Edge-coupled beam splitter and fiber optic communication system

Abstract

The application provides an edge-coupling beam splitter and an optical fiber communication system. The edge-coupling beam splitter comprises a silicon substrate layer, a silicon dioxide buried oxygen layer and a silicon dioxide wrapping layer. The silicon substrate layer is arranged below the silicon dioxide wrapping layer, and the silicon dioxide wrapping layer is arranged above the silicon dioxide buried oxygen layer. A beam splitter layer is arranged in the silicon dioxide wrapping layer. A first end of the beam splitter layer is provided with an input end. The input end is connected with an optical fiber. Input light is transmitted to the input end from the optical fiber. The beam splitter layer is provided with a plurality of silicon waveguide layers and a plurality of silicon nitride layers. The silicon waveguide layers and the silicon nitride layers are arranged alternately. The width of each silicon nitride layer is continuously and monotonously decreased along the input light transmission direction. The width of each silicon waveguide layer is continuously and monotonously increased along the input light transmission direction. Each silicon waveguide layer is provided with an output end. The application can reduce the occupied area of the chip and reduce the influence of the angle on the performance of the chip.

Description

Technical Field

[0001] This invention relates to the technical field of integrated photonic devices, specifically to an edge-coupled beam splitter and an optical fiber communication system. Background Technology

[0002] In a typical photonic integrated architecture for a silicon photonic transmitter, four lasers are typically used as the light source. Each laser couples light to a silicon waveguide within the chip via an edge coupler, and then a 1:2 multimode interference (MMI) beam splitter divides the optical power into two paths, ultimately forming eight optical channels. However, this architecture requires four independent lasers, resulting in high costs, and the four edge couplers and four MMI beam splitters occupy chip area, increasing the chip's manufacturing cost.

[0003] An optical device is provided for coupling external optical signals to multiple on-chip photonic circuits disposed on a substrate. The device includes: a planar waveguide layer on the substrate; a diverging grating coupler configured to couple the external optical signal to the planar waveguide layer, thereby generating an on-chip diverging beam in the planar waveguide layer; and multiple channel waveguides formed in the waveguide layer. Each of the multiple channel waveguides includes a waveguide transition structure having a waveguide aperture facing the diverging grating coupler. This optical device uses the grating coupler to achieve beam splitting through in-plane diffraction. Periodic ripples diffract the perpendicularly incident beam into guided wave modes, and the split output is collected by the aperture located within the diffraction fan-shaped region. However, this mechanism is inherently narrowband because the phase-matching condition depends on the wavelength; that is, the narrowband characteristic stems from the fact that its fixed grating period can only strictly match the wave vector of a specific wavelength. Any wavelength shift introduces a linearly increasing phase mismatch, causing the interference enhancement condition to rapidly fail. Furthermore, this mechanism is angle-sensitive. If the fiber deviates from the design angle, and the angle sensitivity is due to the fact that the incident angle directly determines the value of the transverse component in the wave vector matching equation, the angle change is equivalent to changing the resonant wavelength that satisfies phase matching. Therefore, a small angle deviation (such as within the range of 8°~12°) is enough to break the coherent accumulation condition, causing the coupling efficiency to decrease exponentially. Thus, when the angle changes, the coupling efficiency will drop sharply, thereby affecting the chip performance. Summary of the Invention

[0004] The first objective of this invention is to provide an edge-coupled beam splitter that reduces the influence of angle.

[0005] A second objective of this invention is to provide an optical fiber communication system that utilizes the aforementioned edge-coupled beam splitter.

[0006] To achieve the first objective of this invention, the edge-coupled beam splitter provided by this invention includes a silicon substrate layer, a silicon dioxide buried oxide layer, and a silicon dioxide cladding layer. The silicon substrate layer is disposed below the silicon dioxide cladding layer, and the silicon dioxide cladding layer is disposed above the silicon dioxide buried oxide layer. A beam splitter layer is disposed within the silicon dioxide cladding layer. An input terminal is disposed at the first end of the beam splitter layer, and the input terminal is connected to an optical fiber. The input light is transmitted to the input terminal by the optical fiber. The beam splitter layer is provided with multiple silicon waveguide layers and multiple silicon nitride layers, which are alternately arranged. The width of each silicon nitride layer continuously and monotonically decreases along the direction of input light transmission, and the width of each silicon waveguide layer continuously and monotonically increases along the direction of input light transmission. Each silicon waveguide layer is provided with an output terminal.

[0007] As can be seen from the above scheme, the width of the silicon nitride layer decreases monotonically along the direction of input light propagation, while the width of the silicon waveguide layer increases monotonically. Therefore, the end of the silicon waveguide layer near the optical fiber is far below the mode confinement threshold, making it unable to confine the optical field. The input light directly diffuses into the surrounding material. The silicon nitride layer, with its wider initial width, allows its superimposed attenuation field to occupy a sufficient cross-sectional area, thus overlapping with the optical fiber mode of the incident light and receiving the output light. As the input light propagates, the width of the silicon waveguide layer gradually increases, exceeding approximately 200 to 300 nanometers. Due to the higher refractive index of silicon than silicon nitride, the silicon waveguide layer becomes the dominant optical localization structure. The input light received by the silicon nitride layer is transmitted to the silicon waveguide layer via evanescent coupling. Simultaneously, the width of the silicon nitride layer decreases monotonically along the direction of input light propagation, gradually reducing its own light-guiding capability. This invention receives input light through each silicon nitride layer and then evanescently couples it with a silicon waveguide layer to achieve beam splitting. This structure reduces the use of edge couplers and beam splitters, thus reducing the chip's footprint. Furthermore, since diffraction splitting is not performed through a grating coupler, this invention effectively avoids the stringent limitations of the Bragg condition on the incident angle, thereby reducing the impact of angle and increasing chip performance.

[0008] In a further embodiment, the number of silicon waveguide layers is one more than the number of silicon nitride layers, and both ends along the stacking direction of the silicon nitride layers and silicon waveguide layers are silicon waveguide layers.

[0009] Therefore, by using one more silicon waveguide layer than silicon nitride layer, all the input light received by the silicon nitride layer is received by the silicon waveguide layer, reducing transmission loss.

[0010] In a further design, the length of each silicon nitride layer is less than the length of the silicon waveguide layer.

[0011] In a further embodiment, the first end of each silicon nitride layer and the first end of each silicon waveguide layer are positioned close to the optical fiber; the first end of each silicon nitride layer and the first end of each silicon waveguide layer are aligned; and the first end of each silicon nitride layer and the first end of each silicon waveguide layer form an input end.

[0012] In a further embodiment, the second end of each silicon waveguide layer is closer to the output end face of the silicon dioxide encapsulation layer than the second end of each silicon nitride layer.

[0013] Therefore, the function of the silicon nitride layer is to transmit the input light to the silicon waveguide layer. If the lengths of the silicon nitride layer and the silicon waveguide layer are the same, meaning that the silicon nitride layer is still transmitting light at the output end, it will increase transmission loss. Therefore, the length of each silicon nitride layer is shorter than the length of the silicon waveguide layer to reduce transmission loss, and the silicon waveguide layer can be connected to other components at the output end.

[0014] In a further design, a heater is placed above the output end of each silicon waveguide layer.

[0015] This demonstrates that power balance and fine-tuning can be achieved through the thermo-optical effect of the heater, compensating for manufacturing deviations.

[0016] In a further embodiment, the width of the first end of each silicon waveguide layer ranges from 50 nanometers to 80 nanometers; and the width of the second end of each silicon waveguide layer ranges from 420 nanometers to 450 nanometers.

[0017] It can be seen that when the silicon waveguide layer is between 50 nanometers and 80 nanometers, it cannot receive input light signals and the input light will directly dissipate into the surrounding material. However, when the silicon waveguide layer gradually widens to 200 nanometers to 300 nanometers, the input light received by the silicon nitride layer is transmitted to the silicon waveguide layer through evanescent coupling.

[0018] In a further embodiment, the width of the first end of each silicon nitride layer ranges from 0.8 micrometers to 1.1 micrometers; and the width of the second end of each silicon nitride layer ranges from 350 nanometers.

[0019] Therefore, it can be seen that the silicon nitride layer has a relatively wide initial width, and its superimposed attenuation field can occupy a sufficient cross-sectional area, thus overlapping with the fiber mode of the incident light and receiving the output light. The width of the silicon nitride layer continuously and monotonically decreases along the direction of input light transmission, thereby gradually reducing the light guiding capability of the silicon nitride layer itself.

[0020] To achieve the second objective, the optical fiber communication system provided by the present invention is characterized in that the optical fiber communication system uses the aforementioned edge-coupled beam splitter. Attached Figure Description

[0021] Figure 1 This is a structural diagram of an embodiment of the edge-coupled beam splitter of the present invention.

[0022] Figure 2 This is a top view of an embodiment of the edge-coupled beam splitter of the present invention.

[0023] The present invention will be further described below with reference to the accompanying drawings and embodiments. Detailed Implementation

[0024] The edge-coupled beam splitter provided by this invention uses alternating layers of silicon waveguides and silicon nitrides. The width of each silicon nitride layer decreases continuously and monotonically along the direction of input light transmission, while the width of each silicon waveguide layer increases continuously and monotonically along the direction of input light transmission. This allows a single device to perform the functions of an edge coupler and a beam splitter, reducing device losses and angular influences, and increasing chip performance.

[0025] Example of an edge-coupled beam splitter: The edge-coupled beam splitter in this embodiment includes a silicon substrate layer 11, a silicon dioxide buried oxide layer 12, and a silicon dioxide cladding layer 13. The silicon substrate layer 11 is disposed below the silicon dioxide cladding layer 13, and the silicon dioxide cladding layer 13 is disposed above the silicon dioxide buried oxide layer 12.

[0026] The beam splitter layer is disposed within the silica coating layer 13. An input end 1311 is located at the first end of the beam splitter layer and is connected to the optical fiber 14. Input light is transmitted from the optical fiber 14 to the input end 1311. The beam splitter layer comprises multiple silicon waveguide layers 132 and multiple silicon nitride layers 131, which are alternately arranged. The optical fiber 14 is a standard SMF-28 optical fiber.

[0027] The width of each silicon nitride layer 131 decreases continuously and monotonically along the input light transmission direction, while the width of each silicon waveguide layer 132 increases continuously and monotonically along the input light transmission direction. Each silicon waveguide layer 132 is provided with an output terminal 1321. Further, the width of each silicon nitride layer 131 decreases monotonically in a parabolic manner along the input light transmission direction, while the width of each silicon waveguide layer 132 increases monotonically in a parabolic manner along the input light transmission direction.

[0028] The beam splitter layer has a height of 4.8 micrometers, with the silicon nitride layer 131 having a height of 300 nanometers and the silicon waveguide layer 132 having a height of 200 nanometers. The spacing between each silicon nitride layer 131 and each silicon waveguide layer 132 is 80 nanometers. The mode field diameter of the optical fiber 14 at a wavelength of 1310 nm is approximately 9.2 μm, and its emitted mode field spot can completely cover the coupling entrance of the beam splitter layer, ensuring that the beam splitter layer can receive optical signals. Furthermore, since the beam splitting function is performed by the beam splitter layer and is decoupled from the diffraction direction of the grating coupler, fluctuations in the fiber incident angle will not affect the beam splitting uniformity.

[0029] The first end of optical fiber 14 is connected to the input end 1311, and the second end of optical fiber 14 is connected to the laser. The laser emits input light, enabling the edge-coupled beam splitter to receive the input light. The input light enters through the input end 1311. Because the end of the silicon waveguide layer 132 closest to the optical fiber is far below the mode confinement threshold, the silicon waveguide layer 132 cannot confine the optical field, and the input light directly diffuses into the surrounding material. The silicon nitride layer 131, with its relatively wide initial width, can accumulate a sufficient cross-sectional area due to its superimposed attenuation field, thus overlapping with the optical fiber mode of the incident light and receiving the output light. As the input light propagates, the width of the silicon waveguide layer 132 gradually increases, exceeding approximately 200 to 300 nanometers. Since silicon has a higher refractive index than silicon nitride, the silicon waveguide layer 132 becomes the dominant optical localization structure. The input light received by the silicon nitride layer 131 is transmitted to the silicon waveguide layer 132 via evanescent coupling. Meanwhile, the width of the silicon nitride layer 131 decreases continuously and monotonically along the direction of input light transmission, thereby gradually reducing the light guiding capability of the silicon nitride layer 131 itself. Then, the input light is output from each output terminal 1321.

[0030] Furthermore, the number of silicon waveguide layers 132 is two or more, and the number of silicon waveguide layers 132 can be two, four, eight, or more than eight. Among them, the number of silicon waveguide layers 132 is one more than the number of silicon nitride layers 131, and both ends along the stacking direction of silicon nitride layers 131 and silicon waveguide layers 132 are silicon waveguide layers 132.

[0031] The width of the first end of each silicon waveguide layer 132 ranges from 50 nanometers to 80 nanometers, and the width of the second end of each silicon waveguide layer 132 ranges from 420 nanometers to 450 nanometers. The width of the first end of each silicon nitride layer 131 ranges from 0.8 micrometers to 1.1 micrometers, and the width of the second end of each silicon nitride layer 131 ranges from 350 nanometers.

[0032] The length of each silicon nitride layer 131 is less than the length of the silicon waveguide layer 132. The first ends of both silicon nitride layers 131 and silicon waveguide layers 132 are positioned close to the optical fiber 14, and are flush with each other. The first ends of each silicon nitride layer 131 and silicon waveguide layer 132 form an input terminal 1311, such that the second ends of each silicon waveguide layer 132 extend outward relative to the second ends of each silicon nitride layer 131. The outwardly extending portions of each silicon waveguide layer 132 form an output terminal 1321. The second ends of each silicon waveguide layer 132 are closer to the output end face 133 of the silicon dioxide cladding layer 13 than the second ends of each silicon nitride layer 131. The output end face 133 of the silicon dioxide cladding layer 13 is positioned away from the optical fiber 14. The function of silicon nitride layer 131 is to transmit the input light to the silicon waveguide layer. If the length of silicon nitride layer 131 is the same as that of silicon waveguide layer 132, that is, at the output end, silicon nitride layer 131 is still transmitting light, which will increase transmission loss. Therefore, the length of each silicon nitride layer 131 is shorter than the length of silicon waveguide layer 132 to reduce transmission loss, and silicon waveguide layer 132 can be connected to other components at the output end.

[0033] Furthermore, the widths of each silicon nitride layer 131 and each silicon waveguide layer 132 are inconsistent, and the coupling efficiency is adjusted by regulating the widths of each silicon nitride layer 131 and silicon waveguide layer 132. For example, when the number of silicon waveguide layers 132 is greater than two, the width of the first end of the outermost silicon waveguide layer is the widest, for example, 60 nm to 100 nm, while the width of the first end of the silicon waveguide layer 132 located in the middle of the beam splitter layer is narrower, for example, 40 nm to 80 nm, thereby reducing the power attenuation of the fiber Gaussian mode field in the edge region.

[0034] A heater 15 is provided above the output terminal 1321 of each silicon waveguide layer 132. The power is finely adjusted by the thermo-optical effect of the heater 15 to compensate for manufacturing deviations.

[0035] In this invention, after receiving the input light through each silicon nitride layer 131, the light is subsequently evanescently coupled to a silicon waveguide layer, thereby achieving beam splitting of the input light. This structure reduces the use of edge couplers and beam splitters, reducing the chip's footprint. Furthermore, since beam splitting is not performed through a grating coupler, the influence of angle is reduced, increasing chip performance.

[0036] Example of an optical fiber communication system: The optical fiber communication system in this embodiment uses the edge-coupled beam splitter described above.

[0037] The above are merely preferred embodiments of the present invention, but the design concept of the invention is not limited thereto. Without departing from the concept of the present invention, many other equivalent embodiments may be included. Those skilled in the art can make various obvious changes, readjustments and substitutions without departing from the protection scope of the present invention.

Claims

1. An edge-coupled beam splitter, comprising a silicon substrate, a silicon dioxide buried oxide layer, and a silicon dioxide cladding layer, wherein the silicon substrate is disposed below the silicon dioxide cladding layer, and the silicon dioxide cladding layer is disposed above the silicon dioxide buried oxide layer; Its features are: A beam splitter layer is disposed within the silica coating layer. An input end is provided at the first end of the beam splitter layer. The input end is connected to an optical fiber, and the input light is transmitted to the input end by the optical fiber. The beam splitter layer is provided with multiple silicon waveguide layers and multiple silicon nitride layers, and the multiple silicon waveguide layers and multiple silicon nitride layers are alternately arranged; The width of each silicon nitride layer decreases continuously and monotonically along the direction of input light transmission, and the width of each silicon waveguide layer increases continuously and monotonically along the direction of input light transmission. Each of the silicon waveguide layers is provided with an output terminal.

2. The edge-coupled beam splitter according to claim 1, characterized in that: The number of silicon waveguide layers is one more than the number of silicon nitride layers, and both ends of the stacking direction of the silicon nitride layers and the silicon waveguide layers are silicon waveguide layers.

3. The edge-coupled beam splitter according to claim 1, characterized in that: The length of each silicon nitride layer is less than the length of the silicon waveguide layer.

4. The edge-coupled beam splitter according to claim 3, characterized in that: The first end of each silicon nitride layer and the first end of each silicon waveguide layer are disposed close to the optical fiber; The first end of each silicon nitride layer is flush with the first end of each silicon waveguide layer; the first end of each silicon nitride layer and the first end of each silicon waveguide layer form the input terminal.

5. The edge-coupled beam splitter according to claim 4, characterized in that: The second end of each silicon waveguide layer is closer to the output end face of the silicon dioxide wrapping layer than the second end of each silicon nitride layer.

6. The edge-coupled beam splitter according to claim 5, characterized in that: A heater is provided above the output end of each of the silicon waveguide layers.

7. The edge-coupled beam splitter according to claim 5, characterized in that: The width of the first end of each of the silicon waveguide layers ranges from 50 nanometers to 80 nanometers; The width of the second end of each of the silicon waveguide layers ranges from 420 nanometers to 450 nanometers.

8. The edge-coupled beam splitter according to claim 5, characterized in that: The width of the first end of each of the silicon nitride layers ranges from 0.8 micrometers to 1.1 micrometers; The width of the second end of each of the silicon nitride layers is in the range of 350 nanometers.

9. An optical fiber communication system, characterized in that, The optical fiber communication system uses the edge-coupled beam splitter as described in any one of claims 1 to 8.

Images