Design method of silicon-based mold splitter and silicon-based mold splitter

Abstract

The invention discloses a design method of a silicon-based mode splitter, which is based on a bus waveguide, n branch waveguides and a functional area positioned between the bus waveguide and the branch waveguides on a silicon-on-insulator platform, and is characterized in that the number of the branch waveguides is n and is determined according to the linear combination number of an input mode and an input bandwidth of the bus waveguide, defining initial structures of the bus waveguide, the branch waveguide and the functional area; m times of forward simulation and m times of accompanying simulation need to be carried out in each iteration by utilizing an FDTD solver; calculating a gradient field, optimizing dielectric constant parameters of the functional region according to a gradient field calculation result in combination with a gradient descent method, and updating the structure; iteration is continued, and iteration is stopped when a design target is achieved or optimization cannot be continued; carrying out binarization processing on the discrete dielectric constant, and converting the dielectric constant of the functional region into a structure only defined by the functional region; and outputting the final structure chart. The invention further discloses the silicon-based mold splitter designed through the design method.

Description

Technical Field

[0001] The invention relates to the technical field of integrated photonics, and in particular to a design method for a silicon-based mode splitter and the silicon-based mode splitter. Background Art

[0002] With the rapid development of the information age, people's demand for data transmission has increased dramatically. On this basis, silicon-based photonics has received widespread attention due to its high speed, broadband, and efficient integration in the field of information transmission. Multiplexing technology, as an effective means of expanding information transmission capacity, has received unprecedented attention. Among them, the mode division multiplexing technology (MDM) that uses only one light source to transmit multiple modes in a multimode waveguide, and the wavelength division multiplexing technology (WDM) widely used in optical fiber communication systems have become the key technologies for realizing information expansion on the silicon-on-insulator platform (SOI). However, a single mode division multiplexing technology or a single wavelength division multiplexing (WDM) technology can no longer meet people's needs. Seeking a combination of multiple multiplexing technologies can multiply the bandwidth to further increase the capacity of the current information communication system.

[0003] Silicon-based mode splitters are used to separate modes of different orders into different signals without changing the mode order, and are key components for realizing stable on-chip MDM systems. Since the effective refractive index of high-order modes is always smaller than that of low-order modes in the same waveguide, most mode demultiplexing systems are always accompanied by changes in mode order. Its working mechanism is mainly to achieve mode separation by converting high-order modes into fundamental modes and mapping them to specific channels through a phase matching mechanism. However, in mode-sensitive systems, as the mode order changes, mode conversion will lead to system instability and performance degradation. Silicon-based mode splitters that achieve mode separation without changing the mode order can significantly improve the flexibility of MDM systems. However, it is very challenging to directly split modes with different effective indices in the same waveguide, and there are currently few reports on mode splitters that do not change the mode order. Some people use symmetrical directional couplers (DC) to achieve TE0 / TE1 mode separation. Some people have proposed a mode splitter based on an asymmetric directional coupler (ADC), which consists of two slits and a strip waveguide, which can separate TE0 and TE1 modes within a bandwidth of 100nm, with crosstalk (CT) <-8dB and insertion loss (IL) <1dB. Some people separate the two basic modes TE0 / TM0 by controlling the coupling area size of the tapered directional coupler, achieving high extinction ratio (ER <10dB) and low insertion loss (IL <1.6dB) mode separation within a bandwidth of 90nm. In addition, some people use bridged subwavelength grating assisted (BSWG) DC and mode filters to develop a dual-mode splitter for separating TM0 and TM1 modes, with CT <-15dB and IL <1.8dB at a bandwidth of 84nm. However, these designs consume a huge footprint of hundreds of square microns, and the device structure is also complex. In order to realize a stable and efficient on-chip MDM system, an ultra-compact single-integrated wide-spectrum mode separation device is urgently needed. In order to combine with the on-chip WDM system, the designed photonic device is also required to have wavelength separation function, which makes the device design extremely difficult. Summary of the invention

[0004] In view of the above-mentioned deficiencies in the prior art, the technical problem to be solved by the present invention is: how to provide a design method for a silicon-based mold splitter that can effectively improve device performance and enhance device design efficiency.

[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0006] A method for designing a silicon-based mold splitter comprises the following steps:

[0007] (1) Based on a bus waveguide, a branch waveguide, and a functional region located between the bus waveguide and the branch waveguide on a silicon-on-insulator platform, the number of the branch waveguides is n and is determined according to the number of linear combinations of input modes and input bandwidths of the bus waveguide, and the initial structures of the bus waveguide, the branch waveguide, and the functional region are defined;

[0008] (2) Using the FDTD solver, each iteration requires m forward simulations and m adjoint simulations, where m equals n;

[0009] (3) Calculate the gradient field and optimize the dielectric constant parameters of the functional area based on the gradient field calculation results combined with the gradient descent method to update the structure;

[0010] (4) Continue to iterate and stop iterating when the design goal is reached or optimization cannot be continued;

[0011] (5) Binarizing the discrete dielectric constants, and converting the dielectric constants of the functional regions into a structure defined only by the functional regions in step (1);

[0012] (6) Output the final structure diagram.

[0013] As an optimization, the initial structures of the bus waveguide, branch waveguide and functional area are defined including the width of the bus waveguide and the branch waveguide, the spacing between two adjacent branch waveguides, the shape and size of the functional area, and the design materials and background materials of the functional area.

[0014] As an optimization, the shape of the functional area is any one of a rectangle, a Y shape, a triangle or an ellipse.

[0015] As an optimization, the design material of the functional area is Si or Si3N4, and the background material is SiO2 or air.

[0016] A silicon-based mold splitter is designed according to the design method of the silicon-based mold splitter described above.

[0017] Compared with the prior art, the present invention has the following beneficial effects: the present invention adopts an inverse design algorithm with accompanying optimization, broadens the wavelength range of the traditional mode splitter, and greatly reduces the area of ​​the device, and can achieve wavelength separation and mode separation without changing the mode order and the wide spectrum input light source; this photonic integrated device with integrated wavelength and mode mixing can be used as the basis of a stable on-chip multiplexing system and has broad application prospects in optical communication networks; at the same time, by adjusting the design parameters, the design scheme can be used to design an integrated hybrid multiplexing device with more transmission channels and a larger bandwidth range. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a diagram showing the initialization structure of the silicon-based mold splitter in the present invention;

[0019] Figure 2 are the waveguide dispersion curves at 1310nm and 1550nm in the present invention;

[0020] Figure 3 is a design flow chart of the dual-wavelength mode splitter in the present invention;

[0021] Figure 4 This is a diagram showing the output structure of the dual-wavelength silicon-based mode splitter in the present invention;

[0022] Figure 5 It is a linear graph of the number of FOM iterations and quality factor of each channel in the present invention;

[0023] Figure 6 It is a transmission efficiency diagram of each channel in the present invention;

[0024] Figure 7 It is a performance diagram of each output in the present invention when the input light source bandwidth is ±20nm;

[0025] Figure 8 This is a magnetic field transmission diagram of different signals input into the bus waveguide in the present invention. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical scheme and advantages of the embodiments of the present invention clearer, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention.

[0027] The design method of the silicon-based mold splitter in this specific embodiment includes the following steps:

[0028] (1) Based on a bus waveguide 1, a branch waveguide 2, and a functional region 3 located between the bus waveguide 1 and the branch waveguide 2 on a silicon-on-insulator platform, the number of the branch waveguides 2 is n and is determined according to the number of linear combinations of the input mode and the input bandwidth of the bus waveguide 1, and the initial structures of the bus waveguide 1, the branch waveguide 2, and the functional region 3 are defined;

[0029] (2) Using the FDTD solver, each iteration requires m forward simulations and m adjoint simulations, where m equals n;

[0030] (3) Calculate the gradient field and optimize the dielectric constant parameters of the functional area 3 according to the calculation results of the gradient field combined with the gradient descent method to update the structure;

[0031] (4) Continue to iterate and stop iterating when the design goal is reached or optimization cannot be continued;

[0032] (5) Binarizing the discrete dielectric constants, and converting the dielectric constant of the functional region 3 into a structure defined by only the functional region 3 in step (1);

[0033] (6) Output the final structure diagram.

[0034] In this specific implementation, the initial structures of the bus waveguide 1, the branch waveguide 2 and the functional area 3 are defined, including the width of the bus waveguide 1 and the branch waveguide 2, the spacing between two adjacent branch waveguides 2, the shape and size of the functional area 3, and the design materials and background materials of the functional area 3.

[0035] In this specific implementation, the shape of the functional area 3 is any one of a rectangle, a Y-shape, a triangle or an ellipse.

[0036] In this specific implementation, the design material of the functional area 3 is Si or Si3N4, and the background material is SiO2 or air.

[0037] A silicon-based mold splitter is designed according to the design method of the silicon-based mold splitter described above.

[0038] The structure of the silicon-based splitter in this specific embodiment is based on a silicon-on-insulator platform, the etch layer depth is 220nm, and SiO2 is used to cover the functional area to protect the etch layer structure. Figure 1 As shown in the figure, it consists of a bus waveguide, four branch waveguides and a designed functional area. Both the bus waveguide and the branch waveguide are required to support only TE0 mode transmission and TE1 mode transmission under the input of 1310nm and 1550nm light sources. The full vector finite element method (FV-FEM) is used to find the optimal width of the waveguide. The results are shown in Figure 2As shown. In the width range of 0.75μm to 1.1μm, the waveguide can support TE0 mode transmission and TE1 mode transmission under 1310nm and 1550nm input light sources, and will not support higher-order TE modes, thereby avoiding the influence of mode crosstalk to the greatest extent. Considering that the device area is not easy to be too large, the bus waveguide and branch waveguide widths are finally set to 0.8μm. In the bus waveguide, the bandwidth range of the input signal is 1310±20nm and 1550±20nm, the mode is TE0 mode and TE1 mode, and the input mode is a linear combination of the input bandwidth. The four branch waveguides form four output channels, from bottom to top, channel 1 (output TE0 / 1310nm), channel 2 (output TE1 / 1310nm), channel 3 (output TE0 / 1550nm) and channel 4 (output TE1 / 1550nm). In order to minimize the crosstalk between channels, the spacing between channels needs to be as large as possible, but too large a spacing will affect the integration of the device. Taking all factors into consideration, the spacing between each branch channel is 1μm. Therefore, the optimized functional area can be designed as a rectangular structure of 6.2μm×6.2μm, and of course it can also be set to any shape such as Y-shape, triangle or ellipse. By changing the material distribution in the functional area through the adjoint method topology optimization algorithm, the effective refractive index of the functional area can be changed, thereby realizing the optimized design of the dual-wavelength mode splitter.

[0039] For the designed dual-wavelength silicon-based mode splitter, the optimization space is Figure 1 The square area shown. The gradient of the objective function for each point in the optimization space is calculated by the adjoint method, and combined with the gradient descent method, the filling of the design material (Si) and the background material (SiO2) in the optimization area is guided (Si3N4 can also be used as the design material and air as the background material, or Si3N4 as the design material and SiO2 as the background material, or Si as the design material and air as the background material). The dielectric constants are ε Si =3.48 and ε SiO2 =1.44, thereby changing the effective refractive index of the optimized area, and optimizing the device performance through continuous iteration, so as to achieve the effect of wavelength separation and mode separation. The optimization goal is to separate the light of different wavelengths and modes input into the bus waveguide into different branch waveguides for output, and the transmission efficiency of each channel is required to be the highest.

[0040] The objective function FOM of each channel can be defined as:

[0041]

[0042] Where FOM is the optimization objective function, which is defined as the average transmittance of the target mode at the output port under the design bandwidth, where λ1 and λ2 represent the minimum and maximum wavelengths of the target bandwidth, respectively, i represents the number of the output waveguide under the current incident light source supported, and T cur / T i It represents the normalized transmission efficiency when the light source supported by the current channel is used as input, which can be given by the Pointing vector. The brackets indicate the distance between the current transmission efficiency and the ideal transmission efficiency under each different input condition. The FOM value of each channel will gradually decrease with the iteration of reverse design optimization, which means that it is gradually approaching the optimal transmission efficiency.

[0043] The gradient of the FOM with respect to the dielectric constant can be efficiently calculated using the adjoint method. The forward propagating field E is obtained by forward simulation. forward Then adjust the reverse field by the phase factor to obtain the accompanying field E adjoint The phase factor A is given by:

[0044]

[0045] Where * represents the conjugate operation, ε0 represents the dielectric constant in vacuum, and E m and H m represents the electric and magnetic fields of the desired output mode m. ΔV represents the volume of the divided pixel point. S represents the output waveguide cross section. By calculating the integral on the output cross section S, the accompanying field E can be obtained through the phase factor. adjoint At this time, the gradient of FOM relative to the dielectric constant distribution can be expressed as:

[0046]

[0047] The subscript x here represents the xth element of the calculated functional area. The forward field and the adjoint field can be obtained using the finite difference time domain method. In addition, for a wide spectrum input light source, a wavelength scaling factor C(λ) is required to adjust the gradient field, and the formula is as follows:

[0048]

[0049] Where ω(λ) represents the angular frequency at the wavelength, P source (λ) represents the input source power, j is an imaginary unit, and A(λ) represents the phase factor at the wavelength. At this time, the FOM gradient affected by wavelength can be expressed as:

[0050]

[0051] Here Ω represents the design space. According to formula (5), the influence of each point in the functional area on the final result is calculated, and the dielectric constant of each point is guided according to the gradient in εSi and ε SiO2 The FOM value is calculated by FDTD and continuous iterative optimization is performed until the design goal is achieved and the final device structure diagram is generated.

[0052] The overall reverse design process of the dual-wavelength mode splitter in this specific implementation is as follows: Figure 3 As shown, in the specific design process:

[0053] 1. Determine the initial structure and the initial dielectric constant ε of the functional area old ;

[0054] 2. Four forward simulations to obtain the electric field in the functional area

[0055] 3. Four companion simulations (the light source of the companion simulation is the same size as the light source in the forward simulation, and the position is changed to four branch waveguides for back transmission) to obtain the reverse electric field in the functional area

[0056] 4. Calculate formula (2) to obtain A0, A1, A2, and A3 respectively;

[0057] 5. Use A to adjust the reverse electric field and obtain the accompanying field. Similarly;

[0058] 6. The FOM value is obtained from formula (1), and the gradient is obtained from formula (3)

[0059] 7. Add the gradients to get the overall gradient Here, a, b, c, and d are the weights of the gradient of each branch waveguide. In this embodiment, the default value is 1. The weight ratio can be changed arbitrarily to achieve any ratio of branch waveguide output.

[0060] 8. The above steps are for a single wavelength light source. Using formula (4) and formula (5), we can calculate the input of a multi-wavelength light source, for example, 1550±20nm (band can be expanded). The specific idea is to divide the input wide-band light source into 10 points (assuming), calculate the gradient and gradient scaling factor C(λ) of each point, and use formula (5) to obtain the gradient of the new wide-spectrum light source input.

[0061] 9. Combined with the gradient descent algorithm, update the gradient: α is the step size, which controls the structure update speed and is a value between 0 and 1. The default value is 1 and is generally 0.9 to 0.99. This is the default value in this embodiment.

[0062] 10. Repeat steps (2), (3), (4), (5), (6) to obtain a new FOM value (can be a wide spectrum input) and compare the FOM new , see if the performance meets the requirements. If not, continue with steps (7), (8), and (9) until the maximum number of iterations is reached or the requirements are met, then stop the iteration and enter the binarization stage. The final output structure refractive index is only Si and SiO2.

[0063] The final dual-wavelength silicon-based mode splitter structure is as follows: Figure 4 As shown. The TE0 signal with a wavelength of 1310nm in the bus waveguide is output from the branch waveguide channel numbered 1, the TE1 signal with a wavelength of 1310nm in the bus waveguide is output from the branch waveguide channel numbered 2, the TE0 signal with a wavelength of 1550nm in the bus waveguide is output from the branch waveguide channel numbered 3, and the TE1 signal with a wavelength of 1550nm in the bus waveguide is output from the branch waveguide channel numbered 4. The designed and generated structure realizes the separation of TE0 mode and TE1 mode without changing the mode order and the composite effect of wavelength separation of 1310nm and 1550nm. The input light sources are all wide-spectrum light sources, ensuring a bandwidth range of at least ±20nm.

[0064] The linear graph of the number of iterations and the FOM of each channel is as follows: Figure 5 As shown in Figure 1, after 139 iterations, the optimization objectives of each channel meet the expected results. The subsequent binarization process will cause the performance of the device after the final convergence to decrease.

[0065] When the device acts on mode separation, the transmission efficiency of TE0 channel (channel 1 and channel 3) is as follows Figure 6 (a) and (c) show that TE1 channels (channel 2 and channel 4) are Figure 6 (b) and (d). Obviously, they maintain low crosstalk while ensuring low insertion loss. When the device acts on wavelength separation, the transmission efficiency of the 1310nm channel (channel 1 and channel 2) is shown as follows: Figure 6 As shown in (a) and (b), the transmission efficiency of the 1550nm channels (channel 3 and channel 4) is as follows Figure 6 As shown in (c) and (d), it is clear that the device achieves high transmission efficiency in a wide range with 1310nm and 1550nm as the center wavelength.

[0066] The linear graph of the number of iterations and the FOM of each channel is as follows: Figure 5 As shown in Figure 1, after 139 iterations, the optimization objectives of each channel meet the expected results. The subsequent binarization process will cause the performance of the device after the final convergence to decrease.

[0067] When the device acts on mode separation, the transmission efficiency of TE0 channel (channel 1 and channel 3) is as follows Figure 6 As shown in (a) and (c), (a) input is 1310nm TE0 mode, (c) input is 1550nm TE0 mode, TE1 channels (channel 2 and channel 4) are as follows Figure 6 As shown in (b) and (d), (b) input is 1310nm TE1 mode, (d) input is 1550nm TE1 mode. Obviously, they maintain low crosstalk while ensuring low insertion loss. When the device acts on wavelength separation, the transmission efficiency of the 1310nm channel (channel 1 and channel 2) is shown in Figure 6 As shown in (a) and (b), the transmission efficiency of the 1550nm channels (channel 3 and channel 4) is as follows Figure 6 Obviously, the device achieves high transmission efficiency in a wide range with 1310nm and 1550nm as the center wavelength.

[0068] Figure 7 The performance of each output when the input light source bandwidth is ±20nm, (a) input is 1310nm TE0 mode; (b) input is 1310nm TE1 mode; (c) input is 1550nm TE0 mode; (d) input is 1550nm TE1 mode. It can be seen that each channel can maintain a low IL in a wide wavelength range. At an operating wavelength centered at 1310nm, when the TE0 mode is input, the IL and CT of the structure are 0.15dB and -30.54dB respectively, when the TE1 mode is input, the IL and CT are less than 0.19dB and -27.67dB respectively; at an operating wavelength centered at 1550nm, when the TE0 mode is input, the IL and CT are 0.11dB and -26.35dB respectively, when the TE1 mode is input, the IL and CT are less than 0.11dB and -24.83dB. When the input light source is a wide spectrum, although IL and CT are reduced compared to when the tolerance is 0, the designed dual-wavelength mode splitter can still maintain a large working bandwidth and the performance is still close to the design target.

[0069] The magnetic field distribution of the TE0 mode and TE1 mode of the dual-wavelength silicon-based mode splitter after optimization design is shown in the figure. Figure 8 As shown. (a) and (b) are the propagation of TE1 signal and TE0 signal with a center wavelength of 1550nm in the bus waveguide and the branch waveguide, respectively, and (c) and (d) are the propagation of TE1 signal and TE0 signal with a center wavelength of 1310nm in the bus waveguide and the branch waveguide, respectively. It can be seen that the designed device supports the separation of TE0 and TE1 modes without changing the mode order in the dual wavelength range.

[0070] <![CDATA[Occupied area (μm 2 )]]> aisle Loss / dB Crosstalk / dB Mode change Related references 50×2 1λand 2TE <1 <-8 NO Reference 1 40×12 1λand 2TE <1.47 <-19.6 NO Reference 2 5×3 1λand 2TE <0.54 <-18.06 NO Reference 3 4.8×4.8 2λand 2TE <0.29 <-17.23 YES Reference 4 6.2×6.2 2λand 2TE <0.19 <-24.83 NO This embodiment

[0071] Table 1

[0072] Table 1 shows the development of similar functional devices in recent years. It can be seen from the table that the designed silicon-based splitter has good performance in terms of compactness. In terms of performance, IL and CT have reached the best values ​​of this type of literature. The device in reference 4 can achieve wavelength and mode mixing and multiplexing integration, but it cannot achieve mode separation without changing the order. Other work can achieve stable mode separation, but cannot achieve wavelength multiplexing effect. Finally, the dual-wavelength silicon-based splitter device designed in this embodiment has a size of only 6.2×6.2μm 2 , it is possible to achieve wavelength separation of 1310nm±20nm and 1550nm±20nm and mode separation of TE0 and TE1 without changing the mode order and wide spectrum input light source. At the same time, through electromagnetic simulation experiments, it was found that the IL and CT of the device were less than 0.19dB and -24.83dB respectively, reaching the advanced level of such devices.

[0073] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit the technical solution. Those skilled in the art should understand that those modifications or equivalent substitutions of the technical solution of the present invention that do not depart from the purpose and scope of the technical solution should be included in the scope of the claims of the present invention.

Claims

1. A method for designing a silicon-based mold splitter, characterized in that: The following steps are involved: (1) Based on a bus waveguide, a branch waveguide, and a functional region located between the bus waveguide and the branch waveguide on a silicon-on-insulator platform, the number of the branch waveguides is n and is determined according to the number of linear combinations of the input mode and the input bandwidth of the bus waveguide, and the initial structures of the bus waveguide, the branch waveguide, and the functional region are defined; (2) Using the FDTD solver, each iteration requires m forward simulations and m adjoint simulations, where m equals n. (3) Calculate the gradient field and optimize the dielectric constant parameters of the functional area based on the gradient field calculation results combined with the gradient descent method to update the structure; (4) Continue to iterate and stop iterating when the design goal is reached or optimization cannot be continued; (5) Binarizing the discrete dielectric constants, converting the dielectric constants of the functional areas into a structure defined only by the functional areas in step (1); (6) Output the final structure diagram.

2. The design method of a silicon-based mold splitter according to claim 1, characterized in that: The initial structures of the bus waveguide, branch waveguide and functional area are defined, including the width of the bus waveguide and the branch waveguide, the spacing between two adjacent branch waveguides, the shape and size of the functional area, and the design materials and background materials of the functional area.

3. The design method of a silicon-based mold splitter according to claim 2, characterized in that: The shape of the functional area is any one of a rectangle, a Y shape, a triangle or an ellipse.

4. The method for designing a silicon-based mold splitter according to claim 2, characterized in that: The design material of the functional area is Si or Si3N4, and the background material is SiO2 or air.

5. A silicon-based mold splitter, characterized in that: The design is carried out according to the design method of the silicon-based mold splitter according to any one of claims 1 to 4.

Images