On-chip heterogeneous material waveguide large-dislocation tolerance end face coupler

Abstract

The invention discloses an on-chip heterogeneous material waveguide large-dislocation tolerance end face coupler. Coupling is achieved through two conical waveguide structures. The structure can be applied to end face couplers made of various materials, during coupling, light in an optical fiber is firstly transmitted to the cone of the waveguide b through the waveguide b, and is coupled at the tail ends of the cone a and the cone b to transmit a mode to the cone of the waveguide a along with continuous advancing change of the cone, then is transmitted to the waveguide a through the cone of the waveguide a, and is finally output through the waveguide a. In the coupling area, the diameter of the conical tail end mode field of the waveguide b is larger than the diameter of the conical tail end mode field of the waveguide a, due to the fact that the mode fields are large enough and wide in distribution, even if longitudinal deviation y, a and b occur, the change of the overlapping area of the conical tail end mode fields is relatively smooth, it can be guaranteed that the coupling loss change is not large, and therefore large-tolerance coupling can be achieved in the mode. The process compatibility is good, and the structure is manufactured by adopting a standard CMOS (complementary metal oxide semiconductor) process.

Description

Technical Field

[0001] This invention relates to an end-face coupler in the field of integrated optical chips, and more specifically, to an on-chip heterogeneous material waveguide large dislocation tolerant end-face coupler. Background Technology

[0002] With the rapid expansion and popularization of technologies such as 5G optical communication networks and data transmission systems, as well as the development and application of technologies such as cloud computing and artificial intelligence, people have higher requirements for the computing power and data rate of semiconductor chips. Among them, integrated optical chips have received more attention due to their faster computing speed and lower power consumption. In integrated optical chips, heterogeneous integration technology can integrate chips with different technologies, materials, and functions into a single system, taking into account the advantages of multiple materials to achieve more complex and multifunctional systems.

[0003] Common coupling methods include end-face coupling and grating coupling. An end-face coupler is a structure that directly connects light from an optical fiber to a chip; coupling typically occurs at the chip's edge. Compared to grating couplers, it has the following advantages: First, low coupling loss. A well-designed end-face coupler typically has a coupling loss of less than 1 dB / end-face, while grating couplers have losses of 3-5 dB, enabling efficient energy transmission. Second, very wide bandwidth. End-face coupling loss mainly depends on mode field matching, and the mode field size is relatively insensitive to wavelength changes. Therefore, it can maintain low loss over a very wide wavelength range. Third, wavelength insensitivity. Due to its wide bandwidth, end-face couplers are suitable for wavelength division multiplexing (WDM) systems, allowing multiple different optical signals to be processed at a single port.

[0004] However, the core problem of traditional end-face couplers lies in the mode field mismatch between the waveguide and the fiber. For example, the mode field diameter of ordinary single-mode fiber is about 6-10 μm, while that of SOI waveguide is only about 0.5 μm, a difference of 20 times. Currently, most couplers employ the following solutions: an inverted conical structure that slowly expands the mode field through an insulating waveguide, but this requires precise control of the conical angle; a multilayer waveguide structure that uses the cladding material to adjust the effective refractive index, but this places high demands on the manufacturing process; and couplers using subwavelength grating structures that have high requirements for waveguide dimensional accuracy and rely on electron beam lithography. All of these methods have high dimensional accuracy requirements and are difficult to meet the requirements for large tolerances. Summary of the Invention

[0005] To address, or at least partially address, the problems existing in the aforementioned background technology, the present invention aims to provide an on-chip heterogeneous material waveguide large dislocation-tolerant end-face coupler. This coupler can effectively compensate for lateral and longitudinal misalignments caused by process variations, achieving efficient, stable, and broadband optical coupling between waveguides of different materials, while simultaneously reducing the requirements for manufacturing precision and improving production yield. The present invention is achieved through the following technical solutions:

[0006] This invention discloses an on-chip heterogeneous material waveguide large dislocation tolerance end-face coupler. The coupler includes a silicon dioxide lower cladding, waveguide a and waveguide b disposed on the silicon dioxide lower cladding, and a low refractive index material is coated on the surface of the fabricated waveguides a, b, and lower cladding as an upper cladding. Waveguide a includes a main body and a tapered portion connected to the main body, the width of which gradually decreases towards the chip end-face 5. Waveguide b includes a main body and a tapered portion connected to the main body, the width of which gradually increases towards the end-face 5. The tapered ends of waveguide a and waveguide b are allowed to have some process errors. The tapered ends have large mode field dimensions; even if there is a certain offset y and spacing x during fabrication, the overlap area of ​​the mode fields of the two tapered ends does not change significantly. Therefore, this structure can achieve large dislocation tolerance. At end face 5, the optical fiber is coupled to waveguide b to enable light to be transmitted from the optical fiber into waveguide b; a second coupling is formed at the tapered end of waveguide a and the tapered end of waveguide b to enable light to be coupled from waveguide b into waveguide a.

[0007] As a further improvement, the width of the waveguide a-tip of the present invention ranges from 100 nm to 1000 nm.

[0008] As a further improvement, the width of the waveguide b-tip of the present invention ranges from 200 nm to 2500 nm.

[0009] As a further improvement, the cone angles of both the conical portion a and the conical portion b of the waveguide described in this invention are less than 10°.

[0010] As a further improvement, in the coupling region, the present invention allows for a certain error between the tapered ends of waveguide a and waveguide b, with an offset y of 0-0.5 μm and a spacing x of 0-2 μm.

[0011] As a further improvement, the mode field diameter of the conical end of waveguide a in this invention is 0.5-3 μm, and the mode field diameter of the conical end of waveguide b is 1-5 μm.

[0012] As a further improvement, the conical ends of waveguide a and waveguide b are indirectly contacted through an upper cladding layer, which is a low-refractive-index dielectric layer.

[0013] As a further improvement, the thickness of waveguide a in this invention is 200 nm to 500 nm, and the thickness of waveguide b is 200 nm to 800 nm.

[0014] As a further improvement, the operating wavelength of the coupler described in this invention covers 1200 nm to 2000 nm.

[0015] The present invention discloses the following technical effects:

[0016] This invention discloses an on-chip heterogeneous material waveguide large dislocation-tolerant end-face coupler, which achieves coupling through two tapered waveguide structures. This structure can be applied to end-face couplers made of various materials. During coupling, the light in the optical fiber first propagates through waveguide b to the tapered section of waveguide b, and changes as the tapered section advances. Coupling occurs at the ends of tapered sections a and b, transferring the mode to the tapered section of waveguide a, and then from the tapered section of waveguide a back to waveguide a, finally outputting through waveguide a. In the coupling region, the mode field diameter at the end of the tapered section b is larger than that at the end of the tapered section a. Due to the sufficiently large and wide mode field distribution, even with a longitudinal deviation y, the change in the overlap area of ​​the mode fields at the ends of tapered sections a and b is relatively gradual, ensuring minimal change in coupling loss. Therefore, this method can achieve large-tolerance coupling.

[0017] The present invention has good process compatibility. The structure is manufactured using standard CMOS process. The fabrication of silicon and silicon nitride waveguides and the etching of tapered structure are both mature processes that do not require additional complex steps. Attached Figure Description

[0018] Figure 1 This is a three-dimensional structural schematic diagram of the end-face coupler according to an embodiment of the present invention;

[0019] Figure 2 yes Figure 1 The example shown is a top view of the light propagation direction when there is no misalignment.

[0020] Figure 3 yes Figure 1 The example shown is a top view of the light propagation direction when there is misalignment.

[0021] Figure 4 This is a graph showing the effect of the present invention in a specific example, with the coupling efficiency changing when the spacing between the silicon and silicon nitride conical portions is 0.5 μm and 2 μm, respectively.

[0022] In the diagram, 1 is the main body of waveguide a, 2 is the tapered part of waveguide a, 3 is the main body of waveguide b, 4 is the tapered part of waveguide b, 5 is the chip end face, and 6 is the silicon dioxide undercoating. Detailed Implementation This invention discloses an on-chip heterogeneous material waveguide large dislocation tolerance end-face coupler. The coupler includes a silicon dioxide lower cladding 6, waveguide a and waveguide b disposed on the silicon dioxide lower cladding 6. A deposited silicon dioxide cladding layer is deposited on the surfaces of the fabricated waveguide a, waveguide b, and lower cladding for protection. Waveguide a includes a main body portion 1 and a tapered portion 2 connected to the main body portion 1, the width of which gradually decreases towards the chip end-face 5. Waveguide b includes a main body portion 3 and a tapered portion 4 connected to the main body portion 3, the width of which gradually increases towards the end-face. A coupling region exists at the chip end-face 5 to enable light coupling from the optical fiber into waveguide b. The end portion of the tapered portion 2 of waveguide a and the end portion of the tapered portion 4 of waveguide b form another coupling region to enable light coupling from the end of the tapered portion 4 of waveguide b into the end of the tapered portion 2 of waveguide a. The tapered coupling regions of waveguides a and b are allowed to have some manufacturing errors. The tapered ends have large mode field sizes, and even with certain manufacturing errors, the integral of the overlap between the two mode fields during coupling can be kept small, thus ensuring that the energy loss is not large. Therefore, the structure is allowed to have a certain offset y and spacing x during fabrication.

[0023] The end width of the conical section 2 of waveguide a ranges from 100 nm to 1000 nm; the end width of the conical section 4 of waveguide b ranges from 200 nm to 2500 nm. The cone angles of both conical sections 2 and 4 are less than 10°. In the coupling region, a certain offset (y) of 0-0.5 μm and a spacing (x) of 0-2 μm are allowed between the ends of conical sections 2 and 4 of waveguide a. The mode field diameter at the end of conical section 2 of waveguide a is 0.5-3 μm, and the mode field diameter at the end of conical section 4 of waveguide b is 1-5 μm. The ends of conical sections 2 and 4 of waveguide a are indirectly contacted through an upper cladding layer, which is a low-refractive-index dielectric layer. The thickness of waveguide a is 200 nm to 500 nm, and the thickness of waveguide b is 200 nm to 800 nm. The operating wavelength of the coupler covers 1200 nm to 2000 nm.

[0024] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way.

[0025] like Figures 1 to 3 As shown, a preferred embodiment of the present invention provides an end-face coupler for coupling silicon waveguides and silicon nitride waveguides, where waveguide a is a silicon waveguide and waveguide b is a silicon nitride waveguide. This structure is fabricated on a silicon dioxide underlayer 6.

[0026] A silicon waveguide is formed on a silicon dioxide underlayer 6, comprising a main waveguide portion 1 (waveguide a) and a conical waveguide portion 2 (waveguide a tapered shape), i.e., the main silicon waveguide portion 1 and the conical waveguide portion 2. The typical dimensions of the main silicon waveguide portion are: thickness 220 nm and width 450 nm. The conical waveguide portion is connected to the main waveguide portion, and its width linearly decreases from 450 nm to an extremely narrow end. In this embodiment, the width of this end is designed to be 160 nm, the length of the cone is 200 μm, and the corresponding cone angle is extremely small, less than 10°, ensuring adiabatic transformation of the optical mode field.

[0027] A silicon nitride waveguide is formed on a silicon dioxide underlayer 6. It includes a main waveguide portion 3 (waveguide b) and a conical waveguide portion 4 (waveguide b tapered section 4), which are the main silicon nitride waveguide portion 3 and the conical waveguide portion 4. The typical dimensions of the main silicon nitride waveguide portion are: a thickness of 400 nm and a width of 1.0 μm. The conical waveguide portion is connected to the main waveguide portion, and its width gradually decreases from 1.0 μm to an extremely narrow end. In this embodiment, the width of this end is designed to be 0.25 μm, and the length of the conical portion is 400 μm.

[0028] A low-refractive-index upper cladding layer is applied to the surfaces of the fabricated silicon waveguide, silicon nitride waveguide, and silicon dioxide lower cladding 6. The optical signal enters from the main body of the silicon nitride waveguide via optical fiber, first passing through the silicon nitride waveguide tapered section. During this process, the mode field size gradually expands as the tapered waveguide narrows. Upon reaching the end regions of both waveguide tapered sections, optical energy coupling occurs, transferring energy from the silicon nitride tapered end to the silicon waveguide tapered end. Subsequently, the silicon waveguide tapered section, through its gradually widening structure, thermally transfers the optical mode field to the main body of the silicon waveguide, completing the optical signal transmission. The reverse transmission principle is the same.

[0029] Figure 4 The simulation comparison of the output coupling efficiency of this embodiment under different conditions is shown. Figure 4 The detailed results are shown in Table 1 below:

[0030] Table 1:

[0031]

[0032] As can be seen from the table, the output coupling efficiency gradually decreases with the increase of the offset y. The increase in offset y leads to a smaller overlap area of ​​the optical field modes at the tapered ends of the two waveguides, resulting in a decrease in coupling. It can be seen that the decrease in coupling efficiency is not significant when the offset y is less than 500 nm. Therefore, the offset y in this invention should be controlled within 500 nm. Simultaneously, it can be seen that the coupling efficiency decreases with the increase of the spacing x, but the change is not significant. Therefore, the spacing x in this invention should be controlled within 2 μm. Compared to the high process requirements required for perfect alignment, this scheme, even with some overlay deviations causing a certain deviation between x and y, does not significantly reduce the coupling efficiency, thus achieving a low-cost, high-tolerance end-face coupler.

[0033] In summary, this invention successfully solves the process tolerance problem in manufacturing silicon and silicon nitride end-face couplers through a conical structure design. This invention features a simple structure, excellent performance, and ease of manufacture, and is of great significance for promoting the development of hybrid silicon / silicon nitride photonic integration platforms.

[0034] The present invention has been described in detail above, but it is not limited to the described embodiments. It will be apparent to those skilled in the art that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the invention. Therefore, the scope of protection of the present invention should be defined by the appended claims and their equivalents.

Claims

1. A large dislocation-tolerant end-face coupler for on-chip heterogeneous material waveguides, characterized in that, The coupler includes a silicon dioxide lower cladding, waveguide a and waveguide b disposed on the silicon dioxide lower cladding, and a layer of low refractive index material is coated on the surface of the fabricated waveguides a, b and the lower cladding as an upper cladding. Waveguide a includes a main body and a tapered portion connected to the main body, the width of which gradually decreases towards the chip end face 5. Waveguide b includes a main body and a tapered portion connected to the main body, the width of which gradually increases towards the end face 5. The tapered ends of waveguide a and waveguide b are allowed to have some manufacturing process errors. The tapered ends have large mode field dimensions. Even with a certain offset y and spacing x during fabrication, the overlap area of ​​the mode fields at the two tapered ends does not change significantly. Therefore, this structure can achieve large dislocation tolerance. Optical fiber is coupled to waveguide b at end face 5 to allow light to be transmitted from the optical fiber into waveguide b. A second coupling is formed at the tapered ends of waveguide a and waveguide b to allow light to be coupled from waveguide b into waveguide a.

2. The end-face coupler according to claim 1, characterized in that, The width of the tapered end of the waveguide a ranges from 100 nm to 1000 nm.

3. The end-face coupler according to claim 1, characterized in that, The width of the tapered end of the waveguide b ranges from 200 nm to 2500 nm.

4. The end-face coupler according to claim 1, characterized in that, The cone angles of both the conical portion of waveguide a and the conical portion of waveguide b are less than 10°.

5. The end-face coupler according to any one of claims 1 to 4, characterized in that, In the coupling region, a certain error is allowed between the tapered ends of waveguide a and waveguide b, with an offset y of 0-0.5 μm and a spacing x of 0-2 μm.

6. The end-face coupler according to claim 5, characterized in that, The mode field diameter at the tapered end of waveguide a is 0.5-3 μm, and the mode field diameter at the tapered end of waveguide b is 1-5 μm.

7. The end-face coupler according to claim 6, characterized in that, The conical end of waveguide a and the conical end of waveguide b are indirectly in contact through an upper cladding layer, which is a low refractive index dielectric layer.

8. The end-face coupler according to claim 1, 2, 3, 4, 6, or 7, characterized in that, The thickness of waveguide a is 200 nm to 500 nm, and the thickness of waveguide b is 200 nm to 800 nm.

9. The end-face coupler according to claim 8, characterized in that, The coupler operates in the wavelength range of 1200 nm to 2000 nm.

Images