A 1×4 thermo-optic switch based on silica / polymer hybrid waveguides and its manufacturing method

Abstract

A multimode interferometer cascaded optical switch based on a silica / polymer hybrid waveguide and a preparation method thereof belong to the technical field of silica / polymer hybrid waveguide optical integrated chips. It consists of an Si substrate, an SiO2 lower cladding layer, a polymer core layer, a polymer upper cladding layer, and metal electrodes. The polymer core layer is coated within the polymer upper cladding layer. Along the light transmission direction, the polymer core layer is composed of an input straight waveguide, a 1×4 multimode interferometer, four input S-bent waveguides, two input phase shifters, four modulation arm waveguides, two output phase shifters, four output S-bent waveguides, a 4×4 multimode interferometer, and four output straight waveguides. When a modulation voltage is applied to the metal electrodes, the signal light can be output from the four output straight waveguides respectively, realizing the switching function of four channels. The 1×4 thermo-optic switch of the present invention realizes the free switching of four channels, and is more flexible in switch ports and more compact in structure compared with traditional 1×2 and 2×2 switches.

Description

Technical Field

[0001] The present invention belongs to the technical field of silica / polymer hybrid waveguide optical integrated chips, and particularly relates to a multimode interferometer cascaded optical switch based on a silica / polymer hybrid waveguide and a preparation method thereof. Background Art

[0002] An optical switch is an important device for optical switching. With the rapid growth of the demand for information volume in the information society, the number of interactive ports for optical switching has increased rapidly. At this time, a higher demand is placed on the power consumption of the optical switch. The common silica planar waveguide (Plannar Lightwave Circuit, PLC) devices on the market now have advantages such as low loss, good stability, and high coupling efficiency. However, due to the low thermo-optic coefficient of the silica material (1.19×10 -5 K -1 ), when preparing active devices, a relatively high power consumption is often required, often dozens of milliwatts. At the same time, since the silica waveguide is prepared by doping to form the core layer, the refractive index difference between the core and the cladding is small, Δn = 0.75%, resulting in a relatively large size of the prepared device and difficulty in achieving large-scale integration. The polymer material has a thermo-optic coefficient that is an order of magnitude higher than that of silica (-1.86×10 -4 K -1 ), and is suitable for making active photonic chips such as thermo-optic switches and tunable filters. At the same time, the polymer as the core layer often has a higher refractive index. With silica as the lower cladding, the refractive index difference can reach Δn = 2.5% - 10%, making the device size and bending radius much smaller than those of silica-based planar optical waveguide devices, and it is easy to realize the preparation of large-scale photonic integrated devices.

[0003] In the commonly used optical switches, the mainstream 1×2 and 2×2 low-port switching systems can only achieve switching between two ports, and the integration degree and chip utilization rate are not high. Summary of the Invention

[0004] In order to solve the problems described in the background art, the present invention proposes a 1×4 thermo-optic switch based on a silica / polymer hybrid waveguide and a preparation method thereof.

[0005] As Figure 1 and Figure 4 shown, a 1×4 thermo-optic switch based on a silica / polymer hybrid waveguide according to the present invention is composed of an Si substrate (1), SiO 2 lower cladding (2), polymer core layer (3), polymer upper cladding (4), and metal electrode (5) from bottom to top; the polymer core layer (3) and the polymer upper cladding (4) are located on SiO 2Above the lower cladding layer (2), and the polymer core layer (3) is encapsulated in the polymer upper cladding layer (4); along the optical transmission direction, the polymer core layer is composed of an input straight waveguide (101), a 1×4 multimode interferometer (300), a first input S-bend waveguide (102), a second input S-bend waveguide (103), a third input S-bend waveguide (103’), a fourth input S-bend waveguide (102’), a first input phase shifter (104), a second input phase shifter (104’), a first modulation arm waveguide (105), a second modulation arm waveguide (106), a third modulation arm waveguide (107), a fourth modulation arm waveguide (108), a first output phase shifter (109), a second output phase shifter (109’), a first output S-bend waveguide (110), a second output S-bend waveguide (111), a third output S-bend waveguide (111’), a fourth output S-bend waveguide (110’), a 4×4 multimode interferometer (400), a first output straight waveguide (112), a second output straight waveguide (113), a third output straight waveguide (114), and a fourth output straight waveguide (115).

[0006] According to whether the interference is general, the self-imaging law of the multimode interferometer can be divided into two categories: general interference and restricted interference. Among them, general interference means that all modes in the multimode waveguide are excited, that is, there is no restriction on the excited modes; restricted interference means that some modes in the multimode waveguide are not excited, and thus some other self-imaging laws are obtained. In this sense, restricted interference is just a special type of general interference. According to the different modes with restricted excitation, restricted interference can be further divided into two categories, namely: paired interference and symmetric interference. Symmetric interference excites odd-order modes, while paired interference excites even-order modes. The multimode interferometer with symmetric interference can only be input at the input position on the symmetry axis parallel to the long side of the multimode interferometer, and it will generate the first N-fold image point at where n eff is the effective refractive index of the waveguide, W MMI is the effective width of the multimode interferometer, and λ is the wavelength of the signal light; the multimode interferometer with paired interference has two symmetric input positions with respect to the symmetry axis parallel to the long side of the multimode interferometer, located on both sides of the symmetry axis respectively, and it will generate the first N-fold image point at ; when input at any other position except the above three positions, a multimode interferometer with general interference will be generated, and it will generate the first N-fold image point at In a 1×4 thermo-optic switch based on a silica / polymer hybrid waveguide according to the present invention, the 1×4 multimode interferometer (300) adopts the interference type of symmetric interference, and the 4×4 multimode interferometer (400) adopts the interference type of general interference.

[0007] Such asFigure 2 As shown, a 1×4 thermo-optic switch based on a silica / polymer hybrid waveguide according to the present invention, wherein the 1×4 multimode interferometer (300) is composed of a first input Taper waveguide (310), a first multimode interference region (320) and four first output Taper waveguides (330). The first input Taper waveguide (310) and the first output Taper waveguide (330) have the same structure and are symmetrically arranged at the input end and the output end of the first multimode interference region (320), and are both formed by cascading two sine-shaped Taper waveguides and a rectangular three-segment Taper waveguide (311, 312 and 313, 331, 332 and 333); the signal light transmitted in the optical fiber is coupled into the input straight waveguide (101), and after passing through the first input Taper waveguide (310), it is transmitted to the first multimode interference region (320) to generate a self-imaging effect. When the transmission length of the signal light in the multimode interference region is, the multimode interferometer based on symmetric interference will generate four-fold images for the first time, where n eff is the effective refractive index of the waveguide, W MMI is the effective width of the multimode interferometer, and λ is the wavelength of the signal light; after generating four signal lights with the same light intensity from the position where the four-fold images are generated for the first time, they are respectively output into the four first output Taper waveguides (330), and then transmitted to the first modulation arm waveguide (105), the second modulation arm waveguide (106), the third modulation arm waveguide (107) and the fourth modulation arm waveguide (108) through the first input S-bending waveguide (102), the first input phase shifter (104), the second input S-bending waveguide (103), the third input S-bending waveguide (103’), the fourth input S-bending waveguide (102’) and the second input phase shifter (104’); by designing the size parameters of the first input Taper waveguide (310) and the first output Taper waveguide (330), the mode output at the self-imaging point of the multimode interference region can be made more matched with the mode transmission in the Taper, reducing the insertion loss introduced by the multimode interferometer; by adopting the form of cascading Taper waveguides, a more uniform splitting ratio can be achieved, thereby improving the extinction ratio of the switch.

[0008] As Figure 5As shown in the figure, a 1×4 thermo-optic switch based on a silica / polymer hybrid waveguide according to the present invention, wherein the 4×4 multimode interferometer (400) is composed of 4 second input Taper waveguides (410), a second multimode interference region (420), and 4 second output Taper waveguides (430). The second input Taper waveguides (410) and the second output Taper waveguides (430) have the same structure and are symmetrically arranged at the input end and the output end of the second multimode interference region (420). They are each formed by cascading two sine-shaped Taper waveguides and a rectangular 3-segment Taper waveguide (411, 412, and 413, 431, 432, and 433). The output signal light passing through the first modulation arm waveguide (105), the first output phase shifter (109), the second modulation arm waveguide (106), the third modulation arm waveguide (107), the fourth modulation arm waveguide (108), and the second output phase shifter (109') respectively enters the 4 second input Taper waveguides (410) through the first output S-bending waveguide (110), the second output S-bending waveguide (111), the third output S-bending waveguide (111'), and the fourth output S-bending waveguide (110'), and then enters the second multimode interference region (420) to produce a self-imaging effect. When the transmission length of the signal light in the multimode interference region is such that, based on general interference, the multimode interferometer will first produce four images, where n eff is the effective refractive index of the waveguide, W MMI is the effective width of the multimode interferometer, and λ is the wavelength of the signal light; after generating four signal lights with the same light intensity (for the details of the phase, see the imaginary part of the transmission matrix of the 4×4 multimode interferometer) from the position where the four images are first generated, they are respectively output to the 4 output Taper waveguides (430) (one signal light generates 1 quadruple image, four channels are input simultaneously, and each output channel has the superposition of four wave functions), and then are respectively output from the first output straight waveguide (112), the second output straight waveguide (113), the third output straight waveguide (114), and the fourth output straight waveguide (115); by designing the size parameters of the second input Taper waveguide (410) and the second output Taper waveguide (430), the mode output at the self-imaging point of the multimode interference region can be made more matched with the mode transmission in the Taper, reducing the insertion loss introduced by the multimode interferometer. By adopting the form of cascading the Taper, a more uniform splitting ratio can be achieved, thereby improving the extinction ratio of the switch. The signal light is input from the input straight waveguide (101). When passing through the 1×4 multimode interferometer (300), the input light will be split into four signal lights with the same amplitude. The transmission matrix of its 1×4 multimode interferometer is

[0009]

[0010] Four input light beams are connected to a 4×4 multimode interferometer through the middle connection part and modulation part. The transmission matrix of the 4×4 multimode interferometer is:

[0011]

[0012] And the transmission matrix of its connection part is:

[0013]

[0014] Where is the phase change of the nth connection part.

[0015] According to the transmission matrix method, the transmission matrix of the entire switch is:

[0016] T = T 4×4 T M T 1×4 (4)

[0017] From formulas (1) to (4), it can be obtained that:

[0018]

[0019] Through the optimized design of the S-bend waveguide and the phase shifter (each S-bend waveguide is composed of arcs with the same horizontal offset and radius at both ends but different rotation directions, and its input and output positions are horizontal. The purpose is to increase the spacing between the modulation arms to reduce the influence of thermal crosstalk. The radii of the large and small S-bends are 2500 μm, and the horizontal offsets are 30 μm and 70 μm respectively), it can be made that at the wavelength of 1550 nm, the phase of the central two connection parts, that is, the transmission path without the phase shifter, leading the phase of the transmission path with the phase shifter when reaching the 4×4 multimode interferometer (where the role of the phase shifter is to offset the drastic change of the phase with the wavelength brought by the large S-bend, so as to avoid the influence of high crosstalk when the wavelength deviates from the central wavelength within a short range); the light in the optical fiber is coupled into the input straight waveguide (101), transmitted into the 1×4 multimode interferometer (300), and the light is evenly divided into four optical signals with equal power, and respectively transmitted to the four modulation arm waveguides (105, 106, 107, and 108) through the first input S-bend waveguide (102) and the first input phase shifter (104), the second input S-bend waveguide (103), the third input S-bend waveguide (103’), the fourth input S-bend waveguide (102’) and the second input phase shifter (104’) (at the wavelength of 1550 nm, the phase of the transmission path without the phase shifter leads the phase of the transmission path with the phase shifter );The four modulation arm waveguides are connected to the 4×4 multimode interferometer (400) through the first output phase shifter (109), the first output S-bend waveguide (110), the second output S-bend waveguide (111), the third output S-bend waveguide (111’), the second output phase shifter (109’), and the fourth output S-bend waveguide (110’). When no modulation voltage is applied to the metal electrodes, the four output straight waveguides (112, 113, 114, 115) output equal power.

[0020] When voltages are simultaneously applied to the second metal electrode (202) and the third metal electrode (203) to change the electric power applied to them, the second modulation arm waveguide (106) and the third modulation arm waveguide (107) are heated. Due to the thermo-optic effect, the refractive index of the polymer modulation arm waveguide changes with the increase in temperature, thereby causing a change in the phase of the light. Denote the phase difference before and after modulation as where n is the nth modulation arm. When the light will output from output port ①; while when the light will output from port ④; Voltages are simultaneously applied to the first metal electrode (201) and the fourth metal electrode (204) to change the electric power applied to them, and the first modulation arm waveguide (105) and the fourth modulation arm waveguide (108) are heated. When the light will output from port ②; while when the light will output from port ③. Through the above four modulation methods, the switching functions of four channels are realized.

[0021] The effective refractive index difference between the phase shifter and the straight waveguide changes with the wavelength, which is opposite to the trend of the effective refractive index difference of the large S-bend and the small S-bend changing with the wavelength. Therefore, the phase shifter can offset a part of the phase difference change with the wavelength introduced by the S-bends with different radii on the path, thereby expanding the working bandwidth of the switch. At the same time, compared with the straight waveguide, the phase shifter will bring a phase change, so the phase difference that needs to be modulated will also be different.

[0022] The polymer upper cladding material is one of polymethyl methacrylate (PMMA), polyethylene (PE), polyester (PET), polystyrene (PS), and EpoClad. In this patent, PMMA is used, and its refractive index is 1.476 at a wavelength of 1550 nm.

[0023] The polymer core layer material is one of the polymer materials SU-8 2002, SU-8 2005, and EpoCore with a negative thermo-optic coefficient. In this patent, SU-8 2002 is used, and its refractive index is 1.5802 at a wavelength of 1550 nm.

[0024] The described metal electrode material can be an alloy material composed of one or more of gold, silver, copper, and aluminum.

[0025] The preparation method of the 1×4 optical switch based on the polymer / silica hybrid waveguide according to the present invention is as follows:

[0026] 1) On the silicon wafer substrate (1), a layer of silica with a dense structure and a thickness of 10 - 20 μm is grown by thermal oxidation as the lower cladding (2);

[0027] 2) The polymer core layer material is spin-coated on the silica lower cladding (2) using a vacuum spin coater, pre-baked to remove the excess solvent in the core layer, and then naturally cooled to obtain the polymer core layer film (31);

[0028] 3) Using ultraviolet exposure, development, and post-baking, the pattern designed on the mask I and identical or complementary to the polymer core layer structure is transferred to the polymer core layer film (31), so as to obtain the polymer core layer (3) on the silica lower cladding (2). Except for the multimode interferometer, the thickness and width of the remaining parts of the polymer core layer are the same, being 2 - 5 μm respectively;

[0029] 4) The polymer upper cladding material is spin-coated on the silica lower cladding (2) and the polymer core layer (3) using a vacuum spin coater, and after baking treatment, it is naturally cooled to obtain the polymer upper cladding (4) with a thickness of 3 - 6 μm;

[0030] 5) A metal thin film (51) with a thickness of 80 - 120 nm is deposited on the polymer upper cladding (4) using a coating machine;

[0031] 6) A photoresist layer (61) is spin-coated on the metal surface using a vacuum spin coater, and after pre-baking to remove the photoresist solvent, it is naturally cooled and cured;

[0032] 7) Through ultraviolet exposure, development, and post-baking, the pattern on the mask II identical to the metal electrode structure is transferred to the photoresist layer (61). After development, the film is hardened and naturally cooled to obtain the photoresist pattern (62) identical to the metal electrode structure;

[0033] 8) Corrode the metal covered by the photoresist-free pattern (62) using a metal corrosion solution corresponding to the metal to obtain a metal electrode (5), and then remove the photoresist layer (61) covering the metal electrode (5); the metal electrode (5) is located above the polymer upper cladding (4) at the position directly above the first modulation arm waveguide, the second modulation arm waveguide, the third modulation arm waveguide, and the fourth modulation arm waveguide, and the symmetry center of the metal electrode in the width and length directions is aligned with the symmetry center of the modulation arm waveguide in the width and length directions. The length of the metal electrode (5) is equal to the length of the modulation arm waveguide, and the width of the metal electrode (5) is greater than the width of the modulation arm waveguide, thereby fabricating a 1×4 optical switch based on a silica / polymer hybrid waveguide.

[0034] Compared with the prior art, the innovations of the present invention are as follows:

[0035] 1. The waveguide is an organic-inorganic hybrid integrated composite waveguide structure, using silica as the lower cladding, which reduces the device loss.

[0036] 2. Adopting an organic-inorganic hybrid integrated composite waveguide structure with a polymer as the core layer, the refractive index difference Δn between the core and the cladding is 2.5% - 10%, which can achieve more compact end face dimensions and bending radii, and fabricate large-scale, low-power, and fast planar optical waveguide integrated circuits;

[0037] 3. The waveguide is an organic-inorganic composite waveguide structure, and the device can be fabricated by simple contact exposure, with extremely low processing costs;

[0038] 4. Using a polymer material as the modulation arm, the power consumption required to achieve the switching function is small;

[0039] 5. Adopting a sinusoidal Taper structure on the MMI reduces the device loss;

[0040] 6. Adopting a Taper structure with multiple Taper cascades can make the power distribution of the power divider more uniform and achieve a high extinction ratio;

[0041] 7. Adopting a phase shifter structure design on the modulation arm to compensate for the phase difference shift with wavelength caused by the S-bend waveguide can effectively broaden the working bandwidth;

[0042] 8. The 1×4 thermo-optic switch realizes the free switching of 4 channels, which is more flexible and has a more compact structure compared to traditional 1×2 and 2×2 switches. Description of the Drawings

[0043] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0044] Figure 1 : Schematic diagram of the 1×4 thermo-optic switch structure of the present invention;

[0045] Figure 2 : Schematic diagram of the structure of the 1×4 multimode interferometer (300) of the present invention;

[0046] Figure 3 : Transmission spectrum diagram of the 1×4 multimode interferometer (300) of the present invention;

[0047] Figure 4 : Cross-sectional schematic diagram of the waveguide position of the modulation arm of the 1×4 thermo-optic switch of the present invention;

[0048] Figure 5 : Schematic diagram of the structure of the 4×4 multimode interferometer (400) of the present invention;

[0049] Figure 6 : Transmission spectrum diagram of the 4×4 multimode interferometer (400) of the present invention;

[0050] Figure 7 : Preparation process of the 1×4 thermo-optic switch of the present invention;

[0051] Figure 8 : Optical field simulation diagrams (a) when the 1×4 thermo-optic switch of the present invention applies a voltage to the second modulation arm waveguide (106) to generate a temperature change of 3.7K and to the third modulation arm waveguide (107) to generate a temperature change of 1.5K; optical field simulation diagrams (b) when a voltage is applied to the first modulation arm waveguide (105) to generate a temperature change of 3.9K and to the fourth modulation arm waveguide (108) to generate a temperature change of 1.7K; optical field simulation diagrams (c) when a voltage is applied to the first modulation arm waveguide (105) to generate a temperature change of 1.7K and to the fourth modulation arm waveguide (108) to generate a temperature change of 3.9K; optical field simulation diagrams (d) when a voltage is applied to the second modulation arm waveguide (106) to generate a temperature change of 1.5K and to the third modulation arm waveguide (107) to generate a temperature change of 3.7K;

[0052] Figure 9 : Transmission spectrum diagram when the 1×4 thermo-optic switch of the present invention applies 3.7K to the second modulation arm waveguide (106) and 1.5K to the third modulation arm waveguide (107);

[0053] Figure 10 : Transmission spectrum diagram of the 1×4 thermo-optic switch of the present invention when a temperature change of 3.9K is applied to the first modulation arm waveguide (105) and a temperature change of 1.7K is applied to the fourth modulation arm waveguide (108);

[0054] Figure 11 : Transmission spectrum diagram of the 1×4 thermo-optic switch of the present invention when a temperature change of 1.7K is applied to the first modulation arm waveguide (105) and a temperature change of 3.9K is applied to the fourth modulation arm waveguide (108);

[0055] Figure 12 : Transmission spectrum diagram of the 1×4 thermo-optic switch of the present invention when a temperature change of 1.5K is applied to the second modulation arm waveguide (106) and a temperature change of 3.7K is applied to the third modulation arm waveguide (107); Detailed implementation mode

[0056] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0057] Embodiment 1

[0058] As shown in the attached Figure 1 figure, it is a structural schematic diagram of a 1×4 thermo-optic switch. It is composed of an input straight waveguide (101), a 1×4 multimode interferometer (300), a first input S-bending waveguide (102), a second input S-bending waveguide (102'), a third input S-bending waveguide (103), a fourth input S-bending waveguide (103'), a first input phase shifter (104), a second input phase shifter (104'), a first modulation arm waveguide (105), a second modulation arm waveguide (106), a third modulation arm waveguide (107), a fourth modulation arm waveguide (108), a first output phase shifter (109), a second output phase shifter (109'), a first output S-bending waveguide (110), a second output S-bending waveguide (110'), a third output S-bending waveguide (111), a fourth output S-bending waveguide (111'), a 4×4 multimode interferometer (400), a first output straight waveguide (112), a second output straight waveguide (113), a third output straight waveguide (114), and a fourth output straight waveguide (115). Considering the relationship between modulation efficiency and size, the length of the modulation arm is set to 2000μm. Each phase shifter is composed of a waveguide that becomes wider first and then narrower, with the width of the wide part being 6μm, the width of the narrow part being 3μm, and the length being 820μm; the radii of the large and small S-bends are 2500μm, and the horizontal offset amounts are 30μm and 70μm respectively.

[0059] The signal light of the optical fiber is coupled into the input straight waveguide (101) and transmitted to the 1×4 multimode interferometer (300). The input optical signal is evenly divided into four optical signals with equal power through the 1×4 multimode interferometer (300). The signal light is transmitted to four mutually parallel modulation arm waveguides (105, 106, 107, and 108) through the connected first input S-bending waveguide (102), first input phase shifter (104), second input S-bending waveguide (103), third input S-bending waveguide (103’), fourth input S-bending waveguide (102’), and second input phase shifter (104’). Then, the signal light is transmitted to the 4×4 multimode interferometer (400) through the first output S-bending waveguide (110), first output phase shifter (109), second output S-bending waveguide (111), third output S-bending waveguide (111’), fourth output S-bending waveguide (110’), and second input phase shifter (109’), and finally output from the four output straight waveguides (112, 113, 114, and 115). When no modulation voltage is applied to the metal electrodes on the four modulation arms, the light is output from the four output straight waveguides (112, 113, 114, and 115) with nearly equal power. When voltages are simultaneously applied to the second metal electrode (202) and the third metal electrode (203) on the second modulation arm waveguide (106) and the third modulation arm waveguide (107), the electric power loaded on the second metal electrode (202) and the third metal electrode (203) will be changed, thereby heating the waveguide. Due to the thermo-optic effect, when the external environmental temperature changes, the refractive index of the material will change, resulting in a change in the phase of the light propagating therein. Denote the phase difference after heating and before heating as where n is the nth modulation arm. When the second modulation arm waveguide (106) and the third modulation arm waveguide (107) are heated to make the light will be output from output channel ①. And when making the light will be output from output channel ④. Similarly, using the thermo-optic effect, voltages are simultaneously applied to the first metal electrode (201) and the fourth metal electrode (204) on the first modulation arm waveguide (105) and the fourth modulation arm waveguide (108) to heat these two modulation arms. When the heating makes the light will be output from output channel ②. And when the light will be output from output channel ③. Through the above modulation method, the switching functions of the four channels are realized.

[0060] As shown in the appendix Figure 2As shown in the figure, it is a schematic structural diagram of a 1×4 multimode interferometer (300). From left to right, there are 1 first input Taper waveguide (310), 1 first multimode interference region (320), and 4 second output Taper waveguides (330). In this example, considering the device size and the actual fabrication difficulty comprehensively, both the 1 first input Taper waveguide (310) and the 4 second output Taper waveguides (330) of the 1×4 multimode interferometer adopt the form of cascading 2 equal-length sine-shaped Taper waveguides and 1 rectangular-shaped 3-segment Taper waveguide. The total length of the cascaded Taper waveguides is 132.2 μm. The starting widths of the cascaded Taper waveguides (311, 312, and 313, 331, 332, and 333) from far to near the first multimode interference region (320) are 3 μm, 6.8 μm, and 7 μm respectively, and the termination width of the third-segment cascaded Taper waveguide (313 or 333) is 7 μm. The width of the first multimode interference region (320) is 40 μm and the length is 410 μm. The center spacings of the 4 second output Taper waveguides (330) from top to bottom are 10.18 μm, 10.24 μm, and 10.18 μm respectively. Among them, the sine Taper uses the difference between half of the wide-side width and half of the narrow-side width as the amplitude of the sine, and experiences the first 1 / 4 sine period, and the Taper width changes from the narrow-side width to the wide-side width.

[0061] As shown in the appendix Figure 3 As shown in the figure, it is the transmission spectrum diagram of the 1×4 multimode interferometer (300). From 1500 nm to 1630 nm wavelength, the outputs of the four output channels are all of good uniformity. At 1550 nm, the losses are all -6.05 dB.

[0062] As shown in the appendix Figure 4As described, it is a cross-sectional view of an inorganic / polymer organic material hybrid waveguide. From bottom to top, there are a silicon substrate (1), a silica lower cladding (2), a polymer core layer (3), a polymer upper cladding (4), and a metal electrode (5). In this example, a dense layer of silica with a refractive index of 1.4448 is grown by thermal oxidation. The polymer core layer material (3) used is a negative photoresist of SU-8 2002, with a refractive index of 1.5802 at a wavelength of 1550 nm. The polymer upper cladding material used is polymethyl methacrylate (PMMA), with a refractive index of 1.47606 at a wavelength of 1550 nm. In this example, in order to reduce crosstalk between modes and at the same time reduce the difficulty of process manufacturing, we design the width and thickness dimensions of the input straight waveguide (101), the first input S-bent waveguide (102), the second input S-bent waveguide (103), the third input S-bent waveguide (103’), the fourth input S-bent waveguide (102’), the first input phase shifter (104), the second input phase shifter (104’), the first modulation arm waveguide (105), the second modulation arm waveguide (106), the third modulation arm waveguide (107), the fourth modulation arm waveguide (108), the first output phase shifter (109), the second output phase shifter (109’), the first output S-bent waveguide (110), the second output S-bent waveguide (111), the third output S-bent waveguide (111’), the fourth output S-bent waveguide (110’), the first output straight waveguide (112), the second output straight waveguide (113), the third output straight waveguide (114), and the fourth output straight waveguide (115) to be 3 μm × 3 μm.

[0063] As attached Figure 5As shown in the figure, it is a schematic structural diagram of a 4×4 multimode interferometer (400). From left to right, there are 4 second input Taper waveguides (410), 1 second multimode interference region (420), and 4 second output Taper waveguides (430). In this example, considering the device size and the actual fabrication difficulty, both the 4 first input Taper waveguides (410) and the 4 second output Taper waveguides (430) of the 4×4 multimode interferometer adopt the form of cascading 2 equi-length sinusoidal Taper waveguides and 1 rectangular 3-segment Taper waveguide. The total length of the cascaded Taper waveguides is 132.2 μm. The starting widths of the cascaded Taper waveguides (411, 412, and 413, 431, 432, and 433) from far to near the second multimode interference region (420) are 3 μm, 6.8 μm, and 7 μm respectively, and the termination width of the third-segment cascaded Taper waveguide (413 or 433) is 7 μm; the width of the second multimode interference region (420) is 40 μm and the length is 1690 μm. The center-to-center spacings of the 4 second input Taper waveguides (410) and the 4 second output Taper waveguides (430) from top to bottom are 10.18 μm, 10.24 μm, and 10.18 μm respectively.

[0064] As shown in the appendix Figure 6 As shown in the figure, it is the transmission spectrum diagram of the 4×4 multimode interferometer (400). Throughout the C band from 1535 nm to 1565 nm, the outputs of the four output channels all have good uniformity. At 1550 nm, the losses of the four output ports are all less than -6.1 dB.

[0065] As shown in the appendix Figure 7 As shown in the figure, the preparation method of the 1×4 thermo-optic switch based on a silica / polymer hybrid waveguide according to the present invention is as follows:

[0066] 1) On a silicon wafer substrate (1), a dense 15-μm-thick silica undercladding (2) is grown by thermal oxidation.

[0067] 2) Spin-coat SU-8 2002 photoresist on the surface of the silica undercladding using a vacuum spin coater. First, pre-bake at 60 °C for 10 minutes and at 90 °C for 20 minutes and then cool and cure naturally. Then, by controlling the rotation speed at 600 revolutions per minute and the spin-coating time at 20 s, a 3-μm-thick SU-8 photoresist layer (31) is formed.

[0068] 3) Place the device obtained in step 2) under a 365-nm ultraviolet light lithography machine with a light power of 23 mW / cm 2, For alignment lithography, the structure and shape of the mask I used are complementary to the structure and shape of the SU-8 core layer waveguide to be fabricated. The exposure time is 3.5 s. Then, post-baking is carried out at 65 °C for 10 minutes and at 95 °C for 20 minutes. After cooling to room temperature, it is developed in a PGMEA (Propyleneglygol-monomethylether-acetate) developer, and then rinsed in isopropyl alcohol to remove the remaining glue, and the reaction solution is washed with deionized water; then, at 120 °C, the film is hardened for 30 minutes to form the SU-8 2002 core layer. For polarization insensitivity, the width and thickness of the strip-shaped SU-8 core layer (3) are the same, both being 3 μm;

[0069] 4) Use a vacuum spin coater to spin coat a PMMA polymer upper cladding (4) with a thickness of 7 μm on the chip, where the thickness of the PMMA polymer upper cladding (4) on the SU-8 core layer waveguide is 4 μm. Then, curing is carried out at 120 °C for 30 minutes and cooled to room temperature;

[0070] 5) Evaporate a metal Al film (51) with a thickness of 100 nm on the PMMA polymer upper cladding (4);

[0071] 6) Adopt a spin coating process to spin coat a positive photoresist BP212 (61) with a thickness of 1.5 μm on the metal Al film (51), and bake at 87 °C for 20 minutes;

[0072] 7) Place the sample under an ultraviolet lithography machine, and closely contact it with the mask II for alignment lithography. The structure and shape of the mask II used are the same as the structure and shape of the metal electrode to be fabricated. Expose for 2 s, remove the mask II, and after developing with a NaOH solution with a mass concentration of 5 wt‰, bake at 90 °C for 20 minutes to transfer the pattern on the mask II that is the same as the structure of the metal electrode to be fabricated to the BP212 photoresist layer (62);

[0073] 8) Use a NaOH solution with a mass concentration of 5 wt‰ to remove the metal Al film not masked by the photoresist, expose again for 10 s, and use an ethanol solution to remove the remaining BP212 to expose the Al metal electrode (5), thereby fabricating a 1×4 thermo-optic switch based on a polymer / silica hybrid waveguide. The width of each metal modulation electrode (201, 202, 203, 204) is 21 μm, the distance between the first metal modulation electrode (201) and the second metal modulation electrode (202) is 29 μm, the distance between the second metal modulation electrode (202) and the third metal modulation electrode (203) is 49 μm, and the distance between the third metal modulation electrode (203) and the fourth metal modulation electrode (204) is 29 μm.

[0074] As shown in the appendix Figure 8As shown in (a), when a voltage is applied to the second modulation arm waveguide (106) to generate a temperature change of 3.7K and a voltage is applied to the third modulation arm waveguide (107) to generate a temperature change of 1.5K, the optical signal will be output from output channel ①; as shown in the appendix Figure 8 As shown in (b), when a voltage is applied to the first modulation arm waveguide (105) to generate a temperature change of 1.7K and a voltage is applied to the fourth modulation arm waveguide (108) to generate a temperature change of 3.9K, the optical signal will be output from output channel ②; as shown in the appendix Figure 8 As shown in (c), when a voltage of 3.9K is applied to the first modulation arm waveguide (105) and a voltage is applied to the fourth modulation arm waveguide (108) to generate a temperature change of 1.7K, the optical signal will be output from output channel ③; as shown in the appendix Figure 8 As shown in (d), when a voltage is applied to the second modulation arm waveguide (106) to generate a temperature change of 1.5K and a voltage is applied to the third modulation arm waveguide (107) to generate a temperature change of 3.7K, the optical signal will be output from output channel ④.

[0075] As shown in the appendix Figure 9 As shown, the transmission spectrum diagram of the 1×4 thermo-optical switch of the present invention when a voltage is applied to the second modulation arm waveguide (106) to generate a temperature change of 3.7K and a voltage is applied to the third modulation arm waveguide (107) to generate a temperature change of 1.5K. It can be found that at wavelengths from 1535nm to 1565nm, the loss of output channel ① is less than 0.5dB and the crosstalk is less than -15dB.

[0076] As shown in the appendix Figure 10 As shown, the transmission spectrum diagram of the 1×4 thermo-optical switch of the present invention when a voltage is applied to the first modulation arm waveguide (105) to generate a temperature change of 1.7K and a voltage is applied to the fourth modulation arm waveguide (108) to generate a temperature change of 3.9K. It can be found that at wavelengths from 1535nm to 1565nm, the loss of output channel ② is less than 0.4dB and the crosstalk is less than -21dB.

[0077] As shown in the appendix Figure 11 As shown, the transmission spectrum diagram of the 1×4 thermo-optical switch of the present invention when a voltage is applied to the first modulation arm waveguide (105) to generate a temperature change of 3.9K and a voltage is applied to the fourth modulation arm waveguide (108) to generate a temperature change of 1.7K. It can be found that at wavelengths from 1535nm to 1565nm, the loss of output channel ③ is less than 0.4dB and the crosstalk is less than -21dB.

[0078] As shown in the appendix Figure 12As shown in the figure, the transmission spectrum diagram of the 1×4 thermo-optic switch of the present invention is obtained when a voltage is applied to the second modulation arm waveguide (106) to generate a temperature change of 1.5K and a voltage is applied to the third modulation arm waveguide (107) to generate a temperature change of 3.7K. It can be found that at wavelengths from 1535nm to 1565nm, the loss of output channel ④ is less than 0.5dB and the crosstalk is less than -15dB.

Claims

1. A 1×4 thermo-optic switch based on a silica / polymer hybrid waveguide, Characterized in that: From bottom to top, it is composed of an Si substrate (1), SiO 2 lower cladding (2), a polymer core layer (3), a polymer upper cladding (4), and a metal electrode (5); the polymer core layer (3) and the polymer upper cladding (4) are located above the SiO 2 lower cladding (2), and the polymer core layer (3) is coated in the polymer upper cladding (4); along the light transmission direction, the polymer core layer is composed of an input straight waveguide (101), a 1×4 multimode interferometer (300), a first input S-bend waveguide (102), a second input S-bend waveguide (103), a third input S-bend waveguide (103’), a fourth input S-bend waveguide (102’), a first input phase shifter (104), a second input phase shifter (104’), a first modulation arm waveguide (105), a second modulation arm waveguide (106), a third modulation arm waveguide (107), a fourth modulation arm waveguide (108), a first output phase shifter (109), a second output phase shifter (109’), a first output S-bend waveguide (110), a second output S-bend waveguide (111), a third output S-bend waveguide (111’), a fourth output S-bend waveguide (110’), a 4×4 multimode interferometer (400), a first output straight waveguide (112), a second output straight waveguide (113), a third output straight waveguide (114), and a fourth output straight waveguide (115); Among them, the 1×4 multimode interferometer (300) consists of a first input Taper waveguide (310), a first multimode interference region (320), and four first output Taper waveguides (330). The first input Taper waveguide (310) and the first output Taper waveguide (330) have the same structure and are symmetrically arranged at the input end and the output end of the first multimode interference region (320). They are both formed by cascading two sine-shaped Taper waveguides and a rectangular-shaped three-segment Taper waveguide (311, 312, and 313, 331, 332, and 333). The signal light transmitted in the optical fiber is coupled into the input straight waveguide (101), and after passing through the first input Taper waveguide (310), it is transmitted to the first multimode interference region (320) where the self-imaging effect occurs. When the transmission length of the signal light in the multimode interference region is, the multimode interferometer based on symmetric interference will generate four-fold images for the first time, where n eff is the effective refractive index of the waveguide, W MMI is the effective width of the multimode interferometer, and λ is the wavelength of the signal light. After generating four signal lights with the same light intensity from the position where the four-fold images are generated for the first time, they are respectively output to the four first output Taper waveguides (330), and then transmitted to the first modulation arm waveguide (105), the second modulation arm waveguide (106), the third modulation arm waveguide (107), and the fourth modulation arm waveguide (108) through the first input S-bending waveguide (102), the first input phase shifter (104), the second input S-bending waveguide (103), the third input S-bending waveguide (103’), the fourth input S-bending waveguide (102’), and the second input phase shifter (104’) respectively; The 4×4 multimode interferometer (400) consists of 4 second input Taper waveguides (410), a second multimode interference region (420), and 4 second output Taper waveguides (430). The second input Taper waveguides (410) and the second output Taper waveguides (430) have the same structure and are symmetrically arranged at the input end and the output end of the second multimode interference region (420). They are each formed by cascading two sine-shaped Taper waveguides and a rectangular-shaped 3-segment Taper waveguide (411, 412, and 413, 431, 432, and 433). The output signal lights passing through the first modulation arm waveguide (105), the first output phase shifter (109), the second modulation arm waveguide (106), the third modulation arm waveguide (107), the fourth modulation arm waveguide (108), and the second output phase shifter (109') are respectively input into the 4 second input Taper waveguides (410) after passing through the first output S-bending waveguide (110), the second output S-bending waveguide (111), the third output S-bending waveguide (111'), and the fourth output S-bending waveguide (110'), and then input into the second multimode interference region (420) to produce the self-imaging effect. When the transmission length of the signal light in the multimode interference region is such that, for a multimode interferometer based on general interference, a quadruple image is generated for the first time, where n eff is the effective refractive index of the waveguide, W MMI is the effective width of the multimode interferometer, and λ is the wavelength of the signal light. After generating four signal lights with the same light intensity from the position where the quadruple image is generated for the first time, they are respectively output into the 4 output Taper waveguides (430), and then respectively output from the first output straight waveguide (112), the second output straight waveguide (113), the third output straight waveguide (114), and the fourth output straight waveguide (115). When no modulation voltage is applied to the metal electrode, the four output straight waveguides (112, 113, 114, 115) output equal power. When a modulation voltage is applied to the metal electrode, the signal light can be output from the four output straight waveguides (112, 113, 114, 115) respectively, thereby realizing the switching function of four channels.

2. The 1×4 thermo-optic switch based on a silica / polymer hybrid waveguide according to claim 1, Characterized in that: Through the optimized design of the S-bending waveguide and the phase shifter, at a wavelength of 1550 nm, the phase of the signal light reaching the 4×4 multimode interferometer through the transmission path without the phase shifter leads the phase of the signal light reaching the 4×4 multimode interferometer through the transmission path with the phase shifter. When voltages are simultaneously applied to the second metal electrode (202) and the third metal electrode (203), causing the phase of the light to change, when this occurs, the light is output from the first output straight waveguide (112); while when this occurs, the light is output from the fourth output straight waveguide (115); when voltages are simultaneously applied to the first metal electrode (201) and the fourth metal electrode (204), causing the phase of the light to change, when this occurs, the light is output from the second output straight waveguide (113); when this occurs, the light is output from the third output straight waveguide (114).

3. The 1×4 thermo-optic switch based on a silica / polymer hybrid waveguide according to claim 1, Characterized in that: The polymer upper cladding material is one of polymethyl methacrylate, polyethylene, polyester, polystyrene, EpoClad.

4. The 1×4 thermo-optic switch based on a silica / polymer hybrid waveguide according to claim 1, Characterized in that: The polymer core material is one of SU-8 2002, SU-8 2005, EpoCore.

5. The 1×4 thermo-optic switch based on a silica / polymer hybrid waveguide according to claim 1, Characterized in that: The metal electrode material is an alloy composed of one or more of gold, silver, copper, and aluminum.

6. The 1×4 thermo-optic switch based on a silica / polymer hybrid waveguide according to claim 1, Characterized in that: The lengths of the modulation arm waveguides (105, 106, 107, and 108) are 2000 μm. Each phase shifter is composed of a waveguide that first narrows and then widens and then narrows again. The width of the wide part is 6 μm, the width of the narrow part is 3 μm, and the length is 820 μm. The total lengths of the first input Taper waveguide (310) and the 4 second output Taper waveguides (330) are 132.2 μm respectively. The starting widths of the cascaded Taper waveguides (311, 312, and 313, 331, 332, and 333) from far to near the first multimode interference region (320) are 3 μm, 6.8 μm, and 7 μm respectively. The termination width of the third-stage cascaded Taper waveguide (313 or 333) is 7 μm. The width of the first multimode interference region (320) is 40 μm and the length is 410 μm. The center spacings of the 4 second output Taper waveguides (330) from top to bottom are 10.18 μm, 10.24 μm, and 10.18 μm in sequence. Among them, for the sine Taper waveguide, the amplitude of the sine is the difference between half of the wide-side width and half of the narrow-side width, and it experiences the first 1 / 4 sine period, and the width of the Taper waveguide changes from the narrow-side width to the wide-side width. The widths and thicknesses of the input straight waveguide (101), the first input S-bend waveguide (102), the second input S-bend waveguide (103), the third input S-bend waveguide (103’), the fourth input S-bend waveguide (102’), the first input phase shifter (104), the second input phase shifter (104’), the first modulation arm waveguide (105), the second modulation arm waveguide (106), the third modulation arm waveguide (107), the fourth modulation arm waveguide (108), the first output phase shifter (109), the second output phase shifter (109’), the first output S-bend waveguide (110), the second output S-bend waveguide (111), the third output S-bend waveguide (111’), the fourth output S-bend waveguide (110’), the first output straight waveguide (112), the second output straight waveguide (113), the third output straight waveguide (114), and the fourth output straight waveguide (115) are 3 μm × 3 μm. The lengths of the 4 first input Taper waveguides (410) and the 4 second output Taper waveguides (430) are 132.2 μm respectively. The starting widths of the cascaded Taper waveguides (411, 412, and 413, 431, 432, and 433) from far to near the second multimode interference region (420) are 3 μm, 6.8 μm, and 7 μm respectively. The termination width of the third-stage cascaded Taper waveguide (413 or 433) is 7 μm. The width of the second multimode interference region (420) is 40 μm and the length is 1690 μm. The center spacings of the 4 second input Taper waveguides (410) and the 4 second output Taper waveguides (430) from top to bottom are 10.18 μm, 10.24 μm, and 10.18 μm in sequence.

7. A method for preparing a 1×4 thermo-optic switch based on a silica / polymer hybrid waveguide according to any one of claims 1 to 6, comprising the following steps: 1) On a silicon wafer substrate (1), a layer of silica with a dense structure and a thickness of 10 - 20 μm is grown by thermal oxidation as the lower cladding (2); 2) A polymer core layer material is spin-coated on the silica lower cladding (2) using a vacuum spin coater, pre-baked to remove excess solvent in the core layer, and then allowed to cool naturally to obtain a polymer core layer thin film (31); 3) Using ultraviolet exposure, development, and post-baking, the pattern designed on the mask I that is the same as or complementary to the polymer core layer structure is transferred to the polymer core layer thin film (31), so as to obtain a polymer core layer (3) on the silica lower cladding (2). Except for the multimode interferometer, the thickness and width of the remaining parts of the polymer core layer are the same, being 2 - 5 μm respectively; 4) A polymer upper cladding material is spin-coated on the silica lower cladding (2) and the polymer core layer (3) using a vacuum spin coater, and after baking treatment, it is allowed to cool naturally to obtain a polymer upper cladding (4) with a thickness of 3 - 6 μm; 5) A metal thin film (51) with a thickness of 80 - 120 nm is deposited on the polymer upper cladding (4) using a coating machine; 6) A photoresist layer (61) is spin-coated on the metal surface using a vacuum spin coater, and after pre-baking to remove the photoresist solvent, it is allowed to cool naturally and cure; 7) Through ultraviolet exposure, development, and post-baking, the pattern on the mask II that is the same as the metal electrode structure is transferred to the photoresist layer (61). After development, the film is hardened and allowed to cool naturally to obtain a photoresist pattern (62) that is the same as the metal electrode structure; 8) The metal covered by the non-photoresist pattern (62) is etched using a metal-corresponding metal etching solution to obtain a metal electrode (5), and then the photoresist layer (61) covering the metal electrode (5) is removed; The metal electrode (5) is located above the polymer upper cladding (4) at the positions directly above the first modulation arm waveguide, the second modulation arm waveguide, the third modulation arm waveguide, and the fourth modulation arm waveguide. The symmetry center of the metal electrode in the width and length directions is aligned with the symmetry center of the modulation arm waveguide in the width and length directions. The length of the metal electrode (5) is equal to the length of the modulation arm waveguide, and the width of the metal electrode (5) is greater than the width of the modulation arm waveguide, thereby preparing a 1×4 optical switch based on a silica / polymer hybrid waveguide.

Images