JP6106071B2 - Polarization control element - Google Patents

Description

  The present invention relates to a polarization control element, and more particularly to a polarization control element capable of operating at a high speed and in a wide wavelength band.

  With the increase in capacity of optical communication systems, high-speed optical modulators that are compatible with advanced optical modulation schemes are being demanded. In particular, multi-level optical modulators using digital coherent technology play a major role in realizing large capacity transceivers exceeding 100 Gbps. In these multilevel optical modulators, optical modulators (hereinafter referred to as MZM) capable of Mach-Zehnder interference type zero chirp drive are incorporated in parallel multistages in order to add independent signals to the amplitude and phase of light. ing.

  MZM using a lithium niobate (hereinafter referred to as LN) substrate has been put to practical use mainly in backbone networks because of its excellent low optical loss and low chirp operation, and is commercially introduced in 100 Gbps coherent optical transmission. The multilevel modulator also uses LN material.

  In recent years, with the expectation of further increase in communication capacity, it has been studied that coherent transmission technology can be incorporated not only in long-distance communication but also in medium- and short-distance communication. Therefore, further cost reduction, power consumption, and size reduction of the optical device are required. MZM based on compound semiconductors centered on InP materials can utilize high-efficiency electro-optic effects using quantum effects such as strong interaction between light waves and electric fields and quantum confined Stark effect (QCSE). Has attracted attention as a material.

  Furthermore, in order to cope with future increases in transmission capacity, research and development is being actively conducted to further increase the frequency utilization efficiency by controlling not only the amplitude and phase of light but also the polarization state at high speed. In order to control the polarization at high speed, waveguide type optical devices mainly using a crystal having an electro-optic effect as a substrate material are used. These waveguide type optical devices include the above-described LN and compound semiconductors. It is used as a polarization control element device together with materials.

JP-A-62-284331

Farnoosh Rahmatian et al., “An Ultrahigh-Speed AlGaAs-GaAs Polarization Converter Using Slow-Wave Coplanar Electrodes”, IEEE PHOTONICS TECHNOLOGY LETTERS, May 1998, VOL.10, NO.5, pp.675-677 Farnoosh Rahmatian et al., “An Ultrahigh-Speed AlGaAs-GaAs Polarization Converter Using Slow-Wave Coplanar Electrodes”, ELECTRONICS LETTERS, September 1990, VOL.26, NO.19, pp.1560-1561

  However, in order to control the polarization at high speed using the LN material having a large dielectric constant in the polarization control element, there is a problem that the applied voltage becomes high or the element length becomes long. Further, monolithically integrating elements having an optical amplitude control function, a phase modulation function, and the like still has problems in crystal plane orientation.

  For example, as shown in Non-Patent Document 1, research has been made on polarization control elements of compound semiconductors such as GaAs, but most of these devices have a substrate sliced in the (100) plane direction. In order to rotate the polarization, it is necessary to apply an electric field in the horizontal direction of the substrate to the waveguide. Since this may increase the element capacitance C due to the structure, a special electrode structure (for example, a capacitance loaded electrode) is required for high-speed response, resulting in an unnecessary increase in element size. .

  As shown in Patent Document 1 and Non-Patent Document 2, studies have been made to rotate the polarization by applying an electric field in the direction perpendicular to the substrate by using a substrate sliced in the (110) plane. Because of the large birefringence caused by the above, it has been difficult to achieve sufficient polarization rotation unless a high voltage bias is applied. Furthermore, since these devices are formed of buried waveguides or ridge waveguides, which require relatively low waveguide processing accuracy, parasitic capacitance tends to increase, and cannot be said to be a structure suitable for high-speed operation. . In addition, there are no reports on high-speed operation when a (110) plane orientation substrate is used, including Patent Document 1 and Non-Patent Document 2.

  The present invention has been made in view of such problems, and an object of the present invention is to realize a high-speed polarization control element capable of controlling a low driving voltage in a small size. Furthermore, the polarization control element is integrated in series, and a reverse sign voltage is applied to each element to improve the polarization conversion efficiency, thereby providing a polarization control element capable of controlling the driving voltage at a higher speed and a lower speed. For the purpose.

In order to solve the above-described problem, the polarization control element according to claim 1 is laminated in a (110) plane orientation substrate and a [−110] stripe direction on the (110) plane orientation substrate. High-mesa waveguide structure composed of a type semiconductor crystal, including a lower cladding layer, a core layer, and an upper cladding layer, wherein the core layer is sandwiched between the lower cladding layer and the upper cladding layer A high-mesa waveguide structure, a signal electrode formed on the waveguide structure and applying an electric field in the [110] direction to the core layer, and a ground electrode formed on the lower cladding layer , An electron barrier layer provided between the core layer and the lower clad layer and between the core layer and the upper clad layer is further provided .

  The polarization control element according to claim 2 is the polarization control element according to claim 1, wherein a difference in refractive index between the material of the core layer and the material of the lower cladding layer and the upper cladding layer is 4. % Or less.

According to a third aspect of the present invention , in the inversion type polarization control element, at least two polarization control elements according to the first or second aspect are integrated in series on the same straight waveguide, Each is a p-type layer, and the lower clad layer and the upper clad layer are n-type layers.

  According to the present invention, it is possible to manufacture a small polarization control element with high polarization conversion efficiency and capable of operating in a wide wavelength band with high yield. Further, by arranging the polarization control elements in series, it is possible to further improve the polarization conversion efficiency and to achieve a high speed and low driving voltage. Furthermore, in the polarization rotation, since a phase change accompanying amplitude displacement, that is, modulation chirp does not occur in principle, application to a multilevel modulator using the polarization control element can be expected.

It is sectional drawing of the polarization control element which concerns on Example 1 of this invention. It is sectional drawing of the other example of the polarization control element which concerns on Example 1 of this invention. It is sectional drawing of the further another example of the polarization control element which concerns on Example 1 of this invention. It is sectional drawing of the polarization control element which concerns on Example 2 of this invention. It is a figure which shows the change of the refractive index ellipsoid by the presence or absence of the application of an electric field of a [110] direction when a waveguide is formed in the predetermined stripe direction on a (110) plane orientation board | substrate in a zinc blende type crystal. It is a figure which shows the waveguide width dependence of birefringence (DELTA) n (TE-TM) by the difference in a core / clad specific refractive index difference. It is a figure which shows the waveguide refractive index difference dependence of fabrication processing tolerance, and the waveguide refractive index dependence of an optical confinement coefficient. It is a figure which shows the electrode inversion type | mold polarization control element comprised by arrange | positioning a polarization control element in series It is a figure which shows that conversion efficiency increases by electrode inversion phase matching. It is a figure which shows the high-speed operation characteristic in the waveguide structure from which element capacitance differs.

  Embodiments of the present invention will be described below with reference to the accompanying drawings. Embodiment described below is an illustration and this invention is not restrict | limited to the following embodiment. In addition, in this specification and drawing, the component with the same code | symbol shall show the mutually same thing.

  In the wide wavelength band operation of the polarization control element, it is required to match the phase velocities of the propagating TE mode light and TM mode light in the wide wavelength band. Further, in order to reduce the size of the polarization control element, a strong interaction between light and an electric field is required. In order to satisfy this requirement, in the polarization control element according to the present invention, an electric field is applied in the [110] direction perpendicular to the substrate using a substrate with (110) plane orientation, and a high mesa structure is used as a waveguide structure. Thus, since a high electric field can be applied to the waveguide core layer, the device capacitance C does not increase and a high-speed response is possible, and sufficient polarization rotation can be realized.

  The dependence of the high frequency response characteristics on the waveguide structure (parasitic capacitance) will be described with reference to FIG. FIG. 10 shows high-speed operation characteristics in waveguide structures with different element capacities. In FIG. 10, high-frequency response characteristics were calculated using traveling-wave electrodes (50Ω termination) in a high-mesa waveguide structure and a buried waveguide structure having the same core layer thickness and upper cladding width.

  In general, in a semiconductor optical device, the characteristic impedance of an element is often less than 50Ω. Therefore, in order to satisfy impedance matching, it is necessary to increase the characteristic impedance (lower the element capacitance C). In a high-mesa waveguide structure having a smaller element capacitance C compared to a ridge waveguide structure or a buried waveguide structure, the characteristic impedance of the element approaches 50Ω, so that RF reflection is reduced as shown in FIG. As a result, the transmission characteristics are improved. Therefore, the high mesa waveguide structure is excellent in high-speed response compared with the ridge waveguide structure or the buried waveguide structure.

Example 1
FIG. 1 is a cross-sectional view of a polarization control element according to Embodiment 1 of the present invention. FIG. 1 shows a substrate 101 made of a substrate crystal with a (110) plane orientation, an n-cladding layer 102 laminated on the substrate 101, and an i-core layer 103 laminated on the n-cladding layer 102. Two ground electrodes 104 formed on both sides of the i-core layer 103 on the n-cladding layer 102, a p-cladding layer 105 laminated on the i-core layer 103, and a p-cladding layer 105 A polarization control element including a p-contact layer 106 laminated thereon and a signal electrode 107 formed on the p-contact layer 106 is shown. As shown in FIG. 1, the n-cladding layer 102, the i-core layer 103, the p-cladding layer 105, and the p-contact layer 106 constitute a high mesa waveguide structure using a zinc blende type semiconductor crystal. Yes. In the polarization control device according to the first embodiment, the waveguide structure has a (110) plane orientation on the substrate 101.

An i-core layer 103 serving as a core waveguide is formed in a stripe direction (hereinafter referred to as [−110] stripe direction), and the i-core layer 103 is vertically moved by at least the n-cladding layer 102 and the p-cladding layer 105. A high mesa waveguide structure composed of a zinc blende type semiconductor crystal is adopted, and an electric field is applied from the signal electrode 107 in the [110] direction perpendicular to the substrate.

  In the polarization control element according to the present invention, crystal growth can be performed by a metal organic vapor phase epitaxy (MOVPE) method suitable for a crystal regrowth process. The i-core layer 103 in the light modulation region is a non-doped layer, and the bandgap wavelength and the structure can be determined so that the electro-optic effect works effectively at the operating light wavelength and the light absorption does not become a problem.

  In addition, as will be described later, the semiconductor that constitutes the i-core layer 103 and the upper and lower cladding layers 102 and 105 so that the refractive index difference between the i-core layer 103 and the upper and lower cladding layers 102 and 105 is 4% or less. The material can be selected. For example, when a device in the 1.55 μm band is used, InGaAsP having a band gap wavelength of 1.2 μm can be used for the core layer, and InP material can be used for the upper and lower cladding layers. In this example, InGaAsP was used as the material constituting the i-core layer 103, and InP was used as the material constituting the upper and lower cladding layers 102 and 105. In this example, Si—InP was used as a material constituting the substrate 101, and InGaAsP was used as a material constituting the p-contact layer 106.

  It is obvious that the semiconductor material of the i-core layer 103 and the upper and lower cladding layers 102 and 105 is not limited to InGaAsP / InP, and for example, InGaAlAs, InGaAs, InAlAs or the like may be used.

It is desirable that the doping concentration of the n-cladding layer 102 and the p-cladding layer 105 be 5 × 10 ¹⁷ cm ⁻³ or more so that a voltage drop is efficiently generated in the i-core layer 103 which is a non-doped layer. Moreover, it is desirable that the doping concentration of the p-contact layer 106 for contact of the signal electrode 107 is 1 × 10 ¹⁹ cm ⁻³ or more so that sufficient ohmic contact can be obtained. For example, in this embodiment, the doping concentration of the n-cladding layer 102 and the p-cladding layer 105 is 1 × 10 ¹⁸ cm ^−3, and the doping concentration of the p-contact layer 106 is 2 × 10 ¹⁹ cm ⁻³ .

After depositing the p-i-n layer, in order to make the polarization control element function as an optical waveguide, the n-type is formed by dry etching using, for example, a SiO ₂ mask whose waveguide shape is simulated in the [−110] stripe direction. Processing is performed until the cladding layer 102 is exposed, and a high mesa structure is manufactured. After removing the SiO ₂ mask, as shown in FIG. 1, the ground electrode 104 and the signal electrode 107 are formed on the p-contact layer 106 and the n-cladding layer 102 by, for example, forming Au through Ti. Even if an n-InGaAs contact layer is provided between the n-cladding layer 102 and the ground electrode 104, the usefulness of the present invention is not lost. The same applies to the polarization control element according to Example 2 described later.

  FIG. 2 illustrates another example of the polarization control element according to the first embodiment. In the polarization control element shown in FIG. 1, the n-cladding layer 102 and the p-cladding layer 105 are provided in the upper layer and the lower layer of the i-core layer 103, respectively. Even if the n-cladding layer 102 and the p-cladding layer 105 are provided in the upper layer and the lower layer of the i-core layer 103, respectively, and the n-contact layer 108 is laminated on the n-cladding layer 102, there is a problem. Absent. The same applies to the polarization control element according to Example 2 described later.

  FIG. 3 shows still another example of the polarization control element according to the first embodiment. The polarization control element shown in FIG. 3 has a configuration in which a p-electron barrier layer 109 is formed between the i-core layer 103 and the n-cladding layer 102 of the polarization control element shown in FIG. The p-electron barrier layer 109 can be made of, for example, InAlAs. As in the polarization control element shown in FIG. 3, an electric field can be applied to the core waveguide of the i-core layer 103 even when the n-pn type is stacked in order from the upper layer. Obviously, the usefulness of the invention is not lost. Further, as a configuration in which the p-electron barrier layer 109 is provided between the i-core layer 103 and the p-cladding layer 105 of the polarization control element shown in FIG. It is good also as an element. The same applies to the polarization control element according to Example 2 described later.

  The principle of the polarization control element according to the present invention will be described in more detail with reference to FIGS. FIG. 5 shows changes in the refractive index ellipsoid depending on whether or not an electric field is applied in the [110] direction when a waveguide is formed in a predetermined stripe direction on a (110) plane orientation substrate in a zinc blende type crystal. .

  In a zinc blende crystal, when a waveguide is formed in the [−110] stripe direction of a (110) plane substrate, the refractive index ellipsoid of the waveguide cross section is as shown in FIG. As shown in FIG. 5A, it is known that a zinc blende type crystal is an isotropic crystal unless voltage is applied to the waveguide from the outside.

  On the other hand, it is also known that when an electric field is applied to the waveguide in the [110] direction, the main axis of the refractive index ellipsoid is inclined by 45 degrees and distorted as shown in FIG. That is, by utilizing this principal axis inclination, for example, TE mode light in which the optical electric field is polarized in the horizontal direction of the substrate can be converted into TM mode light in the vertical direction of the substrate on the same principle as the half-wave plate. The following three equations are equations that represent the ratio of polarization conversion to the TM mode with respect to TE polarized light incidence based on the mode coupling theory.

Here, κ is a coupling coefficient determined by the electric field strength, and represents the ease of power transfer between the TE mode and the TM mode. P ₀ is the total optical output power, _PTM is the TM mode optical output power, Δn (TE−TM) is the effective refractive index difference (birefringence) between the TE mode and the TM mode, z is a variable in the propagation direction, alpha is a coefficient determined by the overlap of the optical electric field (1 ≧ α ≧ 0), r is the Pockels constant, n _o is the medium refractive index, E is an electric field intensity is there.

  From the above three formulas, it is understood that Δn (TE−TM) needs to be reduced or κ needs to be increased in order to increase the TM mode conversion efficiency. In order to increase κ, it is effective to increase the electric field intensity E that interacts with light. In order to increase the electric field strength E, it is conceivable to increase the applied voltage. However, since the applied voltage is required to be kept as low as possible in operation, the applied voltage cannot be increased.

  One of the most effective means for increasing the electric field strength E is to form an optical waveguide by a pn junction. Furthermore, in order to increase the electric field strength E, it is desirable to make the electric field drop region as narrow as possible. However, since the waveguide shape is generally an oblong shape with a large aspect ratio, a voltage is applied from the horizontal direction of the substrate. When the voltage is applied from the substrate stacking direction, the voltage drop region is narrower, so that a high κ can be obtained even at a low applied voltage. In this respect, the present invention employs a high-mesa waveguide structure with a pn junction and substrate stacking direction voltage application, and therefore has a superior configuration as a polarization control element.

  On the other hand, in order to reduce Δn (TE-TM), it is straightforward to adjust the aspect ratio of the waveguide core shape. However, as shown in FIG. 6, it is derived from the calculation that the range of the waveguide width in which Δn (TE-TM) is reduced differs depending on the difference in the core / clad relative refractive index. FIG. 6 shows the waveguide width dependence of the birefringence Δn (TE-TM) due to the difference in the refractive index difference of the core / cladding ratio. As shown in FIG. 6, the allowable processing error is larger when the core / cladding relative refractive index difference is 2.5% Δ than when 4% Δ.

  Here, when the polarization control element is actually manufactured, it is assumed that the waveguide width varies due to a manufacturing error. For this reason, when an element having a smaller influence of the manufacturing error is to be manufactured, it is necessary to form a waveguide having a small core / clad relative refractive index difference. For example, from the above formula 3, in order to obtain a conversion efficiency of 85% or more at an element length of 3 mm (z = 3 mm), Δn (TE-TM) needs to be 0.0001 or less.

  When the core / clad relative refractive index difference is set to 2.5% Δ, as shown in FIG. 6, a processing error of ± 0.1 μm or more is allowed when Δn (TE-TM) is 0.0001 or less. Therefore, it is shown that a sufficiently high yield can be produced by making full use of the stepper exposure accuracy (± 0.05 μm). When the core / cladding relative refractive index difference is set to 4% Δ, a processing error of ± 0.05 μm is allowed when Δn (TE-TM) is 0.0001 or less as shown in FIG. This processing error is the limit of the stepper exposure accuracy (± 0.05 μm).

  FIG. 7 is a plot of the dependence of the fabrication tolerance estimated from FIG. 6 on the difference in waveguide refractive index and the dependence of the optical confinement factor on the waveguide refractive index. As shown in FIG. 7, it can be seen that there is a trade-off relationship between the two. From the viewpoint of high modulation efficiency (interaction between light and electricity), the optical confinement factor is desirably 80% or more, and from the viewpoint of manufacturing tolerance (stepper exposure accuracy limit), the manufacturing tolerance of the waveguide width should be 50 nm or more. Is desirable. That is, as shown in FIG. 7, the refractive index difference Δ between the material of the i-core layer 103 and the material of the lower cladding layer and the upper cladding layer is 4% or less, more specifically in the range of 2 to 4%. A device with high efficiency and wide manufacturing tolerance can be manufactured.

(Example 2)
FIG. 8 is a diagram showing an electrode inversion type polarization control element configured by arranging polarization control elements in series. When Δn (TE-TM) is generated due to the slight anisotropy of the waveguide structure, the polarization conversion efficiency may be reduced. Further, it is not easy in terms of manufacturing technology to make Δn (TE−TM) completely zero. When it is desired to convert the polarization with higher efficiency by suppressing such a decrease in polarization conversion efficiency, the polarization control elements are arranged in series like the electrode inversion polarization control element shown in FIG. Then, a voltage whose sign is inverted may be applied to each adjacent polarization control element via each signal electrode. In this case, since it is necessary to apply a positive voltage or a negative voltage to each signal electrode, a structure in which a voltage drop occurs in the core layer (no current flows) regardless of the polarity of the voltage is required.

  FIG. 4 is a cross-sectional view of the polarization control element according to the second embodiment of the present invention. FIG. 4 shows a substrate 201 made of a substrate crystal with a (110) orientation, a p-cladding layer 202 laminated on the substrate 201, and a p-lower electron barrier laminated on the p-cladding layer 202. Layer 208, i-core layer 203 stacked on p-lower electron barrier layer 208, two ground electrodes 204 formed on both sides of i-core layer 203 on p-cladding layer 202, i A p-upper electron barrier layer 209 stacked on the core layer 203, an n-cladding layer 205 stacked on the p-upper electron barrier layer 209, and an n-contact stacked on the n-cladding layer 205. A polarization control element comprising a layer 206 and a signal electrode 207 formed on the n-contact layer 206 is shown.

  In the polarization control element according to the second embodiment shown in FIG. 4, as in the polarization control element according to the first embodiment, the zinc blende in the [−110] stripe direction on the substrate 201 having the (110) plane orientation. A high mesa waveguide structure composed of a type semiconductor crystal is formed, and an electric field is applied from the signal electrode 207 in the [110] direction perpendicular to the substrate. Further, the high mesa waveguide structure of the polarization control element according to the second embodiment has an n-p-i- structure in which the p-lower electron barrier layer 208 and the p-upper electron barrier layer 209 are provided above and below the i-core layer 203. It has a pn structure.

  In the polarization control element according to the second embodiment, in order to cause a voltage drop in the i-core layer 203, a p-lower electron barrier layer 208 and a p-upper electron barrier layer 209 are provided above and below the i-core layer 203, respectively. Provided. Accordingly, in the electrode inversion type polarization control element as shown in FIG. 8, it is possible to apply a positive voltage or a negative voltage whose signs are inverted from each other to the adjacent polarization control element.

In addition, since it is necessary to electrically separate each polarization control element, a semi-insulating or non-doped cladding layer having a high resistivity is removed after etching the upper layer cladding in a region that does not contribute to voltage application between each element. Need to be backfilled with. In this example, as shown in FIG. 8, the elements were electrically separated by a semi-insulating resistance Si-cladding layer 801 using Fe (density 5 × 10 ¹⁶ cm ⁻³ ) as a dopant. The Si-cladding layer 801 can be made of InP, for example.

  FIG. 9 is a diagram showing that conversion efficiency increases due to electrode inversion phase matching. By configuring the electrode inversion type polarization control element shown in FIG. 8, the conversion efficiency can be made 100% by artificially matching the phase between TE and TM as shown in FIG. This quasi-phase matching technique has the same principle as the technique that has been used in the past to improve the efficiency of a directional optical coupler.

  In the polarization control element according to the second embodiment, it is desirable to use the npnppn structure, but the pninip structure may be used. The conductivity types of the layers provided above and below the core layer 203 may be symmetrical with respect to the i-core layer 203.

  As described above, according to the present invention, it is possible to manufacture a small polarization control element with high polarization conversion efficiency and capable of operating in a wide wavelength band with high yield. Further, by forming an n-pnp-n structure capable of applying a positive voltage or a negative voltage whose signs are inverted with respect to adjacent polarization control elements, and arranging the polarization control elements in series, Thus, it is possible to improve the polarization conversion efficiency and to achieve high speed and low driving voltage.

  Here, in the polarization control element according to the first and second embodiments, the substrate material has the same structure as InP, for example, even when GaAs, GaP, ZnS, ZnSe is used, its usefulness does not change. It is clear. In the polarization control element according to the first and second embodiments, the light modulation element corresponding to the 1.55 μm wavelength band is used. However, an element corresponding to the 1.3 μm wavelength band may be used. For example, if a GaAs substrate is used, it can respond to a 0.6 to 1.3 μm wavelength band.

  In the polarization control element according to the first and second embodiments, since there is no problem if the electrode contact layer can secure sufficient conductivity, the semiconductor doped with the p-type impurity is not limited to the InGaAs. For example, InGaAsP Etc. may be used.

  In the polarization control elements according to the first and second embodiments, it is desirable that the signal electrode has a traveling wave type electrode structure. However, for example, a lumped constant type and a resonance type electrode structure may be used.

Substrate 101, 201
n-cladding layer 102, 205
i-core layer 103, 203
Ground electrode 104,204
p-cladding layer 105, 202
p-contact layer 106
n-contact layer 206
Signal electrode 107, 207
p-lower electron barrier layer 208
p-upper electron barrier layer 209
Si-clad layer 801

Claims (3)

(110) plane orientation substrate;
A high-mesa waveguide structure laminated in a [−110] stripe direction on the (110) plane-oriented substrate and composed of a zinc blende type semiconductor crystal;
Lower cladding layer,
A core layer, and an upper cladding layer,
A high mesa waveguide structure in which the core layer is sandwiched between the lower clad layer and the upper clad layer;
A signal electrode formed on the waveguide structure and applying an electric field in a [110] direction to the core layer;
A ground electrode formed on the lower cladding layer;
Equipped with a,
The polarization control element further comprising an electron barrier layer provided between the core layer and the lower clad layer and between the core layer and the upper clad layer .   2. The polarization control element according to claim 1, wherein a difference in refractive index between the material of the core layer and the material of the lower cladding layer and the upper cladding layer is 4% or less. element. At least two or more polarization control elements according to claim 1 or 2 are integrated in series on the same straight waveguide,
Each of the electron barrier layers is a p-type layer, and the lower clad layer and the upper clad layer are n-type layers.

Images