Low-loss heterogeneously integrated electro-optical switch and method of making the same

Abstract

The application discloses a low-loss heterogeneous integrated electro-optical switch and a preparation method thereof, and belongs to the technical field of planar optical waveguide optical switches and preparation thereof. The electro-optical switch is composed of a silicon substrate, a silicon dioxide oxide layer, a lead zirconate titanate flat plate layer, a polymer waveguide and a discrete underpad straight waveguide. The polymer waveguide and the discrete underpad straight waveguide are jointly prepared on the lead zirconate titanate flat plate layer, and a grounding electrode and a signal electrode are prepared on the discrete underpad straight waveguide. The application significantly reduces the complexity of the process and the loss, and fully utilizes the advantage of a large electro-optical coefficient of the lead zirconate titanate, so that the electro-optical switch prepared by the process has the advantages of low loss and high modulation efficiency of the switch, and has a wide application prospect in satisfying low-loss high-efficiency data transmission and the like. In addition, the application uses a polymer material as a part of a waveguide core layer, fully utilizes the advantages of easy processing and compatibility with a semiconductor process of the polymer material, and makes the device have important practical application value.

Description

Technical Field

[0001] This invention belongs to the field of planar optical waveguide optical switches and their fabrication technology, specifically relating to a low-loss heterogeneous integrated electro-optic switch and its fabrication method. Background Technology

[0002] With the rapid development of technologies such as the Internet, big data, and cloud computing, global data traffic is experiencing explosive growth, and the demand for data transmission rates will continue to increase in the future. In particular, the widespread adoption of technologies such as artificial intelligence places higher demands on information interconnection technologies. As a core component of optical interconnection, the performance of electro-optical switches directly affects the operating efficiency and data transmission capabilities of data centers. Therefore, electro-optical switches need to continuously improve their performance to achieve more efficient and lower-loss electro-optical signal conversion.

[0003] To achieve high-performance electro-optic switches, various integrated electro-optic platforms have been reported and developed, mainly including silicon-based platforms, thin-film lithium niobate (LNOI) platforms, organic electro-optic polymer platforms, and lead zirconate titanate (PZT) platforms. Silicon-based electro-optic switches have matured considerably after years of development, but their high power consumption and slow switching speed prevent them from meeting the current demands of optical communication. Thin-film lithium niobate is currently the most widely used crystalline thin-film electro-optic material; however, its electro-optic coefficient limits the switching efficiency, and the complex waveguide fabrication and etching processes slow its development. In contrast, organic electro-optic polymers can be fabricated using spin-coating methods, providing a flexible and scalable alternative for integrated electro-optic switches. However, the fabrication of electro-optic polymer materials is complex, and their low stability leads to a certain degree of degradation in their electro-optic performance, posing significant challenges for their application in the electro-optic switch field.

[0004] PZT material has a large electro-optic coefficient, which significantly improves the efficiency of PZT electro-optic switches, thus attracting widespread attention and rapid development in the industry. However, the current PZT waveguide fabrication process requires dry etching, which greatly increases the complexity and cost of the device fabrication. Summary of the Invention

[0005] To further reduce switching losses and simplify the manufacturing process, this invention proposes a low-loss heterogeneous integrated electro-optic switch and its fabrication method.

[0006] This invention uses a lead zirconate titanate (PZT) thin film and a polymer waveguide on top of it as the waveguide core layer. No etching of the PZT is required. By designing the dimensions of the polymer strip waveguide, the position of the optical field is controlled, ensuring that most of the optical field is distributed within the PZT thin film. A polymer pad-type straight waveguide is added between the metal electrode and the PZT thin film. Through dimensional design, plasmonic mode coupling is suppressed, reducing the increased loss of the electro-optic switch due to the plasmonic effect. This invention significantly reduces the complexity of the process and also reduces losses. It fully utilizes the high electro-optic coefficient of lead zirconate titanate, resulting in an electro-optic switch with low loss and high modulation efficiency, showing broad application prospects for low-loss, high-efficiency data transmission. Furthermore, the use of a polymer material as part of the waveguide core layer fully leverages the ease of processing and compatibility with semiconductor processes, giving the device significant practical application value.

[0007] The low-loss heterogeneous integrated electro-optic switch of the present invention comprises, from bottom to top, a silicon substrate 21, a silicon dioxide oxide layer 22 deposited on the upper surface of the substrate, an unetched lead zirconate titanate plate layer 23 spin-coated on the upper surface of the silicon dioxide oxide layer 22, a polymer waveguide spin-coated on the upper surface of the lead zirconate titanate plate layer 23, and discrete first ground electrode pad straight waveguide 17, signal electrode pad straight waveguide 18, and second ground electrode pad straight waveguide 19; a first ground electrode 25, a signal electrode 26, and a second ground electrode 27 are evaporated on the upper surfaces of the lead zirconate titanate plate layer 23 and the first ground electrode pad straight waveguide 17, signal electrode pad straight waveguide 18, and second ground electrode pad straight waveguide 19, respectively; the first ground electrode pad straight waveguide 17, signal electrode pad straight waveguide 18, and second ground electrode pad straight waveguide 19 are respectively covered by the first ground electrode 25, signal electrode 26, and second ground electrode 27.

[0008] As attached Figure 2 As shown, the polymer waveguide, from left to right along the signal light transmission direction, consists of an input straight waveguide 1, an input tapered waveguide 2, a first rectangular multimode interference waveguide 3 for beam splitting, a first beam splitting output tapered waveguide 4, a second beam splitting output tapered waveguide 5, a first beam splitting output curved waveguide 6, a second beam splitting output curved waveguide 7, a first modulation arm straight waveguide 8, a second modulation arm straight waveguide 9, a first beam combining input curved waveguide 10, a second beam combining input curved waveguide 11, a first beam combining input tapered waveguide 12, a second beam combining input tapered waveguide 13, a second rectangular multimode interference waveguide 14 for beam combining, an output tapered waveguide 15, and an output straight waveguide 16.

[0009] The first rectangular multimode interference waveguide 3 has one input terminal and two output terminals. The input straight waveguide 1 and the input tapered waveguide 2 are connected sequentially to serve as the input terminal of the first rectangular multimode interference waveguide 3. The first beam-splitting output tapered waveguide 4 and the second beam-splitting output tapered waveguide 5 serve as the two output terminals of the first rectangular multimode interference waveguide 3, respectively. The second rectangular multimode interference waveguide 14 has two input terminals and one output terminal. The first beam-combining input tapered waveguide 12 and the second beam-combining input tapered waveguide 13 serve as the second rectangular multimode interference waveguide 14, respectively. The two input terminals, output tapered waveguide 15 and output straight waveguide 16, are connected in sequence to serve as the output terminals of the second rectangular multimode interference waveguide 14; the first beam splitting output tapered waveguide 4, the first beam splitting output curved waveguide 6, the first modulation arm straight waveguide 8, the first beam combining input curved waveguide 10 and the first beam combining input tapered waveguide 12 are connected in sequence; the second beam splitting output tapered waveguide 5, the second beam splitting output curved waveguide 7, the second modulation arm straight waveguide 9, the second beam combining input curved waveguide 11 and the second beam combining input tapered waveguide 13 are connected in sequence.

[0010] The input straight waveguide 1 and the output straight waveguide 16 have the same structure and dimensions, as do the input tapered waveguide 2 and the output tapered waveguide 15. The first beam splitter output tapered waveguide 4 and the second beam splitter output tapered waveguide 5, the first beam splitter output curved waveguide 6 and the second beam splitter output curved waveguide 7, the first modulation arm straight waveguide 8 and the second modulation arm straight waveguide 9, the first beam combiner input curved waveguide 10 and the second beam combiner input curved waveguide 11, the first beam combiner input tapered waveguide 12 and the second beam combiner input tapered waveguide 13 have the same structure and dimensions, and are symmetrically arranged left and right along the signal light propagation direction in the input straight waveguide 1 and the output straight waveguide 16. The first ground electrode pad straight waveguide 17, the signal electrode pad straight waveguide 18, and the second ground electrode pad straight waveguide 19 have the same structure and dimensions. The first ground electrode pad straight waveguide 17 is located at the first... Outside the modulation arm straight waveguide 8, the signal electrode pad straight waveguide 18 is located between the first modulation arm straight waveguide 8 and the second modulation arm straight waveguide 9, and the second ground electrode pad straight waveguide 19 is located outside the second modulation arm straight waveguide 9. The first ground electrode pad straight waveguide 17 and the second ground electrode pad straight waveguide 19 are symmetrically arranged about the signal electrode pad straight waveguide 18. The first ground electrode 25 and the second ground electrode 27 have the same structure and size and are symmetrically arranged about the signal electrode 26. The first modulation arm straight waveguide 8 and the second modulation arm straight waveguide 9 are parallel to each other and are symmetrically arranged about the signal electrode pad straight waveguide 18. The first rectangular multimode interference waveguide 3 and the second rectangular multimode interference waveguide 14 have the same structure and size and are symmetrically arranged front and back about the signal electrode pad straight waveguide 18 together with their input and output ends.

[0011] This invention uses a Mach-Zehnder interferometer (MZI) as its core structure. GSG electrodes (a Ground-Signal-Ground electrode, corresponding to the first ground electrode 25, signal electrode 26, and second ground electrode 27) are placed on both sides of the first modulation arm straight waveguide 8 and the second modulation arm straight waveguide 9 to achieve electric field modulation of the optical signal. The input light is split into two beams of equal amplitude and phase by a beam splitter, which then enter the two modulation arms respectively. When there is no electrical signal at the signal electrode 26, the optical fields of the two modulation arms are of equal amplitude and phase, and are not modulated in the modulation region. The GSG electrodes are driven by a push-pull mechanism (because the electric field directions between the signal electrode 26 and the first ground electrode 25 and the second ground electrode 27 are opposite, the refractive index changes (phase changes) of the first and second modulation arm straight waveguides 8 and 9 are equal in magnitude but opposite in sign, and the phase difference is superimposed by beam combining in the second rectangular multimode interference waveguide 14). After an electrical signal is applied, the electro-optic effect of PZT material is used to change the refractive index of the two arms, thereby changing the phase difference. The two beams interfere at the beam combiner: when the phase difference is... When the phase difference is an integer multiple of the phase difference, the phase interference is long and the switch is turned on; when the phase difference is a half-integer multiple of the phase difference, the phase interference is short and the switch is turned off.

[0012] The silicon substrate 21 has a thickness of 480~520 μm, the silicon dioxide oxide layer 22 has a thickness of 1~5 μm, the lead zirconate titanate plate layer 23 has a thickness h2 of 0.25~0.35 μm, the first modulation arm straight waveguide 8, the second modulation arm straight waveguide 9, the signal electrode pad straight waveguide 18, the first ground electrode pad straight waveguide 17, and the second ground electrode pad straight waveguide 19 all have a thickness h1 of 0.1~2 μm, the first ground electrode 25, the signal electrode 26, and the second ground electrode 27 all have a thickness h3 of 0.5~5 μm, and the signal electrode pad straight waveguide 18, the first ground electrode pad straight waveguide 17, and the second ground electrode pad straight waveguide 19 all have a width wd of 2~30. The distances wd2 from the left and right edges of the signal electrode pad straight waveguide 18 to the left and right edges of the signal electrode 26, from the right edge of the first ground electrode pad straight waveguide 17 to the right edge of the first ground electrode 25, and from the left edge of the second ground electrode pad straight waveguide 19 to the left edge of the second ground electrode 25 are equal and range from 0.1 to 2.1 μm. The width of the signal electrode 26 is wd + wd2 + wd2. The widths of the first ground electrode 25 and the second ground electrode 27 are the same and range from 100 to 200 μm. The gaps gap between the first ground electrode 25, the second ground electrode 27 and the signal electrode 26 are equal and range from 3 to 15 μm.

[0013] The lengths L1 of the input straight waveguide 1 and the output straight waveguide 16 are equal, ranging from 100 to 5000 μm; the input tapered waveguide 2, the first beam-splitting output tapered waveguide 4, the second beam-splitting output tapered waveguide 5, the first beam-combining input tapered waveguide 12, the second beam-combining input tapered waveguide 13, and the output tapered waveguide 15 have the same structure and dimensions, and their projected lengths Lt along the signal light propagation direction in the input straight waveguide 1 and the output straight waveguide 16 are equal, ranging from 10 to 100 μm; the lengths Lm of the first rectangular multimode interference waveguide 3 and the second rectangular multimode interference waveguide 14 are equal, ranging from 100 to 200 μm; the first beam-splitting output curved waveguide 6, the second beam-splitting output curved waveguide 7, the first beam-combining input curved waveguide 10, and the second beam-combining input curved waveguide 11 have the same structure and dimensions, and their projected distances a along the direction parallel to the signal light propagation direction in the input straight waveguide 1 and the output straight waveguide 16 are equal, ranging from 300 to 600 μm. μm; the lengths L of the first ground electrode 25, signal electrode 26, second ground electrode 27, first modulation arm straight waveguide 8, second modulation arm straight waveguide 9, signal electrode bottom straight waveguide 18, first ground electrode bottom straight waveguide 17, and second ground electrode bottom straight waveguide 19 are all equal to 1000~6000 μm; the center position of the connection between the first beam splitter output tapered waveguide 4, the second beam splitter output tapered waveguide 5 and the first rectangular multimode interference waveguide 3 is 1~3 μm away from the upper and lower edges of the first rectangular multimode interference waveguide 3; the center position of the connection between the first beam combiner input tapered waveguide 12, the second beam combiner input tapered waveguide 13 and the second rectangular multimode interference waveguide 14 is 1~3 μm away from the upper and lower edges of the second rectangular multimode interference waveguide 14.

[0014] The widths of the input straight waveguide 1, the initial width of the input tapered waveguide 2, the termination widths of the first beam-splitting output tapered waveguide 4 and the second beam-splitting output tapered waveguide 5, the widths of the first beam-splitting output curved waveguide 6 and the second beam-splitting output curved waveguide 7, the widths of the first modulation arm straight waveguide 8 and the second modulation arm straight waveguide 9, the widths of the first beam-combining input curved waveguide 10 and the second beam-combining input curved waveguide 11, the initial widths of the first beam-combining input tapered waveguide 12 and the second beam-combining input tapered waveguide 13, the termination width of the output tapered waveguide 15, and the width x1 of the output straight waveguide 16 are all equal to 2~4 μm; the termination widths of the input tapered waveguide 2, the initial widths of the first beam-splitting output tapered waveguide 4 and the second beam-splitting output tapered waveguide 5, the termination widths of the first beam-combining input tapered waveguide 12 and the second beam-combining input tapered waveguide 13, and the initial width xt of the output tapered waveguide 15 are all equal to 3~6 μm. μm, and xt>x1; the width xm of the first rectangular multimode interference waveguide 3 and the second rectangular multimode interference waveguide 14 are equal to 10~20 μm; the projection distance b between the termination position and the starting position of the first beam splitting output curved waveguide 6 and the second beam splitting output curved waveguide 7 along the direction perpendicular to the signal light propagation in the input straight waveguide 1 and the output straight waveguide 16 is equal to 5~20 μm; the projection distance b between the starting position and the termination position of the first beam combining input curved waveguide 10 and the second beam combining input curved waveguide 11 along the direction perpendicular to the signal light propagation in the input straight waveguide 1 and the output straight waveguide 16 is equal to 5~20 μm.

[0015] The present invention describes a method for fabricating a low-loss heterogeneous integrated electro-optic switch, such as... Figure 3 As shown, the steps are as follows:

[0016] A: Lead zirconate titanate wafer cleaning

[0017] First, the surface of the lead zirconate titanate wafer, which consists of a silicon substrate 21, a silicon dioxide oxide layer 22 and a lead zirconate titanate plate layer 23, is cleaned 2-3 times in sequence with acetone, methanol and isopropanol, and then dried with nitrogen to ensure that the surface of the lead zirconate titanate wafer is clean.

[0018] B: Fabrication of polymer waveguides and padding straight waveguides

[0019] A spin-coating process is used to coat a polymer material (including a series of materials such as polyimide (PI), polymethyl methacrylate (PMMA), SU-8 2002, SU-8 2005, EpoCore, and EpoClad, etc., with a refractive index lower than that of the lead zirconate titanate plate layer) onto the clean surface of the lead zirconate titanate plate layer 23. The spin-coating speed is 1000~6000 rpm to obtain a polymer film 24. Then, pre-baking is performed, i.e., heating at 80 ℃~140 ℃ for 3~20 minutes, followed by cooling to 50 ℃~80 ℃. Next, photolithography is performed on the polymer film 24, i.e., photolithography under ultraviolet light with a wavelength of 300~400 nm. The waveguide mask has a design compatible with the polymer waveguide and the underlying straight waveguide (such as...) to be prepared. Figure 2 (As shown) a complementary structure, with an exposure time of 3-30 seconds, so that the polymer thin film 24 within the polymer waveguide and the underlying straight waveguide structure to be prepared is exposed to ultraviolet light; after photolithography, it is heated at 70-130 °C for 3-30 minutes, and then cooled to 20-30 °C; then development is performed, that is, first wet etching in the developer corresponding to the polymer material for 10-60 seconds to remove the unexposed polymer thin film 24, and then washing away the residual polymer thin film 24 and developer in isopropanol solution and deionized water respectively, and finally drying with nitrogen gas; after development, post-baking hardening is performed, that is, heating at 120-160 °C for 30-60 minutes, thereby preparing a polymer waveguide with a thickness of 0.1-2 μm (in Figure 3 The positions correspond to the first modulation arm straight waveguide 8, the second modulation arm straight waveguide 9) and the bottom straight waveguide (in Figure 3 The positions correspond to the signal electrode pad bottom straight waveguide 18, the first ground electrode pad bottom straight waveguide 17, and the second ground electrode pad bottom straight waveguide 19).

[0020] C. Electrode Preparation

[0021] A 0.5–5 µm thick electrode layer 25' (materials Al, Au, or Cr) is evaporated onto the surface of the lead zirconate titanate planar layer 23, the polymer waveguide, and the bottom straight waveguide using vacuum evaporation. Then, photoresist BP212 is spin-coated onto the surface of electrode layer 25' at a spin speed of 1000–8000 rpm, resulting in a BP212 thickness of 1–10 μm. The spin-coated BP212 device is baked at 50–300 °C for 10–50 minutes, followed by room temperature photolithography using a mask with the same structure as the electrode structure to be fabricated; that is, exposure to a 300–500 nm UV lamp for 10–20 minutes. The device is exposed to photoresist BP212 in areas other than the electrodes for 5-20 minutes. Then, the device is immersed in a NaOH solution with a mass concentration of 2-5‰ for 5-20 minutes to remove the exposed photoresist BP212. It is then rinsed with deionized water and dried with nitrogen. The device is baked at 80-300 °C for 5-40 minutes and then cooled to room temperature. The resulting device is then developed in a special developer for 5-15 minutes to remove the electrode layer 25' not covered by the photoresist. Finally, the entire device is exposed for 2-5 seconds, then immersed in ethanol for 1-8 minutes to remove the remaining photoresist, yielding the required structure of the first ground electrode 17, signal electrode 18, and second ground electrode 19. The device is then rinsed with deionized water and dried with nitrogen to obtain the low-loss heterogeneous integrated electro-optic switch described in this invention.

[0022] Compared with existing device structures and technologies, the advantages of this invention are as follows: This invention uses an etch-free lead zirconate titanate (LZT) plate as part of the core layer, which has a larger electro-optic coefficient compared to lithium niobate, thus resulting in a switch with higher modulation efficiency. Simultaneously, the introduced polymer-backed straight waveguide suppresses plasmonic coupling between LZT and the metal electrodes, significantly reducing switch losses. When an external electric field is applied, the electric field is uniformly distributed between the signal and ground electrodes. Furthermore, through the design of the polymer waveguide and polymer-backed straight waveguide dimensions, the optical field is more effectively controlled within the LZT plate layer, further improving the modulation efficiency of the electro-optic switch. In addition, compared with etched LZT electro-optic switches, this invention greatly simplifies the fabrication process, reduces the difficulty of device fabrication, and offers advantages such as low production cost and high efficiency. Attached Figure Description

[0023] Figure 1 : Figure 2 Schematic diagram of the cross-section of switch A-A' in the middle;

[0024] Figure 2 Schematic diagram of a low-loss heterogeneous integrated electro-optic switch;

[0025] Figure 3 Flowchart of fabrication process for low-loss heterogeneous integrated electro-optic switch;

[0026] Figure 4 (a): Figure 2 Light field distribution diagram of section A-A';

[0027] Figure 4 (b): Figure 2 Electric field distribution diagram of section A-A';

[0028] Figure 5 (a): Curves showing the variation of half-wave voltage length product and loss in electro-optical switching as a function of wd2;

[0029] Figure 5 (b): Curves showing the variation of half-wave voltage length product and loss as a function of wd in electro-optical switching.

[0030] Figure 1 Cross-section of the electro-optic switch ( Figure 2 The schematic diagram (Cross section A-A') shows the following components: silicon substrate 21, silicon dioxide oxide layer 22 deposited on the upper surface of the substrate, unetched lead zirconate titanate plate layer 23 spin-coated on the upper surface of the silicon dioxide oxide layer 22, first modulation arm straight waveguide 8, second modulation arm straight waveguide 9, signal electrode pad straight waveguide 18, first ground electrode pad straight waveguide 17, and second ground electrode pad straight waveguide 19 spin-coated on the upper surface of the lead zirconate titanate plate layer 23, and first ground electrode 25, signal electrode 26, and second ground electrode 27 evaporated on the upper surface of the lead zirconate titanate plate layer 23 and all pad straight waveguides.

[0031] Figure 2 This is a schematic diagram of the polymer waveguide structure for a low-loss heterogeneous integrated electro-optic switch. The names of each part are as follows: input straight waveguide 1, input tapered waveguide 2, first rectangular multimode interference waveguide for beam splitting 3, first beam splitting output tapered waveguide 4, second beam splitting output tapered waveguide 5, first beam splitting output curved waveguide 6, second beam splitting output curved waveguide 7, first modulation arm straight waveguide 8, second modulation arm straight waveguide 9, first beam combining input curved waveguide 10, second beam combining input curved waveguide 11, first beam combining input tapered waveguide 12, second beam combining input tapered waveguide 13, second rectangular multimode interference waveguide for beam combining 14, output tapered waveguide 15, output straight waveguide 16, first ground electrode 25, signal electrode 26, and second ground electrode 27.

[0032] Figure 3The process flow diagram for fabricating a low-loss heterogeneous integrated electro-optic switch is shown below. The names of each part are: silicon substrate 21, silicon dioxide oxide layer 22, lead zirconate titanate plate layer 23, first modulation arm straight waveguide 8, second modulation arm straight waveguide 9, signal electrode pad straight waveguide 18, first ground electrode pad straight waveguide 17, second ground electrode pad straight waveguide 19, first ground electrode 25, signal electrode 26, second ground electrode 27, electrode layer 25', and polymer film 24 prepared by spin coating.

[0033] Figure 4 (a) For low-loss heterogeneous integrated electro-optic switches Figure 2 Simulated optical field distribution of the fundamental mode in the A-A' section waveguide; it can be seen from the figure that the optical field is mostly distributed in the lead zirconate titanate planar layer waveguide, which ensures effective light transmission while increasing the overlap area of ​​the optical field and electric field, thus improving the modulation efficiency of the electro-optic switch; the direction of the arrows in the figure represents the direction of the electric field in the optical field, and it can be seen from the figure that there is a transverse electric mode in the waveguide; the size of the arrows represents the intensity of the electric field. The larger the arrow, the stronger the electric field intensity. The arrows will be larger and denser in the strongly confined region of the optical field (waveguide core layer), while the arrows will be smaller or even disappear in the cladding region.

[0034] Figure 4 (b) For low-loss heterogeneous integrated electro-optic switches Figure 2 Simulated electric field distribution of the fundamental mode in the waveguide with cross section A-A'; the grayscale in the figure represents the electric field distribution. It can be seen that the electric field distribution is uniform in the lead zirconate titanate plate, and the electric field is larger at the edge of the electrode.

[0035] like Figure 5 The figure shows the electro-optic modulation efficiency and loss of a low-loss heterogeneous integrated electro-optic switch (electro-optic modulation efficiency is expressed as the half-wave voltage-length product V). π L is used for quantization, and its expression is V. π L= Where λ is the free-space wavelength of light, and gap is the electrode spacing. The refractive index of PZT is... The electro-optic overlap factor measures the degree of overlap between optical modes and electrical signals. Let be the electro-optic coefficient of PZT. Loss is defined as Loss = Where λ is the free-space wavelength of light, Im(n eff) represents the imaginary part of the effective refractive index of the waveguide. ) is a curve showing the variation of the waveguide size parameters, where; (a) is a curve showing the variation of the half-wave voltage-length product and loss of the electro-optic switch in Example 1 with the misalignment distance wd2 between the bottom straight waveguide and the metal electrode. The dashed line is the loss curve (right axis), and the solid line is the modulation efficiency curve (left axis). When wd2 satisfies the phase matching condition, the guided mode and the plasma mode are coupled, resulting in a sharp increase in loss and the formation of a sharp resonance peak; the trend of the modulation efficiency curve is highly consistent with the loss. When resonance is formed, mode energy leakage occurs, the electro-optic overlap factor decreases, the half-wave voltage-length product increases, and the modulation of the electro-optic switch... Efficiency decreases; when wd2 does not meet the phase matching condition, the waveguide strongly binds the guided mode, resulting in low loss and high efficiency; (b) is the curve of the half-wave voltage length product and loss of the electro-optic switch in Example 1 as a function of the waveguide width wd. The dashed line is the loss curve (right axis), and the solid line is the modulation efficiency curve (left axis). When wd meets the phase matching condition, the loss increases sharply, forming a sharp resonance peak; the trend of the modulation efficiency curve is highly consistent with the loss. When resonance is formed, the modulation efficiency of the electro-optic switch decreases; comparing the two size parameters, wd2 has a much stronger influence on mode coupling and loss than wd. Detailed Implementation

[0036] Example 1

[0037] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0038] As attached Figure 1 As shown (for) Figure 2 The cross-sectional view at position A-A' shows a low-loss heterogeneous integrated electro-optic switch, which consists of, from bottom to top, a silicon substrate 21, a silicon dioxide oxide layer 22 deposited on the upper surface of the substrate, an unetched lead zirconate titanate plate layer 23 spin-coated on the upper surface of the silicon dioxide oxide layer 22, a first modulation arm straight waveguide 8, a second modulation arm straight waveguide 9, a signal electrode pad straight waveguide 18, a first ground electrode pad straight waveguide 17, and a second ground electrode pad straight waveguide 19 spin-coated on the upper surface of the lead zirconate titanate plate layer 23, and a first ground electrode 25, a signal electrode 26, and a second ground electrode 27 evaporated on the upper surface of the lead zirconate titanate plate layer 23 and all the pad straight waveguides; the first modulation arm straight waveguide 8, the second modulation arm straight waveguide 9, the signal electrode pad straight waveguide 18, the first ground electrode pad straight waveguide 17, and the second ground electrode pad straight waveguide 19 are made of the same polymer.

[0039] The silicon substrate 21 has a thickness of 500 μm, the silicon dioxide oxide layer 22 has a thickness of 4 μm, the lead zirconate titanate plate layer 23 has a thickness h2 of 0.3 μm, the first modulation arm straight waveguide 8, the second modulation arm straight waveguide 9, the signal electrode pad straight waveguide 18, the first ground electrode pad straight waveguide 17, and the second ground electrode pad straight waveguide 19 all have the same thickness h1 of 0.3 μm, the first ground electrode 25, the signal electrode 26, and the second ground electrode 27 all have the same thickness h3 of 0.6 μm, the signal electrode pad straight waveguide 18, the first ground electrode pad straight waveguide 17, and the second ground electrode pad straight waveguide 19 all have the same width wd of 25.4 μm, and the distance wd2 from the left and right edges of the signal electrode pad straight waveguide 18 to the left and right edges of the signal electrode 26 is equal to 0.7 μm. μm, the right edge of the first ground electrode pad straight waveguide 17 is to the left of the right edge of the first ground electrode 25, and the left edge of the second ground electrode pad straight waveguide 19 is to the right of the left edge of the second ground electrode 25, with a distance of 0.7 μm from wd2. The width of the signal electrode 26 is wd+wd2+wd2=26.8 μm. The widths of the first ground electrode 25 and the second ground electrode 27 are the same at 100 μm. The gaps between the first ground electrode 25, the second ground electrode 27 and the signal electrode 26 are equal at 6 μm.

[0040] As attached Figure 2 As shown, the polymer waveguide of the low-loss heterogeneous integrated electro-optic switch of the present invention comprises, from left to right, an input straight waveguide 1, an input tapered waveguide 2, a first rectangular multimode interference waveguide 3 for beam splitting, a first beam-splitting output tapered waveguide 4, a second beam-splitting output tapered waveguide 5, a first beam-splitting output curved waveguide 6, a second beam-splitting output curved waveguide 7, a first modulation arm straight waveguide 8, a second modulation arm straight waveguide 9, a first beam-combining input curved waveguide 10, a second beam-combining input curved waveguide 11, a first beam-combining input tapered waveguide 12, a second beam-combining input tapered waveguide 13, a second rectangular multimode interference waveguide 14 for beam combining, an output tapered waveguide 15, and an output straight waveguide 16.

[0041] The input straight waveguide 1 and the output straight waveguide 16 have the same structure and dimensions, and their lengths L1 are equal at 4000 μm. The input tapered waveguide 2, the first beam-splitting output tapered waveguide 4, the second beam-splitting output tapered waveguide 5, the first beam-combining input tapered waveguide 12, the second beam-combining input tapered waveguide 13, and the output tapered waveguide 15 have the same structure and dimensions, and their lengths Lt are equal at 40 μm. The first rectangular multimode interference waveguide 3 and the second rectangular multimode interference waveguide 14 have the same structure and dimensions, and their lengths Lm are equal at 150 μm. The first beam-splitting output curved waveguide 6, the second beam-splitting output curved waveguide 7, the first beam-combining input curved waveguide 10, and the second beam-combining input curved waveguide 11 have the same structure and dimensions, and their projected distances a along the direction of signal light propagation parallel to the input straight waveguide 1 and the output straight waveguide 16 are equal at 500 μm. μm; the length L of the first ground electrode 25, signal electrode 26, second ground electrode 27, first modulation arm straight waveguide 8, second modulation arm straight waveguide 9, signal electrode bottom straight waveguide 18, first ground electrode bottom straight waveguide 17, and second ground electrode bottom straight waveguide 19 is equal to 4000 μm; the distance between the center position of the connection between the first beam splitter output tapered waveguide 4, the second beam splitter output tapered waveguide 5 and the first rectangular multimode interference waveguide 3 and the upper and lower edges of the first rectangular multimode interference waveguide 3, and the distance d between the center position of the connection between the first beam combiner input tapered waveguide 12, the second beam combiner input tapered waveguide 13 and the second rectangular multimode interference waveguide 14 and the upper and lower edges of the second rectangular multimode interference waveguide 14 are equal to 1.5 μm;

[0042] The widths of the input straight waveguide 1, the initial width of the input tapered waveguide 2, the termination widths of the first beam-splitting output tapered waveguide 4 and the second beam-splitting output tapered waveguide 5, the widths of the first beam-splitting output curved waveguide 6 and the second beam-splitting output curved waveguide 7, the widths of the first modulation arm straight waveguide 8 and the second modulation arm straight waveguide 9, the widths of the first beam-combining input curved waveguide 10 and the second beam-combining input curved waveguide 11, the initial widths of the first beam-combining input tapered waveguide 12 and the second beam-combining input tapered waveguide 13, the termination width of the output tapered waveguide 15, and the width x1 of the output straight waveguide 16 are all equal to 3 μm; the widths xm of the first rectangular multimode interference waveguide 3 and the second rectangular multimode interference waveguide 14 are all equal to 15. μm; the termination width of the input tapered waveguide 2, the starting width of the first beam-splitting output tapered waveguide 4 and the second beam-splitting output tapered waveguide 5, the termination width of the first beam-combining input tapered waveguide 12 and the second beam-combining input tapered waveguide 13, and the starting width xt of the output tapered waveguide 15 are all equal to 4 μm; the projection distance between the termination position and the starting position of the first beam-splitting output curved waveguide 6 and the second beam-splitting output curved waveguide 7 along the direction of signal light propagation perpendicular to the input straight waveguide 1 and the output straight waveguide 16, and the projection distance b between the starting position and the termination position of the first beam-combining input curved waveguide 10 and the second beam-combining input curved waveguide 11 along the direction of signal light propagation perpendicular to the input straight waveguide 1 and the output straight waveguide 16 are all equal to 10 μm;

[0043] Under the above parameters, the modulation efficiency of the electro-optic switch is calculated to be 0.25 V·cm and the loss is 0.7 dB / cm using the simulation software Comsol Multiphysics. The fabrication method of the low-loss heterogeneous integrated electro-optic switch described in this invention comprises the following steps:

[0044] (1) Cleaning of lead zirconate titanate wafer: First, use acetone, methanol and isopropanol to clean the surface of lead zirconate titanate wafer (which can be purchased directly, and the lead zirconate titanate wafer is composed of silicon substrate 21, silicon dioxide oxide layer 22 and lead zirconate titanate plate layer 23) three times in sequence, and then blow it dry with nitrogen gas to confirm that the surface of lead zirconate titanate wafer is clean, and obtain silicon substrate 21, silicon dioxide oxide layer 22 and lead zirconate titanate plate layer 23;

[0045] (2) Fabrication of polymer waveguides (and bottom straight waveguides); using a spin coating process, polymer material (EpoCore, whose refractive index is lower than that of the lead zirconate titanate plate layer) is coated on the clean surface of the lead zirconate titanate plate layer 23 at a rotation speed of 5000 rpm to obtain a polymer film 24; then pre-baking is performed, i.e., heating at 120 ℃ for 10 minutes, and then cooling to 60 ℃; then photolithography is performed on the polymer film 24, i.e., photolithography is performed under ultraviolet light with a wavelength of 365 nm, and the waveguide mask has the same shape as the polymer waveguide and bottom straight waveguide (e.g., ...) to be prepared. Figure 2The complementary structure (shown) is exposed for 6 seconds, allowing the polymer thin film 24 within the polymer waveguide and the underlying straight waveguide structure to be exposed to ultraviolet light. After photolithography, the structure is heated to 85°C for 10 minutes and then cooled to 25°C. Development is then performed, first by wet etching in the developer corresponding to the polymer strip layer material for 40 seconds to remove the unexposed polymer thin film 24, then by washing away the residual polymer thin film 24 and developer in isopropanol solution and deionized water respectively, and finally by drying with nitrogen. After development, post-bake hardening is performed, i.e., heating at 120°C for 40 minutes, thereby obtaining a polymer waveguide with a thickness of 0.3 μm (in... Figure 3 The positions correspond to the first modulation arm straight waveguide 8, the second modulation arm straight waveguide 9) and the polymer pad bottom straight waveguide (in Figure 3 The positions correspond to the signal electrode pad bottom straight waveguide 18, the first ground electrode pad bottom straight waveguide 17, and the second ground electrode pad bottom straight waveguide 19).

[0046] (3) Electrode fabrication: A 0.8 µm thick electrode layer 25' (material Au) was evaporated on the surface of the lead zirconate titanate plate layer 23, the polymer waveguide, and the bottom straight waveguide using vacuum evaporation. Then, photoresist BP212 was spin-coated onto the surface of electrode layer 25' at a speed of 6000 rpm, resulting in a BP212 thickness of 1.5 μm. The spin-coated device was baked at 87 °C for 10 minutes, and then cooled to room temperature for photolithography. The structure of the photomask was the same as that of the electrode to be fabricated. It was exposed to a 365 nm UV lamp for 2 s to expose the photoresist in areas other than the electrodes. The device was then immersed in a 5‰ NaOH solution for 5 minutes to remove the exposed photoresist. It was then rinsed with deionized water and dried with nitrogen. After baking the device at ℃ for 10 minutes, it was cooled to room temperature. Then, the sample was placed in a gold developer (the developer is iodine and potassium iodide in a mass ratio of 1:8) and developed for 8 minutes to remove the electrode layer 25' that was not covered by the photoresist. Finally, the entire device was exposed for 3 seconds, and then the device was immersed in ethanol for 3 minutes to remove the remaining photoresist, obtaining the signal electrode 18, the first ground electrode 17, and the second ground electrode 19 with the required structure. The device was then rinsed with deionized water and dried with nitrogen gas to obtain the low-loss heterogeneous integrated electro-optic switch described in this invention.

[0047] This results in the fabrication of a low-loss heterogeneous integrated electro-optic switch that meets the requirements. It should be noted that while this invention contains descriptions of many details, it should not be construed as limiting the scope or possible claims of any disclosed technology, but rather as a description of features that may be specific to particular embodiments of the disclosed technology. The invention can also be modified in many ways, such as using electro-optic materials like barium titanate or lithium niobate. What those skilled in the art can derive from the explicit disclosure of this invention or from the written description herein without objection falls within the scope of protection of this invention.

Claims

1. A low-loss heterogeneous integrated electro-optic switch, characterized in that: The structure consists of, from bottom to top, a silicon substrate (21), a silicon dioxide oxide layer (22) deposited on the upper surface of the substrate (21), an unetched lead zirconate titanate plate layer (23) spin-coated on the upper surface of the silicon dioxide oxide layer (22), a polymer waveguide spin-coated on the upper surface of the lead zirconate titanate plate layer (23), and discrete first ground electrode pad waveguide (17), signal electrode pad waveguide (18), and second ground electrode pad waveguide (19). The first ground electrode (25), signal electrode (26), and second ground electrode (27) are evaporated on the upper surfaces of the lead zirconate titanate plate layer (23), the first ground electrode pad waveguide (17), the signal electrode pad waveguide (18), and the second ground electrode pad waveguide (19), respectively. The first ground electrode pad waveguide (17), the signal electrode pad waveguide (18), and the second ground electrode pad waveguide (19) are respectively covered by the first ground electrode (25), the signal electrode (26), and the second ground electrode (27). The polymer waveguide, from left to right along the transmission direction of the signal light, consists of an input straight waveguide (1), an input tapered waveguide (2), a first rectangular multimode interference waveguide (3) for beam splitting, a first beam splitting output tapered waveguide (4), a second beam splitting output tapered waveguide (5), a first beam splitting output curved waveguide (6), a second beam splitting output curved waveguide (7), a first modulation arm straight waveguide (8), a second modulation arm straight waveguide (9), a first beam combining input curved waveguide (10), a second beam combining input curved waveguide (11), a first beam combining input tapered waveguide (12), a second beam combining input tapered waveguide (13), a second rectangular multimode interference waveguide (14) for beam combining, an output tapered waveguide (15), and an output straight waveguide (16). The first rectangular multimode interference waveguide (3) has one input end and two output ends. The input straight waveguide (1) and the input tapered waveguide (2) are connected in sequence to serve as the input end of the first rectangular multimode interference waveguide (3). The first beam splitting output tapered waveguide (4) and the second beam splitting output tapered waveguide (5) serve as the two output ends of the first rectangular multimode interference waveguide (3), respectively. The second rectangular multimode interference waveguide (14) has two input ends and one output end. The first beam combining input tapered waveguide (12) and the second beam combining input tapered waveguide (13) serve as the two output ends of the second rectangular multimode interference waveguide (14), respectively. The first input terminal, the output tapered waveguide (15) and the output straight waveguide (16) are connected in sequence to serve as the output terminal of the second rectangular multimode interference waveguide (14); the first beam splitting output tapered waveguide (4), the first beam splitting output curved waveguide (6), the first modulation arm straight waveguide (8), the first beam combining input curved waveguide (10) and the first beam combining input tapered waveguide (12) are connected in sequence; the second beam splitting output tapered waveguide (5), the second beam splitting output curved waveguide (7), the second modulation arm straight waveguide (9), the second beam combining input curved waveguide (11) and the second beam combining input tapered waveguide (13) are connected in sequence.

2. The low-loss heterogeneous integrated electro-optic switch as described in claim 1, characterized in that: The input straight waveguide (1) and the output straight waveguide (16) have the same structure and size, and the input tapered waveguide (2) and the output tapered waveguide (15) have the same structure and size; the first beam splitting output tapered waveguide (4) and the second beam splitting output tapered waveguide (5), the first beam splitting output curved waveguide (6) and the second beam splitting output curved waveguide (7), the first modulation arm straight waveguide (8) and the second modulation arm straight waveguide (9), the first beam combining input curved waveguide (10) and the second beam combining input curved waveguide (11), the first beam combining input tapered waveguide (12) and the second beam combining input tapered waveguide (13) have the same structure and size, and are symmetrically arranged left and right along the signal light propagation direction in the input straight waveguide (1) and the output straight waveguide (16); the first ground electrode pad straight waveguide (17), the signal electrode pad straight waveguide (18), and the second ground electrode pad straight waveguide (19) have the same structure and size, and the first ground electrode pad straight waveguide (17) is located at Outside the first modulation arm straight waveguide (8), the signal electrode pad straight waveguide (18) is located between the first modulation arm straight waveguide (8) and the second modulation arm straight waveguide (9), and the second ground electrode pad straight waveguide (19) is located outside the second modulation arm straight waveguide (9). The first ground electrode pad straight waveguide (17) and the second ground electrode pad straight waveguide (19) are symmetrically arranged about the signal electrode pad straight waveguide (18). The first ground electrode (25) and the second ground electrode (27) have the same structure and size, and are symmetrically arranged about the signal electrode (26). The first modulation arm straight waveguide (8) and the second modulation arm straight waveguide (9) are parallel to each other, and are symmetrically arranged about the signal electrode pad straight waveguide (18). The first rectangular multimode interference waveguide (3) and the second rectangular multimode interference waveguide (14) have the same structure and size, and together with their input and output ends, are symmetrically arranged about the signal electrode pad straight waveguide (18).

3. The low-loss heterogeneous integrated electro-optic switch as described in claim 2, characterized in that: The thickness of the silicon substrate (21) is 480~520 μm, the thickness of the silicon dioxide oxide layer (22) is 1~5 μm, the thickness h2 of the lead zirconate titanate plate layer (23) is 0.25~0.35 μm, the thickness h1 of the first modulation arm straight waveguide (8), the second modulation arm straight waveguide (9), the signal electrode pad straight waveguide (18), the first ground electrode pad straight waveguide (17), and the second ground electrode pad straight waveguide (19) is the same, which is 0.1~2 μm, the thickness h3 of the first ground electrode (25), the signal electrode (26), and the second ground electrode (27) is the same, which is 0.5~5 μm, and the width wd of the signal electrode pad straight waveguide (18), the first ground electrode pad straight waveguide (17), and the second ground electrode pad straight waveguide (19) is the same, which is 2~30. The distances wd2 from the left and right edges of the bottom straight waveguide (18) of the signal electrode to the left and right edges of the signal electrode (26), from the right edge of the bottom straight waveguide (17) of the first ground electrode to the right edge of the first ground electrode (25), and from the left edge of the bottom straight waveguide (19) of the second ground electrode to the left edge of the second ground electrode (25) are equal and range from 0.1 to 2.1 μm. The width of the signal electrode (26) is wd + wd2 + wd2. The widths of the first ground electrode (25) and the second ground electrode (27) are the same and range from 100 to 200 μm. The gaps gap between the first ground electrode (25), the second ground electrode (27) and the signal electrode (26) are equal and range from 3 to 15 μm.

4. A low-loss heterogeneous integrated electro-optic switch as described in claim 2, characterized in that: The lengths L1 of the input straight waveguide (1) and the output straight waveguide (16) are equal, ranging from 100 to 5000 μm; the input tapered waveguide (2), the first beam-splitting output tapered waveguide (4), the second beam-splitting output tapered waveguide (5), the first beam-combining input tapered waveguide (12), the second beam-combining input tapered waveguide (13), and the output tapered waveguide (15) have the same structure and dimensions, and their projected lengths Lt along the signal light propagation direction in the input straight waveguide (1) and the output straight waveguide (16) are equal, ranging from 10 to 100 μm; the lengths Lm of the first rectangular multimode interference waveguide (3) and the second rectangular multimode interference waveguide (14) are equal, ranging from 100 to 200 μm. μm; the first beam splitting output curved waveguide (6), the second beam splitting output curved waveguide (7), the first beam combining input curved waveguide (10), and the second beam combining input curved waveguide (11) have the same structure and size, and their projection distance a along the direction of signal light propagation parallel to the input straight waveguide (1) and the output straight waveguide (16) is equal to 300~600 μm; the length L of the first ground electrode (25), the signal electrode (26), the second ground electrode (27), the first modulation arm straight waveguide (8), the second modulation arm straight waveguide (9), the signal electrode pad straight waveguide (18), the first ground electrode pad straight waveguide (17), and the second ground electrode pad straight waveguide (19) is equal to 1000~6000 μm; the center position of the connection between the first beam splitting output tapered waveguide (4), the second beam splitting output tapered waveguide (5) and the first rectangular multimode interference waveguide (3) is equal to the distance d between the upper and lower edges of the first rectangular multimode interference waveguide (3) and the center position of the connection between the first beam splitting output tapered waveguide (4) and the second beam splitting output tapered waveguide (5) and the first rectangular multimode interference waveguide (3) is equal to 1~3 μm, the center position of the connection between the first beam input tapered waveguide (12), the second beam input tapered waveguide (13) and the second rectangular multimode interference waveguide (14) is equal to the distance d between the upper and lower edges of the second rectangular multimode interference waveguide (14) and is 1~3 μm.

5. A low-loss heterogeneous integrated electro-optic switch as described in claim 2, characterized in that: The width of the input straight waveguide (1), the starting width of the input tapered waveguide (2), the ending width of the first beam-splitting output tapered waveguide (4) and the second beam-splitting output tapered waveguide (5), the width of the first beam-splitting output curved waveguide (6) and the second beam-splitting output curved waveguide (7), the width of the first modulation arm straight waveguide (8) and the second modulation arm straight waveguide (9), the width of the first beam-combining input curved waveguide (10) and the second beam-combining input curved waveguide (11), the starting width of the first beam-combining input tapered waveguide (12) and the second beam-combining input tapered waveguide (13), the ending width of the output tapered waveguide (15), and the width x1 of the output straight waveguide (16) are all equal to 2~4. μm; the termination width of the input tapered waveguide (2), the starting width of the first beam-splitting output tapered waveguide (4) and the second beam-splitting output tapered waveguide (5), the termination width of the first beam-combining input tapered waveguide (12) and the second beam-combining input tapered waveguide (13) and the starting width xt of the output tapered waveguide (15) are all equal to 3~6 μm, and xt>x1; the width xm of the first rectangular multimode interference waveguide (3) and the second rectangular multimode interference waveguide (14) are all equal to 10~20 μm; the projection distance b between the termination position and the starting position of the first beam-splitting output curved waveguide (6) and the second beam-splitting output curved waveguide (7) along the direction perpendicular to the signal light propagation in the input straight waveguide (1) and the output straight waveguide (16) is all equal to 5~20 μm. μm; The projection distance b between the starting and ending positions of the first and second bundled input curved waveguides (10) and the signal light propagation direction perpendicular to the input straight waveguide (1) and the output straight waveguide (16) is equal to 5~20 μm.

6. A method for fabricating a low-loss heterogeneous integrated electro-optic switch according to any one of claims 1 to 5, comprising the following steps: A: Lead zirconate titanate wafer cleaning First, the surface of the lead zirconate titanate wafer, which consists of a silicon substrate (21), a silicon dioxide oxide layer (22), and a lead zirconate titanate plate layer (23), is cleaned 2-3 times in sequence with acetone, methanol, and isopropanol, and then dried with nitrogen gas to ensure that the surface of the lead zirconate titanate wafer is clean. B: Fabrication of polymer waveguides and padding straight waveguides The polymer material was coated onto the clean surface of a lead zirconate titanate plate (23) using a spin coating process at a speed of 1000-6000 rpm to obtain a polymer film (24). Pre-baking was then performed, i.e., heating at 80 ℃-140 ℃ for 3-20 minutes, followed by cooling to 50 ℃-80 ℃. The polymer film (24) was then photolithographically lithographically lithographically performed under ultraviolet light with a wavelength of 300-400 nm. The waveguide mask had a structure complementary to the polymer waveguide and the underlying straight waveguide to be prepared, with an exposure time of 3-30 seconds, allowing the polymer film (24) within the polymer waveguide and the underlying straight waveguide structure to be exposed to ultraviolet light. After photolithography, the film was heated at 70-130 ℃ for 3-30 minutes, followed by cooling to 20-30 ℃. ℃; then development is performed, that is, wet etching is first performed in the developer corresponding to the polymer material for 10~60 seconds to remove the unexposed polymer film (24), and then the residual polymer film (24) and developer are washed away in isopropanol solution and deionized water respectively, and finally dried with nitrogen gas; After development, post-bake hardening is performed, i.e., heating at 120~160 ℃ for 30~60 minutes, thereby preparing polymer waveguides and bottom straight waveguides; C. Electrode Preparation Electrode layers (25') were evaporated on the surface of lead zirconate titanate plate layer (23) and polymer waveguide and bottom straight waveguide using vacuum evaporation method; then photoresist BP212 was spin-coated onto the surface of electrode layer (25') using spin coating process at a speed of 1000~8000 rpm, and the thickness of BP212 was 1~10 μm. Devices coated with spin-coated BP212 photoresist are baked at 50–300 °C for 10–50 minutes, then cooled to room temperature for mask photolithography. The mask structure is the same as the electrode structure to be prepared. Specifically, the device is exposed to a 300–500 nm UV lamp for 10–20 seconds to expose the BP212 photoresist in areas other than the electrodes. The device is then immersed in a 2–5‰ NaOH solution for 5–20 minutes to remove the exposed BP212 photoresist. It is then rinsed with deionized water and dried with nitrogen. The device is baked at 80–300 °C for 5–40 minutes and then cooled to room temperature. The resulting device is then developed in a dedicated developer for 5–15 minutes to remove the electrode layers not covered by the photoresist (25'). Finally, the entire device is exposed for 2–5 seconds. Then, immerse the device in ethanol for 1-8 minutes to remove the remaining photoresist and obtain the first ground electrode (17), signal electrode (18), and second ground electrode (19) with the required structure. Then rinse with deionized water and dry with nitrogen to obtain a low-loss heterogeneous integrated electro-optic switch.

7. The method for fabricating a low-loss heterogeneous integrated electro-optic switch as described in claim 6, characterized in that: The polymer waveguide and the bottom straight waveguide are made of polyimide, polymethyl methacrylate, SU-8 2002, SU-8 2005, EpoCore or EpoClad; the first ground electrode (25), the signal electrode (26) and the second ground electrode (27) are made of Al, Au or Cr.

Images