Low-loss non-volatile phase shifter based on lateral heating electrode, method and application

Abstract

The invention discloses a low-loss non-volatile phase shifter based on a lateral heating electrode, a method and application. Comprising a substrate and a silicon flat plate, a heating electrode and a hybrid waveguide which are arranged on the substrate, the silicon flat plate is arranged on the upper surface of the substrate, the heating electrode and the hybrid waveguide are both arranged on the upper surface of the silicon flat plate, and the heating electrode and the hybrid waveguide are arranged at an interval; two ends of the hybrid waveguide are respectively used as an input end and an output end of the phase shifter; the hybrid waveguide comprises a silicon waveguide and a phase change material, the silicon waveguide and the phase change material are arranged on the silicon flat plate in the propagation direction, the phase change material is embedded in the silicon waveguide, and a heating electrode is arranged on the side face of the hybrid waveguide. The invention has the advantages of non-volatility, large phase shift range, low insertion loss, strong durability and the like, and can be widely applied to various advanced optical systems.

Description

Technical Field

[0001] The present invention relates to the technical field of phase shifters, and in particular to a low-loss non-volatile phase shifter based on a lateral heating electrode, a method and an application thereof. Background Art

[0002] Neural networks have been increasingly widely used in all aspects of social life and production, and the further development of neural networks requires more powerful computing power support and reduced energy consumption. At present, optical neural networks developed based on optical computing technology have significant advantages over traditional electrical neural networks in terms of computing bandwidth and energy consumption, and are considered to be a solution that can greatly improve computing power and save energy. Phase shifters, as the core components of optical computing technology, are the basic components of optical computing basic units (microring filters, Mach-Zehnder interferometers). The performance of phase shifters is directly related to the performance of optical computing basic units. The development of phase shifters with excellent performance is of great significance to the development of high-performance optical neural networks.

[0003] Silicon has a large thermo-optic coefficient, and the silicon doping process is mature and perfect. Traditional silicon optical phase shifters mainly use thermo-optic effect and carrier dispersion effect mechanisms. Such phase shifters are all volatile phase shifters. On the one hand, they need continuous external energy supply to maintain the phase shift state, which consumes a lot of energy. On the other hand, the effective refractive index of the mode produced by the thermo-optic effect or carrier dispersion effect changes little. Therefore, the size of traditional phase shifters is relatively large, which is not conducive to large-scale integration. Phase change materials have significant refractive index differences between crystalline and amorphous states, and also have a series of nearly continuous intermediate states between crystalline and amorphous states, that is, the refractive index can be adjusted nearly continuously within a range, and its own state does not require external energy to maintain. Therefore, phase shifters made of phase change materials have significant advantages over traditional phase shifters in terms of large-scale integration and energy consumption.

[0004] The phase change material part of the existing phase change material phase shifter with transparent conductive material as the heating electrode is basically in direct contact with the heating electrode part. For example, CN116819844A discloses an MZI structure photonic device based on phase change material and transparent conductive material ITO electrode, in which the phase shifter part electrode is in direct contact with the phase change material. When the heating method of direct electrode contact is adopted, the heat generated by the heating electrode can be directly transferred to the phase change material, and the heating efficiency is high. However, the heating electrode is often absorptive to light. Even for the transparent conductive material ITO, the imaginary part of the refractive index of In2O3 is not completely 0, and it will still absorb a small part of the light. Usually, the absorption loss introduced by the direct contact heating electrode is above 0.2dB. Neural networks often have a multi-layer structure, and the loss introduced by a single device will be superimposed layer by layer. If the superimposed loss is too high, the light energy in the system will not reach the threshold required for nonlinear calculations, and the calculation accuracy of optical calculations will be reduced, making it difficult to realize large-scale neural networks. Non-volatile devices based on phase change materials need to have a high enough durability to support a sufficient number of reconfiguration times. For commercial use, the number of reconfiguration times should reach 100,000 times or even 1 million times. For the heating method in which the heating electrode directly contacts the phase change material, due to the different electronegativity of different elements in the phase change material, the direct contact between the current area and the phase change material will cause electromigration of the phase change material, that is, the negatively charged elements migrate to the anode and the positively charged elements migrate to the cathode, resulting in element segregation and damage to the phase change material. In addition, when the heating electrode is in direct contact, the temperature at the center of the electrode area is often significantly higher than the surrounding temperature. The uneven temperature can easily cause the phase change material to undergo thermophoresis and deform, destroying the morphology of the phase change material film, damaging the phase change material and reducing the working performance of the device. In addition, the phase change material needs to undergo a process of melting from solid to liquid and then rapidly cooling to solid during the amorphization process. During this process, stress will be generated externally, which can easily damage the heating electrode and reduce the service life of the phase shifter. Summary of the invention

[0005] In view of the defects and improvement needs of the existing phase change material phase shifter using transparent conductive material as heating electrode, the present invention proposes a low-loss non-volatile phase shifter based on lateral heating electrode, aiming to reduce the loss of the phase shifter and improve the durability.

[0006] The present invention has the advantages of non-volatility, large phase shift range, low insertion loss, strong durability, etc., and can be widely used in various advanced optical systems.

[0007] In order to solve the above problems, the solution of the present invention is as follows:

[0008] 1. A low-loss non-volatile phase shifter based on lateral heating electrodes:

[0009] It comprises a substrate, a silicon plate, a heating electrode and a hybrid waveguide arranged on the substrate, the silicon plate is arranged on the upper surface of the substrate, the heating electrode and the hybrid waveguide are arranged on the upper surface of the silicon plate, and the heating electrode and the hybrid waveguide are arranged at a distance; two ends of the hybrid waveguide serve as an input end and an output end of a phase shifter respectively;

[0010] The hybrid waveguide includes a phase change material and a silicon waveguide. The silicon waveguide and the phase change material are arranged on a silicon plate along the propagation direction. The phase change material is embedded in the silicon waveguide. A heating electrode is arranged on the silicon plate on the side of the hybrid waveguide. The heating electrode is also located on the side / side of the phase change material.

[0011] For the silicon plate below the heating electrode, only the silicon plate near the heating electrode is retained by etching; for the silicon plate below the hybrid waveguide, only the silicon plate near the hybrid waveguide and within 200nm-500nm from the side wall of the silicon waveguide is retained by etching; the silicon plate below the heating electrode and the silicon plate below the hybrid waveguide are connected as a whole.

[0012] The phase shifter further comprises an oxidized protective upper cladding layer, which covers the outside of the silicon plate, the heating electrode and the hybrid waveguide as a whole and the top of the substrate.

[0013] The heating electrode is one of a metal electrode, a metal compound electrode, a transparent conductive material electrode, and a doped silicon electrode; the phase change material is a metal oxide, a sulfur compound, or an organic phase change material; the oxidation protection upper cladding is an oxide or a nitride, including aluminum oxide (Al2O3), silicon oxide (SiO2), and aluminum nitride (AlN).

[0014] The phase change material has an amorphous state, a crystalline state, and a series of intermediate crystalline states between the amorphous state and the crystalline state within the working wavelength range. Different crystalline states have different refractive indices. These states are self-stable. Joule heat is generated by applying voltage to the heating electrode. The heat is conducted to the phase change material of the hybrid waveguide through the silicon plate, and the refractive index of the phase change material is regulated, thereby changing the propagation constant of the light wave mode in the hybrid waveguide and adjusting the phase of the light wave mode passing through the hybrid waveguide.

[0015] The substrate comprises a silicon substrate layer and a silicon dioxide layer arranged on the silicon substrate layer, and the silicon plate is arranged on the silicon dioxide layer.

[0016] The thickness of the heating electrode is 50nm-300nm;

[0017] The spacing distance between the heating electrode and the hybrid waveguide is 100nm-500nm.

[0018] The spacing distance between the heating electrode and the hybrid waveguide at each location along the propagation direction of the hybrid waveguide may be a constant value, or may vary along the propagation direction of the hybrid waveguide.

[0019] The temperature uniformity of heating the phase change material can be improved by regulating the width of the heating electrode along the voltage drop direction and thereby regulating the heating power of different regions of the heating electrode.

[0020] 2. Control method applied to the phase shifter:

[0021] The phase change material has a lower refractive index n_a in the amorphous state and a higher refractive index n_c in the crystalline state. Since the phase change material has a series of intermediate states in addition to the amorphous state and the crystalline state, the refractive index of the phase change material is any value in the interval [n_a, n_c].

[0022] The initial refractive index of the phase change material is n_1, which is regulated as follows:

[0023] When it is necessary to adjust the refractive index of the phase change material to become higher, a lower voltage is applied to the heating electrode to generate Joule heat to heat the phase change material to a state higher than the crystallization temperature T_c and lower than the melting temperature T_m, and the phase change material changes to a crystalline state, and the refractive index becomes higher; after a period of time, if the heating voltage is removed and heating is stopped, the refractive index of the phase change material reaches a refractive index n_2 higher than the initial refractive index n_1 after cooling, that is, n_2>n_1, and the refractive index n_2 is maintained until the next adjustment voltage appears; if heating is continued, the phase change material continues to change to the crystalline state until it is completely transformed into the crystalline state, until the refractive index reaches the maximum value n_c, at which time the heating voltage is removed and heating is stopped, and the phase change material reaches the refractive index n_c in the crystalline state after cooling, and the refractive index n_c in the crystalline state is maintained until the next adjustment voltage is applied;

[0024] When it is necessary to adjust the refractive index of the phase change material to decrease, a higher voltage is applied to the heating electrode to generate Joule heat to heat the phase change material to a state higher than the melting temperature T_m, so that the phase change material melts from solid to liquid. At this time, the heating voltage is immediately removed, the phase change material cools down, and because silicon has a higher thermal conductivity, the silicon plate has strong thermal conductivity and heat dissipation capabilities, and the silicon plate below the heating electrode away from the hybrid waveguide has been removed by etching and has a higher directional heating efficiency. At the same time, the heating electrode has a thinner thickness and less heat storage, and the heating electrode is close to the hybrid waveguide and the heat transfer is rapid. Therefore, the phase shifter system has a lower thermal time constant, which will allow the phase change material to cool down at a cooling rate that reaches or exceeds the critical cooling rate of 1K / ns, and transform from liquid to solid and then to amorphous. At this time, the refractive index of the phase change material reaches the refractive index n_a of the amorphous state, and maintains the refractive index n_a of the amorphous state until the next application of the adjustment voltage.

[0025] The above-mentioned higher voltage and lower voltage are relative, and the higher voltage is higher than the lower voltage.

[0026] The phase shifter system also has a heating scheme for regulating the refractive index of the phase change material to become higher and lower, and thus has the ability to repeatedly reconstruct the refractive index of the phase change material. The phase shifter system can advance or lag the phase of the light wave mode passing through the hybrid waveguide.

[0027] 3. A filter using the phase shifter:

[0028] The hybrid waveguide of the phase shifter is a section of arc waveguide arranged on the micro-ring waveguide, the heating electrode is a circular arc electrode placed next to the hybrid waveguide, the micro-ring waveguide is placed next to the bus waveguide, the bus waveguide is a straight waveguide, the bus waveguide is also placed on the silicon plate, there is a gap between the bus waveguide and the micro-ring waveguide, the gap is determined by the critical coupling condition, specifically, the micro-ring waveguide can be placed between the bus waveguide and the heating electrode. Only the silicon plate within 200nm-500nm around the bottom of the micro-ring waveguide, the silicon plate within 200nm-500nm around the bottom of the heating electrode, and the silicon plate within 200nm-500nm from the side wall of the silicon waveguide are retained by etching; the two ends of the bus waveguide are used as the input end and the output end of the filter respectively.

[0029] The micro-ring waveguide, hybrid waveguide and bus waveguide form a ring resonator system, whose filtering characteristics are controlled by the refractive index of the phase change material. The refractive index of the phase change material is regulated by the heating electrode to adjust the filtering characteristics of the resonator and change the transmittance of each wavelength signal light in the bus waveguide.

[0030] The light passing through the bus waveguide is coupled into the micro-ring waveguide and then passes through the phase shifter for phase shifting, thereby adjusting the filtering wavelength of the filter.

[0031] The heating electrode of the present invention is located on the side of the hybrid waveguide. The heat generated by the heating electrode is transferred from the side to the hybrid waveguide through the silicon plate, thereby heating the phase change material to change its state. It should be noted that, because among the current mainstream waveguide platforms, only silicon material has a high thermal conductivity (131 [W / (m*K)]), it can quickly transfer the heat generated by the side electrode to the phase change material, while other waveguide platforms cannot achieve rapid heat transfer due to low thermal conductivity. Therefore, in the scheme of the present invention, the plate on the substrate (1) must be made of silicon material. In this scheme, the heating electrode does not directly contact the hybrid waveguide, so the loss caused by the heating electrode absorbing light is almost zero.

[0032] The phase shifter technical solution provided by the present invention may have the following beneficial effects:

[0033] 1. Compared with the existing phase-change material phase shifter using transparent conductive material as the heating electrode, the heating electrode of the phase shifter of the present invention does not produce additional absorption loss. When it is constructed into a Mach-Zehnder interferometer, its insertion loss is smaller; when it is constructed into a microring filter, a higher quality factor can be achieved. Using the Mach-Zehnder interferometer and / or microring filter to construct an optical neural network can reduce the loss of each layer in the network, thereby realizing a larger-scale neural network.

[0034] 2. Compared with the existing phase change material phase shifter using transparent conductive material as the heating electrode, the phase shifter of the present invention has high durability, and the electric field applied by the heater does not pass through the phase change material, thus avoiding element segregation. Since the width of the side electrode along the voltage drop direction can be flexibly adjusted, a suitable resistance distribution can be designed, and the heating power of the side electrodes at different positions can be optimized, the temperature distribution of the phase change material can be adjusted so that its maximum temperature difference does not exceed 100°C, thereby improving temperature uniformity and suppressing the occurrence of thermophoresis. In addition, the use of side electrode heating avoids the stress generated during the phase change of the phase change material directly acting on the heating electrode, which can increase the number of reconstruction times of the phase shifter.

[0035] 3. Compared with the traditional phase shifter, the present invention combines the phase shifter with the phase change material, and utilizes the characteristic of the phase change material having a large refractive index difference between the amorphous state and the crystalline state, which can greatly change the propagation constant of the mode in the waveguide. In particular, when the phase change material is embedded in the waveguide, the interaction between the phase change material and the light is further enhanced, which can greatly regulate the waveguide mode. For a straight waveguide with a waveguide width of 600nm and a shallow etching of 130nm, when the embedded phase change material has a height of 70nm and a width of 300nm, the modulation efficiency is 4.6um / π at 1550nm.

[0036] The present invention aims to provide a high-performance phase shifter. As a basic optical component, the phase shifter is widely used in optical devices or systems such as micro-ring filters and MZI interferometers, and has broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a schematic diagram of the three-dimensional structure of the low-loss non-volatile phase shifter based on the lateral heating electrode of the present invention;

[0038] Figure 2 This is a TE0 mode propagation field distribution diagram of a low-loss non-volatile phase shifter based on a lateral heating electrode of the present invention;

[0039] Figure 3 The temperature change and the corresponding heating and cooling rate results of the amorphization process in the embodiment are shown in FIG.

[0040] Figure 4The present invention is a schematic structural diagram of an all-pass micro-ring filter applied to a low-loss non-volatile phase shifter based on a lateral heating electrode.

[0041] Explanation of the reference numerals: substrate 1, silicon plate 2, heating electrode 3, oxidation protection upper cladding layer 4, hybrid waveguide 5, phase change material 5-1, silicon waveguide 5-2, bus waveguide 6, microring waveguide 7; silicon substrate layer 1-1, silicon dioxide layer 1-2. DETAILED DESCRIPTION

[0042] The present invention will be further described below in conjunction with the accompanying drawings and embodiments, but the protection scope of the present invention shall not be limited thereto.

[0043] like Figure 1 As shown, the phase shifter includes a substrate 1 and a silicon plate 2, a heating electrode 3 and a hybrid waveguide 5 arranged on the substrate 1. The silicon plate 2 is arranged on the upper surface of the substrate 1, and the heating electrode 3 and the hybrid waveguide 5 are both arranged on the upper surface of the silicon plate 2. The heating electrode 3 and the hybrid waveguide 5 are arranged at a distance from each other; two ends of the hybrid waveguide 5 serve as an input end and an output end of the phase shifter respectively;

[0044] The hybrid waveguide 5 includes a phase change material 5-1 and a silicon waveguide 5-2. The phase change material 5-1 and the silicon waveguide 5-2 are arranged on the silicon plate 2 along the propagation direction. The phase change material 5-1 is embedded in the silicon waveguide 5-2 and surrounded by the silicon waveguide 5-2. The heating electrode 3 is arranged on the silicon plate 2 on the side of the hybrid waveguide 5.

[0045] For the silicon flat plate 2 below the heating electrode 3, only the silicon flat plate 2 close to the hybrid waveguide 5 within 200nm-500nm around the heating electrode 3 is retained by etching; for the silicon flat plate 2 below the hybrid waveguide 5, only the silicon flat plate 2 below the silicon waveguide 5-2 within 200nm-500nm from the side wall of the silicon waveguide 5-2 is retained by etching. By adjusting the shape of the silicon flat plate 2, the Joule heat generated by the heating electrode 3 is directionally transferred to the phase change material 5-1 instead of diffusing to the surroundings, thereby improving the heating efficiency of the heating electrode 3.

[0046] The heating electrode 3 is one of a metal electrode, a metal compound electrode (TiN), a transparent conductive material electrode (two-dimensional material, ITO, In2O3, FTO), and a doped silicon electrode; the phase change material 5-1 is a metal oxide, a sulfur compound or an organic phase change material.

[0047] The oxidation protection upper cladding layer 4 covers the outside of the silicon plate 2, the heating electrode 3 and the hybrid waveguide 5 as a whole and the top of the substrate 1. The oxidation protection upper cladding layer 4 is an oxide or a nitride, including aluminum oxide Al2O3, silicon oxide SiO2 and aluminum nitride AlN.

[0048] The phase change material 5 - 1 has low optical loss characteristics in the amorphous state and / or crystalline state within the working wavelength range, and the imaginary part of the refractive index of the material is 0.

[0049] The phase change material 5-1 has an amorphous state, a crystalline state and a series of intermediate states within the working wavelength range. Different crystalline states have different refractive indices. These states are self-stable. Joule heat is generated by applying voltage to the heating electrode 3. The heat is conducted to the hybrid waveguide 5 through the silicon plate 2 and then to the phase change material 5-1. The refractive index of the phase change material 5-1 is regulated, thereby changing the propagation constant of the light wave mode in the hybrid waveguide 5 and adjusting the phase of the light wave mode passing through the hybrid waveguide 5. The substrate 1 includes a silicon substrate layer 1-1 and a silicon dioxide layer 1-2 arranged on the silicon substrate layer 1-1, and the silicon plate 2 is arranged on the silicon dioxide layer 1-2.

[0050] The hybrid waveguide 5 is a straight waveguide or a curved waveguide;

[0051] The shape of the heating electrode 3 is a strip or an arc.

[0052] The thickness of the heating electrode 3 is 50nm-300nm;

[0053] The spacing distance between the heating electrode 3 and the hybrid waveguide 5 is 100nm-500nm.

[0054] The width of the heating electrode 3 along the voltage drop direction may be varied to improve the temperature uniformity of heating the phase change material 5 - 1 .

[0055] The phase change material 5-1 has a lower refractive index in the amorphous state, denoted as n a , has a higher refractive index in the crystalline state, denoted as n c Since the phase change material 5-1 has a series of intermediate states in addition to the amorphous state and the crystalline state, the refractive index of the phase change material 5-1 can be in the range [n a ,n c ] is any exact value.

[0056] Note that the initial refractive index of the phase change material is n1. When the phase change material 5-1 needs to be regulated to increase its refractive index, a lower voltage is applied to the heating electrode 3 to generate Joule heat to heat the phase change material 5-1 to a temperature higher than the crystallization temperature T c (For the materials Sb2Se3, T c 200℃) and below the melting temperature T m (For the materials Sb2Se3, T mAt this time, if the heating voltage is removed and the heating is stopped, the refractive index of the phase change material 5-1 after cooling is n2, then n2>n1, and it remains n2 until the next regulation voltage appears, which is non-volatile. If the heating continues, the phase change material 5-1 continues to transform into the crystalline state until it is completely transformed into the crystalline state, and the refractive index reaches the maximum value n c At this time, the heating voltage is removed to stop heating. After cooling, the refractive index of the phase change material 5-1 is n c , and keep it as n c Until the next voltage regulation occurs.

[0057] When the phase change material 5-1 needs to be regulated to reduce its refractive index, a higher voltage is applied to the heating electrode 3 to generate Joule heat to heat the phase change material 5-1 to a temperature higher than the melting temperature T m The phase change material 5-1 melts from solid to liquid. At this time, the heating voltage is quickly removed, and the phase change material 5-1 will cool down quickly. Since silicon has a high thermal conductivity, the silicon plate 2 has strong heat conduction and heat dissipation capabilities, and the silicon plate 2 below the heating electrode 3, which is far away from the hybrid waveguide 5, has been removed by etching and has a high directional heating efficiency. At the same time, the heating electrode 3 has a thinner thickness and less heat storage. The heating electrode 3 is close to the hybrid waveguide 5 and the heat transfer is rapid. Therefore, the phase shifter system has a lower thermal time constant, and the phase change material 5-1 will cool down at a cooling rate that reaches or exceeds the critical cooling rate of 1K / ns, and transforms from liquid to solid and then to amorphous. At this time, the refractive index of the phase change material is n a , and keep it as n a Until the next voltage regulation occurs.

[0058] The phase shifter system also has a heating scheme for regulating the refractive index of the phase change material 5-1 to become higher or lower, and thus has the ability to repeatedly reconstruct the refractive index of the phase change material 5-1. The phase shifter system can advance or lag the phase of the light wave mode passing through the hybrid waveguide 5.

[0059] The electric field applied by the phase shifter heater of the present application does not pass through the phase change material, thus avoiding element segregation. It is also possible to flexibly adjust the width of the side electrode along the voltage drop direction, design a suitable resistance distribution, optimize the heating power of the side electrodes at different positions, adjust the temperature distribution of the phase change material, and make its maximum temperature difference not exceed 100°C, thereby improving temperature uniformity and suppressing the occurrence of thermophoresis. In addition, the use of side electrode heating avoids the stress generated during the phase change process of the phase change material directly acting on the heating electrode, which can increase the number of reconstruction times of the phase shifter.

[0060] In the phase shifter, light is input from one end of the hybrid waveguide 5 and output from the other end of the hybrid waveguide 5. The present invention adjusts the thickness of the heating electrode 3 and the spacing between the heating electrode 3 and the hybrid waveguide 5 so that the light in the hybrid waveguide 5 will not be absorbed by the heating electrode 3 due to direct contact with the hybrid waveguide 5 or through the evanescent field. At the same time, no light or only a very small amount of light in the hybrid waveguide 5 escapes from the hybrid waveguide 5 and enters the heating electrode 3 through mode coupling to be absorbed. Figure 2 As shown, the energy of the light wave mode TE0 is concentrated in the hybrid waveguide 5, and the component on the heating electrode 3 is extremely small, so that the light energy in the hybrid waveguide 5 can be transmitted in the hybrid waveguide 5 with almost no loss, that is, the loss introduced by the heating electrode 3 is almost zero, achieving low loss.

[0061] The embodiments of the present invention are as follows:

[0062] Example 1

[0063] like Figure 1 As shown, the phase shifter includes a substrate 1 and a silicon plate 2, a heating electrode 3, an oxidation protection upper cladding layer 4 and a hybrid waveguide 5 arranged on the substrate 1. The silicon plate 2 is arranged on the upper surface of the substrate 1, and the heating electrode 3 and the hybrid waveguide 5 are both arranged on the upper surface of the silicon plate 2, and the heating electrode 3 and the hybrid waveguide 5 are arranged at a distance from each other; the hybrid waveguide 5 is a straight waveguide, and the two ends of the hybrid waveguide 5 serve as the input end and the output end of the phase shifter respectively; the oxidation protection upper cladding layer 4 is coated on the outside of the silicon plate 2, the heating electrode 3 and the hybrid waveguide 5 as a whole and on the top of the substrate 1.

[0064] The hybrid waveguide 5 includes a phase change material 5-1 and a silicon waveguide 5-2. The phase change material 5-1 and the silicon waveguide 5-2 are arranged on the silicon plate 2 along the propagation direction. The top surfaces of the phase change material 5-1 and the silicon waveguide 5-2 are flush. The phase change material 5-1 is embedded in the silicon waveguide 5-2 and surrounded by the silicon waveguide 5-2. The heating electrode 3 is arranged on the silicon plate 2 on the side of the phase change material 5-1 of the hybrid waveguide 5.

[0065] The heating electrode 3 and the hybrid waveguide 5 are not in direct contact, but are arranged on the silicon plate 2 with a gap.

[0066] In a specific implementation, the phase change material 5 - 1 is embedded in the silicon waveguide 5 - 2 , and the oxidation protection upper cladding layer 4 covers the phase change material 5 - 1 and the silicon waveguide 5 - 2 .

[0067] In this embodiment, the thickness of the silicon substrate 1-1 is 500um, and the thickness of the bottom silicon dioxide 1-2 is 3um.

[0068] In this embodiment, in order to ensure a fast lateral heat transfer speed, the heat generated by the electrode can be quickly transferred to the hybrid waveguide through the silicon plate 2 and then to the phase change material 5-1. The thickness of the silicon plate 2 is selected to be 90nm, and in order to improve the heating efficiency of the heating electrode 3, for the silicon plate 2 below the heating electrode 3, only the silicon plate 2 close to the hybrid waveguide 5 is retained by etching; for the silicon plate 2 below the hybrid waveguide 5, only the silicon plate 2 within 200nm from the side wall of the silicon waveguide 5-2 is retained by etching, so that the Joule heat generated by the heating electrode 3 is directionally transferred to the phase change material 5-1 instead of diffusing to the surroundings, thereby improving the heating efficiency of the heating electrode 3.

[0069] In this embodiment, the phase change material is Sb2Se3, which has an imaginary part of 0 refractive index in both the crystalline and amorphous states within the 1500nm-1600nm operating band. The phase change material itself does not absorb light, ensuring that the loss of the phase shifter is sufficiently low. The oxidation protection upper cladding material is SiO2, which is compatible with the cladding process of traditional devices.

[0070] In this embodiment, in the hybrid waveguide 5, the silicon waveguide 5-2 on the silicon plate 2 has a thickness of 130nm and a width of 600nm. In order to make the phase shifter have a strong phase shifting capability and to avoid the introduction of mode mismatch loss due to the large difference between the effective refractive index at the phase change material and the effective refractive index in the silicon waveguide, the width of the phase change material 5-1 is selected to be 300nm, the length is 4um, and the thickness is selected to be 70nm. The upper surface of the phase change material is flush with the silicon waveguide and embedded in the hybrid waveguide.

[0071] In this embodiment, to ensure that the heating electrode 3 has a sufficiently low absorption of light, the material of the heating electrode 3 is selected to be In2O3. To make the resistance of the heating electrode 3 low enough to reduce the driving voltage, the thickness of the heating electrode 3 is selected to be 200nm and the width is 800nm. In order to take into account that the absorption of light by the heating electrode 3 is sufficiently low and the heat generated by the heating electrode can be quickly transferred to the phase change material 5-1, so that during the amorphization process, after the phase change material 5-1 is heated to the melting temperature, the heating signal is removed and the phase change material 5-1 can quickly cool down at or above the critical cooling rate (1K / ns) and enter the amorphous state, the spacing between the heating electrode 3 and the hybrid waveguide 5 is selected to be 200nm. At this time, the mode field distribution of the phase shifter is as follows: Figure 2 As shown in Figure 1, it can be found that at this time, the light field energy is almost completely concentrated in the hybrid waveguide, the light field component on the side heating electrode is extremely small, and the side electrode absorbs almost no light. When an 8V, 350ns heating pulse is applied to the heating electrode 3, the heating process is as follows Figure 3As shown, the phase change material 5-1 is heated to nearly 700°C, exceeds the melting temperature and turns into a liquid, and then cools down to a solid at a critical cooling rate (1K / ns) or above, to ensure that amorphization can be achieved. During the crystallization process, as long as the phase change material 5-1 is heated to be kept above the crystallization temperature and below the melting temperature for a period of time, there is no requirement for the cooling speed, so any heating electrode design that can achieve amorphization can achieve crystallization, thereby achieving repeated phase changes of the phase change material 5-1.

[0072] The phase shifter of the present invention uses the lossless phase change material Sb2Se3 embedded in the waveguide, which effectively strengthens the interaction between the phase change material and light. It has a small volume and strong phase shifting ability. Since the heating electrode is not in direct contact with the hybrid waveguide, the loss introduced by the heating electrode is almost zero, and the occurrence of element segregation is avoided. The stress generated during the phase change of the phase change material does not directly act on the heating electrode, and the device has high durability. The state of the phase change material does not require external conditions to maintain, is non-volatile, and reduces energy consumption. Specific embodiment 2

[0074] like Figure 4 As shown, an all-pass microring filter, which is applied with the low-loss non-volatile phase shifter based on the lateral heating electrode of the present invention, includes a substrate 1 and a silicon plate 2, a heating electrode 3, a bus waveguide 6 and a microring waveguide 7 arranged on the substrate 1, the silicon plate 2 is arranged on the upper surface of the substrate 1, and the heating electrode 3, the bus waveguide 6 and the microring waveguide 7 are all arranged on the upper surface of the silicon plate 2.

[0075] The micro-ring waveguide 7 is placed beside the bus waveguide 6. The bus waveguide 7 is a straight waveguide. The gap between the bus waveguide 6 and the micro-ring waveguide 7 is determined according to the critical coupling condition. The two ends of the bus waveguide 6 serve as the input end and the output end of the filter respectively.

[0076] The two ends of the bus waveguide 6 can also be connected to the input waveguide and the output waveguide respectively. The output end of the input waveguide is connected to the input end of the output waveguide after passing through the bus waveguide 6. The phase shifter in the microring waveguide 7 is placed at a position that does not affect the coupling.

[0077] In the above-mentioned micro-ring waveguide, the hybrid waveguide 5 is a section of arc-shaped waveguide arranged on the micro-ring waveguide 7, the heating electrode 3 is a circular arc-shaped electrode placed beside the hybrid waveguide 5, and is located outside the micro-ring waveguide 7, and the heating electrode 3 and the hybrid waveguide 5 are arranged at a distance. The hybrid waveguide 5 includes a phase change material 5-1 and a silicon waveguide 5-2, the phase change material 5-1 is arc-shaped, the phase change material 5-1 is embedded in the silicon waveguide 5-2 and surrounded by the silicon waveguide 5-2, and the heating electrode 3 is arranged on the silicon plate 2 on the side of the phase change material 5-1 hybrid waveguide 5.

[0078] In this embodiment, the radius of the micro-ring waveguide 7 is 5 um, the width is 600 nm, and the height is 130 nm. The waveguide height of the bus waveguide 6 is 130 nm, the width is 610 nm, and the interval between the bus waveguide 6 and the micro-ring waveguide 7 is 200 nm.

[0079] The micro-ring waveguide 7, the hybrid waveguide 5 and the bus waveguide 6 form a ring resonator system, whose filtering characteristics are controlled by the refractive index of the phase change material 5-1. The refractive index of the phase change material 5-1 is regulated by the heating electrode 3, thereby changing the propagation constant of the light wave mode in the hybrid waveguide 5, adjusting the circular optical path of the micro-ring resonant cavity, changing the resonant wavelength, and then adjusting the filtering characteristics of the resonator, and changing the transmittance of each wavelength signal light in the bus waveguide 6. The micro-ring filter system is wrapped by an oxidation protection upper cladding 4 to prevent oxidation of the phase change material 5-1 and the electrode 3.

[0080] The application value of the present invention is that the phase shifter has the characteristics of small size, strong phase shifting ability, low loss, high durability and low energy consumption. It can be used as a basic component in many popular fields such as reconfigurable photonic devices, large-scale optical switch arrays, large-scale optical neural networks, etc.

[0081] The above embodiments are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A low-loss non-volatile phase shifter based on a lateral heating electrode, characterized in that: The invention comprises a substrate (1), a silicon plate (2), a heating electrode (3) and a hybrid waveguide (5) arranged on the substrate (1), wherein the silicon plate (2) is arranged on the upper surface of the substrate (1), the heating electrode (3) and the hybrid waveguide (5) are both arranged on the upper surface of the silicon plate (2), and the heating electrode (3) and the hybrid waveguide (5) are arranged at a distance from each other; and the two ends of the hybrid waveguide (5) serve as the input end and the output end of the phase shifter respectively; The hybrid waveguide (5) comprises a phase change material (5-1) and a silicon waveguide (5-2); the silicon waveguide (5-2) and the phase change material (5-1) are both arranged on a silicon flat plate (2) along a propagation direction; the phase change material (5-1) is embedded in the silicon waveguide (5-2); and a heating electrode (3) is arranged on the silicon flat plate (2) on the side of the hybrid waveguide (5).

2. A low-loss non-volatile phase shifter based on a lateral heating electrode as claimed in claim 1, characterized in that: For the silicon plate (2) below the heating electrode (3), only the silicon plate (2) around the heating electrode (3) is retained by etching; for the silicon plate (2) below the hybrid waveguide (5), only the silicon plate (2) around the hybrid waveguide (5) and below the silicon waveguide (5-2) within 200nm-500nm from the side wall of the silicon waveguide (5-2) is retained by etching; the silicon plate (2) below the heating electrode (3) and the silicon plate (2) below the hybrid waveguide (5) are connected.

3. A low-loss non-volatile phase shifter based on a lateral heating electrode as claimed in claim 1, characterized in that: The phase shifter also includes an oxidized protective upper cladding layer (4), which covers the outside of the silicon plate (2), the heating electrode (3) and the hybrid waveguide (5) as a whole and the top of the substrate (1).

4. A low-loss non-volatile phase shifter based on a lateral heating electrode as claimed in claim 1, characterized in that: The heating electrode (3) is one of a metal electrode, a metal compound electrode, a transparent conductive material electrode, and a doped silicon electrode; The phase change material (5-1) is a metal oxide, a sulfur compound or an organic phase change material; The oxidation protection upper cladding layer (4) is an oxide or a nitride, including aluminum oxide (Al2O3), silicon oxide (SiO2) and aluminum nitride (AlN).

5. A low-loss non-volatile phase shifter based on a lateral heating electrode as claimed in claim 1, characterized in that: The phase change material (5-1) has an amorphous state, a crystalline state, and a series of intermediate crystalline states between the amorphous state and the crystalline state within the working wavelength range, and different crystalline states have different refractive indices. Joule heat is generated by applying a voltage to the heating electrode (3), and the heat is conducted to the phase change material (5-1) of the hybrid waveguide (5) through the silicon plate (2), thereby adjusting the refractive index of the phase change material (5-1).

6. A low-loss non-volatile phase shifter based on a lateral heating electrode as claimed in claim 1, characterized in that: The substrate (1) comprises a silicon substrate layer (1-1) and a silicon dioxide layer (1-2) arranged on the silicon substrate layer (1-1), and a silicon flat plate (2) is arranged on the silicon dioxide layer (1-2).

7. A low-loss non-volatile phase shifter based on a lateral heating electrode as claimed in claim 1, characterized in that: The thickness of the heating electrode (3) is 50nm-300nm; The spacing distance between the heating electrode (3) and the hybrid waveguide (5) is 100nm-500nm.

8. The control method for the phase shifter according to any one of claims 1 to 7, characterized in that: The initial refractive index of the phase change material is n1, which is regulated as follows: When the refractive index of the phase change material (5-1) needs to be adjusted to be higher, a lower voltage is applied to the heating electrode (3) to generate Joule heat to heat the phase change material (5-1) to a temperature higher than the crystallization temperature T c Below the melting temperature T m The phase change material (5-1) changes to a crystalline state, and the refractive index becomes higher; at this time, if the heating voltage is removed and the heating is stopped, the refractive index of the phase change material (5-1) reaches a refractive index n2 higher than the initial refractive index n1 after cooling, and the refractive index n is maintained until the next regulation voltage appears; if the heating is continued, the phase change material (5-1) continues to change to a crystalline state until it is completely transformed into a crystalline state, until the refractive index reaches a maximum value n c At this time, the heating voltage is removed to stop heating, and after cooling, the phase change material (5-1) reaches the refractive index n in the crystalline state. c , and maintain the refractive index n in the crystalline state c Until the next application of the regulated voltage; When the refractive index of the phase change material (5-1) needs to be lowered, a higher voltage is applied to the heating electrode (3) to generate Joule heat to heat the phase change material (5-1) to a temperature higher than the melting temperature T m state, so that the phase change material (5-1) melts from solid to liquid. At this time, the heating voltage is immediately removed, and the phase change material (5-1) cools down, which will make the phase change material (5-1) cool down at a cooling rate that reaches or exceeds the critical cooling rate of 1K / ns, and transforms from liquid to solid and then to amorphous state. At this time, the refractive index of the phase change material reaches the refractive index n of the amorphous state. a , and maintain the refractive index n in the amorphous state a Until the next application of control voltage.

9. A filter using the phase shifter according to any one of claims 1 to 7, characterized in that: The hybrid waveguide (5) of the phase shifter is a section of arc waveguide arranged on a microring waveguide (7); the heating electrode (3) is a circular arc electrode arranged next to the hybrid waveguide (5); the microring waveguide (7) is arranged next to a bus waveguide (6); the bus waveguide (6) is also arranged on a silicon flat plate (2); a gap is provided between the bus waveguide (6) and the microring waveguide (7); only the silicon flat plate (2) within 200nm-500nm around the bottom of the microring waveguide (7), the silicon flat plate (2) within 200nm-500nm around the bottom of the heating electrode (3), and the silicon flat plate (2) within 200nm-500nm from the side wall of the silicon waveguide (5-2) are retained by etching; the two ends of the bus waveguide (6) serve as the input end and the output end of the filter respectively.

10. The filter according to claim 9, characterized in that: The micro-ring waveguide (7), the hybrid waveguide (5) and the bus waveguide (6) form a ring resonator system, the filtering characteristics of which are controlled by the refractive index of the phase change material (5-1). The refractive index of the phase change material (5-1) is regulated by the heating electrode (3) to adjust the filtering characteristics of the resonator and change the transmittance of signal lights of various wavelengths in the bus waveguide (6).

Images