A scalable 1×3 non-volatile optical switch unit and a non-volatile programmable 1×2 multimode optical switch

Abstract

The present invention is an expandable 1×3 non-volatile optical switch unit and a non-volatile programmable 1×2 multimode optical switch. The 1×3 optical switch unit includes a bare silicon multimode bus waveguide and two PCM-Si single-mode hybrid waveguides, wherein the upper and lower arm hybrid waveguides are provided with Sb2Se3 phase change materials of equal length and thickness. By regulating the state of the two waveguide phase change materials, namely the crystalline state (c-Sb2Se3) and the amorphous state (a-Sb2Se3), the two functions of splitting and optical switching can be realized; using adiabatic tapered waveguides, 1×3 optical switch units with different multimode bus waveguide widths can be cascaded into a 1×2 multimode optical switch supporting 6 modes. When the multimode bus waveguide and the PCM-Si hybrid waveguide reach the phase matching condition, the corresponding high-order mode signal will be converted into a fundamental mode signal for data transmission. The optical switch structure proposed by the present invention supports the multiplexing and switching of multiple mode signals, and has the advantages of adjustable switch dynamic state, zero static power consumption, low insertion loss and low crosstalk.

Description

Technical Field

[0001] The present invention belongs to the technical field of photon signal processing, and in particular relates to an expandable 1×3 non-volatile optical switch unit and a non-volatile programmable 1×2 multimode optical switch supporting six-mode multiplexing. Background Art

[0002] The statements in this section only provide background information related to the present disclosure, and these statements may constitute prior art. In the process of implementing the present invention, the inventors found that there are at least the following problems in the prior art.

[0003] With its significant advantages of large bandwidth, high-speed transmission, excellent CMOS compatibility and high-density integration, silicon-based optical interconnect technology provides a promising solution to meet the growing demand for data traffic. In order to expand the capacity of on-chip optical links, researchers have conducted in-depth research on technologies such as wavelength division multiplexing (WDM), mode multiplexing (MDM) and polarization multiplexing (PDM). Due to the constraints of the Shannon limit and the bandwidth limitations of erbium-doped fiber amplifiers, WDM technology is facing a severe capacity bottleneck. To overcome these challenges, MDM has attracted much attention as a viable alternative. MDM treats each spatial mode as an independent channel and significantly improves the link capacity by multiplexing orthogonal modes onto a single wavelength carrier. Integrating spatial mode parallelism into photonic integrated circuits (PICs) is expected to expand the capabilities of existing optical communication systems.

[0004] Recent advances in MDM systems have led to the development of passive mode switches, utilizing a range of architectures, including multimode interferometers, waveguide crossovers, curved waveguides, grating couplers, Y-junctions, adiabatic couplers, and computationally optimized nanophotonic structures. However, fundamental to scalable MDM networks is the ability to dynamically switch channels between these structures and precisely control the propagation of specific modes. Currently, reported dynamic mode switches, including those based on Mach-Zehnder interferometers (MZIs), microring resonators (MRRs), MMIs, and Y-junctions, primarily rely on thermo-optical, electro-optical, or all-optical effects due to their relative simplicity and low cost. However, the refractive index changes induced by these effects are quite small (Δn < 0.01), resulting in device sizes typically on the order of hundreds of microns, limiting their scalability. Furthermore, the implemented switches are unstable and require a continuous power supply to maintain the switching state, inevitably resulting in high power consumption.

[0005] For example, the patent application number 202311695651.2 is titled “A non-volatile optical switch unit and an optical switching network composed thereof”, which is a non-volatile non-blocking optical switch based on Mach-Zehnder interferometer and phase change material. The structure involved in this invention is as follows: First, a non-volatile phase change material Sb2Se3 and MZI are used to form a non-volatile optical switch unit. The phase shift value is changed by switching the state of the phase change material, thereby controlling the state of the switch. In addition, based on the above optical switch unit, 2 n ×2 n Benes topology non-blocking optical switching network. However, this design does not consider improving the refractive index contrast between the two states of the phase change material to shorten the coupling region length and mode (de)multiplexing, and has the following defects:

[0006] 1. The insertion loss is high, affecting the signal transmission quality.

[0007] This patent uses a Benes topology to interconnect 2×2 optical switch units into an 8×8 non-volatile, non-blocking optical switch. While this structure enables non-blocking data transmission, the switch consists of 20 2×2 optical switch units based on Mach-Zehnder interferometers and non-volatile phase-change materials, and 16 cross-waveguides. The excessive use of components inevitably results in higher insertion loss, affecting signal transmission quality. Furthermore, the design of the cross-waveguides must be considered to minimize insertion loss.

[0008] 2. The effective refractive index contrast is low, and the upper and lower arm signals cannot be separated well.

[0009] Although the 8×8 non-volatile non-blocking optical switch uses Sb2Se3 phase change material to achieve non-volatile characteristics and reduce energy consumption, its thickness is 25nm. Figure 1 It can be deduced that its effective refractive index contrast is only between 0.045-0.161. It is usually necessary to set the width of the waveguide very finely so as to achieve phase matching with the hybrid waveguide embedded with the amorphous state, so that the signal can be completely output from the port on the amorphous side.

[0010] 3. If you want to transmit more data, you need to expand to a larger optical switch and more components.

[0011] The waveguide of the 8×8 non-volatile non-blocking optical switch is only a single-mode waveguide, and each port can only transmit one data at a time. If you want to transmit more data, you need to expand it to a larger optical switch, which requires more components. Summary of the Invention

[0012] In view of the above problems, the present invention aims to solve some of the problems in the prior art, or at least alleviate these problems.

[0013] A scalable 1×3 non-volatile optical switch unit includes a multimode bus waveguide located in the middle and an upper arm PCM-Si single-mode hybrid waveguide and a lower arm PCM-Si single-mode hybrid waveguide respectively arranged on both sides of the multimode bus waveguide; the upper arm PCM-Si single-mode hybrid waveguide and the lower arm PCM-Si single-mode hybrid waveguide are provided with a phase change material Sb2Se3 film of equal length and thickness, and the Sb2Se3 films on the upper and lower arm PCM-Si single-mode hybrid waveguides can switch between a crystalline state (c-Sb2Se3) and an amorphous state (a-Sb2Se3); the multimode bus waveguide and the PCM-Si single-mode hybrid waveguide containing a-Sb2Se3 achieve phase matching conditions to convert high-order mode signals input from the input end of the multimode bus waveguide into fundamental mode signals for data transmission, while mode signals that do not meet the phase matching conditions continue to be transmitted forward, from O 12 Port output; the fundamental mode signal in the upper arm PCM-Si single mode hybrid waveguide is output from O 11 The fundamental mode signal in the lower arm PCM-Si single-mode hybrid waveguide is output from O 13 Port output.

[0014] Optionally, the upper and lower arm PCM-Si single-mode hybrid waveguides can control the state of the Sb2Se3 film by laser pulse irradiation to achieve two different functions of light splitting and light switching.

[0015] Furthermore, the coupling region between the upper and lower arm PCM-Si single-mode hybrid waveguides and the multimode bus waveguide is composed of a PCM-Si hybrid waveguide of equal length to the coupling length; the PCM-Si hybrid waveguide in the coupling region is a 450nm wide and 220nm thick PCM-Si hybrid waveguide, including a Sb2Se3 thin film of the same width and 100nm thick and a silicon waveguide of the same width and 120nm thickness.

[0016] Furthermore, the width of the multimode bus waveguide is adjustable; the coupling gaps between the upper and lower arm PCM-Si hybrid waveguides and the middle multimode bus waveguide are both 0.1 um.

[0017] Furthermore, the phase change material Sb2Se3 is doped with GeTe.

[0018] A non-volatile programmable 1×2 multimode optical switch employs six of the aforementioned expandable 1×3 non-volatile optical switch units, dividing the 1×2 multimode optical switch into six levels (including first to sixth levels of multimode bus waveguides). From left to right, the number of modes supported by the multimode bus waveguide decreases by one, and two levels are connected by an adiabatic tapered waveguide to match two waveguides with different geometric cross-sectional parameters. Fundamental mode signals in the plurality of upper-arm PCM-Si single-mode hybrid waveguides are all output from the O1 port, and fundamental mode signals in the plurality of lower-arm PCM-Si single-mode hybrid waveguides are all output from the O2 port.

[0019] Furthermore, the first-level multimode bus waveguide supports a total of 6 modes, namely: TE5, TE4, TE3, TE2, TE1, TE0; the second-level multimode bus waveguide supports a total of 5 modes, namely: TE4, TE3, TE2, TE1, TE0; the third-level multimode bus waveguide supports a total of 4 modes, namely: TE3, TE2, TE1, TE0; the fourth-level multimode bus waveguide supports a total of 3 modes, namely: TE2, TE1, TE0; the fifth-level multimode bus waveguide supports a total of 2 modes, namely: TE1, TE0; the sixth-level bus waveguide only supports 1 mode, namely: TE0.

[0020] Preferably, the widths of the first to sixth level multimode bus waveguides are 2.40μm, 1.99μm, 1.57μm, 1.17μm, 0.77μm, and 0.37μm, respectively, and the optimal coupling length and Sb2Se3 length are 13.7μm, 12.8μm, 11.4μm, 9.8μm, 8.0μm, and 5.3μm, respectively; the upper and lower floating ranges of the multimode bus waveguide width and coupling length are approximately 0.01μm and 0.02μm, respectively.

[0021] The present invention has the following beneficial effects:

[0022] 1. The scalable 1×3 non-volatile optical switch unit of the present invention uses a 450nm wide, 100nm thick Sb2Se3 and a 120nm thick silicon waveguide to form a 450nm wide, 220nm thick hybrid waveguide. Compared with the hybrid waveguide composed of 20nm, 50nm, and 100nm thick Sb2Se3 and a 220nm silicon waveguide, the effective refractive index contrast is increased by nearly 3-9 times. The large refractive index contrast can better separate the upper and lower arm signals and achieve better optical switching. Performance; the a-Sb2Se3 of the hybrid waveguide has the lowest refractive index value, and coupling with the amorphous Sb2Se3 can effectively reduce the width of the multimode bus waveguide, thereby reducing the area of ​​the multimode optical switch and improving the scalability of the optical switch; and the use of zero static power consumption, high refractive index contrast, non-volatile phase change material Sb2Se3, and doping it with GeTe and increasing the crystallization temperature can improve its thermal stability, so that no additional power is required to maintain the switch state, thereby reducing the power consumption of the optical switch;

[0023] 2. The present invention cascades the six scalable 1×3 non-volatile optical switch units to form a 1×2 multimode optical switch that supports six modes. By utilizing the non-volatility and high contrast between the complex refractive index of the crystalline and amorphous states, and by setting the optimal multimode waveguide width, coupling length, and coupling gap, the six different modes and the TE0 fundamental mode signals in the upper and lower arm hybrid waveguides are phase-matched for efficient coupling, ensuring optimal optical switching performance. This allows the structure to perform both splitting and optical switching functions. Compared to non-volatile, non-blocking optical switches based on Mach-Zehnder interferometers and phase-change materials, the insertion loss of the six modes in the 1×2 multimode optical switch proposed by the present invention is less than or equal to 0.5dB within the C-band. By reducing insertion loss, signal transmission quality is significantly improved. Furthermore, due to the incorporation of mode multiplexing technology, parallel transmission of different modes can be achieved by simply designing the appropriate multimode bus waveguide width to support more high-order modes, thereby facilitating the parallel transmission of more data and increasing the capacity of the uplink optical link.

[0024] 3. The present invention uses laser pulse irradiation to achieve rapid and reversible switching of Sb2Se3 between the crystalline and amorphous states, that is, the switching state can change rapidly;

[0025] 4. The optical switch structure proposed in the present invention supports the multiplexing and exchange of multiple mode signals, and has the advantages of adjustable switch dynamic state, zero static power consumption, low insertion loss and low crosstalk, and can realize flexible, dynamic and non-volatile switching of the optical switch. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1Schematic diagram of the effective refractive index of a PCM-Si hybrid waveguide composed of phase change materials of different thicknesses and a silicon-based waveguide according to the present invention, which varies with the waveguide width;

[0027] Figure 2 The present invention provides a 1×3 optical switch unit based on phase change materials; Figure (a) is a schematic diagram of light splitting achieved when all hybrid waveguides of the 1×3 optical switch unit are a-Sb2Se3; Figure (b) is a schematic diagram of optical switching achieved when the upper arm hybrid waveguide of the 1×3 optical switch unit is a-Sb2Se3 and the lower arm hybrid waveguide is c-Sb2Se3; Figure (c) is a schematic diagram of optical switching achieved when the upper arm hybrid waveguide of the 1×3 optical switch unit is c-Sb2Se3 and the lower arm hybrid waveguide is a-Sb2Se3; Figure (d) is a schematic diagram of optical signal transmission when all hybrid waveguides of the 1×3 optical switch unit are c-Sb2Se3.

[0028] Figure 3 is a schematic cross-sectional view of a 1×3 optical switch unit of the present invention;

[0029] Figure 4 The present invention is a 3dB optical power beam splitter based on mode multiplexing and phase change materials;

[0030] Figure 5 、 Figure 6 、 Figure 7 The present invention is a 1×2 multimode optical switch based on mode multiplexing and a combination of different phase change material states;

[0031] Figure 8 Schematic diagram of the effective refractive index of the eigenmode in a 220nm thick bare silicon waveguide calculated in the present invention. DETAILED DESCRIPTION

[0032] The present invention will be further described below in conjunction with the accompanying drawings. The embodiments of the present invention are only used to illustrate the present invention and are not intended to limit the present invention. Without departing from the technical concept of the present invention, various substitutions and modifications can be made based on common technical knowledge and customary means in the field, and all should be included in the scope of the present invention.

[0033] The present invention is a patent application for an optical switch based on mode multiplexing and phase change materials, which aims to expand the capacity of on-chip optical links and enable flexible, dynamic and non-volatile switching of optical switches.

[0034] like Figure 2As shown in (a)-(d), an expandable 1×3 non-volatile optical switch unit includes a multimode bus waveguide located in the middle and an upper arm PCM-Si single-mode hybrid waveguide and a lower arm PCM-Si single-mode hybrid waveguide respectively arranged on both sides of the multimode bus waveguide; the upper arm PCM-Si single-mode hybrid waveguide and the lower arm PCM-Si single-mode hybrid waveguide are provided with a phase change material Sb2Se3 film of equal length and thickness, and the Sb2Se3 films on the upper and lower arm PCM-Si single-mode hybrid waveguides can be switched between a crystalline state (c-Sb2Se3) and an amorphous state (a-Sb2Se3); the multimode bus waveguide and the PCM-Si single-mode hybrid waveguide containing a-Sb2Se3 achieve phase matching conditions to convert the high-order mode signal input from the input end of the multimode bus waveguide into a fundamental mode signal for data transmission, while the mode signal that does not meet the phase matching conditions continues to be transmitted forward, from O 12 Port output; the fundamental mode signal in the upper arm PCM-Si single mode hybrid waveguide is output from O 11 The fundamental mode signal in the lower arm PCM-Si single-mode hybrid waveguide is output from the O 13 Port output.

[0035] The multimode bus waveguide and the single-mode hybrid waveguide achieve phase matching conditions, that is, the effective refractive index values ​​are equal, to achieve the process of converting high-order mode signals into fundamental mode signals for data transmission. The mode signals that do not meet the phase matching conditions will continue to transmit forward. In the coupling region, the upper arm hybrid waveguide and the lower arm hybrid waveguide are provided with Sb2Se3 films of equal length and thickness, such as Figure 3 As shown, the Sb2Se3 films on both arm waveguides can switch between the crystalline state (c-Sb2Se3) and the amorphous state (a-Sb2Se3). This application replaces the traditional dynamic mode optical switch composed of MZI, MRR, MMI and Y-junction that relies on thermo-optical, electro-optical or all-optical effects with zero static power consumption, high refractive index contrast, and non-volatile phase change material Sb2Se3. By forming a hybrid PCM-Si waveguide with Si waveguide and Sb2Se3, its effective refractive index contrast can be increased and the length of the coupling region can be reduced. Therefore, the present invention can reduce the area of ​​the multi-mode optical switch.

[0036] Accordingly, doping GeTe into the phase change material Sb2Se3 and increasing the crystallization temperature can improve its thermal stability, so no additional power is required to maintain the switch state, that is, the power consumption of the optical switch can be reduced.

[0037] like Figure 1 and Figure 3As shown, the coupling region between the upper and lower arm PCM-Si single-mode hybrid waveguides and the multimode bus waveguide is composed of a PCM-Si hybrid waveguide of equal length to the coupling length; the PCM-Si hybrid waveguide in the coupling region is a 450nm wide and 220nm thick PCM-Si hybrid waveguide, including a Sb2Se3 thin film of the same width and 100nm thickness and a silicon waveguide of the same width and 120nm thickness. In the coupling region, the present invention selects a 450nm wide, 100nm thick Sb2Se3 and a 120nm thick silicon waveguide of the same width to form a 450nm wide, 220nm thick hybrid waveguide. Compared with the hybrid waveguide composed of Sb2Se3 and 220nm thick silicon waveguides of the same width of 20nm, 50nm, and 100nm, the hybrid PCM-Si waveguide with the selected parameters has a larger refractive index contrast of about 0.42 at a waveguide width of 450nm, which can effectively shorten the length of the coupling region. The refractive index contrasts of the other three are 0.045, 0.161, and 0.24, respectively. The effective refractive index contrast of the present invention is increased by nearly 3-9 times, which can better separate the upper and lower arm signals and achieve better optical switching performance. On the other hand, by Figure 1 It can be seen that the a-Sb2Se3 of the hybrid waveguide has the lowest refractive index value, so coupling with amorphous Sb2Se3 can effectively reduce the width of the multimode bus waveguide, thereby reducing the area of ​​the structure.

[0038] In the present invention, the upper and lower arm PCM-Si single-mode hybrid waveguides can control the state of the Sb2Se3 thin film by laser pulse irradiation to achieve two different functions: splitting and optical switching. That is, when the Sb2Se3 thin films on the upper and lower arm hybrid waveguides are both a-Sb2Se3, the multimode bus waveguide simultaneously reaches a phase matching condition with the two PCM-Si single-mode hybrid waveguides, and the corresponding high-order mode signal is converted into two equal fundamental mode signals for data transmission. The upper and lower ports will each output half of the optical signal. At this time, the structure is a 3dB optical power splitter to achieve 3dB splitting; when one of the Sb2Se3 thin films on the two arm waveguides is a-Sb2Se3 and the other is c-Sb2Se3, the multimode bus waveguide simultaneously reaches a phase matching condition with the two PCM-Si single-mode hybrid waveguides, and the corresponding high-order mode signal is converted into two equal fundamental mode signals for data transmission. The upper and lower ports will each output half of the optical signal. At this time, the structure is a 3dB optical power splitter to achieve 3dB splitting; when one of the Sb2Se3 thin films on the two arm waveguides is a-Sb2Se3 and the other is c-Sb2Se3, Because the multimode bus waveguide and a-Sb2Se3 have similar effective refractive index values, the high-order mode optical signal carrying data will be completely converted into the fundamental mode signal in the single-mode hybrid waveguide on the amorphous side for output. At this time, the structure is a 1×2 optical switch to realize optical switching; when the Sb2Se3 films on the upper and lower arm hybrid waveguides are both c-Sb2Se3, the multimode bus waveguide and the two PCM-Si single-mode hybrid waveguides do not reach the phase matching condition. At this time, the optical signal carrying data will continue to move forward and be output from the straight-through port.

[0039] The specific working mode of the expandable 1×3 non-volatile optical switch unit is as follows Figure 2 As shown in (a)-(d).

[0040] like Figure 2 (a) shows the multimode bus waveguide input port I of the scalable 1×3 non-volatile optical switch unit. 11 Take the support of TE1 and TE0 modes as an example. When the Sb2Se3 films on the upper and lower arm hybrid waveguides are both a-Sb2Se3, the TE1 high-order mode signal in the multimode bus waveguide simultaneously reaches the phase matching condition with the TE0 fundamental mode signal in the two PCM-Si single-mode hybrid waveguides. At this time, the TE1 mode signal is converted into two equal TE0 mode signals, that is, half of the TE1 mode signal is converted into the TE0 fundamental mode signal in the upper arm hybrid waveguide from O 11 The other half is converted into the TE0 fundamental mode signal in the lower arm hybrid waveguide from O 13 Port output, achieving 3dB splitting in TE1 mode, and I 11 The TE0 mode signal input from the port will continue to move forward because it does not reach the phase matching condition with the TE0 fundamental mode signal in the PCM-Si single-mode hybrid waveguide. 12 Port output;

[0041] like Figure 2 As shown in (b), when the Sb2Se3 film of the upper arm waveguide is a-Sb2Se3 and the Sb2Se3 film of the lower arm waveguide is c-Sb2Se3, at this time, the high-order mode signal in the multimode bus waveguide and the fundamental mode signal in the upper arm PCM-Si single-mode hybrid waveguide reach the phase matching condition, and the TE1 mode signal is converted into the TE0 fundamental mode signal in the upper arm hybrid waveguide completely from O 11 Port output, realize TE1 mode signal from O 11 The output of the port, I 11 The TE0 mode signal of the port still starts from 0 12 Port output;

[0042] like Figure 2 As shown in (c), when the Sb2Se3 film of the upper arm waveguide is c-Sb2Se3 and the Sb2Se3 film of the lower arm waveguide is a-Sb2Se3, at this time, the high-order mode signal in the multimode bus waveguide and the fundamental mode signal in the lower arm PCM-Si single-mode hybrid waveguide reach the phase matching condition, and the TE1 mode signal is completely converted into the TE0 fundamental mode signal in the upper arm hybrid waveguide from O 13 Port output, realize TE1 mode signal from O 13 The output of the port, I 11 The TE0 mode signal of the port still starts from 0 12 Port output;

[0043] like Figure 2As shown in (d), when the Sb2Se3 films on the upper and lower arm hybrid waveguides are both c-Sb2Se3, the TE1 and TE0 mode signals in the multimode bus waveguide and the TE0 fundamental mode signals in the two PCM-Si single-mode hybrid waveguides all reach the phase matching condition. At this time, the TE1 and TE0 mode signals will continue to move forward from O 12 Port output.

[0044] The width of the multimode bus waveguide is adjustable to support different high-order modes; Figure 3 As shown, if the state of the phase change material is not considered, the structure is symmetrical, and the coupling gaps between the upper and lower arm PCM-Si hybrid waveguides and the middle multimode bus waveguide are both 0.1 μm.

[0045] The multimode bus waveguide is a bare silicon multimode bus waveguide, such as Figure 3 shown.

[0046] The present invention also proposes a non-volatile programmable 1×2 multimode optical switch that supports six-mode multiplexing, such as Figure 4-7 As shown, six scalable 1×3 non-volatile optical switch units are used to divide the 1×2 multimode optical switch into six levels (including the first to sixth multimode bus waveguides). From left to right, the number of modes supported by the multimode bus waveguide decreases by 1, and the two levels are connected by an adiabatic tapered waveguide to match the two waveguides with different geometric cross-sectional parameters to reduce energy loss. The fundamental mode signals in the multiple upper arm PCM-Si single-mode hybrid waveguides are all output from the O1 port, and the fundamental mode signals in the multiple lower arm PCM-Si single-mode hybrid waveguides are all output from the O2 port. Compared to the non-volatile, non-blocking optical switch based on Mach-Zehnder interferometers and phase change materials, the insertion loss of the six modes in the 1×2 multimode optical switch proposed by the present invention is less than or equal to 0.5dB within the C-band range. By reducing the insertion loss, the signal transmission quality is greatly improved. Moreover, due to the combination of mode multiplexing technology, it is only necessary to design a suitable multimode bus waveguide width to support more high-order modes, so as to achieve parallel transmission of different modes, so as to facilitate the parallel transmission of more data, thereby improving the capacity of the optical link.

[0047] The optical switch structure proposed in the present invention supports multiplexing and switching of multiple mode signals, and has the advantages of adjustable switch dynamic state, zero static power consumption, low insertion loss and low crosstalk.

[0048] like Figure 4-7As shown, the mode signal is input from the input end of the first-level multimode centerline waveguide. The first-level multimode bus waveguide supports a total of 6 modes, namely: TE5, TE4, TE3, TE2, TE1, TE0; since the TE5 mode signal in the first-level multimode bus waveguide is converted into the TE0 fundamental mode signal of the hybrid waveguide, the second-level multimode bus waveguide supports a total of 5 modes, namely: TE4, TE3, TE2, TE1, TE0; since the TE4 mode signal in the second-level multimode bus waveguide is also converted into the TE0 fundamental mode signal of the hybrid waveguide, the third-level multimode bus waveguide supports a total of 4 modes, namely: TE3, TE2, TE1, TE0; Similarly, since the TE3 mode in the third-level multimode bus waveguide is also converted into the TE0 fundamental mode signal of the hybrid waveguide, the fourth-level multimode bus waveguide supports a total of three modes, namely: TE2, TE1, and TE0; since the TE2 mode signal in the fourth-level multimode bus waveguide is converted into the TE0 fundamental mode signal of the hybrid waveguide, the fifth-level multimode bus waveguide supports a total of two modes, namely: TE1 and TE0; since the TE1 mode signal in the fifth-level multimode bus waveguide is also converted into the TE0 fundamental mode signal of the hybrid waveguide, the sixth-level bus waveguide only supports one mode, namely: TE0.

[0049] When this structure acts as a 3dB optical beam splitter, the first-stage multimode bus waveguide and the two single-mode hybrid waveguides in the upper and lower arms simultaneously achieve phase matching conditions. At this time, the high-order mode signal is converted into two equal fundamental mode signals, namely: the TE5 mode signal is converted into two equal TE0 mode signals. Half of the TE5 mode signal is converted into the TE0 fundamental mode signal in the upper arm hybrid waveguide and output from the upper port, and the other half is converted into the TE0 signal in the lower arm hybrid waveguide and output from the lower port, realizing 3dB splitting of the TE5 mode. The same process is used to achieve 3dB splitting of the TE4 mode in the second-stage multimode bus waveguide, the TE3 mode in the third-stage multimode bus waveguide, the TE2 mode in the fourth-stage multimode bus waveguide, the TE1 mode in the fifth-stage multimode bus waveguide, and the TE0 mode in the sixth-stage single-mode bus waveguide.

[0050] When this structure acts as a 1×2 multimode optical switch, the first-level multimode bus waveguide and the waveguide on the side embedded with a-Sb2Se3 in the two hybrid waveguides reach phase matching conditions. At this time, the high-order mode signal is converted into the fundamental mode signal and output completely from the waveguide port on the amorphous side. That is, when the Sb2Se3 in the upper arm waveguide is a-Sb2Se3, the TE5 mode signal is converted into the TE0 fundamental mode signal in the upper arm hybrid waveguide and output completely from the upper port. When the Sb2Se3 in the lower arm waveguide is a-Sb2Se3, the TE5 mode signal is converted into the TE0 fundamental mode signal in the lower arm hybrid waveguide and output completely from the lower port. The same process applies to the TE4 mode signal in the second-level multimode bus waveguide, the TE3 mode signal in the third-level multimode bus waveguide, the TE2 mode signal in the fourth-level multimode bus waveguide, the TE1 mode signal in the fifth-level multimode bus waveguide, and the TE0 mode signal in the sixth-level bus waveguide to achieve arbitrary port output.

[0051] The specific working mode of the non-volatile programmable 1×2 multimode optical switch is as follows: Figures 4 to 7 shown.

[0052] like Figure 4 As shown, when the structure acts as a 3dB optical beam splitter:

[0053] Stage 1: The TE5 mode signal in the multimode bus waveguide and the TE0 fundamental mode signal in the upper and lower hybrid waveguides simultaneously reach phase matching conditions. At this point, the TE5 mode signal is converted into two equal TE0 fundamental mode signals: one half is converted into the TE0 fundamental mode signal in the upper hybrid waveguide and output from port O1, and the other half is converted into the TE0 fundamental mode signal in the lower hybrid waveguide and output from port O2, achieving 3dB splitting of the TE5 mode.

[0054] Second stage: The TE4 mode signal in the multimode bus waveguide and the TE0 fundamental mode signals in the upper and lower hybrid waveguides reach phase matching conditions at the same time. At this point, the TE4 mode signal is converted into two equal TE0 fundamental mode signals: one half is converted into the TE0 fundamental mode signal in the upper hybrid waveguide and output from port O1, and the other half is converted into the TE0 fundamental mode signal in the lower hybrid waveguide and output from port O2, achieving 3dB splitting of the TE4 mode.

[0055] Level 3: The TE3 mode signal in the multimode bus waveguide and the TE0 fundamental mode signal in the upper and lower hybrid waveguides reach phase matching conditions at the same time. At this point, the TE3 mode signal is converted into two equal TE0 fundamental mode signals: one half is converted into the TE0 fundamental mode signal in the upper hybrid waveguide and output from port O1, and the other half is converted into the TE0 fundamental mode signal in the lower hybrid waveguide and output from port O2, achieving 3dB splitting of the TE3 mode.

[0056] Level 4: The TE2 mode signal in the multimode bus waveguide and the TE0 fundamental mode signals in the upper and lower hybrid waveguides reach phase matching conditions at the same time. At this point, the TE2 mode signal is converted into two equal TE0 fundamental mode signals: one half is converted into the TE0 fundamental mode signal in the upper hybrid waveguide and output from the O1 port, and the other half is converted into the TE0 fundamental mode signal in the lower hybrid waveguide and output from the O2 port, achieving 3dB splitting of the TE2 mode.

[0057] Level 5: The TE1 mode signal in the multimode bus waveguide and the TE0 fundamental mode signals in the upper and lower hybrid waveguides reach phase matching conditions at the same time. At this point, the TE1 mode signal is converted into two equal TE0 fundamental mode signals: one half is converted into the TE0 fundamental mode signal in the upper hybrid waveguide and output from the O1 port, and the other half is converted into the TE0 fundamental mode signal in the lower hybrid waveguide and output from the O2 port, achieving 3dB splitting of the TE1 mode.

[0058] Level 6: The TE0 mode signal in the bus waveguide and the TE0 fundamental mode signals in the upper and lower hybrid waveguides reach phase matching conditions at the same time. At this point, the TE0 mode signal is converted into two equal TE0 fundamental mode signals: one half is converted into the TE0 fundamental mode signal in the upper hybrid waveguide and output from the O1 port, and the other half is converted into the TE0 fundamental mode signal in the lower hybrid waveguide and output from the O2 port, achieving 3dB splitting of the TE0 mode.

[0059] like Figure 5 As shown, when the structure acts as a 1×2 multimode optical switch:

[0060] First stage: The TE5 mode signal in the multimode bus waveguide and the TE0 fundamental mode signal in the upper arm hybrid waveguide embedded with a-Sb2Se3 reach phase matching conditions. At this point, the TE5 mode signal is converted into the TE0 fundamental mode signal in the upper arm hybrid waveguide and is completely output from the O1 port.

[0061] Second stage: The TE4 mode signal in the multimode bus waveguide and the TE0 fundamental mode signal in the hybrid waveguide embedded with a-Sb2Se3 in the lower arm reach phase matching conditions. At this point, the TE4 mode signal is converted into the TE0 fundamental mode signal in the hybrid waveguide in the lower arm and is completely output from the O2 port.

[0062] Stage 3: The TE3 mode signal in the multimode bus waveguide and the TE0 fundamental mode signal in the upper arm hybrid waveguide embedded with a-Sb2Se3 reach phase matching conditions. At this point, the TE3 mode signal is converted into the TE0 fundamental mode signal in the upper arm hybrid waveguide and is completely output from the O1 port.

[0063] Level 4: The TE2 mode signal in the multimode bus waveguide and the TE0 fundamental mode signal in the upper arm hybrid waveguide embedded with a-Sb2Se3 reach phase matching conditions. At this point, the TE2 mode signal is converted into the TE0 fundamental mode signal in the upper arm hybrid waveguide and is completely output from the O1 port.

[0064] Level 5: The TE1 mode signal in the multimode bus waveguide and the TE0 fundamental mode signal in the hybrid waveguide embedded with a-Sb2Se3 in the lower arm reach phase matching conditions. At this point, the TE1 mode signal is converted into the TE0 fundamental mode signal in the hybrid waveguide in the lower arm and is completely output from the O2 port.

[0065] Level 6: The TE0 mode signal in the bus waveguide and the TE0 fundamental mode signal in the upper arm hybrid waveguide embedded with a-Sb2Se3 reach the phase matching condition. At this time, the TE0 mode signal is converted into the TE0 fundamental mode signal of the upper arm hybrid waveguide and is completely output from the O1 port.

[0066] Figure 6 、 Figure 7 The working principle and Figure 2 (b) Figure 2 (c) and Figure 5 Consistent, due to Figure 6 The phase change material embedded in the upper arm hybrid waveguide of each 1×3 optical switch unit is a-Sb2Se3, and the lower arm hybrid waveguide is c-Sb2Se3, so the TE5, TE4, TE3, TE2, TE1, and TE0 mode signals are all output from the O1 port; Figure 7 The phase change material embedded in the upper arm hybrid waveguide of each 1×3 optical switch unit is c-Sb2Se3, and the lower arm hybrid waveguide is a-Sb2Se3, so the TE5, TE4, TE3, TE2, TE1, and TE0 mode signals are all output from the O2 port.

[0067] The key to achieving optimal optical switching performance in the 1×2 multimode optical switch of the present invention, which is composed of 1×3 optical switch units, is that the multimode bus waveguide and the PCM-Si waveguides containing a-Sb2Se3 on both sides can achieve phase matching conditions, that is, the effective refractive index values ​​are equal to achieve conversion of high-order mode signals to fundamental mode signals. Figure 5 As shown in the figure, the approximate waveguide width when the TE5-TE1 and the TE0 fundamental mode signals in the hybrid waveguide embedded with a-Sb2Se3 are coupled is marked. In this structure, the coupling gap between the fixed bus waveguide and the upper and lower arm hybrid waveguides is 0.1μm. Figure 5The results shown in the figure were scanned and optimized using the Ansys Lumerical simulation platform to obtain the specific parameter values ​​of the multimode bus waveguide width, coupling length and phase change material length to achieve the optimal optical switching performance. The obtained multimode bus waveguide widths supporting TE5, TE4, TE3, TE2, TE1 and TE0 modes are 2.40μm, 1.99μm, 1.57μm, 1.17μm, 0.77μm and 0.37μm, respectively. The optimal coupling lengths of these modes are the same as the Sb2Se3 lengths, which are 13.7μm, 12.8μm, 11.4μm, 9.8μm, 8.0μm and 5.3μm, respectively. Since fine-tuning the bus waveguide width and coupling length can change the performance of optical switching, the upper and lower floating ranges of the bus waveguide width and coupling length to ensure the optimal optical switching performance are approximately 0.01μm and 0.02μm, respectively.

[0068] Table 1 summarizes the specific coupling length values, Sb2Se3 length values, coupling gap values, and bus waveguide width values ​​for the six different modes to achieve optimal optical switching conditions:

[0069] Table 1 1×2 multimode optical switch parameters

[0070]

[0071] The present invention utilizes a 450nm wide, 100nm thick Sb2Se3 and a 120nm thick silicon waveguide to construct a 450nm wide, 220nm thick hybrid PCM-Si waveguide. Compared to hybrid waveguides constructed with Sb2Se3 and 220nm thick silicon waveguides of 20nm, 50nm, and 100nm widths, the hybrid PCM-Si waveguide with the selected parameters exhibits a high refractive index contrast of approximately 0.42 at a 450nm waveguide width, effectively shortening the length of the coupling region. Furthermore, by combining mode multiplexing, a 1×2 multimode optical switch supporting six modes was constructed by cascading optical switch units embedded with phase change materials. Using the Ansys Lumerical simulation platform, parameters such as TE width and coupling length were scanned and optimized. At a wavelength of 1550nm, the transmittance of each mode of this multimode optical switch at the corresponding output port exceeded 94%. By combining non-volatile phase-change materials with mode multiplexing, dynamic non-volatile multimode optical switches can precisely control the transmission path of optical signals, ensuring efficient and stable network operation. They can also be combined with other multiplexing methods to further expand on-chip link capacity to meet growing bandwidth demands.

[0072] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A non-volatile programmable 1×2 multimode optical switch, characterized in that: The multimode optical switch includes an input port and two output ports; the input port is a mode demultiplexer part, and the two output ports are upper and lower mode multiplexer parts; the mode demultiplexer part is composed of 6 expandable 1×3 non-volatile optical switch units, and each two optical switch units are connected by an adiabatic tapered waveguide to match two waveguides with different geometric cross-sectional parameters. From left to right, the number of modes supported by the multimode bus waveguide decreases by 1. The mode multiplexer part, except for the TE0 mode signal, consists of 5 multimode bus waveguides of different widths and a single-mode access waveguide with a width of 450nm. When the PCM of the upper arms of the 6 non-volatile optical switch units of the mode demultiplexer part are all amorphous (a-Sb2Se3), then the PCM-Si single-mode hybrid waveguides in the multiple upper arms are The fundamental mode signal is transmitted forward and passes through the upper multiplexer part to realize the conversion of the fundamental mode to the high-order mode, and is output from the O1 port; on the contrary, when the PCM of the lower arms of the 6 non-volatile optical switch units of the mode demultiplexer part are all a-Sb2Se3, the fundamental mode signals in the multiple lower arm PCM-Si single-mode hybrid waveguides are transmitted forward and pass through the lower multiplexer part to realize the conversion of the fundamental mode to the high-order mode, and are output from the O2 port; when the PCM of the upper and lower arms of the 6 non-volatile optical switch units of the mode demultiplexer part are all a-Sb2Se3, the fundamental mode signals in the multiple upper and lower arm PCM-Si single-mode hybrid waveguides are transmitted forward and pass through the upper and lower multiplexer parts to realize the conversion of the fundamental mode to the high-order mode, and 1 / 2 of the optical power is output from the O1 and O2 ports respectively, that is, the state of the PCM controls the output of the mode signal from different ports and the output transmittance; The expandable 1×3 non-volatile optical switch unit includes a multimode bus waveguide located in the middle and an upper arm PCM-Si single-mode hybrid waveguide and a lower arm PCM-Si single-mode hybrid waveguide respectively arranged on both sides of the multimode bus waveguide; the upper arm PCM-Si single-mode hybrid waveguide and the lower arm PCM-Si single-mode hybrid waveguide are provided with a phase change material Sb2Se3 film of equal length and thickness, and the Sb2Se3 films on the upper and lower arm PCM-Si single-mode hybrid waveguides can be switched between a crystalline state (c-Sb2Se3) and an amorphous state (a-Sb2Se3); the multimode bus waveguide and the PCM-Si single-mode hybrid waveguide containing a-Sb2Se3 achieve phase matching conditions to convert the high-order mode signal input from the input end of the multimode bus waveguide into a fundamental mode signal for data transmission, while the mode signal that does not meet the phase matching conditions continues to be transmitted forward, from O 12 Port output; when the upper arm PCM-Si waveguide Sb2Se3 is amorphous, the fundamental mode signal in the single-mode hybrid waveguide from O 11 Port output; when the Sb2Se3 of the lower arm PCM-Si waveguide is amorphous, the fundamental mode signal in the single-mode hybrid waveguide is from O 13 Port output.

2. The non-volatile programmable 1×2 multimode optical switch according to claim 1, wherein: The upper and lower arm PCM-Si single-mode hybrid waveguides can control the state of the Sb2Se3 thin film through laser pulse irradiation to achieve two different functions: light splitting and light switching.

3. The non-volatile programmable 1×2 multimode optical switch according to claim 1, wherein: The coupling region between the upper and lower arm PCM-Si single-mode hybrid waveguides and the multimode bus waveguide is composed of a PCM-Si hybrid waveguide of equal length to the coupling length; the PCM-Si hybrid waveguide in the coupling region is a 450nm wide and 220nm thick PCM-Si hybrid waveguide, including a Sb2Se3 thin film of the same width and 100nm thickness and a silicon waveguide of the same width and 120nm thickness.

4. The non-volatile programmable 1×2 multimode optical switch according to claim 1, wherein: The width of the multimode bus waveguide is adjustable; the coupling gaps between the upper and lower arm PCM-Si hybrid waveguides and the middle multimode bus waveguide are both 0.1 μm.

5. The non-volatile programmable 1×2 multimode optical switch according to claim 1, wherein: The phase change material Sb2Se3 is doped with GeTe.

6. The non-volatile programmable 1×2 multimode optical switch according to claim 1, wherein: The first-level multimode bus waveguide supports a total of 6 modes, namely: TE5, TE4, TE3, TE2, TE1, TE0; the second-level multimode bus waveguide supports a total of 5 modes, namely: TE4, TE3, TE2, TE1, TE0; the third-level multimode bus waveguide supports a total of 4 modes, namely: TE3, TE2, TE1, TE0; the fourth-level multimode bus waveguide supports a total of 3 modes, namely: TE2, TE1, TE0; the fifth-level multimode bus waveguide supports a total of 2 modes, namely: TE1, TE0; the sixth-level bus waveguide only supports 1 mode, namely: TE0.

7. The non-volatile programmable 1×2 multimode optical switch according to claim 6, wherein: The widths of the first to sixth level multimode bus waveguides are 2.40μm, 1.99μm, 1.57μm, 1.17μm, 0.77μm, and 0.37μm, respectively; the optimal coupling length and Sb2Se3 length are 13.7μm, 12.8μm, 11.4μm, 9.8μm, 8.0μm, and 5.3μm, respectively; the upper and lower floating ranges of the multimode bus waveguide width and coupling length are approximately 0.01μm and 0.02μm, respectively.

Images