JP6926499B2 - Light modulator - Google Patents

Description

The present invention relates to an optical modulator used in an optical communication system or an optical information processing system, and relates to a structure for providing an optical modulator having excellent frequency characteristics, particularly when performing an optical modulation operation with a high-speed electric signal. ..

A Mach-Zehnder (MZ) type light modulator branches incident light into two optical waveguides at a ratio of 1: 1 and propagates the branched light for a certain length to perform phase modulation and then joins the waves. It has a basic structure to output. The modulation electrical signal applied to the phase modulation section provided in each of the two branched optical waveguides changes the phase of the two lights to change the interference condition of the light when they are combined. The intensity and phase of the output light can be modulated.

_{Dielectrics such as LiNbO 3} and semiconductors such as InP, GaAs, and Si are used as materials constituting the optical waveguide, and a modulated electric signal is input to electrodes arranged in the vicinity of these optical waveguides to apply a voltage to the optical waveguide. By applying it, the phase of the light propagating in the optical waveguide is changed. As the principle of changing the phase of light, LiNbO ₃ mainly uses the Pockels effect, InP and GaAs mainly use the Pockels effect and the Quantum Confined Stark Effect (QCSE), and Si mainly uses the carrier plasma effect. Be done.

In order to perform high-speed and low-power-consumption optical communication, an optical modulator having a high modulation speed and a low drive voltage is required. In order to perform optical modulation with a voltage amplitude of several volts at a high speed of 10 Gbps or more, the high-speed modulated electric signal and the speed of light propagating through the optical waveguide are matched, and the interaction is performed while propagating. A wave electrode is required. An optical modulator with a traveling wave electrode whose length is several millimeters to several tens of millimeters has been put into practical use. (For example, Non-Patent Document 1)

In the light modulator of a traveling wave electrode, an electrode structure and an optical waveguide structure with low loss and low reflection are required so that the electric signal propagating in the electrode and the light propagating in the waveguide can interact with each other without reducing the intensity. Be done.

Further, the MZ type optical modulator includes a Si optical modulator in which an optical waveguide is composed of Si. The Si optical modulator is composed of an SOI (Silicon on Insulator) substrate in which a Si thin film is attached on an oxide film (BOX) layer in which the surface of the Si substrate is thermally oxidized, and the Si thin film is formed so that light can be waveguideed through the SOI layer. After processing into a thin wire, a dopant is injected so as to form a p-type / n-type semiconductor, and SiO ₂ as a clad layer of light is deposited, electrodes are formed, and the like to produce the semiconductor. At this time, the light waveguide needs to be designed and processed so that the light loss is small, and the p-type / n-type doping and the fabrication of the electrodes suppress the occurrence of light loss to a small value and reflect high-speed electric signals. It is necessary to design and process so as to keep the loss and loss small.

(Cross-sectional structure of optical waveguide)
FIG. 1 shows a cross-sectional structure diagram of an optical waveguide which is a basis of a conventional Si optical modulator. In FIG. 1, light is assumed to propagate in the direction perpendicular to the paper surface. The optical waveguide of this Si optical modulator is _{composed of a Si layer 2 sandwiched between upper and lower SiO 2} clad layers 1 and 3, and the thin Si wire for confining light in the center of FIG. 1 has ribs having different thicknesses. It has a cross-sectional structure called a waveguide.

An optical waveguide 7 that traps light propagating in the direction perpendicular to the paper surface by utilizing the difference in refractive index from the _{surrounding SiO 2} clad layers 1 and 3 with the thick Si layer 201 in the center of the Si layer 2 in FIG. 1 as the core. To configure.

In order to realize the optical modulation function, a high-concentration p-type semiconductor layer 211 and a high-concentration n-type semiconductor layer 214 are provided in the slab regions 202 on both sides of the optical waveguide 7, and a medium concentration is further provided in the central portion of the core of the optical waveguide 7. A pn junction structure is formed by the p-type semiconductor layer 212 and the medium-concentration n-type semiconductor layer 213, and modulated electrical signals and biases are applied from the left and right ends of FIG. The pn junction structure formed by the medium-concentration p-type semiconductor layer 212 and the medium-concentration n-type semiconductor layer 213 may be a pin structure in which an undoped i-type (intrinsic) semiconductor is sandwiched between them.

Although not shown in FIG. 1, metal electrodes connected to the high-concentration semiconductor layers 211 and 214 at both ends are provided, and a reverse bias electric field (right in FIG. 1) is provided at the pn junction from the metal electrode together with an RF (high frequency) modulated electric signal. By applying (from left to left), the carrier density inside the optical waveguide core 201 is changed (carrier plasma effect), and by changing the refractive index of the optical waveguide, the phase of light can be modulated.

Since the waveguide size depends on the refractive index of the core / clad material, it cannot be uniquely determined, but it has a rib-type silicon waveguide structure having an optical waveguide core portion 201 and slab regions 202 on both sides as shown in FIG. As an example of this case, the width of the waveguide core is 400 to 600 (nm) × the height is 150 to 300 (nm) × the slab thickness is 50 to 200 (nm) × the number of lengths (mm).

(MZ type Si optical modulator with single electrode structure)
FIG. 2 shows a plan view of a Si optical modulator constituting an MZ-type modulator having a structure called a conventional single electrode, and FIG. 3 shows a cross-sectional structure diagram of AA'in FIG. (See, for example, Non-Patent Document 3)

In the plan view of FIG. 2, the light input from the left side is branched into two optical waveguides 7a and 7b, and between the two upper and lower RF electrodes 5a and 5b arranged in parallel as traveling wave electrodes and the central DC electrode 6. Is phase-modulated by a modulated electrical signal (RF signal, differential electrical signal) applied as a pair of differential signals to the RF electrodes 5a and 5b. After that, the two phase-modulated optical signals are combined and interfered with each other, and the light is output from the right end to form an MZ type Si optical modulator having a single electrode structure.

In the cross-sectional structure diagram of FIG. 3 (corresponding to the AA'part of FIG. 2), the Si layer 2 has a basic structure in which two optical waveguides having the same cross-sectional structure as that of FIG. 1 are arranged symmetrically. , Two high frequency lines (RF electrodes 5a and 5b) for inputting a pair of differential modulated electric signals (RF signals) on both sides, and a DC electrode 6 for applying a common bias voltage between them. It is provided.

Two optical waveguides 7a and 7b are provided between the two RF electrodes 5a and 5b with the DC electrode 6 interposed therebetween, and a pn junction structure is symmetrically formed in the optical waveguides 7a and 7b. There is. The RF electrodes 5a and 5b are connected to the high-concentration p-type semiconductor layer 211 of the Si layer 2 via one or a plurality of vias (connection electrodes), respectively. The same applies hereinafter, but the vias (connection electrodes) are not shown in the plan view of FIG. 2 for the sake of simplicity.

Similarly, the DC electrode 6 is connected to the central high-concentration n-type semiconductor layer 214, and by applying a positive voltage to the RF electrodes 5a and 5b to the DC electrode 6, two optical waveguides 7a and 7b on the left and right are applied. A reverse bias can be applied to the pn junction of.

In such a Si optical modulator having a single electrode structure, the RF electrodes 5a and 5b and the DC electrode 6 are electrically independent due to the reverse bias of the pn junction, and a positive bias voltage is applied to the RF electrode. No longer needed. For this reason, it is possible to simplify the configuration, such as eliminating the need for bias tees for superimposing and applying a bias voltage on the RF electrodes and capacitors for the DC block installed between the driver IC and the RF electrodes. Has merit.

Here, the example in which the RF electrodes 5a and 5b are connected to the p-type semiconductor layer and the DC electrode 6 is connected to the n-type semiconductor layer has been described, but the conductive type of the semiconductor layer is replaced and the RF electrode is changed to the n-type semiconductor layer. , The DC electrode may be connected to the p-type semiconductor layer.

At this time, the bias voltage applied to the DC electrode 6 can be a reverse bias at the pn junction by applying a negative voltage to the RF electrode.

The same applies hereinafter, but when the p-type semiconductor layer or the n-type semiconductor layer is used as the first conductive type semiconductor layer without losing generality, the corresponding n-type semiconductor layer or p-type semiconductor layer is used as the second conductive type semiconductor layer. It can be called a type semiconductor layer.

According to such a notation, the MZ type optical modulator having a single electrode structure shown in FIGS. 2 and 3 has two RF electrodes to which a pair of differential electric signals are applied and the two RFs. The two first conductive semiconductor layers to which the electrodes are connected, the second conductive semiconductor layer formed between the two first conductive semiconductor layers and connected to a fixed potential, and the first and second conductive semiconductor layers. It can be said that the optical modulator includes two optical waveguides formed at the pn junction boundary of the type semiconductor layer.

In the conventional MZ type optical modulator having a single electrode structure shown in FIGS. 2 to 3, the RF electrodes 5a and 5b acting as traveling wave electrodes are on the SiO ₂ clad layer 3 of the optical modulator substrate which is a dielectric layer. In addition, it is formed by a structure called a Coplanar Stripline (CPS line).

FIG. 4 conceptually shows the state of propagation of a high-frequency electric signal on such a CPS line. In the CPS line, a pair of differential RF electric signals V + and V- are input to and propagated on the two high frequency electrodes 5a and 5b. A high-frequency signal that propagates on a CPS line as a traveling wave electromagnetically has positive and negative signs on the paired electrodes of the differential line due to Coulomb interaction when a densely charged part moves like a wave on the high-frequency propagation line. It can be understood by the model that the opposite charge is induced in the dense part and moves in the same way as the high frequency signal. The CPS line is one of the balanced lines in which the positive and negative charges are balanced by two electrodes paired with the differential line.

Po Dong, Long Chen, and Young-kai Chen, "High-speed low-voltage single-drive push-pull silicon Mach-Zehnder modulators", OPTICS EXPRESS Vol.20, No.6, pp.6163-6169, 12 March 2012. R. Ding et al., "Design and characterization of a 30-GHz bandwidth low-power silicon traveling-wave modulator", Optics. Communication, vol. 321, pp. 124-133, 7 February 2014. David Patel, Samir Ghosh, Mathieu Chagnon, Alireza Samani, Venkat Veerasubramanian, Mohamed Osman, and David V. Plant, "Design, analysis, and transmission system performance of a 41 GHz silicon photonic modulator," OPTICS EXPRESS Vol.23, 14263- 142871, 22 May 2015

In order to perform large-capacity optical communication, an optical modulator capable of modulating light at high speed is required. In order to perform high-speed optical modulation, a frequency characteristic capable of operating over a wide frequency band of several hundred kHz to several tens of GHz is required. In particular, in order to increase the signal baud rate (modulation rate) and perform a larger capacity communication, the optical modulation band (EO band), which is the signal frequency characteristic of the optical signal with respect to the modulated electric signal, is widened. is important.

There are two main factors that limit the EO band of the MZ type optical modulator, one is the mismatch between the signal propagation speeds of light and electricity, and the other is the propagation loss of the high frequency signal.

In the MZ type optical modulator, the phase of the optical signal propagating through the two optical waveguides is modulated by a pair of high-frequency electric signals propagating through the two traveling wave type RF electrodes provided above the optical waveguide. .. Therefore, when the signal propagation speed of light and the signal propagation speed of electricity are different, the phase shift between the two gradually occurs while the light and electricity propagate through the modulator. This phase shift reduces the conversion efficiency from the electric signal to the optical signal. In particular, in the high frequency region, the modulation efficiency drops significantly due to the phase shift. Therefore, it is necessary to sufficiently match the electric signal speed of the RF electrode and the optical signal speed of the optical waveguide.

Further, the MZ type optical modulator can be regarded as a kind of the above-mentioned CPS transmission line from the viewpoint of propagation of high frequency electric signals. An ideal transmission line can propagate a high-frequency electric signal without loss, but an actual line causes loss due to resistance and scattering. Even in the MZ type optical modulator, the signal amplitude is reduced while the modulated electrical signal propagates through the RF electrode, which deteriorates the EO band.

(Distribution constant circuit model of Si optical modulator)
FIG. 5 shows a distributed constant circuit model of the Si optical modulator. Based on this model, the electric velocity and propagation loss of the MZ type Si optical modulator will be described. _{Here, C pn} and R _pn of the transmission line element in FIG. 5 are the capacitance and resistance per unit length of the pn junction, R _tl is the resistance per unit length of the transmission line, and L _tl is the inductance per unit length of the line. C _tl is the capacitance per unit length of the line other than the pn junction.

This model represents a single-phase drive, but even in the case of differential drive, a similar model can be considered by extracting effective parameters for the differential signal. The voltage amplitude of the differential signal is represented as the amplitude on the signal line in the figure.

Using this model, the attenuation of the radio frequency (RF) electrical signal of the Si light modulator is represented by the following equation (1) (Non-Patent Document 2).

However,

Is.

Here, α on the left side of equation (1) is the attenuation of the RF signal (propagation loss, unit is Np / m), f is the signal frequency, Z _dev of equation (2) is the characteristic impedance, and _frc of equation (3) is. The RC band, v _p of equation (4), is the propagation velocity of the electric signal.

In the formula (1), the first term on the right side _{represents the loss due to the resistance R tl} of the RF electrode, and the second term represents the loss due to _{the resistance R pn in the pn doping region. The R tl of} the first term _{is R tl to} √f due to the skin effect of the RF electrode, which is proportional to the square root of f, while the second term is proportional to ^{f 2.} Therefore, the second term is a factor that limits the propagation loss of the electric signal in the high frequency region and limits the EO band. To reduce this effect, the second term by utilizing the fact that depend on Z _dev, it is effective to reduce the Z _dev.

_{As a design to reduce Z dev} in the conventional CPS single electrode type Si optical modulator, it is considered to increase the _{effective C tl} by widening the width of the RF electrode or reducing the distance between the two RF electrodes. Be done. However, such a design reduces the _{effective L tl at the same time.} Therefore, there is a problem that the electric velocity v _p represented by the equation (4) increases, and the velocity matching between light and electricity cannot be obtained.

In addition, although speed matching can be achieved with a structure like a capacitively loaded modulator (Non-Patent Document 3), it is difficult to set _{Z dev low due to the design restriction that the electrode area is small.} rice field.

The present invention is characterized by providing the following configurations in order to achieve such an object.

(Structure 1 of the invention)
Between the two RF electrodes to which a pair of differential electric signals are applied, the two first conductive semiconductor layers to which the two RF electrodes are connected, and the two first conductive semiconductor layers. An optical modulator comprising a second conductive semiconductor layer formed and connected to a fixed electrode, and two optical waveguides formed at the pn junction boundary of the first and second conductive semiconductor layers.
It further comprises a plurality of conductive regions spaced apart along the RF electrode.
One end of the plurality of conductive regions is formed in a lower layer or an upper layer of the RF electrode and is capacitively coupled to a part of the RF electrode, and at least a part of the other end different from the one end has common wiring. Connected to each other through
At least one or more of the plurality of conductive regions is a conductive region having both a region in which the RF electrode is arranged in the upper layer or the lower layer and a region in which the RF electrode is not arranged in the upper layer or the lower layer.
An optical modulator, wherein the plurality of conductive regions are a plurality of elongated electrodes extending in a direction perpendicular to an RF electrode.

(Structure 2 of the invention)
The light modulator according to the configuration 1 of the present invention.
A light modulator characterized in that the plurality of conductive regions are formed in a region sandwiched between the two RF electrodes including a portion that is multilayered with the two RF electrodes.

(Structure 3 of the invention)
The light modulator according to the configuration 1 or 2 of the present invention.
A light modulator characterized in that the plurality of conductive regions are formed toward the outside of a region sandwiched between the two RF electrodes.

(Structure 4 of the invention)
The light modulator according to any one of the configurations 1 to 3 of the present invention.
An optical modulator, characterized in that at least a part of the plurality of conductive regions is connected to a specific potential.

(Structure 5 of the invention)
The light modulator according to any one of the configurations 1 to 4 of the present invention.
An optical modulator characterized in that the arrangement interval of the plurality of conductive regions and the length in the high-frequency signal propagation direction are each ½ or less of the high-frequency signal operating wavelength.

In the light modulator according to the present invention, the RF electrode and a plurality of conductive regions arranged apart from each other along the RF electrode are capacitively coupled. Due to the effect of this capacitive coupling, in this modulator, the capacitance C _{tl of the} transmission line can be increased while maintaining the photoelectric velocity matching, so that the characteristic impedance Z _dev can be reduced. This makes it possible to reduce the propagation loss of the high frequency signal in the high frequency region. Therefore, it is possible to provide an optical modulator having excellent high frequency characteristics and good waveform quality.

It is sectional drawing of the optical waveguide of the conventional Si light modulator. It is a top view of the Si light modulator which constitutes the Mach Zenda (MZ) type Si light modulator of the conventional single electrode structure. It is sectional drawing of the conventional MZ type Si optical modulator of FIG. It is a schematic diagram which conceptually showed the state of propagation of a high frequency electric signal in a CPS line. It is explanatory drawing of the distributed constant circuit model of a Si optical modulator. It is a top view which shows the structure of the MZ type optical modulator according to Example 1 of this invention. It is sectional drawing which shows the structure of the MZ type optical modulator according to Example 1 of this invention. This is a calculation result showing the frequency characteristics of (a) attenuation and (b) velocity of the MZ type optical modulator according to the first embodiment of the present invention in comparison with the conventional example. This is a calculation result showing (a) the characteristic impedance and (b) the frequency characteristic of the capacitance of the MZ type optical modulator according to the first embodiment of the present invention in comparison with the conventional example. This is a calculation result showing (a) inductance and (b) frequency characteristics of the EO band of the MZ type optical modulator according to the first embodiment of the present invention in comparison with the conventional example. This is a calculation result showing the frequency characteristics of (a) band limitation due to speed mismatch and (b) band limitation due to propagation loss of the MZ type optical modulator according to the first embodiment of the present invention in comparison with the conventional example. It is a top view which shows the structure of the MZ type optical modulator according to Example 2 of this invention. It is sectional drawing which shows the structure of the MZ type optical modulator according to Example 2 of this invention. It is a top view which shows the structure of the MZ type optical modulator according to Example 3 of this invention. It is sectional drawing which shows the structure of the MZ type optical modulator according to Example 3 of this invention.

(Example 1)
Hereinafter, the form of the optical modulator of the present invention will be described in detail with reference to preferred examples.

FIG. 6 is a plan view showing the configuration of the MZ type optical modulator according to the first embodiment of the present invention.

As shown in FIG. 6, the optical modulator according to the first embodiment of the present invention is arranged separately along the RF electrode in a Si optical modulator having a single electrode structure similar to the conventional one, and is capacitively different from the RF electrode. It is characterized in that a plurality of conductive regions 9a and 9b which are coupled to and act as floating electrodes are provided. As the conductive region, various shapes can be applied, but for the sake of simplicity, the conductive region is oriented in the direction perpendicular to the RF electrode toward the outside of the region sandwiched between the two RF electrodes. As examples of elongated linear orthogonal electrodes 9a and 9b extending apart from each other, a configuration in which a plurality of elongated linear orthogonal electrodes 9a and 9b are formed along the RF electrodes is illustrated.

In the light modulator of the first embodiment of FIG. 6, the plurality of orthogonal electrodes 9a and 9b extend from the lower part of the RF electrodes 5a and 5b toward the outside of the light modulator, respectively, and the signal propagation direction in the RF electrode ( It is an elongated linear conductive region arranged perpendicular to the longitudinal direction of the RF electrode), and a plurality of orthogonal electrodes are exemplified in a configuration in which they are arranged at equal lengths and at equal intervals.

As shown in the cross-sectional view of Example 1 of FIG. 7, the orthogonal electrodes 9a and 9b, which are conductive regions, are produced in the same process on the electrode layer at the same level as the DC electrode 6. In the first embodiment, the orthogonal electrodes 9a and 9b, which are conductive regions, are formed at the same level as the DC electrode 6, but if a capacitive coupling can be formed with the RF electrode, they may be formed in a layer of another level. good.

For example, the same effect as in the first embodiment can be obtained by forming an insulating layer such as a dielectric on the RF electrodes 5a and 5b and arranging the orthogonal electrodes 9a and 9b on the layer. However, in a general silicon photonics process, since the top electrode is the thickest electrode, a signal electrode is usually formed in the top layer. Therefore, in the first embodiment, the configuration in which the orthogonal electrodes 9a and 9b are formed in multiple layers below the electrode layer of the RF electrode is exemplified.

Further, all of such a plurality of conductive regions (orthogonal electrodes) do not necessarily have to be formed in the lower layer or the upper layer of the RF electrode, and at least one or more of the plurality of conductive regions has the RF electrode in the upper layer. Alternatively, a capacitive coupling can be formed as long as it is a conductive region having both a region in which the RF electrodes are arranged in multiple layers in the lower layer and a region in which the RF electrode is not arranged in the upper layer or the lower layer.

In the first embodiment, the plurality of conductive regions constituting the orthogonal electrodes 9a and 9b are floating electrodes that are DC-independent and DC-electrically not connected to other parts. Further, the length, width, density, etc. of each linear electrode constituting the orthogonal electrode can be freely designed within a range that can be manufactured. For example, the length can be designed to be 200 μm, the interval to be 30 μm, and the electrode width to be 2 μm. ..

Here, the length (electrode width) in the signal propagation direction and the electrode spacing of the plurality of conductive regions constituting the orthogonal electrodes 9a and 9b are set to 1/2 or less of the operating wavelength of the RF electrode of the high-frequency signal. It is designed to. This is a design for preventing resonance in a plurality of conductive regions constituting the RF electrode and the orthogonal electrode.

Electromagnetic wave propagation velocity in a vacuum is approximately 3 × 10 ⁸ m / s, when approximately n _eff = 3 the effective refractive index of the high-frequency signal propagating through the RF electrodes of the Si modulators, wavelength of 10GHz of the high frequency signal The wavelength of the high frequency signal of 10 mm and 40 GHz is about 2.5 mm. Therefore, if the length (electrode width) and electrode spacing of each orthogonal electrode in the signal propagation direction is 1/2 of the wavelength of the RF electrode of the high-frequency signal, which is 5 mm or less, up to 10 GHz. If it is 25 mm or less, it is possible to realize an optical modulator that does not cause resonance up to a signal of 40 GHz.

Here, the manifestation of the effect of widening the EO band in the present invention will be described. FIG. 7 shows a cross-sectional view taken along the line BB'of Example 1 of FIG. Consider the propagation of electrical signals using the charge model of FIG. 4 described above.

In the cross-sectional view of Example 1 of FIG. 7, it is assumed that the RF electrode 5a is given a negative charge and the RF electrode 5b is given a positive charge. At this time, consider the potential created by the positive charge of the RF electrode 5b on the right side of FIG. Considering the distance from the RF electrode 5a, the potentials (potential A and potential B) at the right and left ends of the orthogonal electrode 9b are potential A> potential B, and a potential difference is formed inside the orthogonal electrode 9b. Here, since the orthogonal electrode 9b is a conductor, an electric charge is induced on the surface so as to cancel the internal potential difference. This state can be considered to be a state in which the RF electrode 5b and the orthogonal electrode 9b form a capacitor in a multi-layered portion where they are vertically overlapped and are capacitively coupled.

Similarly, the RF electrode 5a and the orthogonal electrode 9a on the left side of FIG. 7 also form a capacitor. By strengthening the AC electrical coupling between the RF electrode and the orthogonal electrode in this way, the capacitance C _tl of the transmission line in the circuit model of FIG. 5 increases.

Further, since the orthogonal electrodes 9a and 9b are divided in the signal traveling direction of the RF electrode (the longitudinal direction of the RF electrode), the return current induced in the signal potential hardly flows. _{Therefore, the value of the inductance L tl} in the circuit model of FIG. 5 hardly changes with or without the orthogonal electrodes 9a and 9b.

Therefore, the presence of the orthogonal electrodes 9a and 9b makes it possible to increase only _{C tl , and the Z dev} represented by the equation (2) is reduced. As described above, since the propagation loss in the high frequency region is _{reduced by reducing Z dev} , the EO band can be designed wider when there are orthogonal electrodes.

Further, the propagation speed of the electric signal (v _{p of the} equation (4)) can also be adjusted by _{adjusting the L tl} of the modulator by designing the electrode widths of the RF electrodes 5a and 5b, the electrode spacing, and the like. _{Since C tl} increases due to the effect of the orthogonal electrodes 9a and 9b, L _tl can be designed to decrease. _{Since this adjustment is an adjustment in the direction of reducing Z dev} from the equation (2), the propagation loss does not increase by this adjustment.

That is, according to the present invention, it is possible to design to match the photoelectric velocity at the same time as expanding the EO band.

Here, FIGS. 8 to 11 show the case where there are orthogonal electrodes 9a and 9b of the MZ type optical modulator according to the present invention (With Cross Electrode, solid line) and the case where there is no orthogonal electrode of the conventional MZ type optical modulator (with cross electrode). A design example of Without Cross Electrode (dotted line) will be described by comparing the calculation results of the characteristics of various optical modulators by the electromagnetic field simulator HFSS.

In the case of differentially driving two single-electrode Si optical modulators of the same design, one was calculated with orthogonal electrodes 9a and 9b, and the other was calculated without orthogonal electrodes 9a and 9b. Further, the group index of the optical signal is set to 3.8 (optical signal rate ^{0.79 × 10 8 m / s)} , and a modulator having a modulation section of 3 mm. The orthogonal electrode was a floating electrode.

As shown in the capacitance of FIG. 9B and the inductance of FIG. 10A, in the present invention, the addition of the orthogonal electrodes 9a and 9b increases only the effective capacitance without lowering the effective inductance. You can see that it is possible. As a result, as shown in FIG. 9A, the characteristic impedance can be reduced, and as shown in FIG. 8A, the attenuation in the high frequency region (> 13 GHz) can be reduced. In FIG. 8A, the attenuation in the low frequency region (<13 GHz) is slightly large, which is due to the increase in the first term of the equation (1).

As described above, the EO band is determined by the band limitation due to the inconsistency of the optical electric velocity and the band limitation due to the propagation loss. As shown in FIG. 11B, the limitation of the EO band is relaxed by reducing the propagation loss. I can do it. Further, in this design, by attaching orthogonal electrodes 9a and 9b, the design is made so that the speed can be almost matched. Therefore, the band limitation due to the speed mismatch can be relaxed (Fig. 11 (a)).

This effect is due to the fact that if there is a conductive region in the vicinity of the RF electrodes 5a and 5b, a capacitive coupling between the RF electrodes 5a and 5b and the conductive region is realized due to the potential difference in the conductor. Therefore, the shapes of the orthogonal electrodes 9a and 9b do not have to be linear or rectangular, and may be tapered, bent, or wave-shaped so as to widen or narrow toward the outside of the modulator. The same effect can be obtained even in the case of a shape extending from a direction perpendicular to the RF electrodes 5a and 5b in a direction inclined.

In this embodiment, RF electrodes 5a and 5b having a simple shape have been discussed, but these also do not have to be rectangular, and depending on the convenience of design, deformed electrodes such as a capacitive loading type as described in Non-Patent Document 3. It may be.

(Example 2)
FIG. 12 shows a plan view showing the configuration of the MZ type optical modulator according to the second embodiment of the present invention. In the second embodiment, with respect to the features described in the first embodiment, the parts that perform the same function are designated by the same reference numerals, and duplicate description will be omitted.

As shown in FIG. 12, the light modulators according to the second embodiment of the present invention include a plurality of Si optical modulators having a single electrode structure similar to the conventional one and orthogonal electrodes 9a and 9b having the same structure as the first embodiment. Is composed of bus electrodes 10a and 10b that electrically connect the orthogonal electrodes 9a and 9b of the above to each other at an end portion away from the RF electrode.

The plurality of conductive regions constituting the orthogonal electrodes 9a and 9b described in the first embodiment of FIG. 6 were independent floating electrodes, but in the second embodiment, as shown in the cross-sectional view of FIG. The orthogonal electrodes 9a and 9b are connected to the bias potential via bus electrodes 10a and 10b formed on the upper portion of the end portion of the orthogonal electrodes 9a and 9b on the side away from the RF electrode. Here, the bias potential is a specific potential applied to the DC electrode 6 for bias.

FIG. 13 shows a cross-sectional view of CC ′ of FIG. 12 of Example 2. The bus electrodes 10a and 10b are formed in the same electrode layer as the RF electrodes 5a and 5b by the same process, and are connected to the orthogonal electrodes 9a and 9b by short vias (connection electrodes) at the ends on the side away from the RF electrodes. NS. In the plan view of FIG. 12, vias are not displayed for the sake of simplicity.

In the second embodiment, the bus electrodes 10a and 10b are realized by using the electrode layers at the same level as the RF electrodes 5a and 5b, but they may be manufactured by using a layer having another conductivity. The bus electrode is connected to a plurality of conductive regions constituting the orthogonal electrode via vias. The bus electrodes do not have to be rectangular and may be split. When splitting, all bus electrodes must be connected to the bias potential.

In the floating orthogonal electrodes 9a and 9b of Example 1, the amount of charge induced in the orthogonal electrodes is proportional to the potential difference between the right end and the left end of the orthogonal electrodes 9a and 9b. Therefore, _{if it is desired to increase C tl} , it is necessary to lengthen the orthogonal electrode. However, at this time, there is a problem that the footprint (planar size) of the modulator becomes large.

In the second embodiment, a plurality of conductive regions constituting the orthogonal electrodes are connected to the bias potential by the bus electrodes 10a and 10b. In this case, the potential difference between the right end and the left end in the orthogonal electrode becomes equal to the absolute value of the signal potential, and a larger potential difference than in Example 1 can be obtained. Since more charges can be induced by this potential difference, in the second embodiment, a sufficient C _tl increase effect can be obtained even when the orthogonal electrode length is short.

In the second embodiment, an example in which all the orthogonal electrodes 9a and 9b on each side are connected by the bus electrodes 10a and 10b has been described, but it is not necessary to connect all the orthogonal electrodes, and it is necessary. At least a part of the orthogonal electrodes (conductive regions) may be connected to the bus electrodes. Further, it is not necessary to have two bus electrodes, and the same effect can be obtained even if the respective orthogonal electrodes 9a and 9b are connected by a plurality of wirings. Further, in the second embodiment, all the orthogonal electrodes are connected to the bias potentials, but if necessary, at least a part of the plurality of conductive regions (orthogonal electrodes) may be connected to different, arbitrary specific potentials. A similar effect can be obtained.

(Example 3)
Example 3 of the present invention shown in the plan view of FIG. 14 and the cross-sectional view of DD'of FIG. 15 is an example of an MZ type optical modulator in which an orthogonal electrode is connected to a bias potential as in Example 2. Is. The difference from the second embodiment is that in the third embodiment, the DC electrode 6 for bias driving, which is located between the two RF electrodes, is partially extended (widened) vertically in the signal propagation direction in a branch shape at a plurality of places as it is. ), And it is used as a plurality of orthogonal electrodes 9a and 9b. Therefore, in the plan view of FIG. 14, a plurality of orthogonal electrodes 9a are formed in a region sandwiched between the two RF electrodes, including a portion between the two RF electrodes 5a and 5b that overlaps the two RF electrodes and forms a multilayer. , 9b will be formed.

In FIGS. 14 and 15 of the third embodiment, the orthogonal electrodes 9a and 9b are described as two electrodes according to the other embodiments, but the DC electrode 6 is an electrode formed by extending (widening) the DC electrode 6 as it is. Therefore, they have the same electric potential and may be configured as an integral electrode including the DC electrode 6.

In Examples 1 and 2, since the orthogonal electrodes are not arranged in the vicinity of the optical waveguide, there is an advantage that the operation is stable, but since the orthogonal electrodes extend toward the outside of the two RF electrodes, the light modulator There was a possibility that the overall flat size (footprint) would increase. In the third embodiment, the orthogonal electrode is formed between the two RF electrodes and does not spread to the outside, which is advantageous for miniaturization.

As shown in the cross-sectional view of Example 3 of FIG. 15, even orthogonal electrodes 9a and 9b having such a structure are arranged apart from each other along the RF electrodes 5a and 5b, and are capacitively different from the RF electrodes. It can be said that there are a plurality of conductive regions to be bonded.

Therefore, as in Examples 1 and 2, in Example 3, the capacitance C _{tl of the} transmission line can be increased, Z _dev can be reduced, and the propagation loss in the high frequency region is also reduced, so that the EO band can be increased. Can be widely designed.

As described above, in the optical modulator according to the present invention, it is possible to increase the optical modulation efficiency in the high frequency region. Therefore, it is possible to improve adverse effects such as deterioration of waveform quality during high-speed modulation due to deterioration of the frequency response characteristics of the optical modulator, and provide an optical modulator having excellent high-frequency characteristics and good waveform quality. Is possible.

1,3 SiO2 clad layer 2 Si layer 201 Optical waveguide core part 202 Slab region 211 High-concentration p-type semiconductor layer 212 Medium-concentration p-type semiconductor layer 213 Medium-concentration n-type semiconductor layer 214 High-concentration n-type semiconductor layer 5a, 5b RF electrodes 6 DC electrodes 7, 7a, 7b Optical waveguides 9a, 9b Orthogonal electrodes (conductive region)
10a, 10b bus electrodes

Claims (5)

Between the two RF electrodes to which a pair of differential electric signals are applied, the two first conductive semiconductor layers to which the two RF electrodes are connected, and the two first conductive semiconductor layers. An optical modulator comprising a second conductive semiconductor layer formed and connected to a fixed electrode, and two optical waveguides formed at the pn junction boundary of the first and second conductive semiconductor layers.
It further comprises a plurality of conductive regions spaced apart along the RF electrode.
One end of the plurality of conductive regions is formed in a lower layer or an upper layer of the RF electrode and is capacitively coupled to a part of the RF electrode, and at least a part of the other end different from the one end has common wiring. Connected to each other through
At least one or more of the plurality of conductive regions is a conductive region having both a region in which the RF electrode is arranged in the upper layer or the lower layer and a region in which the RF electrode is not arranged in the upper layer or the lower layer.
An optical modulator, wherein the plurality of conductive regions are a plurality of elongated electrodes extending in a direction perpendicular to an RF electrode. The optical modulator according to claim 1.
A light modulator characterized in that the plurality of conductive regions are formed in a region sandwiched between the two RF electrodes including a portion that is multilayered with the two RF electrodes. The light modulator according to claim 1 or 2.
A light modulator characterized in that the plurality of conductive regions are formed toward the outside of a region sandwiched between the two RF electrodes. The optical modulator according to any one of claims 1 to 3.
An optical modulator, characterized in that at least a part of the plurality of conductive regions is connected to a specific potential. The optical modulator according to any one of claims 1 to 4.
An optical modulator characterized in that the arrangement interval of the plurality of conductive regions and the length in the high-frequency signal propagation direction are each ½ or less of the high-frequency signal operating wavelength.

Images