first Si waveguides 32A, two first transition sections 32B, two LN waveguides 32C, two second transition sections 32D, and two second Si waveguides 32E. The first Si waveguide 32A is, for example, a channel-type Si waveguide. The first Si waveguide 32A is a tapered waveguide having a waveguide width at one end that is tapered. The LN waveguide 32C is, for example, a ridge-type LN waveguide. The second Si waveguide 32E is, for example, a channel-type Si waveguide. The second Si waveguide 32E is a tapered waveguide whose waveguide width at one end is tapered. The first Si waveguide 32A and the second Si waveguide 32E are formed in the same layer, and the LN waveguide 32C is formed in a layer different from the first Si waveguide 32A and the second Si waveguide 32E.

[0031] The first transition section 32B is an interlayer transition section that optically transitions the signal light between the first Si waveguide 32A and the LN waveguide 32C. The first transition section 32B is a different layer from the first Si waveguide 32A and the LN waveguide 32C. The RF electrode 34 has a signal electrode 34A arranged in parallel for each waveguide in the two optical waveguide arms 32, and a ground electrode 34B arranged in parallel for each waveguide. When a high-frequency signal is input from the driver circuit 35, the signal electrode 34A modulates the signal light propagating through the LN waveguide 32C arranged between the signal electrode 34A and the ground electrode 34B.

[0032] The second transition section 32D is an interlayer transition section that optically transitions the modulated signal light between the LN waveguide 32C and the second Si waveguide 32E. The second Si waveguide 32E and the LN waveguide 32C are in different layers. The multiplexing section 33 is, for example, a Si multiplexing section that multiplexes the modulated signal lights from the second Si waveguides 32E and outputs the multiplexed signal lights to the parent DC phase shifter 22C (23C).

[0033] 3 is an explanatory diagram showing an example of a first transition section 32B in the child MZM 22B. The first transition section 32B has a Si substrate 41 and a cladding layer 42 formed on the Si substrate 41. The cladding layer 42 is, for example, a SiO2 layer having a thickness of 4 μm and a refractive index lower than that of the Si substrate 41. The cladding layer 42 has a first cladding layer 42A formed on the Si substrate 41 and a second cladding layer 42B formed on the first cladding layer 42A. The optical transceiver 1 also has a Si substrate 41 and a cladding layer 42.

[0034] The first cladding layer 42A is, for example, a SiO2 layer made of SiO2. The second cladding layer 42B is, for example, a SiO2 layer made of SiO2. The first Si waveguide 32A and the second Si waveguide 32E are lower layer waveguides arranged in the first cladding layer 42A on the Si substrate 41. The LN waveguide 32C is an upper layer waveguide formed on the second cladding layer 42B.

[0035] Fig. 4 is a schematic cross-sectional view showing an example of the cross section taken along line AA shown in Fig. 3. The first Si waveguide 32A shown in Fig. 4 is disposed in the cladding layer 42, and the LN waveguide 32C is disposed on the cladding layer 42. The first transition portion 32B performs interlayer transition of the signal light from the first Si waveguide 32A to the LN waveguide 32C.

[0036] Fig. 5 is a schematic cross-sectional view showing an example of the cross section along the line BB shown in Fig. 3. The first Si waveguide 32A shown in Fig. 5 is disposed in the cladding layer 42, and the LN waveguide 32C is disposed on the cladding layer 42. The first Si waveguide 32A has a tapered structure in which the waveguide width gradually tapers toward the LN waveguide 32C. The first transition section 32B performs interlayer transition of the signal light from the first Si waveguide 32A to the LN waveguide 32C.

[0037] Fig. 6 is a schematic cross-sectional view showing an example of the CC cross section shown in Fig. 3. The first Si waveguide 32A has a waveguide width that gradually narrows from the AA cross section shown in Fig. 3 toward the CC cross section shown in Fig. 3, and ends at the CC cross section as shown in Fig. 6. In the first transition section 32B, the signal light confined in the first Si waveguide 32A makes an interlayer transition from the first Si waveguide 32A toward the LN waveguide 32C to the LN waveguide 32C.

[0038] For convenience of explanation, the first transition section 32B has been explained based on Fig. 4 to Fig. 6, but the second transition section 32D has almost the same configuration. The second Si waveguide 32E is, for example, a channel-type Si waveguide having a tapered structure in which the waveguide width gradually narrows toward the LN waveguide 32C. In the second transition section 32D, the signal light confined in the LN waveguide 32C makes an interlayer transition from the LN waveguide 32C toward the second Si waveguide 32E to the second Si waveguide 32E.

[0039] The first Si waveguide 32A and the second Si waveguide 32E are in the same layer. The LN waveguide 32C is in a layer different from the first Si waveguide 32A and the second Si waveguide 32E.

[0040] Fig. 7 is a schematic plan view showing an example of a capacitance-loaded electrode of the daughter MZM 22B. The RF electrode 34 is a capacitance-loaded electrode shown in Fig. 7. The RF electrode 34 has a signal electrode 34A and a ground electrode 34B.

[0041] The signal electrode 34A includes a portion arranged in parallel with the LN waveguide 32C and is made up of multiple T-shaped rails. The ground electrode 34B includes a portion arranged in parallel with the LN waveguide 32C and is made up of multiple T-shaped rails. The capacitively loaded electrode is made up of multiple T-shaped rails, and the current is distributed over a wide area of ​​the electrode, increasing the effective electrode size, which is advantageous for reducing high-frequency loss and contributing to a wider bandwidth. However, the refractive index of the electrical signal increases due to the influence of the capacitively loaded electrode in addition to the influence of the refractive index of the Si substrate 41. As a result, when comparing the speed of the electrical signal of the RF electrode 34 and the signal light propagating through the LN waveguide 32C, the electrical signal is slightly slower, and the bandwidth may be limited due to a mismatch in speed.

[0042] Fig. 8 is a schematic cross-sectional view showing an example of the cross section taken along the line AA shown in Fig. 2. The two optical waveguide arms 32 shown in Fig. 8 have a Si substrate 41, a cladding layer 42 formed on the Si substrate 41, and a first Si waveguide 32A formed in the cladding layer 42. The two optical waveguide arms 32 have an LN waveguide 32C formed on the cladding layer 42, and an RF electrode 34 arranged in parallel to the LN waveguide 32C. The RF electrode 34 has a signal electrode 34A and a ground electrode 34B, and the ridge portion of the LN waveguide 32C is sandwiched between the signal electrode 34A and the ground electrode 34B.

[0043] 9 is a schematic cross-sectional view showing an example of the cross section along the line BB shown in FIG. 2, and FIG. 10 is a schematic cross-sectional view showing an example of the cross section along the line CC shown in FIG. 2. The cross section along the line CC is a cross section in the waveguide direction of two optical waveguide arms 32. In the two optical waveguide arms 32 shown in FIG. 9 and FIG. 10, a recessed portion 41A is formed in the Si substrate 41 below the LN waveguide 32C. Although the recessed portion 41A is formed in the Si substrate 41 below the LN waveguide 32C as an example, it may be formed in the Si substrate 41 below the LN waveguide 32C and in a part of the cladding layer 42 in contact with the Si substrate 41, and can be appropriately changed. The recessed portion 41A is a space removed by dry etching in the Si substrate 41. The opening width W1 of the recessed portion 41A is the substrate removal width.

[0044] In the recess 41A, the Si substrate 41 is removed by a width W1 from the entire lower portion of the LN waveguide 32C by a method such as dry etching, but the cladding layer 42 having a thickness of 4 μm remains under the LN waveguide 32C. As a result, the LN waveguide 32C maintains high rigidity and strength sufficient to withstand vibration tests and impact tests. This type of structure is also called a membrane structure, and has a proven track record in the market, for example, in tunable Si etalon filters for external cavity lasers, and is highly reliable.

[0045] The dielectric constant of the air in the recess 41A is "1", the dielectric constant of the SiO2 layer that is the cladding layer 42 is about "4", and the dielectric constant of Si is "12", so that the dielectric constants of the recess 41A and the cladding layer 42 are smaller than the dielectric constant of the Si substrate 41. This means that the refractive index of high frequencies felt by the high-frequency electrical signal becomes lower, and the speed of the traveling high-frequency signal becomes faster. As a result, the two optical waveguide arms 32 can match the speed of the signal light propagating through the LN waveguide 32C with the speed of the high-frequency signal.

[0046] FIG. 11 is an explanatory diagram showing an example of the relationship between the substrate removal width and the high-frequency refractive index in the daughter MZM 22B. For convenience of explanation, the substrate removal width is set to 0 μm when there is no recess 41A of the conventional Si substrate 41. When the substrate removal width is 0 μm, as shown in FIG. 11, the high-frequency refractive index of the high-frequency signal is 2.45, and the refractive index of the signal light propagating through the LN waveguide 32C is 2.2. As a result, the high-frequency refractive index of the high-frequency signal is larger than the refractive index of the light. In contrast, when the substrate removal width of the recess 41A of the Si substrate 41 of this embodiment is 3 μm to 12 μm, for example, 5 μm to 10 μm, the high-frequency refractive index of the high-frequency signal is about 2.2, which is almost the same as the refractive index of the signal light propagating through the LN waveguide 32C.

[0047] 12 is an explanatory diagram showing an example of the frequency dependence of the EO response of the daughter MZM 22B. As shown in FIG. 12, when the substrate removal width of the recess 41A is 10 μm, the EO response to a high frequency signal can be significantly increased compared to when the substrate removal width is 0 μm.

[0048] In the child MZM 22B (23B) of the first embodiment, a recess 41A is provided in the Si substrate 41 below the LN waveguide 32C in the two optical waveguide arms 32. As a result, the speed of the signal light propagating through the LN waveguide 32C matches the speed of the high-frequency signal, thereby realizing a broadband optical modulator element 2.

[0049] In addition, although the RF electrode 34 of the daughter MZM 22B (23B) in the first embodiment is, for example, a capacitively loaded electrode, a normal electrode may be used and changes can be made as appropriate.

[0050] In the embodiment 1, the recessed portion 41A of the child MZM 22B (23B) is formed for one LN waveguide 32C in each of the two optical waveguide arms 32. However, the recessed portion 41A may be formed for two LN waveguides 32C. This embodiment will be described below as embodiment 2. EXAMPLES

[0051] Fig. 13 is a schematic cross-sectional view showing an example of the daughter MZM22B1 (23B1) of the second embodiment. The schematic cross-sectional view shown in Fig. 13 is a cross-sectional view corresponding to the cross section of the line BB shown in Fig. 2. The same components as those of the optical transceiver 1 of the first embodiment are given the same reference numerals, and the description of the overlapping components and operations is omitted. The daughter MZM22B1 (23B1) of the second embodiment differs from the daughter MZM22B (23B) of the first embodiment in that a recessed portion 41B is formed in the Si substrate 41 below the two LN waveguides 32C for each pair of LN waveguides 32C.

[0052] 13 is formed by dry etching in the Si substrate 41 below the two LN waveguides 32C in the two optical waveguide arms 32. The opening width W2 of the recess 41B is wider than the opening width W1 of the recess 41A in the first embodiment.

[0053] Fig. 14 is a cross-sectional view showing an example of a daughter MZM 22B1 (23B1) of Example 2. The cross-sectional view shown in Fig. 14 is a cross-sectional view corresponding to the cross section taken along line CC shown in Fig. 2. Recesses 41B shown in Fig. 14 are formed in a Si substrate 41 at predetermined intervals in the waveguide direction of an LN waveguide 32C.

[0054] The Si substrate 41 below the two LN waveguides 32C has a recess 41B and a remaining portion 41C. The recess 41B has an opening width W2 in the width direction of the two optical waveguide arms 32 and an opening width La in the waveguide direction. The opening width W2 of the recess 41B is, for example, 80 μm. The remaining portion 41C has a width Ls in the waveguide direction of the remaining portion 41C between the recesses 41B. The thickness of the remaining portion 41C below the two LN waveguides 32C is, for example, 4 μm.

[0055] Since the two LN waveguides 32C have a plurality of remaining portions 41C at their lower portions, the LN waveguides 32C can ensure high rigidity and strength sufficient to withstand, for example, vibration tests and impact tests.

[0056] When the opening width W2 of the recess 41B is as wide as 80 μm or more, the refractive index of high frequency is 2.0 or less in the structure in which the entire Si substrate 41 is removed in the waveguide direction (=structure with 100% substrate removal rate) as in Example 1. Therefore, it is conceivable that a speed mismatch occurs because the speed of the high frequency signal is too fast compared to the refractive index of light, 2.2.

[0057] In contrast, in a structure in which only a portion of the Si substrate 41 is removed in the direction along the LN waveguide 32C, the refractive index of the high-frequency signal is adjusted by changing the proportion of the Si substrate 41 that is removed, thereby achieving an appropriate removal rate. As a result, the speed of the signal light propagating through the LN waveguide 32C can be matched with the speed of the high-frequency signal.

[0058] 15 is an explanatory diagram showing an example of the relationship between the substrate removal width W2 and the high-frequency refractive index in the daughter MZM 22B. When the opening width W2 of the recess 41B is 40 μm to 80 μm, if the Si substrate 41 below the two LN waveguides 32C is entirely removed and the substrate removal rate is set to 100%, the refractive index of the high-frequency signal is 1.9 and the refractive index of the signal light propagating through the LN waveguide 32C is 2.2.

[0059] Fig. 16 is an explanatory diagram showing an example of the relationship between the substrate removal rate for each substrate removal width and the high-frequency refractive index, and Fig. 17 is an explanatory diagram showing an example of the frequency dependence of the EO characteristics for each substrate removal rate of the daughter MZM 22B1 (23B1). When the substrate removal rate is 60% by arranging the recessed portion 41B of La = 30 μm and the remaining portion 41C of Ls = 20 μm at a predetermined interval in the lower portion of the two LN waveguides 32C as in Example 2, the refractive index of the high-frequency signal is about 2.2 as shown in Fig. 16. As a result, the refractive index of the signal light propagating through the LN waveguide 32C and the refractive index of the high-frequency signal are approximately the same.

[0060] Even when the opening width W2 of the recessed portion 41B is set to 40 μm to 80 μm, the EO response band can be expanded as in the first embodiment. When the arm interval is 40 μm and the substrate removal rate defined by La / (La+Ls) is set to 100%, even when the removal width of the Si substrate 41 in the lower portion of the LN waveguide 32C in the arm is changed, the refractive index of the high frequency signal does not change when the removal width is 40 μm or more. Therefore, it is preferable to set the removal width of the Si substrate 41 in the lower portion of the LN waveguide 32C in the arm to be equal to or larger than the arm interval, and set the substrate removal rate defined by La / (La+Ls) to 40% (0.4) to 80% (0.8) as shown in FIG. 16.

[0061] The opening width W2, the opening width La, and the width Ls are not limited to the above, and can be changed as appropriate as long as the EO band can be expanded by approximately matching the refractive index of the high frequency signal with the refractive index of the signal light. In addition, the Si substrate 41 in the lower part of the two LN waveguides 32C in the second embodiment has the recessed parts 41B and the remaining parts 41C arranged at equal intervals. However, as long as the refractive index of the high frequency signal and the refractive index of the signal light can be approximately matched and the EO band can be expanded, the intervals do not have to be equal and can be changed as appropriate.

[0062] In the child MZM 22B1 (23B1) of the second embodiment, a part of the Si substrate 41 located under the two LN waveguides 32C is removed to form the recessed part 41B and the remaining part 41C. However, the recessed part and the remaining part may be formed by removing a part of the Si substrate 41 located under one LN waveguide 32C of the child MZM 22B (23B) of the first embodiment, and can be modified as appropriate.

[0063] In the LN waveguide 32C of the child MZM 22B, 22B1 (23B, 23B1) in Examples 1 and 2, the Pockels coefficient, which indicates the electro-optic effect, is small at about 30 pm / V. Therefore, in order to increase the modulation efficiency so that the driver can be driven with a practical amplitude voltage, the modulator arm length is long, at 10 mm or more, as shown in FIG. 2. As a result, the economies of scale, which are the advantages of SiPh elements, cannot be utilized. Therefore, an embodiment that can deal with such a situation will be described below as Example 3. EXAMPLES

[0064] 18 is a schematic plan view showing an example of the daughter MZM22B2 (23B2) of the third embodiment. For ease of explanation, the same components as those in the optical transceiver 1 of the second embodiment are given the same reference numerals, and explanations of the overlapping components and operations are omitted. The daughter MZM23B2 has the same configuration as the daughter MZM22B2, and therefore the same reference numerals are given to the daughter MZM23B2, and explanations of the overlapping components and operations are omitted. The difference between the daughter MZM22B1 (23B1) of the second embodiment and the daughter MZM22B2 (23B2) of the third embodiment is that the two optical waveguide arms 32 are folded back.

[0065] The child MZM 22B2 has a branching section 31, two optical waveguide arms 32, a multiplexing section 33, and an RF electrode 34. The branching section 31 is, for example, a Si branching section that branches and outputs the signal light from the second branching section 22A (23A) to each of the first Si waveguides 32A in the two optical waveguide arms 32. The two optical waveguide arms 32 have two first Si waveguides 32A, two first transition sections 32B, and two first LN waveguides 32C1. The two optical waveguide arms 32 have two third transition sections 51A, two third Si waveguides 51B, two folded waveguides 51C, two fourth Si waveguides 51D, and two fourth transition sections 51E. The two optical waveguide arms 32 have two second LN waveguides 32C2, two second transition portions 32D, and two second Si waveguides 32E.

[0066] The first Si waveguide 32A is, for example, a channel-type Si waveguide. The first Si waveguide 32A is a tapered waveguide with a waveguide width at one end tapered. The first LN waveguide 32C1 is, for example, a ridge-type LN waveguide. The third Si waveguide 51B is, for example, a channel-type Si waveguide connected to the folded waveguide 51C. The third Si waveguide 51B is a tapered waveguide with a waveguide width at one end tapered. The folded waveguide 51C is a channel-type Si waveguide with a folded structure that is folded back into a U-shape with a small bending radius due to strong confinement.

[0067] The fourth Si waveguide 51D is, for example, a channel-type Si waveguide connected to the folded waveguide 51C. The fourth Si waveguide 51D is a tapered waveguide having a tapered waveguide width at one end. The second LN waveguide 32C2 is, for example, a ridge-type LN waveguide. The second Si waveguide 32E is, for example, a channel-type Si waveguide. The second Si waveguide 32E is a tapered waveguide having a tapered waveguide width at one end.

[0068] The first transition section 32B is an interlayer transition section that optically transitions the signal light between the first Si waveguide 32A and the first LN waveguide 32C1. The RF electrode 34 has a signal electrode 34A arranged in parallel for each waveguide in the two optical waveguide arms 32, and a ground electrode 34B arranged in parallel for each waveguide. When a high-frequency signal is input from the driver circuit 35A, the signal electrode 34A modulates the signal light propagating through the first LN waveguide 32C1 arranged between the signal electrode 34A and the ground electrode 34B.

[0069] The third transition section 51A is an interlayer transition section that causes interlayer optical transition of the signal light between the first LN waveguide 32C1 and the third Si waveguide 51B. The fourth transition section 51E is an interlayer transition section that causes interlayer optical transition of the signal light between the fourth Si waveguide 51D and the second LN waveguide 32C2. When a high-frequency signal is input from the driver circuit 35A, the signal electrode 34A modulates the signal light propagating through the second LN waveguide 32C2 arranged between the signal electrode 34A and the ground electrode 34B.

[0070] The second transition section 32D is an interlayer transition section that performs interlayer optical transition of the signal light between the second LN waveguide 32C2 and the second Si waveguide 32E. The multiplexing section 33 is, for example, a Si multiplexing section that multiplexes the signal lights from the second Si waveguides 32E and outputs the multiplexed signal lights to the parent DC phase shifter 22C (23C).

[0071] The two optical waveguide arms 32 are connected between the first LN waveguide 32C1 and the second LN waveguide 32C2 by the folded waveguide 51C, so that the longitudinal size can be reduced to half that of the LN waveguide 32C of the first embodiment, thereby achieving miniaturization.

[0072] The child MZM22B2 has a first heater electrode 52A arranged in the third Si waveguide 51B, a second heater electrode 52B arranged in the fourth Si waveguide 51D, and an electrode wiring 53. The electrode wiring 53 is a metal wiring that injects a current into the first heater electrode 52A and also injects a current into the second heater electrode 52B.

[0073] Fig. 19 is a schematic cross-sectional view showing an example of the cross section taken along line AA shown in Fig. 18. The two optical waveguide arms 32 shown in Fig. 19 have a Si substrate 41, a cladding layer 42 formed on the Si substrate 41, a first Si waveguide 32A formed in the cladding layer 42, and a second Si waveguide 32E formed in the cladding layer 42. The first Si waveguide 32A and the second Si waveguide 32E are in the same layer.

[0074] The two optical waveguide arms 32 include a first LN waveguide 32C1 formed on the cladding layer 42, a second LN waveguide 32C2 formed on the cladding layer 42, and an RF electrode 34 arranged in parallel to the first LN waveguide 32C1 and the second LN waveguide 32C2. The first LN waveguide 32C1 and the second LN waveguide 32C2 are in the same layer. The first LN waveguide 32C1 is arranged in parallel between the signal electrode 34A and the ground electrode 34B. The second LN waveguide 32C2 is arranged in parallel between the signal electrode 34A and the ground electrode 34B. The first LN waveguide 32C1 and the second LN waveguide 32C2 are in a different layer from the first Si waveguide 32A and the second Si waveguide 32E.

[0075] Fig. 20 is a schematic cross-sectional view showing an example of the cross section along the line BB shown in Fig. 18. The Si substrate 41 below the first LN waveguide 32C1 and the second LN waveguide 32C2 has a recessed portion 41D and a remaining portion 41E. Since the first LN waveguide 32C1 and the second LN waveguide 32C2 have a plurality of remaining portions 41E below, the first LN waveguide 32C1 and the second LN waveguide 32C2 can ensure high rigidity and strength sufficient to withstand, for example, a vibration test or an impact test.

[0076] The dielectric constant of the air in the recess 41D is "1", the dielectric constant of the SiO2 layer that is the cladding layer 42 is about "4", and the dielectric constant of Si is "12", so that the dielectric constants of the recess 41D and the cladding layer 42 are smaller than the dielectric constant of the Si substrate 41. This means that the refractive index experienced by the electric high-frequency signal becomes lower, and the speed of the traveling high-frequency signal becomes faster. As a result, the speed of the signal light propagating through the first LN waveguide 32C1 and the second LN waveguide 32C2 can be matched with the speed of the high-frequency signal.

[0077] The signal electrode 34A has a first signal electrode arranged near the first LN waveguide 32C1, a second signal electrode arranged near the second LN waveguide 32C2, and a U-shaped signal electrode electrically connecting the first signal electrode and the second signal electrode. The ground electrode 34B has a first ground electrode arranged near the first LN waveguide 32C1, a second ground electrode arranged near the second LN waveguide 32C2, and a U-shaped ground electrode electrically connecting the first ground electrode and the second ground electrode.

[0078] The child MZM22B2 has a differential drive type driver circuit 35A that is connected to a first signal electrode in the signal electrode 34A and is also connected to a second signal electrode in the signal electrode 34A. The driver circuit 35A applies a high frequency signal in the same electric field direction to the first signal electrode and the first LN waveguide 32C1, and also applies a high frequency signal in the same electric field direction to the second signal electrode and the second LN waveguide 32C2, so that it is possible to prevent a situation in which the phases of the modulation are offset.

[0079] When an open collector type driver circuit available on the market is used as the differential drive type driver circuit 35A and a DC voltage is applied to the first LN waveguide 32C1 and the second LN waveguide 32C2, it is possible that a DC drift of the operating point voltage of the LN modulator occurs. However, in the child MZM22B2 of the third embodiment, a phase shifter including a heater electrode 52 that compensates for the DC drift is disposed. The first phase shifter has a third Si waveguide 51B and a first heater electrode 52A formed below the third Si waveguide 51B. The second phase shifter has a fourth Si waveguide 51D and a second heater electrode 52B formed below the fourth Si waveguide 51D.

[0080] Fig. 21 is a schematic cross-sectional view showing an example of the cross section taken along the line CC shown in Fig. 18. The cross-sectional portion shown in Fig. 21 is a cross-sectional portion of the first phase shifter. For convenience of explanation, the second phase shifter has the same configuration as the first phase shifter, and therefore the same reference numerals are used to denote the same components and operations, and explanations of the same components and operations will be omitted. The first phase shifter shown in Fig. 21 includes a Si substrate 41, a cladding layer 42, a third Si waveguide 51B formed in the cladding layer 42, and a first heater electrode 52A formed in the cladding layer 42 and below the third Si waveguide 51B.

[0081] The first phase shifter heats the first heater electrode 52A by passing a current through the first heater electrode 52A, and the heating of the first heater electrode 52A heats the third Si waveguide 51B. By heating the third Si waveguide 51B, the refractive index of the third Si waveguide 51B increases, and the phase of the signal light propagating through the third Si waveguide 51B is adjusted. By adjusting the phase of the signal light, the operating point voltage can be controlled by an external ABC circuit (Auto Bias Control), so that DC drift can be compensated. Since the Si waveguide itself does not generate DC drift, it is relatively easy to ensure a lifespan in terms of DC drift, and since the refractive index of Si changes greatly with temperature, the power consumption of the heater can be suppressed. The heater electrode 52 can be formed in the manufacturing process of SiPh, and metal nitride is generally used. EXAMPLES

[0082] FIG. 22 is a cross-sectional schematic diagram showing an example of a daughter MZM22B3 (23B3) of the fourth embodiment. The same components as those of the optical transceiver 1 of the first embodiment are given the same reference numerals, and the description of the overlapping components and operations is omitted. The optical transceiver 1 of the fourth embodiment has an optical modulator element 2 mounted in 2.5 dimensions and an optical receiver element 3 mounted in 2.5 dimensions. The optical transceiver 1 of the fourth embodiment differs from the optical transceiver 1 of the first embodiment in that a daughter MZM22B3 (23B3) is disposed instead of the daughter MZM22 (23). The cross-sectional portion shown in FIG. 22 is a cross-sectional portion of the first transition portion 32B and the second transition portion 32D in the daughter MZM22B3 (23B3) and the PD14.

[0083] The first transition portion 32B in the child MZM 22B3 has a Si substrate 41, a cladding layer 42, a first Si waveguide 32A formed in the cladding layer 42, a first LN waveguide 32C1 formed on the cladding layer 42, a signal electrode 34A, and a ground electrode 34B. The second transition portion 32D in the child MZM 22B3 has a Si substrate 41, a cladding layer 42, a second Si waveguide 32E formed in the cladding layer 42, a second LN waveguide 32C2 formed on the cladding layer 42, and a signal electrode 34A and a ground electrode 34B.

[0084] The child MZM22B3 has a TSV (Through Si Via) 34D that penetrates between the surface of the cladding layer 42 and the surface of the Si substrate 41 and electrically connects between a signal electrode 34A arranged on the surface of the cladding layer 42 and an electrode pad arranged on the surface of the Si substrate 41.

[0085] PD14 has a Si substrate 41, a cladding layer 42, a Si layer 62 formed in the cladding layer 42, a Ge layer 63 formed in the cladding layer 42 and in contact with the Si layer 62, and a via 64. The via 64 penetrates between the Si layer 62 and the surface of the cladding layer 42, and electrically connects between the Si layer 62 and an electrode pad on the surface of the cladding layer 42. PD14 has a via 65 that penetrates between the electrode pad on the surface of the Si substrate 41 and the electrode pad on the surface of the cladding layer 42, and electrically connects between the electrode pad of the Si substrate 41 and the electrode pad of the cladding layer 42.

[0086] The surface of the cladding layer 42 on which the first LN waveguide 32C1 and the second LN waveguide 32C2 are integrated is preferably subjected to CMP (Chemical Mechanical Polishing) to reduce surface roughness. The height distance between the first LN waveguide 32C1 and the first Si waveguide 32A in the first transition section 32B is set to, for example, 1500 nm or less. The height distance between the second LN waveguide 32C2 and the second Si waveguide 32E in the second transition section 32D is set to, for example, 1500 nm or less.

[0087] 23 is a cross-sectional schematic diagram showing an example of the daughter MZM 22B3 (23B3) of the fourth embodiment. The Si substrate 41 below the first LN waveguide 32C1 and the second LN waveguide 32C2 has a recessed portion 41F and a remaining portion 41G. Since the first LN waveguide 32C1 and the second LN waveguide 32C2 have a plurality of remaining portions 41G below them, the first LN waveguide 32C1 and the second LN waveguide 32C2 can ensure high rigidity and strength sufficient to withstand, for example, a vibration test or an impact test.

[0088] The dielectric constant of the air in the recess 41F is "1", the dielectric constant of the SiO2 layer that is the cladding layer 42 is about "4", and the dielectric constant of Si is "12", so that the dielectric constants of the recess 41F and the cladding layer 42 are smaller than the dielectric constant of the Si substrate 41. This means that the refractive index of high frequencies felt by the high-frequency electric signal becomes lower, and the speed of the traveling high-frequency signal becomes faster. As a result, the speed of the signal light propagating through the first LN waveguide 32C1 and the second LN waveguide 32C2 can be matched with the speed of the high-frequency signal.

[0089] However, in a SiPh wafer, the distance from the Si waveguide to the surface is 7000 nm or more because of a multi-layered structure including metal wiring layers, insulating layers, etc. Therefore, it is not possible to perform CMP polishing to reduce the distance from the first transition portion 32B made of Si to within 1500 nm from the surface of the cladding layer 42.

[0090] The SiPh wafer is once bonded to another Si substrate as a handle substrate, and the Si substrate (not shown) on the SiPh wafer side is removed by grinding and polishing from the back side. At this time, the PD141 including the Ge layer 63A is turned upside down, resulting in a vertical relationship as shown in Figure 22. The first Si waveguide 32A in the first transition portion 32B and the second Si waveguide 32E in the second transition portion 32D, and the doped conductive Si layer 62A abutting the Ge layer 63A of the PD141 are all Si layers of the SOI that is the basis of the SiPh wafer, and are at the same height.

[0091] Here, since the via 64A electrically connected to the Si layer 62A of the PD 141 is buried in the Si substrate 41, it is necessary to draw the electrode to the surface of the Si substrate 41 by the TSV 61D. When the TSV 61D is fabricated, it is necessary to open a hole in the Si substrate 41 by dry etching, so the partial recess 41F of the Si substrate 41 in Fig. 23 can be formed by dry etching at the same time as the TSV is formed. Therefore, the recess 41F of the Si substrate 41 does not significantly increase the cost.

[0092] The optical transceiver 1 of the fourth embodiment has a 2.5-dimensionally implemented optical modulator element 2 and a 2.5-dimensionally implemented optical receiver element 3. The optical modulator element 2 and the optical receiver element 3 can be realized in a 2.5-dimensionally implemented manner. EXAMPLES

[0093] FIG. 24 is a cross-sectional schematic diagram showing an example of a daughter MZM22B4 (23B4) of the fifth embodiment. The same components as those of the optical transceiver 1 of the first embodiment are given the same reference numerals, and the description of the overlapping components and operations is omitted. The optical transceiver 1 of the fifth embodiment differs from the optical transceiver 1 of the first embodiment in that the daughter MZM22B4 (23B4) and PD141 are arranged instead of the daughter MZM22 (23) and PD14. The cross-sectional portion shown in FIG. 24 is a cross-sectional portion of the first transition portion 32B and the second transition portion 32D in the daughter MZM22B4 (23B4) and PD141.

[0094] The first transition portion 32B in the child MZM 22B4 has a Si substrate 41, a cladding layer 42, a first Si waveguide 32A formed in the cladding layer 42, and a recessed portion 42C in the cladding layer 42. The first transition portion 32B has a first LN waveguide 32C1 formed in the recessed portion 42C, a signal electrode 34A, and a ground electrode 34B. The recessed portion 42C is formed by digging a part of the cladding layer 42 into a groove shape by dry etching. The second transition portion 32D in the child MZM 22B4 has a Si substrate 41, a cladding layer 42, a second Si waveguide 32E formed in the cladding layer 42, and a recessed portion 42C. The second transition section 32D includes a second LN waveguide 32C2 formed in the recess 42C, a signal electrode 34A and a ground electrode 34B.

[0095] PD141 has a Si substrate 41, a cladding layer 42, a Si layer 62A formed in the cladding layer 42, and a Ge layer 63A formed in the cladding layer 42 and disposed on the Si layer 62A. PD141 has a via 64A that penetrates between the Si layer 62A and an electrode pad on the surface of the cladding layer 42 and electrically connects between the Si layer 62A and the electrode pad of the cladding layer 42. PD141 has a TSV 61D that penetrates between the electrode pad on the surface of the Si substrate 41 and the electrode pad on the surface of the cladding layer 42 and electrically connects between the electrode pad of the Si substrate 41 and the electrode pad of the cladding layer 42.

[0096] The height distance between the first LN waveguide 32C1 and the first Si waveguide 32A in the first transition section 32B is, for example, within 1500 nm. The height distance between the second LN waveguide 32C2 and the second Si waveguide 32E in the second transition section 32D is, for example, within 1500 nm.

[0097] 25 is a cross-sectional schematic diagram showing an example of the daughter MZM 22B4 of the fifth embodiment. The Si substrate 41 below the first LN waveguide 32C1 and the second LN waveguide 32C2 has a recessed portion 41H and a remaining portion 41J. Since the first LN waveguide 32C1 and the second LN waveguide 32C2 have a plurality of remaining portions 41J below, the first LN waveguide 32C1 and the second LN waveguide 32C2 can ensure high rigidity and strength sufficient to withstand, for example, a vibration test or an impact test.

[0098] In the daughter MZM22B4 of the embodiment 5, it is not necessary to perform CMP polishing on the entire surface of the cladding layer 42, as compared with the daughter MZM22B3 of the embodiment 4. The height distance between the first LN waveguide 32C1 and the first Si waveguide 32A in the first transition portion 32B and the height distance between the second LN waveguide 32C2 and the second Si waveguide 32E in the second transition portion 32D can be set to within 1500 nm, and the distance from the Si waveguide layer to the surface can be set to 7000 nm or more, which is determined by the multi-layered structure of the metal wiring layer, insulating layer, etc. of the SiPh wafer.

[0099] Moreover, in the slave MZM22B4 of the fifth embodiment, it is not necessary to bond the SiPh wafer to another Si substrate as a handle substrate, as compared with the slave MZM22B3 of the fourth embodiment. In this case, the PD141 including the Ge layer 63A is not upside down, but has the upside-down relationship shown in FIG. 24. Here, the first Si waveguide 32A, the second Si waveguide 32E, and the Si layer 62A are all made of the Si layer of the SOI that is the source of the SiPh wafer, and are at the same height.

[0100] Also, the electrode of PD141 can use the via 64A that is electrically connected to the doped conductive Si layer 62A of PD141, which is fabricated from a normal SiPh wafer, as it is, and is connected to a TIA (Trans Impedance Amplifier). On the other hand, the electrode of the child MZM22B4 is connected to the driver circuit 35A through the TSV 34D that creeps up from the signal electrode 34A on the bottom surface of the recessed portion 42C to the surface. Also, the TSV 61D that penetrates the Si substrate 41 to perform 2.5-dimensional mounting is used to connect to an interposer wiring substrate (not shown) on the back side of the Si substrate 41. However, when fabricating the TSV 61D, it is necessary to open a hole in the Si substrate 41 by dry etching, so the partial recessed portion 41H of the Si substrate 41 in FIG. 25 can be simultaneously formed by dry etching when forming the TSV. Therefore, the formation of the recessed portion 41H does not significantly increase the cost. EXAMPLES

[0101] 26 is a schematic plan view showing an example of a modulator according to a sixth embodiment. The same components as those of the optical transceiver 1 according to the third embodiment are given the same reference numerals, and the description of the overlapping components and operations will be omitted. The modulator has an X-polarized wave modulation unit 22 consisting of two single-type MZMs with a 200G baud rate, and a Y-polarized wave modulation unit 23 consisting of two single-type MZMs. The modulator is a DP-QAM (Dual Polarization-Quadrature Amplitude Modulation) type MZM.

[0102] 26 is an IQ modulator having a second branch 22A, two child MZMs 22B5, and one parent DC phase shifter 22C. The Y-polarized modulation unit 23 is an IQ modulator having a second branch 23A, two child MZMs 23B5, and one parent DC phase shifter 23C.

[0103] The modulator has a differential drive type driver circuit 35A for DP-QAM that inputs a high frequency signal to the RF electrode 34 in the child MZM 22B5 and also inputs a high frequency signal to the RF electrode 34 in the child MZM 23B5.

[0104] The modulator of the sixth embodiment is applied to a DP-QAM type MZM, and the speed of the signal light propagating through the first LN waveguide 32C1 and the second LN waveguide 32C2 is matched with the speed of the high frequency signal to realize a wide bandwidth. Moreover, since the two optical waveguide arms 32 of the child MZM 22B5 (23B5) have a folded structure, the longitudinal size of the modulator can be made compact, and the optical transceiver 1 can be made compact.

[0105] In addition, the two optical waveguide arms 32 of the child MZM22B2 (23B2) in the third embodiment are folded back once to reduce the size of the modulator in the longitudinal direction. However, the present invention is not limited to this, and an embodiment that achieves further reduction in size will be described below as the seventh embodiment. The same components as those in the optical transceiver 1 in the third embodiment are denoted by the same reference numerals, and the description of the overlapping components and operations will be omitted. EXAMPLES

[0106] 27 is a cross-sectional schematic diagram showing an example of a daughter MZM 22B6 (23B6) of Example 7. The daughter MZM 22B2 of Example 3 and the daughter MZM 22B6 of Example 7 differ in that two optical waveguide arms 32 are configured with two folded back structures.

[0107] The two optical waveguide arms 32 each have two first Si waveguides 32A, two first transition portions 32B, and two first LN waveguides 32C11. The two optical waveguide arms 32 each have two third transition portions 51A1, two third Si waveguides 51B1, two folded waveguides 51C1, two fourth Si waveguides 51D1, two fourth transition portions 51E1, and two second LN waveguides 32C12. The two optical waveguide arms 32 have two fifth transition portions 51G1, two fifth Si waveguides 51F1, two folded waveguides 51H1, two sixth Si waveguides 51J1, two sixth transition portions 51K1, and two third LN waveguides 32C13. The two optical waveguide arms 32 have two second transition portions 32D, and two second Si waveguides 32E.

[0108] The first Si waveguide 32A is, for example, a channel-type Si waveguide. The first Si waveguide 32A is a tapered waveguide with a waveguide width at one end tapered. The first LN waveguide 32C11 is, for example, a ridge-type LN waveguide. The third Si waveguide 51B1 is, for example, a channel-type Si waveguide connected to the folded waveguide 51C1. The third Si waveguide 51B1 is a tapered waveguide with a waveguide width at one end tapered. The folded waveguide 51C1 is a channel-type Si waveguide with a folded structure that is folded back into a U-shape with a small bending radius because of strong confinement.

[0109] The fourth Si waveguide 51D1 is, for example, a channel-type Si waveguide that is connected to the folded waveguide 51C1. The fourth Si waveguide 51D1 is a tapered waveguide whose waveguide width at one end is tapered. The second LN waveguide 32C12 is, for example, a ridge-type LN waveguide.

[0110] The fifth Si waveguide 51F1 is, for example, a channel-type Si waveguide that is connected to the folded waveguide 51H1. The fifth Si waveguide 51F1 is a tapered waveguide whose one end has a tapered waveguide width. The folded waveguide 51H1 is a channel-type Si waveguide that has a folded structure in which the bending radius is small and the waveguide is folded back into a U shape because of strong confinement.

[0111] The sixth Si waveguide 51J1 is, for example, a channel-type Si waveguide connected to the folded waveguide 51H1. The sixth Si waveguide 51J1 is a tapered waveguide having a tapered waveguide width at one end. The third LN waveguide 32C13 is, for example, a ridge-type LN waveguide. The second Si waveguide 32E is, for example, a channel-type Si waveguide. The second Si waveguide 32E is a tapered waveguide having a tapered waveguide width at one end.

[0112] The first transition section 32B is an interlayer transition section that optically transitions the signal light between the first Si waveguide 32A and the first LN waveguide 32C11. The RF electrode 34 has a signal electrode 34A arranged in parallel for each waveguide in the two optical waveguide arms 32, and a ground electrode 34B arranged in parallel for each waveguide. When a high-frequency signal is input from the driver circuit 35A, the signal electrode 34A modulates the signal light propagating through the first LN waveguide 32C11 arranged between the signal electrode 34A and the ground electrode 34B.

[0113] The third transition section 51A1 is an interlayer transition section that causes interlayer optical transition of the signal light between the first LN waveguide 32C11 and the third Si waveguide 51B1. The fourth transition section 51E1 is an interlayer transition section that causes interlayer optical transition of the signal light between the fourth Si waveguide 51D1 and the second LN waveguide 32C12. When a high-frequency signal is input from the driver circuit 35A, the signal electrode 34A modulates the signal light propagating through the second LN waveguide 32C12 arranged between the signal electrode 34A and the ground electrode 34B.

[0114] The fifth transition section 51G1 is an interlayer transition section that causes interlayer optical transition of the signal light between the second LN waveguide 32C12 and the fifth Si waveguide 51F1. The sixth transition section 51K1 is an interlayer transition section that causes interlayer optical transition of the signal light between the sixth Si waveguide 51J1 and the third LN waveguide 32C13. When a high-frequency signal is input from the driver circuit 35A, the signal electrode 34A modulates the signal light propagating through the third LN waveguide 32C13 arranged between the signal electrode 34A and the ground electrode 34B.

[0115] The second transition section 32D is an interlayer transition section that performs interlayer optical transition of the signal light between the third LN waveguide 32C13 and the second Si waveguide 32E. The multiplexing section 33 is, for example, a Si multiplexing section that multiplexes the signal lights from the second Si waveguides 32E and outputs the multiplexed signal lights to the parent DC phase shifter 22C (23C).

[0116] In the two optical waveguide arms 32, the first LN waveguide 32C11 and the second LN waveguide 32C12 are connected by a folded waveguide 51C1, and the second LN waveguide 32C12 and the third LN waveguide 32C13 are connected by a folded waveguide 51H1. The size of the modulator in the longitudinal direction is reduced to 1 / 3 compared to the LN waveguide 32C of the first embodiment, making it possible to achieve miniaturization.

[0117] The child MZM22B6 has a first heater electrode 52A1 disposed on the fifth Si waveguide 51F1, a second heater electrode 52B1 disposed on the sixth Si waveguide 51J1, and an electrode wiring 53A. The electrode wiring 53A is a metal wiring that injects a current into the first heater electrode 52A1 and also into the second heater electrode 52B1.

[0118] Fig. 28 is a schematic cross-sectional view showing an example of the cross section taken along the line AA shown in Fig. 27. The two optical waveguide arms 32 shown in Fig. 28 have a Si substrate 41, a cladding layer 42 formed on the Si substrate 41, and a first Si waveguide 32A formed in the cladding layer 42. The two optical waveguide arms 32 have a fifth Si waveguide 51F1 formed in the cladding layer 42, and a sixth Si waveguide 51J1 formed in the cladding layer 42. The first Si waveguide 32A, the third Si waveguide 51B1, the fourth Si waveguide 51D1, the fifth Si waveguide 51F1, the sixth Si waveguide 51J1, and the second Si waveguide 32E are in the same layer.

[0119] The two optical waveguide arms 32 have a first LN waveguide 32C11 formed on the clad layer 42, a second LN waveguide 32C12 formed on the clad layer 42, and a third LN waveguide 32C13 formed on the clad layer 42. The two optical waveguide arms 32 have RF electrodes 34 arranged in parallel to the first LN waveguide 32C11, the second LN waveguide 32C12, and the third LN waveguide 32C13. The first LN waveguide 32C11, the second LN waveguide 32C12, and the third LN waveguide 32C13 are in the same layer. The first LN waveguide 32C11 is arranged in parallel between the signal electrode 34A and the ground electrode 34B. The second LN waveguide 32C12 is disposed in parallel between the signal electrode 34A and the ground electrode 34B. The third LN waveguide 32C13 is disposed in parallel between the signal electrode 34A and the ground electrode 34B.

[0120] Fig. 29 is a schematic cross-sectional view showing an example of the cross section along the line BB shown in Fig. 27. The Si substrate 41 below the first LN waveguide 32C11, the second LN waveguide 32C12, and the third LN waveguide 32C13 has a recessed portion 41K and a remaining portion 41L. The first LN waveguide 32C11, the second LN waveguide 32C12, and the third LN waveguide 32C13 have a plurality of remaining portions 41L below. As a result, the first LN waveguide 32C11, the second LN waveguide 32C12, and the third LN waveguide 32C13 can ensure high rigidity and strength that can sufficiently withstand, for example, a vibration test or an impact test.

[0121] The dielectric constant of the air in the recess 41K is "1", the dielectric constant of the SiO2 layer which is the cladding layer 42 is about "4", and the dielectric constant of Si is "12", so that the dielectric constants of the recess 41K and the cladding layer 42 are smaller than the dielectric constant of the Si substrate 41. This means that the refractive index of high frequencies felt by the high frequency electrical signal becomes lower, and the speed of the traveling high frequency signal becomes faster. As a result, the speed of the signal light propagating through the first LN waveguide 32C11, the second LN waveguide 32C12, and the third LN waveguide 32C13 can be matched with the speed of the high frequency signal.

[0122] The signal electrode 34A has a first signal electrode arranged near the first LN waveguide 32C11, a second signal electrode arranged near the second LN waveguide 32C12, and a third signal electrode arranged near the third LN waveguide 32C13. The ground electrode 34B has a first ground electrode arranged near the first LN waveguide 32C11, a second ground electrode arranged near the second LN waveguide 32C12, and a third ground electrode arranged near the third LN waveguide 32C13. The child MZM 22B6 has a differential drive type driver circuit 35A connected to the first signal electrode, the second signal electrode, and the third signal electrode. In the driver circuit 35A, high-frequency signals are applied to the first signal electrode and the first LN waveguide 32C11 in the same electric field direction, the second signal electrode and the second LN waveguide 32C12 in the same electric field direction, and the third signal electrode and the third LN waveguide 32C13 in the same electric field direction. As a result, it is possible to prevent the phases of modulation from being offset.

[0123] In the child MZM22B6, a phase shifter including a heater electrode 52 for compensating for DC drift is disposed. The third phase shifter has a fifth Si waveguide 51F1 and a first heater electrode 52A1 formed below the fifth Si waveguide 51F1. The fourth phase shifter has a sixth Si waveguide 51J1 and a second heater electrode 52B1 formed below the sixth Si waveguide 51J1.

[0124] The third phase shifter heats the first heater electrode 52A1 by passing a current through the first heater electrode 52A1, and the heating of the first heater electrode 52A1 heats the fifth Si waveguide 51F1. By heating the fifth Si waveguide 51F1, the refractive index of the fifth Si waveguide 51F1 increases, and the phase of the signal light propagating through the fifth Si waveguide 51F1 is adjusted. By adjusting the phase of the signal light, the operating point voltage can be controlled by an external ABC circuit (Auto Bias Control), and therefore DC drift can be compensated for.

[0125] The fourth phase shifter heats the second heater electrode 52B1 by passing a current through the second heater electrode 52B1, and heats the sixth Si waveguide 51J1 by heating the second heater electrode 52B1. By heating the sixth Si waveguide 51J1, the refractive index of the sixth Si waveguide 51J1 increases, and the phase of the signal light propagating through the sixth Si waveguide 51J1 is adjusted. By adjusting the phase of the signal light, the operating point voltage can be controlled by an external ABC circuit (Auto Bias Control), so that DC drift can be compensated for.

[0126] In the two optical waveguide arms 32 of the seventh embodiment, the first LN waveguide 32C11 and the second LN waveguide 32C12 are connected by a folded waveguide 51C1, and the second LN waveguide 32C12 and the third LN waveguide 32C13 are connected by a folded waveguide 51H1. As a result, the longitudinal size of the modulator is reduced to 1 / 3 compared to the LN waveguide 32C of the first embodiment, thereby achieving miniaturization. EXAMPLES

[0127] 30 is a schematic plan view showing an example of a modulator according to an eighth embodiment. The modulator has an X-polarized wave modulation section 22 made up of two single-type MZMs with a 200 G baud rate, and a Y-polarized wave modulation section 23 made up of two single-type MZMs. The modulator according to the sixth embodiment differs from the modulator according to the eighth embodiment in that a DP-QPSK (Dual Polarization-Quadrature Phase Shift Keying) MZM is arranged instead of a DP-QAM MZM.

[0128] 30 is one of the IQ modulators having a second branch 22A, two child MZMs 22B7, and one parent DC phase shifter 22C. The Y-polarized modulation unit 23 is the other of the IQ modulators having a second branch 23A, two child MZMs 23B7, and one parent DC phase shifter 23C.

[0129] The modulator has a differential drive type driver circuit 35A for DP-QPSK that inputs a high frequency signal to the RF electrode 34 in the child MZM 22B7 and also inputs a high frequency signal to the RF electrode 34 in the child MZM 23B7.

[0130] The modulator of the eighth embodiment is applied to a DP-QPSK type MZM, and the speed of the signal light propagating through the first LN waveguide 32C11, the second LN waveguide 32C12, and the third LN waveguide 32C13 is matched with the speed of the high frequency signal to realize a wide band. Moreover, since the two optical waveguide arms 32 of the child MZM 22B7 (23B7) have a folded structure at two points, the longitudinal size of the modulator can be made compact, and the optical transceiver 1 can be made compact.

[0131] The electro-optical material of the optical modulator element 2 is exemplified as LN (LiNbO3). For example, perovskite oxide may be used, and may be changed as appropriate. As perovskite oxide, (Pb)(Zr,Ti)O3 (hereinafter, PZT), (Pb,La)(Zr,Ti)O3, (hereinafter, PLZT), BaTiO3 (hereinafter, BTO), (Sr,Ba)TiO3 (hereinafter, SBT), LiNbO3 (hereinafter, LN), etc. may be used. However, other perovskite oxides having electro-optical effects may be used.

[0132] In addition, in this embodiment, the case where the LN waveguide 32C is used is illustrated, but any material having a lower dielectric constant than Si and a higher electro-optic effect than Si may be used, and may be changed as appropriate. EXAMPLES

[0133] Next, an optical transceiver 100 employing the optical transceiver 1 of this embodiment will be described. FIG. 31 is a block diagram showing an example of the optical transceiver 100 of this embodiment. The optical transceiver 100 shown in FIG. 31 includes an optical transceiver 110 and a DSP (Digital Signal Processor) 120. The optical transceiver 110 includes an optical modulator element 111, a driver circuit 112, an optical receiver element 113, and a TIA (Transimpedance Amplifier) ​​114. The DSP 120 controls the entire optical transceiver 110. The DSP 120 is an electrical component that performs digital signal processing, such as IQ modulation processing of a transmission signal and demodulation processing of a reception signal.

[0134] The DSP 120 executes processing such as encoding of transmission data, generates an electrical signal including the transmission data, and outputs the generated electrical signal to the driver circuit 112. The driver circuit 112 drives the optical modulator element 111 in response to the electrical signal from the DSP 120. The optical modulator element 111 optically modulates the signal light. The optical modulator element 111 is the optical modulator element of Examples 1 to 8.

[0135] The optical receiver element 113 converts the signal light into an electrical signal. The TIA 114 amplifies the electrical signal after electrical conversion, and outputs the amplified electrical signal to the DSP 120. The DSP 120 performs processing such as decoding the electrical signal obtained from the TIA 114 to obtain received data. The optical modulator element 111 and the optical receiver element 113 constitute an optical transceiver 1. This is not limited to the optical transceiver 1, but may also be applied to an optical transmitter incorporating only the optical modulator element 111.

[0136] The following supplementary notes are further disclosed regarding the above-described embodiments including the present example. (Note 1) An optical modulator element having, on a substrate, an optical branching section and an optical multiplexing section each containing a first material, two optical waveguide arms connecting the optical branching section and the optical multiplexing section, and electrodes for applying electric signals to the two optical waveguide arms, Each optical waveguide of the two optical waveguide arms is a first optical waveguide including the first material; a second optical waveguide including a second material having a higher electro-optic effect than the first material; a transition section that performs optical transition between the first optical waveguide and the second optical waveguide, The substrate is 13. An optical modulator element comprising: a recessed portion in which all or a part of the substrate located under the second optical waveguide in a plan view is removed. (Additional Note 2) The electrode is 2. The optical modulator element according to claim 1, which has a capacitance-loaded electrode. (Appendix 3) The optical modulator element according to appendix 1, wherein the first material includes Si, and the second material includes LiNbO3. (Appendix 4) The recessed portion is 2. The optical modulator element of claim 1, wherein the recess is formed by removing all or a part of the substrate below the two optical waveguide arms. (Additional Note 5) The substrate below the second optical waveguide is The recessed portion and a remaining portion other than the recessed portion are included, 2. The optical modulator element according to claim 1, wherein the recess occupies 40% to 80% of the area of ​​the substrate below the second optical waveguide. (Note 6) The opening width of the recessed portion is 2. The optical modulator element according to claim 1, wherein the thickness is within a range of at least 3 μm to 12 μm. (Appendix 7) The opening width of the recessed portion is 7. The optical modulator element according to claim 6, wherein the thickness is within a range of 5 μm to 10 μm within a range of 3 μm to 12 μm. (Additional Note 8) The second optical waveguide is a first straight waveguide; a second straight waveguide; a folded waveguide connecting the first straight waveguide and the second straight waveguide; The folded waveguide is 2. The optical modulator element of claim 1, comprising the first material. (Supplementary Note 9) The folded waveguide is a heater electrode disposed adjacent to the folded waveguide for heating the folded waveguide; 9. The optical modulator element according to claim 8, wherein a phase of a signal light propagating through the folded waveguide is adjusted in response to a current injection into the heater electrode. (Appendix 10) A clad layer formed on the substrate; the first optical waveguide formed in the cladding layer; the second optical waveguide formed on the cladding layer; the electrode formed on the cladding layer; a via that penetrates the substrate and the cladding layer from front to back and is electrically connected to the electrode; 2. The optical modulator element according to claim 1, comprising: (Appendix 11) A clad layer formed on the substrate; the first optical waveguide formed in the cladding layer; the second optical waveguide formed in a recess in the cladding layer; the electrode formed in a recess in the cladding layer; a via that penetrates the substrate and the cladding layer from front to back and is electrically connected to the electrode; 2. The optical modulator element according to claim 1, comprising: (Additional Note 12) The second optical waveguide is a first straight waveguide; a second straight waveguide; a third straight waveguide; and a first folded waveguide connecting the first straight waveguide and the second straight waveguide; a second folded waveguide connecting between the second straight waveguide and the third straight waveguide; The first folded waveguide and the second folded waveguide are 2. The optical modulator element of claim 1, comprising the first material. (Appendix 13) An optical transmitter including an optical modulator element having an optical branching section and an optical multiplexing section, each containing a first material, two optical waveguide arms connecting the optical branching section and the optical multiplexing section, and electrodes for applying electric signals to the two optical waveguide arms, on a substrate, Each optical waveguide of the two optical waveguide arms is a first optical waveguide including the first material; a second optical waveguide including a second material having a higher electro-optic effect than the first material; a transition section that performs optical transition between the first optical waveguide and the second optical waveguide, The substrate is 13. An optical transmitter comprising: a recessed portion in which all or a part of the substrate below the second optical waveguide in a plan view is removed. (Appendix 14) An optical modulator element including an optical modulator; an optical receiver element including an optical receiver; a processor for executing signal processing of the optical modulator and the optical receiver; The optical modulator element includes: a substrate, on which an optical branching section and an optical multiplexing section each containing a first material are provided, two optical waveguide arms connecting the optical branching section and the optical multiplexing section, and electrodes for applying electric signals to the two optical waveguide arms; Each optical waveguide of the two optical waveguide arms is a first optical waveguide including the first material; a second optical waveguide including a second material having a higher electro-optic effect than the first material; a transition section that performs optical transition between the first optical waveguide and the second optical waveguide, The substrate is 13. An optical transceiver comprising: a recessed portion in which all or a part of the substrate below the second optical waveguide in a plan view is removed. [Explanation of symbols]

[0137] 1 Optical Transmitter / Receiver 2. Optical Modulator Element 3. Optical receiver element 31 Branch 32 Two optical waveguide arms 32A First Si waveguide 32B First Transition 32C LN waveguide 33 Multiplexing section 34 RF electrode 41 Silicon substrate 41A Recess 42 Cladding layer

Claims

1. An optical modulator element having, on a substrate, an optical branching section and an optical multiplexing section each containing a first material, two optical waveguide arms connecting the optical branching section and the optical multiplexing section, and electrodes for applying electric signals to the two optical waveguide arms, Each optical waveguide of the two optical waveguide arms is a first optical waveguide comprising the first material; a second optical waveguide including a second material having a higher electro-optic effect than the first material; a transition section that performs optical transition between the first optical waveguide and the second optical waveguide, The substrate is 13. An optical modulator element comprising: a recessed portion in which all or a part of said substrate located under said second optical waveguide in a plan view is removed.

2. The electrode is 2. The optical modulator element according to claim 1, which is a capacitance-loaded electrode.

3. The first material is Si, and the second material is LiNbO ３ 2. The optical modulator element of claim 1, comprising:

4. The recessed portion is 2. The optical modulator element according to claim 1, wherein the substrate located below the two optical waveguide arms is a recessed portion in which all or a part of the substrate is removed.

5. The substrate below the second optical waveguide includes: The recessed portion and a remaining portion other than the recessed portion are included, 2. The optical modulator element according to claim 1, wherein the proportion of the area of ​​the substrate below the second optical waveguide that is occupied by the recess is within a range of 40% to 80%.

6. The second optical waveguide is a first straight waveguide; a second straight waveguide; a folded waveguide connecting the first straight waveguide and the second straight waveguide; The folded waveguide is 2. The optical modulator element of claim 1, comprising the first material.

7. a cladding layer formed on the substrate; the first optical waveguide formed in the cladding layer; the second optical waveguide formed on the cladding layer; the electrode formed on the cladding layer; a via that penetrates the substrate and the cladding layer from front to back and is electrically connected to the electrode; 2. The optical modulator element according to claim 1, further comprising:

8. The second optical waveguide is A first straight waveguide; a second straight waveguide; a third straight waveguide; and a first folded waveguide connecting between the first straight waveguide and the second straight waveguide; a second folded waveguide connecting between the second straight waveguide and the third straight waveguide; The first folded waveguide and the second folded waveguide are 2. The optical modulator element of claim 1, comprising the first material.

9. An optical transmitter including an optical modulator element having an optical branching section and an optical multiplexing section each including a first material, two optical waveguide arms connecting the optical branching section and the optical multiplexing section, and electrodes for applying electric signals to the two optical waveguide arms, on a substrate, Each optical waveguide of the two optical waveguide arms is a first optical waveguide comprising the first material; a second optical waveguide including a second material having a higher electro-optic effect than the first material; a transition section that performs optical transition between the first optical waveguide and the second optical waveguide, The substrate is 13. An optical transmitter comprising: a recessed portion in which all or a part of said substrate is removed below said second optical waveguide in a plan view.

10. an optical modulator element including an optical modulator; an optical receiver element including an optical receiver; a processor for executing signal processing of the optical modulator and the optical receiver; The optical modulator element includes: a substrate, on which an optical branching section and an optical multiplexing section each containing a first material are provided, two optical waveguide arms connecting the optical branching section and the optical multiplexing section, and electrodes for applying electric signals to the two optical waveguide arms; Each optical waveguide of the two optical waveguide arms is a first optical waveguide comprising the first material; a second optical waveguide including a second material having a higher electro-optic effect than the first material; a transition section that performs optical transition between the first optical waveguide and the second optical waveguide, The substrate is 11. An optical transceiver comprising: a recess formed by removing all or a part of the substrate located under the second optical waveguide in a plan view.

Images