A nonvolatile phase shifter based on phase change material

Abstract

The present invention discloses a non-volatile phase shifter based on phase change material, comprising a substrate, a first silicon waveguide fixedly disposed on the substrate, and a hybrid waveguide, wherein the hybrid waveguide comprises a second silicon waveguide and a Ge2Sb2Te5 layer. The second silicon waveguide is fixedly disposed on the substrate, spaced apart from the first silicon waveguide, and the length direction of the second silicon waveguide is parallel to the length direction of the first silicon waveguide. The Ge2Sb2Te5 layer is fixedly disposed on the second silicon waveguide layer, and the length direction of the Ge2Sb2Te5 layer is parallel to the length direction of the second silicon waveguide. The present invention has the following advantages: the phase shifter of the present invention has the characteristics of non-volatility, small size, wide operating bandwidth, low insertion loss and crosstalk, and can be constructed into a Mach-Zehnder interferometer 2x2 optical switch with low switch insertion loss and crosstalk.

Description

Technical Field

[0001] The present invention relates to the technical field of phase shifters, and in particular to a non-volatile phase shifter based on phase change materials. Background Art

[0002] As people's requirements for information processing speed, data transmission rate, data storage capacity, etc. continue to increase, information networks are growing at an explosive rate. In order to meet the needs of large-capacity, ultra-high-speed information interconnection, superior electro-optical signal processing devices are often required. Phase shifters, as the core components for information modulation and transmission in optical communication transmission, have broad development and application prospects and have aroused great interest among scientific researchers.

[0003] Phase shifts in optical switches in silicon photonics are primarily achieved through free-carrier injection and thermo-optic effects. These effects lead to small changes in the refractive index, which in turn increases the device length required to achieve the desired phase transition. Another option is to use resonant structures to achieve devices with smaller footprints, but at the expense of low bandwidth and high sensitivity. Compact hybrid plasmonic-photonic switches based on three-waveguide directional couplers have also been reported, but the associated insertion loss is high.

[0004] Since its discovery, the GST (Ge2Sb2Te5) phase-change material has found widespread application in electronics, physics, materials science, and other fields. It has achieved particularly successful commercialization in optical storage. GST possesses excellent optical and electrical properties. Its significant difference in properties between its crystalline and amorphous states, nanosecond-scale phase transition speeds, and its ability to maintain phase stability without the need for additional energy make it an ideal candidate for new electro-optical modulators. However, no research reports on low-loss, non-volatile phase shifters based on phase-change materials have been published, either domestically or internationally. Summary of the Invention

[0005] The object of the present invention is to provide a non-volatile phase shifter based on phase change material with small size for easy on-chip integration, low energy consumption, wide operating bandwidth and low insertion loss.

[0006] To address the above-mentioned problems, the present invention provides a non-volatile phase shifter based on phase change material, comprising a substrate, a first silicon waveguide fixedly disposed on the substrate, and a hybrid waveguide. The hybrid waveguide comprises a second silicon waveguide and a Ge2Sb2Te5 layer. The second silicon waveguide is fixedly disposed on the substrate, spaced apart from the first silicon waveguide, and the length direction of the second silicon waveguide is parallel to the length direction of the first silicon waveguide. The Ge2Sb2Te5 layer is fixedly disposed on the second silicon waveguide layer, and the length direction of the Ge2Sb2Te5 layer is parallel to the length direction of the second silicon waveguide.

[0007] Preferably, the first silicon waveguide has a height range of 200-240nm and a width range of 400-500nm, the second silicon waveguide has a height equal to that of the first silicon waveguide, a width range of 350-400nm, and a length range of 25-30μm, the Ge2Sb2Te5 layer has a length equal to that of the second silicon waveguide, a width range of 250-300nm, and a height range of 15-25nm, and the distance between the second silicon waveguide and the first silicon waveguide is 100-200nm. To ensure efficient transmission of light in the silicon waveguide under TE mode, the silicon waveguide width is generally selected to be 400nm-500nm, and the waveguide thickness is generally set to be 200-240nm. Based on the phase matching conditions, simulation can calculate the dimensions of the hybrid waveguide when the effective refractive index of the first silicon waveguide and the hybrid waveguide are similar. Due to the limiting effect of the Ge2Sb2Te5 in the hybrid waveguide, the hybrid waveguide width is generally smaller than the first silicon waveguide width. Therefore, the dimensions of the first silicon waveguide and the hybrid waveguide are not fixed and can be adjusted appropriately to meet the waveguide width constraints on light and phase matching conditions.

[0008] Preferably, the substrate includes a silicon substrate layer and a silicon dioxide layer fixedly disposed on the silicon substrate layer, and the first silicon waveguide and the hybrid waveguide are fixedly disposed on the silicon dioxide layer.

[0009] Preferably, the height of the silicon substrate layer is in the range of 200-300 μm, and the height of the silicon dioxide layer is in the range of 2-4 μm.

[0010] Preferably, when the Ge2Sb2Te5 layer in the phase shifter is in a crystalline state, the phase difference of π radians is: in, represents the effective refractive index of the odd supermode when the Ge2Sb2Te5 layer is in the crystalline state, represents the effective refractive index of the Ge2Sb2Te5 layer in the even supermode in the amorphous state, represents the effective refractive index of the odd supermode of the Ge2Sb2Te5 layer in the amorphous state, and π represents the π radian phase difference between the odd supermode of the waveguide when the Ge2Sb2Te5 layer is in the crystalline state and the odd supermode and even supermode in the amorphous state.

[0011] Preferably, when the Ge2Sb2Te5 layer in the phase shifter is in an amorphous state, the length calculation formula of the hybrid waveguide is: Where Lp represents the length of the hybrid waveguide, and λ represents the wavelength of the incident light.

[0012] Another object of the present invention is to provide a Mach-Zehnder interferometer optical switch, wherein the Mach-Zehnder interferometer optical switch uses the above-mentioned non-volatile phase shifter based on phase change material.

[0013] Preferably, the Mach-Zehnder interferometer optical switch includes two input waveguides, two output waveguides, and a silicon waveguide arm fixedly arranged on the substrate, and the two input waveguides are fixedly arranged on one side of the upper surface of the substrate, and the two output waveguides are fixedly arranged on the other side of the upper surface of the substrate. The output ends of the two input waveguides are coupled and connected through a first directional coupler, and the input ends of the two output waveguides are coupled and connected through a second directional coupler. The two ends of the silicon waveguide arm are respectively connected to the first directional coupler and the second directional coupler, and the two ends of the first silicon waveguide are respectively connected to the first directional coupler and the second directional coupler.

[0014] The operating principle of the new phase shifter based on phase-change materials is that the phase-change material Ge2Sb2Te5 has significantly different refractive indices and optical absorption coefficients between its amorphous and crystalline states. When the Ge2Sb2Te5 is in the amorphous state, its refractive index and absorption coefficient are low. The fundamental modes of the hybrid waveguide and the first silicon waveguide are phase-matched when the Ge2Sb2Te5 is in the amorphous state, resulting in strong interaction between the two waveguides. Light is coupled from the silicon waveguide to the hybrid waveguide and then back from the hybrid waveguide to the silicon waveguide, resulting in a phase difference of π radians between the input and output light. When the Ge2Sb2Te5 transitions from the amorphous state to the crystalline state, its refractive index and absorption coefficient increase dramatically, causing a phase mismatch between the silicon waveguide and the hybrid waveguide. The two waveguides do not interact, and light enters and exits directly through the silicon waveguide without passing through the hybrid waveguide. The phase of the light remains unchanged in this state. A key feature of the phase shifter is that light passes through the phase shifter with low loss in both states. Therefore, by applying appropriate electric pulse signals to the phase change material Ge2Sb2Te5, Ge2Sb2Te5 can be converted between the crystalline state and the amorphous state, thereby realizing the modulation of the optical signal.

[0015] Preferably, the silicon waveguide arm and the first silicon waveguide have the same length and the same cross-sectional dimensions.

[0016] Preferably, the thickness of the two input waveguides is the same as the thickness of the two output waveguides, and the thickness of the two input waveguides and the thickness of the two output waveguides are in the range of 200-240 nm, and the width of the two input waveguides is the same as the width of the two output waveguides, and the width of the two input waveguides and the width of the two output waveguides are in the range of 400-500 nm.

[0017] Compared with the prior art, the advantages of the present invention are:

[0018] First, the phase shifter of the present invention has the characteristics of non-volatility, small size, wide operating bandwidth, low insertion loss and crosstalk, etc., and is constructed into a Mach-Zehnder interferometer 2x2 optical switch with low switch insertion loss and crosstalk.

[0019] Second, compared with traditional phase shifters, the new phase shifter based on phase change material has a larger contrast in the imaginary part of the refractive index, which greatly reduces the device size, makes the device structure more compact, and facilitates on-chip integration. Since the phase change material is designed to be on one side of the low-loss Mach-Zehnder interferometer arm, regardless of the state of the low-loss Mach-Zehnder interferometer, its loss is very low compared with the traditional low-loss Mach-Zehnder interferometer. In addition, the entire device is simple to manufacture and is compatible with CMOS technology, which is conducive to industrialization.

[0020] Third, the new phase shifter based on phase change material, Ge2Sb2Te5, can ensure phase stability in both crystalline and amorphous states without the need for additional energy supply. Energy is consumed only when the phase change material transitions, and the energy consumption is extremely low, at the sub-nJ / bit level. The extremely low energy consumption of the device is in line with the trend of low power consumption.

[0021] Fourth, the phase change material Ge2Sb2Te5 has a fast conversion speed between different phases (nanosecond level), which makes the new phase shifter have a higher modulation rate.

[0022] Fifth, the new phase shifter based on phase change materials has an operating bandwidth greater than 100nm and extremely low insertion loss and power consumption in the wavelength range of 1500nm-1600nm, and has broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 Schematic diagram of the three-dimensional structure of the low-loss non-volatile phase shifter based on phase change material of the present invention;

[0024] Figure 2TE mode propagation field distribution diagram of the low-loss non-volatile phase shifter based on phase change material of the present invention, where (a) is the Ge2Sb2Te5 layer in a crystalline state and (b) is the Ge2Sb2Te5 layer in an amorphous state;

[0025] Figure 3 A schematic structural diagram of a low-loss Mach-Zehnder interferometer optical switch using a non-volatile phase shifter based on phase change material according to the present invention;

[0026] Figure 4 Transmission spectrum of the low-loss Mach-Zehnder interferometer optical switch using a non-volatile phase shifter based on phase change material according to the present invention, wherein (a) is the Ge2Sb2Te5 layer in a crystalline state, and (b) is the Ge2Sb2Te5 layer in an amorphous state.

[0027] Description of reference numerals:

[0028] Substrate 1, first silicon waveguide 2, second silicon waveguide 3, Ge2Sb2Te5 layer 4, silicon dioxide layer 12, silicon substrate layer 11, input waveguide 5, output waveguide 6, first directional coupler 7, second directional coupler 8, silicon waveguide arm 9. DETAILED DESCRIPTION

[0029] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0030] Example 1

[0031] like Figure 1 As shown: A non-volatile phase shifter based on phase change material includes a substrate 1, the substrate 1 includes a silicon substrate layer 11 and a silicon dioxide layer 12 fixedly disposed on the silicon substrate layer 11, a first silicon waveguide 2 and a hybrid waveguide fixedly disposed on the silicon dioxide layer 12, the hybrid waveguide including a second silicon waveguide 3 and a Ge2Sb2Te5 layer 4, the second silicon waveguide 3 is fixedly disposed on the substrate 1, the second silicon waveguide 3 is spaced apart from the first silicon waveguide 2, the length direction of the second silicon waveguide 3 is parallel to the length direction of the first silicon waveguide 2, and the Ge2Sb2Te5 layer 4 is fixedly disposed on the second silicon waveguide layer 3 and the length direction of the Ge2Sb2Te5 layer 4 is parallel to the length direction of the second silicon waveguide 3.

[0032] In the phase shifter of the present invention, since the first silicon waveguide 2 and the hybrid waveguide are in an amorphous phase-matched state near selected parameters, power is coupled from the first silicon waveguide 2 to the hybrid waveguide. To ensure that power is coupled back to the first silicon waveguide 2, the length of the phase shifter must meet the coupling conditions. Furthermore, a phase difference of π radians must exist between the crystalline odd supermodes of the phase-change material Ge2Sb2Te5 layer 4 and the amorphous even and odd supermodes. This requires the following conditions to be met: We can get: Its due to and It can be obtained that: n1≈n2, where n1 is the effective refractive index of the first silicon waveguide 2; n2 is the effective refractive index of the hybrid waveguide when the phase change material is in an amorphous state. That is, when the effective refractive indices of the first silicon waveguide 2 and the hybrid waveguide are similar, the phase matching condition can be met.

[0033] The parameters in the above formula change with the distance between the first silicon waveguide 2 and the hybrid waveguide. For different distance values, the effective index of the supermode can be used to calculate the coupling length of the phase shifter: in, Represents the effective refractive index of the phase change material Ge2Sb2Te5 in the odd supermode in the crystalline state, is the effective refractive index of the phase change material Ge2Sb2Te5 in the even supermode in the amorphous state and It is the effective refractive index of the odd supermode of the phase change material Ge2Sb2Te5 in the amorphous state, Lp is the coupling length or the hybrid waveguide length, and λ is the wavelength of the incident light.

[0034] The spacing between the first silicon waveguide 2 and the hybrid waveguide and the coupling length are not fixed. When the spacing Wg between the first silicon waveguide 2 and the hybrid waveguide changes, the parameters described in the formula will change, including: the effective refractive index of the odd supermode of the phase change material Ge2Sb2Te5 in the crystalline state The effective refractive index of the even supermode of the phase change material Ge2Sb2Te5 in the amorphous state The effective refractive index of the odd supermode of the phase change material Ge2Sb2Te5 in the amorphous state The change of the value of leads to the change of coupling length. When the wavelength of the incident light remains unchanged, generally and The value of will decrease with the increase of the interval Wg. Therefore, the coupling length Lp will increase with the increase of the interval Wg. When it increases, the coupling length will increase, and the two are in a positive proportional relationship.

[0035] To ensure efficient transmission of light in the silicon waveguide under TE mode, the width of the first silicon waveguide 2 is typically selected to be between 400nm and 500nm, with a height typically ranging from 200nm to 240nm. Based on phase matching conditions, simulations can be performed to determine the dimensions of the hybrid waveguide when the effective refractive index of the first silicon waveguide 2 and the hybrid waveguide are similar. Due to the limiting effect of Ge2Sb2Te5 in the hybrid waveguide, the hybrid waveguide width is generally smaller than that of the first silicon waveguide 2. Therefore, the dimensions of the first silicon waveguide 2 and the hybrid waveguide are not fixed and can be adjusted appropriately to meet the waveguide width constraints and phase matching requirements. In this embodiment, the first cabinet waveguide and the hybrid waveguide can adopt the following dimensions: the thickness (HS) of the first silicon waveguide 2 is 220nm and the width (W0) is 400-500nm; the total thickness of the hybrid waveguide is 240nm, of which the thickness (HGST) of the upper phase change material Ge2Sb2Te5 layer 4 is 20nm, the width (WGST) is 281nm, and the length is 29.41μm; the thickness of the lower second silicon waveguide 3 (HS) is 220nm, the width (W1) is 381nm, and the length (Lp) is 29.41μm. The gap Wg between the second silicon waveguide 3 and the first silicon waveguide 2 is 150nm. The thickness of the silicon substrate is 250μm, and the thickness of the silicon dioxide layer 12 is 3μm. When the phase change material Ge2Sb2Te5 layer 4 is in a crystalline state, the first silicon waveguide 2 is phase-mismatched with the hybrid waveguide, and light passes directly through the first silicon waveguide 2, such as Figure 2 (a) When the phase-change material Ge2Sb2Te5 is in an amorphous state, the first silicon waveguide 2 achieves phase matching with the hybrid waveguide under the conditions of a hybrid waveguide length Lp of 29.41 μm, a second silicon waveguide 3 width of 381 nm, and a first silicon waveguide 2 width of 400 nm. Light is input from the first silicon waveguide 2, coupled into the hybrid waveguide, and then coupled back from the hybrid waveguide to the first silicon waveguide 2 with minimal loss, as shown in Figure 2. Figure 2 (b) shown.

[0036] The phase shifter device of the present invention has a small size, low crosstalk and can achieve low-loss transmission. The introduction of the phase change material Ge2Sb2Te5 layer 4 effectively strengthens the interaction between light and waveguide, and reduces the size of the device while changing the coupling strength. Ge2Sb2Te5 has a refractive index and absorption coefficient that are quite different in the crystalline and amorphous states. The conversion of Ge2Sb2Te5 between the crystalline and amorphous phases can be achieved by applying an electric pulse or a light pulse excitation, and Ge2Sb2Te5 has a phase transition speed of nanoseconds. In addition, Ge2Sb2Te5 has good non-volatility and can maintain phase stability without the need for additional energy. The change in the phase state of Ge2Sb2Te5 causes the refractive index and light absorption coefficient of the hybrid waveguide to change, which in turn causes the change in the output optical signal, thereby achieving modulation of the optical signal coupling strength. Specific embodiment 2

[0038] like Figure 3 As shown, a Mach-Zehnder interferometer optical switch, which uses the non-volatile phase shifter based on phase change material described in the present invention, includes two input waveguides 5, two output waveguides 6, and a silicon waveguide arm 9 fixedly disposed on a substrate 1. The two input waveguides 5 are fixedly disposed on one side of the upper surface of the substrate 1, and the two output waveguides 6 are fixedly disposed on the other side of the upper surface of the substrate 1. The output ends of the two input waveguides 5 are coupled and connected via a first directional coupler 7, and the input ends of the two output waveguides 6 are coupled and connected via a second directional coupler 8. The two ends of the silicon waveguide arm 9 are respectively connected to the first directional coupler 7 and the second directional coupler 8, and the two ends of the first silicon waveguide 2 are respectively connected to the first directional coupler 7 and the second directional coupler 8.

[0039] In this embodiment, the silicon waveguide arm 9 is equal to the first silicon waveguide 2 in length and has the same cross-sectional dimensions. The thickness of the input waveguide 5 and the thickness of the output waveguide 6 are both 220 nm and their widths are both 450 nm.

[0040] Figure 4 The figure plots the relationship between the insertion loss IL and crosstalk CT of the low-loss Mach-Zehnder interferometer optical switch in two states as a function of wavelength. It can be seen that both states have smaller loss, crosstalk and flatter bandwidth.

[0041] When the phase change material undergoes a phase change, the effective refractive index of the hybrid waveguide in the phase shifter changes, causing phase matching and mismatching between the first silicon waveguide 2 and the hybrid waveguide. This allows for optical phase modulation. The application value of this invention: This phase shifter features small size for easy on-chip integration, low loss, low crosstalk, high operating bandwidth, and high speed. It is ideally suited for a wide range of promising research areas, including photonic information processing, reconfigurable photonic devices, and photonic neural networks.

[0042] Although the present disclosure is disclosed as above, the protection scope of the present disclosure is not limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present disclosure, and these changes and modifications will fall within the protection scope of the present invention.

Claims

1. A nonvolatile phase shifter based on phase change material, characterized in that: The invention comprises a substrate (1) and a first silicon waveguide (2) and a hybrid waveguide fixedly arranged on the substrate (1), wherein the hybrid waveguide comprises a second silicon waveguide (3) and a Ge2Sb2Te5 layer (4), wherein the second silicon waveguide (3) is fixedly arranged on the substrate (1), the second silicon waveguide (3) and the first silicon waveguide (2) are spaced apart, and the length direction of the second silicon waveguide (3) is parallel to the length direction of the first silicon waveguide (2), the Ge2Sb2Te5 layer (4) is fixedly arranged on the second silicon waveguide layer (3), and the length direction of the Ge2Sb2Te5 layer (4) is parallel to the length direction of the second silicon waveguide (3), the length of the Ge2Sb2Te5 layer (4) is equal to the length of the second silicon waveguide (3), the width of the Ge2Sb2Te5 layer (4) is in the range of 250-300nm, the height is in the range of 15-25nm, and the distance between the second silicon waveguide (3) and the first silicon waveguide (2) is in the range of 100- 200nm; when Ge2Sb2Te5 is in an amorphous state, its refractive index and absorption coefficient are low, and phase matching is achieved between the hybrid waveguide and the first silicon waveguide (2), and light is effectively coupled from the first silicon waveguide (2) into the hybrid waveguide loaded with Ge2Sb2Te5, and is coupled back to the first silicon waveguide (2) again after propagating for a specific length; in this process, the light mainly interacts with the low-loss hybrid waveguide loaded with amorphous Ge2Sb2Te5, so the overall light transmission loss is small; when Ge2Sb2Te5 is transformed into a crystalline state, its refractive index and absorption coefficient increase significantly, resulting in a phase mismatch between the hybrid waveguide and the first silicon waveguide (2), which suppresses the coupling behavior. At this time, the light mainly propagates along the first silicon waveguide (2) to avoid interacting with the high-loss hybrid waveguide; by precisely designing the coupling length, the light accumulates a phase difference of π radians in the amorphous path relative to the crystalline path; therefore, non-volatile phase modulation of about π can be achieved between the two states, and low insertion loss is maintained during the entire switching process.

2. The nonvolatile phase shifter based on phase change material according to claim 1, wherein: The first silicon waveguide (2) has a height range of 200-240 nm and a width range of 400-500 nm.

3. The nonvolatile phase shifter based on phase change material according to claim 2, wherein: The height of the second silicon waveguide (3) is equal to the height of the first silicon waveguide (2).

4. The nonvolatile phase shifter based on phase change material according to claim 3, wherein: The width of the second silicon waveguide (3) is in the range of 350-400 nm, and the length is in the range of 25-30 μm.

5. The nonvolatile phase shifter based on phase change material according to claim 1, wherein: The substrate (1) comprises a silicon substrate layer (11) and a silicon dioxide layer (12) fixedly arranged on the silicon substrate layer (11); the first silicon waveguide (2) and the hybrid waveguide are fixedly arranged on the silicon dioxide layer (12).

6. The nonvolatile phase shifter based on phase change material according to claim 5, wherein: The height of the silicon substrate layer (11) is in the range of 200-300 μm, and the height of the silicon dioxide layer (12) is in the range of 2-4 μm.

7. The nonvolatile phase shifter based on phase change material according to claim 1, wherein: There is a phase difference of π radians between the odd supermodes in the crystalline state of the phase change material Ge2Sb2Te5 layer (4) and the even supermodes and odd supermodes in the amorphous state. The phase difference of π radians is: , in the above formula, represents the effective refractive index of the odd supermode when the Ge2Sb2Te5 layer (4) is in the crystalline state, represents the effective refractive index of the Ge2Sb2Te5 layer (4) in the even supermode in the amorphous state, represents the effective refractive index of the odd supermode of the Ge2Sb2Te5 layer (4) in the amorphous state, and π represents the π radian phase difference between the odd supermode of the waveguide when the Ge2Sb2Te5 layer (4) is in the crystalline state and the odd supermode and even supermode in the amorphous state.

8. The nonvolatile phase shifter based on phase change material according to claim 7, wherein: When the Ge2Sb2Te5 layer (4) in the phase shifter is in an amorphous state, the length calculation formula of the hybrid waveguide is: , in the above formula, Lp represents the length of the hybrid waveguide, Represents the wavelength of the incident light.

Images