A lithium niobate thin film phase modulator heterogeneously integrated with silicon nitride

Abstract

The present invention discloses a lithium niobate thin film phase modulator heterogeneously integrated with silicon nitride, belonging to the field of integrated optical gyroscopes; comprising a substrate, a silicon dioxide cladding, a lithium niobate thin film-silicon nitride optical waveguide core layer, and a modulation electrode; wherein the optical waveguide core layer is a ridge structure, comprising a lithium niobate thin film layer, a silicon nitride layer, and a silicon nitride carrier strip, satisfying single-mode transmission conditions; the silicon nitride carrier strip constitutes a multimode interference coupler, a bending unit, and a modulation arm; the multimode interference coupler adopts a one-to-two structure with a splitting ratio close to 1:1; the ridge waveguide structure is used for light transmission, and the multimode interference coupler unit is used for light splitting; simultaneously, three modulation electrodes generate an electric field, causing the light transmitted in the two modulation arms to undergo phase changes, thereby outputting two beams of light with a certain phase difference. Polarization is achieved based on the different binding effects of the ridge waveguide on the TE mode and the TM mode, that is, the loss of the TE mode and the TM mode in the waveguide is different, achieving a high polarization extinction ratio.

Description

Technical Field

[0001] The present invention belongs to the field of integrated optical gyroscopes, in particular to a lithium niobate thin film phase modulator heterogeneously integrated with silicon nitride. Background Art

[0002] After decades of research and development since the 20th century, optical gyros have become a mainstream form of inertial instrumentation. Optical gyros offer significant advantages, including solid-state performance, high precision, and reliability. Their applications span a wide range of military and civilian applications, including navigation and positioning, attitude control, and drilling exploration.

[0003] The optical gyroscope is mainly composed of a phase modulator, a coupler, a detector, an angular velocity sensitive unit and a light source. The phase modulator is one of its core components, which plays the role of splitting and phase modulation of the transmitted light field.

[0004] With the emergence and development of the concept of integration, scientists hope to utilize advanced integration technology to integrate all optical components of an optical gyroscope onto a single optical chip, enabling the mass production of miniaturized integrated optical gyroscopes. To further achieve the integration of phase modulators, a phase modulator based on lithium niobate thin film material is required. This allows for a smaller size, lower half-wave voltage, and a certain degree of cost reduction.

[0005] Among integrated optical gyroscope sensing solutions, the resonant integrated optical gyroscope has an advantage in sensing principle. Through the resonant principle, the sensitivity is improved by multiple transmissions of light in a single-circle sensitive ring, which can achieve the principle advantage of small size and high precision.

[0006] Currently in resonant integrated optical gyroscope technology, the traditional silicon dioxide resonant cavity has polarization errors that are difficult to suppress, while the silicon nitride resonant cavity theoretically has lower losses. At the same time, the extremely low aspect ratio structure can achieve single polarization through bending polarization, thereby suppressing polarization noise from a mechanistic perspective.

[0007] The novel silicon nitride resonant cavity, capable of achieving single-polarization characteristics, has attracted extensive attention and research due to its unique advantages. Furthermore, both the silicon nitride resonant cavity and the lithium niobate thin-film modulator possess high polarization extinction ratios. Monolithic integration of these two elements is crucial for achieving high precision and chip-scale integration of integrated optical gyros.

[0008] Based on the above considerations, it is particularly important to study lithium niobate thin film phase modulators that can be heterogeneously integrated with silicon nitride resonant cavities and have mode field matching. Summary of the Invention

[0009] To address the above problems, the present invention provides a lithium niobate thin film phase modulator heterogeneously integrated with silicon nitride. The lithium niobate thin film phase modulator is suitable for the field of resonant integrated optical gyroscope technology and is conducive to good matching with the silicon nitride resonant cavity mode field.

[0010] The lithium niobate thin film phase modulator heterogeneously integrated with silicon nitride comprises a substrate, a silicon dioxide cladding, a lithium niobate thin film-silicon nitride optical waveguide core layer, and a modulation electrode.

[0011] The substrate material is silicon or lithium niobate, located at the bottom layer of the device, and plays a supporting and protective role.

[0012] The silica cladding is a double-layer structure, including a first silica layer and a second silica layer. Taking advantage of the fact that the refractive index of silica is lower than that of lithium niobate and silicon nitride, silica is respectively covered on the upper and lower layers of the lithium niobate film-silicon nitride optical waveguide layer to serve as the cladding of the optical waveguide.

[0013] The lithium niobate film-silicon nitride optical waveguide core layer includes a lithium niobate film layer, a silicon nitride layer arranged on the lithium niobate film layer, and a silicon nitride carrier strip arranged on the silicon nitride layer. The three together constitute a ridge optical waveguide core layer; the silicon nitride layer acts as a buffer in the middle to improve the quality of the transmitted light field.

[0014] The lithium niobate film-silicon nitride optical waveguide core layer and the silicon dioxide cladding together form a complete waveguide structure for light transmission.

[0015] The sum of the thickness of the silicon nitride layer and the silicon nitride carrier strip should be slightly smaller than the thickness of the lithium niobate film layer, which is about 2 / 3 of the thickness of the lithium niobate film layer. At the same time, the thickness of the silicon nitride layer and the silicon nitride carrier strip should be comparable.

[0016] The silicon nitride carrier strip constitutes a multimode interference coupler unit, two bending units and two modulation arm units, realizing the splitting and modulation functions; the multimode interference coupler unit includes a tapered gradient input waveguide, a multimode waveguide and two tapered gradient output waveguides.

[0017] The narrow end of the tapered gradient input waveguide receives the light output by the laser, and the wide end is connected to one end of the multimode waveguide. The other end of the multimode waveguide is connected to the wide ends of the two tapered gradient output waveguides. At the same time, the narrow ends of the two tapered gradient output waveguides are respectively connected to the front ends of the two bending units; the rear ends of the two bending units are respectively connected to the two modulation arm units; the two modulation arm units are respectively located in the middle of the three modulation electrodes, forming a push-pull electrode configuration.

[0018] All other ridge waveguides except the multimode interference coupler unit need to meet single-mode transmission conditions to suppress multimode noise; the details are as follows:

[0019] First, the field on the cross section of the silicon nitride-lithium niobate film ridge waveguide is divided into two cases:

[0020] 1) E x 、H y Main model E x mn , that is, the transverse electric field component and the longitudinal magnetic field component, which is equivalent to TE polarization;

[0021] 2) E y 、H x Main model E y mn , that is, the longitudinal electric field component and the transverse magnetic field component, which is equivalent to TM polarization.

[0022] Then, according to Maxwell's equations of time-harmonic electromagnetic field, the vector is expanded according to each component and the Macatili processing method is used to set H x =0 and H y =0, thus determining the modulus E x mn and Model E y mn The characteristic equation of the waveguide mode field distribution is obtained; and after preliminary data estimation based on the waveguide mode field distribution, it is further verified whether the single-mode condition is met.

[0023] for The guided mode equation it satisfies is:

[0024]

[0025] in,

[0026]

[0027] Among them, k x and k y Represent the wave number in the x and y directions respectively; w represents the half width of the ridge waveguide rectangular core layer; t represents the half thickness of the ridge waveguide rectangular core layer; n1, n2, n3, n4, and n5 are all refractive indices after equivalent calculation; n1 represents the refractive index of the ridge waveguide rectangular core layer; n2 represents the refractive index of the left cladding of the ridge waveguide rectangular core layer; n3 represents the refractive index of the right cladding of the ridge waveguide rectangular core layer; n4 represents the refractive index of the upper cladding of the ridge waveguide rectangular core layer; n5 represents the refractive index of the lower cladding of the ridge waveguide rectangular core layer; λ represents the wavelength of light; ω represents the circular frequency; μ represents the magnetic permeability of the medium; ε represents the dielectric constant of the medium;

[0028] To E x mn Guided mode (equivalent to TE polarization), similarly, using the duality principle, just replace t and w, and x and y in the guided mode equation.

[0029] For the above two guided mode equations, when t and w are only valid for m and n = 1 respectively, the waveguide satisfies the TM / TE single-mode condition. The intersection of the two is the single-mode condition of the designed silicon nitride-lithium niobate waveguide.

[0030] The multimode interference coupler unit should adopt a one-to-two structure with a splitting ratio close to 1:1, and its input waveguide and output waveguide should be symmetrical about the center line of the multimode waveguide to ensure 1:1 splitting;

[0031] The relationship between the length and width of the multimode waveguide is:

[0032]

[0033] Among them L M and W M They represent the length and width of the multimode waveguide respectively, n1 represents the refractive index of the ridge waveguide core layer, n2 represents the refractive index of the ridge waveguide cladding, λ0 represents the wavelength of light, for TE mode, σ=0, for TM mode, σ=1.

[0034] The specific working principle of the lithium niobate thin film phase modulator heterogeneously integrated with silicon nitride is as follows:

[0035] The light emitted by the laser enters the multimode waveguide from the input end of the multimode interference coupler unit input waveguide, and is split into two by two tapered gradient output waveguides and a bending unit, generating two identical beams of light that enter the two modulation arm units respectively. At the same time, the three modulation electrodes generate an electric field. Due to the linear electro-optic effect of lithium niobate, the refractive index of lithium niobate changes under the action of the external electric field, which causes the phase of the light transmitted in the two modulation arms to change, thereby outputting two beams of light with a certain phase difference.

[0036] The phase shift introduced by the applied electric field is expressed as:

[0037]

[0038] in, represents the phase shift, n e represents the refractive index of lithium niobate e-light, γ 33 =30.9×10 -12 m / V represents the electro-optic tensor, L z Represents the electro-optical modulation length, E op represents the transmission light field in the waveguide, E ele It is the modulated external electric field generated by the electrode action, and the light transmission direction is defined as the x direction.

[0039] When the phase shift is π, the corresponding voltage is the half-wave voltage V π , then the modulation efficiency is expressed as:

[0040]

[0041] Where V is the applied voltage, and a push-pull electrode configuration is employed. Numerical calculations show that the decisive factor influencing modulation efficiency is the distance between the adjacent electrode and waveguide surfaces. A smaller distance results in higher modulation efficiency, but too small a distance introduces absorption losses of the light field in the waveguide due to the electrode. Therefore, the modulation efficiency should be maximized while avoiding absorption by the metal electrode. Optimally, a distance of 1.8±0.1 μm between the adjacent electrode and waveguide surfaces achieves a modulation efficiency of approximately 2 V-cm. This achieves high modulation efficiency while maintaining a compact device and reducing losses.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] The present invention discloses a lithium niobate thin film phase modulator heterogeneously integrated with silicon nitride. The phase modulator is based on silicon nitride and lithium niobate thin film materials, fully utilizing the characteristics of both materials. The linear phase modulation characteristics of lithium niobate are combined with the low transmission loss characteristics of silicon nitride to achieve the two basic functions of light splitting and phase modulation. The device can reduce the volume and reduce the cost. In addition to the basic light splitting and phase modulation functions, the device can further achieve low loss and high polarization extinction ratio. The device is suitable for the field of integrated optical gyroscope technology and is also conducive to mode field matching with the silicon nitride resonant cavity. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a side view of the lithium niobate film-silicon nitride optical waveguide phase modulator chip structure in an embodiment of the present invention;

[0045] Figure 2 A top view of the lithium niobate film-silicon nitride optical waveguide phase modulator chip structure in an embodiment of the present invention;

[0046] Figure 3 Schematic diagram of the cross section of the input port in an embodiment of the present invention.

[0047] Figure 4 Schematic cross-sectional view of a modulation arm in an embodiment of the present invention.

[0048] Figure 5 Schematic diagram of the fundamental mode light field intensity distribution in the lithium niobate film-silicon nitride optical waveguide in an embodiment of the present invention.

[0049] Figure 6 Schematic diagram of the cross section of an equivalent rectangular waveguide with a ridge structure in an embodiment of the present invention.

[0050] Figure 7 This is a curve diagram of single-mode transmission conditions of lithium niobate film-silicon nitride optical waveguide in an embodiment of the present invention.

[0051] Figure 8 1 is a graph showing how the output electric field intensity of the multimode interference coupler changes with the length of the multimode waveguide when the TE fundamental mode is input in an embodiment of the present invention.

[0052] Figure 9 FIG. 4 is a diagram of light field transmission in a multimode interference coupler according to an embodiment of the present invention.

[0053] Figure 10 Light field transmission diagram in the first bending unit in an embodiment of the present invention.

[0054] In the figure: 1-substrate, 2-first silicon dioxide layer, 3-lithium niobate thin film layer, 4-silicon nitride layer, 5-second silicon dioxide layer, 6-first electrode, 7-second electrode, 8-third electrode, 41-tapered gradient input waveguide, 42-multimode waveguide, 43-first tapered gradient output waveguide, 44-second tapered gradient output waveguide, 45-first bending unit, 46-second bending unit, 47-first modulation arm, 48-second modulation arm.

[0055] In order to more clearly illustrate the core structure of the present invention, Figure 1 and Figure 2 The second silicon dioxide layer 5 is made transparent to better display the lithium niobate film-silicon nitride ridge waveguide structure. DETAILED DESCRIPTION

[0056] The following is a complete and detailed description of the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without inventive effort are also within the scope of protection of the present invention.

[0057] In order to make the above-mentioned objects, features and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0058] The present invention provides a lithium niobate thin film phase modulator that can be heterogeneously integrated with silicon nitride. Because lithium niobate has excellent linear phase modulation characteristics, the present invention utilizes its linear electro-optical effect to achieve phase modulation function. However, this material has a more prominent problem: high loss. Considering this problem, the lithium niobate thin film is heterogeneously integrated with other materials to jointly realize the production of a phase modulator. Silicon nitride is a very ideal material among them. It has a refractive index similar to that of lithium niobate and has extremely low transmission loss. In addition, both lithium niobate and silicon nitride have the characteristics of high polarization extinction ratio. In this way, silicon nitride and lithium niobate films can be heterogeneously integrated, thereby fully utilizing the advantages of both. A silicon nitride-lithium niobate film ridge waveguide structure is designed as the waveguide core layer, and silicon dioxide material is selected to form the upper and lower cladding structures to achieve light transmission. The phase modulator made of the above structure is also more conducive to mode field matching and heterogeneous integration with the silicon nitride resonant cavity; it can reduce the volume and reduce the cost. In addition to the basic splitting and phase modulation functions, it can further achieve low loss and high polarization extinction ratio, which is suitable for the field of integrated optical gyroscope technology. At the same time, it is also conducive to mode field matching with the silicon nitride resonant cavity.

[0059] The lithium niobate thin film phase modulator heterogeneously integrated with silicon nitride, such as Figure 1 As shown, it includes a substrate, a silicon dioxide cladding, a lithium niobate film-silicon nitride optical waveguide core layer, and a modulation electrode.

[0060] The substrate material is silicon or lithium niobate, located at the bottom layer of the device, and plays a supporting and protective role.

[0061] The silica cladding is a double-layer structure, including a first silica layer and a second silica layer. Taking advantage of the fact that the refractive index of silica is lower than that of lithium niobate and silicon nitride, silica is respectively covered on the upper and lower layers of the lithium niobate film-silicon nitride optical waveguide layer to serve as the cladding of the optical waveguide.

[0062] The lithium niobate film-silicon nitride optical waveguide core layer includes a lithium niobate film layer, a silicon nitride layer arranged on the lithium niobate film layer, and a silicon nitride carrier strip arranged on the silicon nitride layer. The three together constitute a ridge optical waveguide core layer; the silicon nitride layer acts as a buffer in the middle to improve the quality of the transmitted light field.

[0063] The lithium niobate film-silicon nitride optical waveguide core layer and the silicon dioxide cladding together form a complete waveguide structure for light transmission.

[0064] The sum of the thickness of the silicon nitride layer and the silicon nitride carrier strip should be slightly less than the thickness of the lithium niobate film layer, which is about 2 / 3 of the thickness of the lithium niobate film layer. At the same time, the thickness of the silicon nitride layer and the silicon nitride carrier strip should be equal. Figure 5As shown, at this time, most of the energy in the fundamental mode of the light field will be concentrated in the lithium niobate layer, achieving a better effect. The purpose is to allow more energy of the light mode field to be transmitted in the lithium niobate layer, thereby improving the electro-optical modulation efficiency of lithium niobate and reducing the half-wave voltage of the device. At this time, compared with a pure lithium niobate thin film modulator, the waveguide transmission loss can be reduced by about 40%.

[0065] In order to realize the splitting and modulation functions, the silicon nitride carrier strip constitutes a multimode interference coupler unit, two bending units and two modulation arm units, realizing the splitting and modulation functions; the multimode interference coupler unit includes a tapered gradient input waveguide, a multimode waveguide and two tapered gradient output waveguides.

[0066] The narrow end of the tapered gradient input waveguide receives the light output by the laser, and the wide end is connected to one end of the multimode waveguide. The other end of the multimode waveguide is connected to the wide ends of the two tapered gradient output waveguides. At the same time, the narrow ends of the two tapered gradient output waveguides are respectively connected to the front ends of the two bending units; the rear ends of the two bending units are respectively connected to the two modulation arm units; the two modulation arm units are respectively located in the middle of the three modulation electrodes.

[0067] Except for the multimode interference coupler unit, the other parts of the ridge waveguide need to meet the single-mode transmission conditions to suppress multimode noise.

[0068] There are three modulation electrodes in total, which are respectively located at the upper, middle and lower positions of the two modulation arm units, forming a push-pull electrode configuration to reduce the size of the device.

[0069] Preferably, in order to make the transition of the light field between the single-mode waveguide and the multi-mode waveguide smoother, improve the quality of the transmitted light field, and reduce the leakage loss at the connection between the two, the tapered gradient input waveguide and the tapered gradient output waveguide have exactly the same structural dimensions, and the two tapered gradient output waveguides are symmetrically positioned with respect to the multi-mode waveguide. The specific dimensions of the tapered gradient structure are designed according to the multi-mode waveguide. The width of the narrow end is the width of the single-mode waveguide, and the width of the wide end is as wide as possible. However, a processing spacing of not less than 1um must be left between the two tapered gradient output waveguides. The longer the length, the longer the transition distance and the smoother the transition. However, the device must also be miniaturized. It is better to take the length of the tapered gradient structure as 2-3 times the length of the multi-mode waveguide.

[0070] Preferably, the two bending units have completely identical structural dimensions, are both composed of two completely identical and tangent arc segments, and meet single-mode transmission conditions.

[0071] Preferably, the two modulation arm units are two straight waveguides with exactly the same structural dimensions.

[0072] Preferably, the modulation electrode material is gold or platinum, the three modulation electrodes have exactly the same structural dimensions, and the electrode-waveguide spacing between each of the adjacent modulation arms is the same.

[0073] The multimode interference coupler unit should adopt a one-to-two structure with a splitting ratio close to 1:1, and its input waveguide and output waveguide should be symmetrical about the center line of the multimode waveguide to ensure 1:1 splitting;

[0074] The specific working principle of the lithium niobate thin film phase modulator heterogeneously integrated with silicon nitride is as follows:

[0075] The light emitted by the laser is transmitted through a ridge waveguide structure formed by a lithium niobate film and silicon nitride, and is split using a multimode interference coupler unit. Specifically, the light enters the multimode waveguide from the input end of the multimode interference coupler unit's input waveguide, and is split into two by passing through two tapered gradient output waveguides and a bending unit, generating two identical beams of light that enter the two modulation arm units respectively. At the same time, three modulation electrodes generate an electric field. Due to the linear electro-optic effect of lithium niobate, the refractive index of lithium niobate changes under the action of the external electric field, which causes the phase of the light transmitted in the two modulation arms to change, thereby outputting two beams of light with a certain phase difference.

[0076] The phase shift introduced by the applied electric field is expressed as:

[0077]

[0078] in, represents the phase shift, n e represents the refractive index of lithium niobate e-light, γ 33 =30.9×10 -12 m / V represents the electro-optic tensor, L z Represents the electro-optical modulation length, E op represents the transmission light field in the waveguide, E ele It is the modulated external electric field generated by the electrode action, and the light transmission direction is defined as the x direction.

[0079] When the phase shift is π, the corresponding voltage is the half-wave voltage V π , then the modulation efficiency is expressed as:

[0080]

[0081] Where V is the applied voltage, and a push-pull electrode configuration is employed. Numerical calculations show that the decisive factor influencing modulation efficiency is the distance between the adjacent electrode and waveguide surfaces. A smaller distance results in higher modulation efficiency, but too small a distance introduces absorption losses of the light field in the waveguide due to the electrode. Therefore, the modulation efficiency should be maximized while avoiding absorption by the metal electrode. Optimally, a distance of 1.8±0.1 μm between the adjacent electrode and waveguide surfaces achieves a modulation efficiency of approximately 2 V-cm. This achieves high modulation efficiency while maintaining a compact device and reducing losses.

[0082] Example:

[0083] The specific manufacturing process of the lithium niobate thin film phase modulator heterogeneously integrated with silicon nitride is as follows:

[0084] Step 1: Determine a single-mode waveguide structure with high modulation efficiency that is conducive to mode field matching with the silicon nitride resonant cavity;

[0085] In order to suppress the multimode noise interference in the waveguide, the silicon nitride-lithium niobate film ridge waveguide structure needs to meet the single-mode transmission conditions. The ridge optical waveguide structure, after first being processed by the equivalent refractive index method, is also a type of rectangular waveguide, such as Figure 6 As shown. Since most of the light field energy is concentrated in the core layer, if the value of t / w in the strip waveguide is large, there are two cases of the field quantity on the cross section: 1) E x 、H y Main model E x mn , that is, the transverse electric field component and the longitudinal magnetic field component, which is equivalent to TE polarization;

[0086] 2) E y 、H x Main model y mn , that is, the longitudinal electric field component and the transverse magnetic field component, which is equivalent to TM polarization.

[0087] The following are analyzed separately:

[0088] According to Maxwell's equations for time-harmonic electromagnetic fields:

[0089]

[0090] Expanding the vector by its components yields:

[0091]

[0092]

[0093] ⑴E x 、H y Main model x mn

[0094] According to Macatili's method, let H in equations (1.2) and (1.3) x =0, we can get:

[0095]

[0096]

[0097] ⑵E y 、H x Main model y mn

[0098] Similarly, let H in equations (1.2) and (1.3) be y =0, we can get:

[0099]

[0100]

[0101] Next, E y mn The wave equation can be used to analyze the mode. Figure 6 The magnetic field distribution in the five regions has the following form:

[0102]

[0103] in

[0104]

[0105] According to H at y = ±t x and E z Continuous, available

[0106]

[0107] From this we can get

[0108]

[0109] Considering that the mode field order of a general rectangular waveguide starts from m = 1 and n = 1, Equation (1.11) can be rewritten as

[0110]

[0111] in

[0112]

[0113] Similarly, according to H at x = ±ω x and E z Continuous, available

[0114]

[0115] in

[0116]

[0117] From this we can determine the module E y mnThe characteristic equations of (k) are (1.12) and (1.14), which can be obtained by numerical methods. x ,k y ); then the waveguide mode field distribution can be obtained. After making a preliminary data estimate based on the waveguide mode field distribution, the simulation software is further used to verify whether the single-mode condition is met, thereby obtaining the optimized value, as shown in the attached figure. Figure 7 shown.

[0118] for The guided mode equation it satisfies is:

[0119]

[0120] in,

[0121]

[0122] Among them, k x and k y Represent the wave number in the x and y directions respectively; w represents the half width of the ridge waveguide rectangular core layer; t represents the half thickness of the ridge waveguide rectangular core layer; n1, n2, n3, n4, and n5 are all refractive indices after equivalent calculation; n1 represents the refractive index of the ridge waveguide rectangular core layer; n2 represents the refractive index of the left cladding of the ridge waveguide rectangular core layer; n3 represents the refractive index of the right cladding of the ridge waveguide rectangular core layer; n4 represents the refractive index of the upper cladding of the ridge waveguide rectangular core layer; n5 represents the refractive index of the lower cladding of the ridge waveguide rectangular core layer; λ represents the wavelength of light; ω represents the circular frequency; μ represents the magnetic permeability of the medium; ε represents the dielectric constant of the medium.

[0123] To E x mn Guided mode (equivalent to TE polarization), similarly, using the duality principle, just replace t and w, and x and y in the guided mode equation.

[0124] For the above two guided mode equations, when t and w are only valid for m and n = 1 respectively, the waveguide satisfies the TM / TE single-mode condition. The intersection of the two is the single-mode condition of the designed silicon nitride-lithium niobate waveguide.

[0125] Preferably, when the thickness of the lithium niobate film layer is 300±50nm and the thickness of the silicon nitride carrier is 80±50nm, the length and width of the silicon nitride carrier can be simplified to satisfy the equation:

[0126] 60.65sin(0.6191L s +2.295)+58.07sin(0.6399L s -0.8487)-D s ≥0

[0127] Among them L sand D s They represent the length and width of the silicon nitride carrier strip, respectively, where the unit of length is um.

[0128] At this time, the waveguide structure meets the single-mode transmission conditions.

[0129] Step 2: Design a one-to-two structure with a splitting ratio close to 1:1;

[0130] The phase modulator used in the integrated optical gyroscope must not only have phase modulation capabilities but also beam splitting capabilities; therefore, a one-to-two structure must be determined. Currently, the more commonly used structure is the Y-branch structure. Considering the requirement that the splitting ratio should be as close to 1:1 as possible, the Y-branch must meet the above requirements, that is, the optical power in its two branches must be equal. This requires that the two branch arms are strictly and completely symmetrical. If the deviation is large, the beam splitting will be poor, which requires very high process requirements. At the same time, the Y-branch structure is usually used when the difference between the core and cladding refractive indices is not high. When the difference between the two is too large, such as in the selected silicon nitride-lithium niobate thin film waveguide, the waveguide cladding refractive index is 1.45 and the core refractive index is about 2.1, resulting in a difference of about 0.7. This value will generate a large amount of backscattering noise at the branch point, which is very unfavorable for the phase modulator.

[0131] Taking all of the above factors into consideration, this example abandons the Y-branch structure and uses a multimode interference coupler structure to achieve the splitting function. The principle of this structure to achieve splitting is the self-image effect, that is, in the multimode waveguide region, the various modes of transmission will interfere with each other in their transmission direction, so that one or more copies of the input light field will appear at specific locations. These specific locations are often spaced periodically. Compared with the Y-branch structure, the multimode interference coupler is easier to process and can avoid back noise. In theory, it is easier to achieve 1:1 splitting. Therefore, it can be seen that within a certain width, it is very important to determine the length of the multimode waveguide.

[0132] The relationship between the width and length of a multimode waveguide is analyzed theoretically below.

[0133] Assume that the refractive index of the core layer of the multimode waveguide is n1, the refractive index of the cladding layer is n2, and the width and length of the multimode waveguide are W M and L M , the input light wavelength is λ0, the waveguide can propagate m waveguide modes, usually m>3, and let ν=0, 1, 2……(m-1) be the order of different propagation modes in the waveguide. Let the propagation constant of the ν-order mode be β v , the longitudinal propagation constant is k zv , calculated by the following formula:

[0134]

[0135] Where W ν It is called the effective width of the νth order mode. To simplify the calculation, it can be considered that W v ≈W M The following relationship is obtained through the dispersion equation:

[0136]

[0137] Where k0 is called the wave number,

[0138]

[0139] From this we can get:

[0140]

[0141] Right now:

[0142]

[0143] Normally From this we can get:

[0144]

[0145] Right now:

[0146]

[0147] Assume that the light field input into the multimode waveguide along the z-axis is Θ(z, 0). Based on the previous assumption that the waveguide can propagate m waveguide modes, and m>3, the input light field is decomposed into a linear superposition of all modes and expanded into a series of characteristic modes, which can be expressed as follows:

[0148]

[0149] The light field is transmitted along the y direction. According to the theory of wave propagation, the light field distribution at y can be expressed as:

[0150]

[0151] For the convenience of analysis, the time variable exp(jωt) is hidden, and the fundamental mode phase exp(-jβ0y) in the summation is extracted as a common factor and hidden. Then the light field can be simplified as follows:

[0152]

[0153] Substituting the base model into it, we can get:

[0154]

[0155] Simplified to:

[0156]

[0157] in

[0158]

[0159] From this we can get the light field distribution at point y:

[0160]

[0161] When the following conditions are met:

[0162]

[0163] Θ(z, y) will be an image of Θ(z, 0), distinguishing between positive and negative images. Let the current position y = L M , then when L M The following conditions are met:

[0164] L M =N(3L π ), N=0, 1, 2,...(2.16)

[0165] At these positions, an input image can be found, either positive or negative. Consider

[0166]

[0167] After substituting the light field distribution, we get:

[0168]

[0169] After decomposing according to the odd-even mode, we can get:

[0170]

[0171] For the characteristic modes propagating in the waveguide, the even-order modes are even-symmetric and the odd-order modes are odd-symmetric, so the above formula can be decomposed into:

[0172]

[0173] After combining like terms:

[0174]

[0175] Right now:

[0176]

[0177] Thus, in: At N=1, 3, 5, ..., two symmetrical images of the input light field will appear, and the amplitudes of the two images are the amplitudes of the input field. The energy is half of the input light field, which can achieve one-to-two.

[0178] In order to further reduce the size of the device and reduce the loss, In the case of only a single input, consider selecting the input position at the center of the multimode waveguide and inputting a symmetrical light field. Then, when the input light field is expanded, there are only even-order terms, that is, the odd-order mode is not excited. At this time, the following relationship is obtained:

[0179] ν(ν+2)mod 4≡0, ν is an even number (2.23)

[0180] Then the image cycle can be shortened to the previous Then in position You can get two outputs at this point, and the size becomes the previous one directly

[0181] From this we can get the relationship between the length and width of the multimode waveguide:

[0182]

[0183] Furthermore, during design, the theoretical dimensions are first calculated according to the above formula, and then optimized through simulation. For the TE mode (transverse electric wave) and the TM mode (transverse magnetic wave), there is a slight difference in the self-image position during transmission, that is, the optimal output position of each is different. This position can be determined by simulating different single-mode light sources. Therefore, the dimension with higher output light intensity for the TE mode and higher loss for the TM mode can be selected. For example, when the multimode waveguide width is 10um, the relationship between the output light energy of the TE mode and the length of the multimode waveguide is shown in the attached figure. Figure 8 As shown, the multimode waveguide length can be selected as 80.46 μm to achieve maximum output in TE mode. At the same time, the maximum TM output position can be simulated to be 70.9 μm, which contributes to improving the polarization extinction ratio to a certain extent.

[0184] The relationship between the length and width of the multimode waveguide is:

[0185]

[0186] Among them L M and W N They represent the length and width of the multimode waveguide respectively, n1 represents the refractive index of the ridge waveguide core layer, n2 represents the refractive index of the ridge waveguide cladding, λ0 represents the wavelength of light, for TE mode, σ=0, for TM mode, σ=1.

[0187] According to the optimization results, the width of the multimode waveguide is selected to be 8±1um.

[0188] Step 3: Design low loss and high polarization extinction ratio;

[0189] In the designed phase modulator, the modulation region occupies the vast majority of the entire device, and thus contributes significantly to losses. This is achieved by heterogeneously integrating the device with silicon nitride, leveraging its low-loss properties. Furthermore, electrodes are required above and below the two modulator arms of the bifurcated structure to activate them. This can be achieved by employing a push-pull configuration, reducing the number of electrodes from four to three, further reducing device size and lowering losses.

[0190] At the same time, considering the influence on the half-wave voltage, assuming that a voltage V is applied between the two electrodes and the electrode spacing is P, the theoretical uniform electric field E0 = V / P, but in reality the electric field is not uniform. Assume that the interaction length between the electric field and the light field is L z , then the phase shift introduced by the external electric field can be expressed as:

[0191]

[0192] G is used to represent the degree of mutual influence between the electric field and the light field, that is, how much the electric field acts to produce a phase shift, which is called the overlap factor. G can be calculated by the following formula:

[0193]

[0194] Among them E op represents the transmission light field in the waveguide, E ele is the modulated external electric field generated by the electrode. From this we can get:

[0195]

[0196] When the phase difference is equal to π, the corresponding voltage is:

[0197]

[0198] The half-wave voltage is an important parameter for evaluating modulators. The smaller this parameter is, the better the modulator performance is. When the modulation arm length is longer, the half-wave voltage is smaller, but the loss will increase accordingly. The two need to be considered comprehensively.

[0199] Furthermore, because the ridge waveguide achieves polarization due to its different confinement effects on the TE (transverse electric) and TM (transverse magnetic) modes, i.e., the losses in the TE and TM modes are different within the waveguide, it is necessary to consider increasing the straight waveguide length to some extent during design, thereby allowing more TM modes to dissipate and improve the polarization extinction ratio. Of course, increasing the length also increases the device size and losses, so the specific dimensions should be determined based on actual needs. For example, in the embodiment, a waveguide length of 10 mm can be simulated to achieve a polarization extinction ratio of approximately 45 dB.

[0200] Step 4: Based on the designs in the first three steps, a lithium niobate thin film phase modulator structure that can be heterogeneously integrated with silicon nitride is constructed;

[0201] In order to better set the electrodes, the overall width of the device is set at 60um.

[0202] Structurally, the lower surface of the first silicon dioxide layer 2 is arranged on the upper surface of the substrate 1; the lithium niobate thin film layer 3 is arranged on the upper surface of the first silicon dioxide layer; the silicon nitride layer 4 is arranged on the upper surface of the lithium niobate thin film layer, and a silicon nitride carrier strip is arranged on the upper surface of the silicon nitride layer 4; the lower surface of the second silicon dioxide layer 5 is arranged on the upper surface of the silicon nitride layer 4; the first modulation electrode 6, the second modulation electrode 7 and the third modulation electrode 8 are arranged on the upper surface of the silicon nitride layer 4.

[0203] The material used for substrate 1 is silicon, with a thickness of 500um;

[0204] The silica cladding layer specifically includes a first silica layer 2 with a thickness of 4.7 μm and a second silica layer 4 with a thickness of 1 μm. The refractive index of silica is lower than that of lithium niobate and silicon nitride, forming the cladding of the optical waveguide. At the same time, it can be deposited on the waveguide layer as a buffer layer to reduce metal absorption loss introduced by the electrode.

[0205] The lithium niobate thin film-silicon nitride optical waveguide layer comprises a lithium niobate thin film layer 3, a silicon nitride layer 4, and a silicon nitride carrier strip disposed on the silicon nitride layer 4. Together, these three form a ridge optical waveguide. Polarization is achieved by leveraging the different confinement effects of this ridge waveguide on TE (transverse electric waves) and TM (transverse magnetic waves). The lithium niobate thin film layer 3 is 300nm thick, while the silicon nitride layer 4 and the silicon nitride carrier strip are both 110µm thick.

[0206] Under the action of an external electric field, the lithium niobate thin film layer 3 can change the phase of the transmitted light field due to the electro-optical effect.

[0207] The silicon nitride layer 4 acts as a buffer, which can improve the quality of the transmitted light field and reduce the light field transmission loss.

[0208] The silicon nitride carrier strip constitutes a multimode interference coupler unit, a bending unit and a modulation arm unit. The multimode interference coupler unit specifically includes a tapered gradient input waveguide 41, a multimode waveguide 42, a first tapered gradient output waveguide 43 and a second tapered gradient output waveguide 44. The bending unit specifically includes a first bending unit 45 and a second bending unit 46. The modulation arm unit specifically includes a first modulation arm 47 and a second modulation arm 48. The narrow end of the tapered gradient input waveguide 41 receives the light output by the laser, and the wide end is connected to the multimode waveguide 42. The wide ends of the first tapered gradient output waveguide 43 and the second tapered gradient output waveguide 44 are connected to the multimode waveguide, and the two have exactly the same structural dimensions and are symmetrical in position with respect to the multimode waveguide, thereby ensuring that after the input end of the multimode interference coupler unit receives the light from the laser, it can output it with a splitting ratio close to 1:1, thereby playing a splitting role.

[0209] The left ends of the first curved unit 45 and the second curved unit 46 are connected to the narrow ends of the first tapered output waveguide 43 and the second tapered output waveguide 44, respectively, and the right ends are connected to the first modulation arm 47 and the second modulation arm 48, respectively. Both are composed of two tangent arcs with the same structure, so that the light field can smoothly transition to the modulation arm unit. The first modulation arm 47 and the second modulation arm 48 are straight waveguides with the same structural dimensions. Figure 3 As shown, the tapered input waveguide 41, the first tapered output waveguide 43, and the second tapered output waveguide 44 are all 100 μm long, with a narrow end of 1.5 μm and a wide end of 3 μm. The multimode waveguide 42 is 10 μm wide and 80.46 μm long. The bending unit and the modulation arm unit are both 1.5 μm wide. The ridge waveguide formed with these dimensions meets the requirements for single-mode transmission. To reduce bending losses, the bending unit uses two arcs with a radius of 500 μm to form the optical path from the multimode interference coupler to the modulation arm. The modulation arm is 10 mm long.

[0210] The modulation electrodes specifically include a first modulation electrode 6, a second modulation electrode 7 and a third modulation electrode 8, and their structural dimensions are exactly the same. The first modulation electrode 6 and the second modulation electrode 7 are respectively located at the upper and lower ends of the first modulation arm 47, and the second modulation electrode 7 and the third modulation electrode 8 are respectively located at the upper and lower ends of the second modulation arm 48, and each has the same electrode-waveguide spacing as the adjacent modulation arms, forming a push-pull electrode configuration. Under the action of an external electric field, the electro-optical effect of lithium niobate is utilized to phase modulate the light field transmitted in the first modulation arm 47 and the second modulation arm 48, thereby outputting two beams of light with a certain phase difference. In order to improve the modulation efficiency, the electrode width used is 14um, the thickness is 1.6um, and the length is 10mm. The electrode-waveguide spacing is designed as follows: Figure 4 The distance between the middle modulation arm 48 and the adjacent surface of the electrode 8 is 1.8 μm.

[0211] The light field transmission in the multimode interference coupler and the bending unit obtained by simulation is shown as follows: Figure 9 and Figure 10 Show.

[0212] In this embodiment, based on the aforementioned structural dimensions, the lithium niobate thin film-silicon nitride optical waveguide phase modulator chip has an overall width of 60 μm, a thickness of 506.71 μm, and a length of 10.40 mm. Specific performance parameters obtained through simulation are summarized in Table 1. These parameters demonstrate a near 1:1 splitting ratio, a half-wave voltage of less than 2 V, a total on-chip optical loss of 3.5 dB, and a polarization extinction ratio of 45 dB.

[0213]

[0214] This document describes the principles and implementation methods of the present invention with reference to specific examples. The above examples are intended only to help understand the methods and core concepts of the present invention. For those skilled in the art, variations may occur in the specific implementation and application based on the concepts of the present invention. In summary, this specification should not be construed as limiting the present invention.

Claims

1. A lithium niobate thin film phase modulator heterogeneously integrated with silicon nitride, characterized in that: The invention comprises a substrate, a silicon dioxide cladding, a lithium niobate film-silicon nitride optical waveguide core layer, and a modulation electrode; The optical waveguide core layer and the silica cladding together form a complete waveguide structure for light transmission; The optical waveguide core layer includes a lithium niobate thin film layer, a silicon nitride layer disposed on the lithium niobate thin film layer, and a silicon nitride carrier strip disposed on the silicon nitride layer, the three together forming a ridge-shaped optical waveguide core layer; The silicon nitride carrier strip constitutes a multimode interference coupler unit, two bending units and two modulation arm units, realizing the functions of splitting and modulation; the multimode interference coupler unit includes a tapered gradient input waveguide, a multimode waveguide and two tapered gradient output waveguides; The narrow end of the tapered gradient input waveguide receives the light output by the laser, and the wide end is connected to one end of the multimode waveguide. The other end of the multimode waveguide is connected to the wide ends of the two tapered gradient output waveguides. At the same time, the narrow ends of the two tapered gradient output waveguides are respectively connected to the front ends of the two bending units; the rear ends of the two bending units are respectively connected to the two modulation arm units; The two modulation arm units are located in the middle of the three modulation electrodes, forming a push-pull electrode configuration; All other ridge waveguides except the multimode interference coupler unit need to meet single-mode transmission conditions to suppress multimode noise, as follows: First, the field on the cross section of the silicon nitride-lithium niobate film ridge waveguide is divided into two cases: 1) E x 、H y Main model E x mn , that is, the transverse electric field component and the longitudinal magnetic field component, which is equivalent to TE polarization; 2) E y 、H x Main model y mn , that is, the longitudinal electric field component and the transverse magnetic field component, which is equivalent to TM polarization; Then, according to Maxwell's equations of time-harmonic electromagnetic field, the vector is expanded according to each component and the Macatili processing method is used to set H x =0 and H y =0, thus determining the modulus E x mn and Model E y mn The characteristic equation of the waveguide mode field distribution is obtained; and after preliminary data estimation based on the waveguide mode field distribution, further verification is made to determine whether the single-mode condition is met; for The guided mode equation it satisfies is: in, Among them, k x and k y Represent the wave number in the x and y directions respectively; w represents the half width of the ridge waveguide rectangular core layer; t represents the half thickness of the ridge waveguide rectangular core layer; n1, n2, n3, n4, and n5 are all refractive indices after equivalent calculation; n1 represents the refractive index of the ridge waveguide rectangular core layer; n2 represents the refractive index of the left cladding of the ridge waveguide rectangular core layer; n3 represents the refractive index of the right cladding of the ridge waveguide rectangular core layer; n4 represents the refractive index of the upper cladding of the ridge waveguide rectangular core layer; n5 represents the refractive index of the lower cladding of the ridge waveguide rectangular core layer; λ represents the wavelength of light; ω represents the circular frequency; μ represents the magnetic permeability of the medium; ε represents the dielectric constant of the medium; To E x mn Guided mode is equivalent to TE polarization. Using the duality principle, we can replace t and w, and x and y in the guided mode equation. For the above two guided mode equations, when t and w only hold true for m and n, respectively, the waveguide satisfies the TM / TE single-mode condition. The intersection of the two is the single-mode condition of the designed silicon nitride-lithium niobate thin film ridge waveguide. The multimode interference coupler unit should adopt a one-to-two structure with a splitting ratio close to 1:1, and its input waveguide and output waveguide should be symmetrical about the center line of the multimode waveguide to ensure 1:1 splitting; The relationship between the length and width of the multimode waveguide is: Among them L M and W M They represent the length and width of the multimode waveguide respectively, n1 represents the refractive index of the ridge waveguide core layer, n2 represents the refractive index of the ridge waveguide cladding, λ0 represents the wavelength of light, for TE mode, σ=0, for TM mode, σ=1.

2. The lithium niobate thin film phase modulator heterogeneously integrated with silicon nitride according to claim 1, characterized in that: The silicon nitride layer acts as a buffer in the middle, improving the quality of the transmitted light field. In order to transmit more energy of the optical mode field in the lithium niobate layer and improve the electro-optical modulation efficiency of lithium niobate, the sum of the thickness of the silicon nitride layer and the silicon nitride carrier strip should be slightly smaller than the thickness of the lithium niobate thin film layer, which is about 2 / 3 of the thickness of the lithium niobate thin film layer. At the same time, the thickness of the silicon nitride layer and the silicon nitride carrier strip are equivalent. At this time, compared with a pure lithium niobate thin film modulator, the waveguide transmission loss is reduced by 40%.

3. The lithium niobate thin film phase modulator heterogeneously integrated with silicon nitride according to claim 1, characterized in that: The tapered input waveguide and the tapered output waveguide have exactly the same structural dimensions, and the two tapered output waveguides are symmetrically positioned about the multimode waveguide, making the transition of the light field between the single-mode waveguide and the multimode waveguide smoother, improving the quality of the transmitted light field, and reducing leakage loss at the connection between the two. The specific dimensions of the tapered structure are designed according to the multimode waveguide. The width of the narrow end is the width of the single-mode waveguide, and the width of the wide end satisfies the processing spacing of not less than 1 μm between the two tapered output waveguides. The longer the length, the longer the transition distance and the smoother the transition. However, the device must also be miniaturized. It is preferably when the length of the tapered structure is 2-3 times the length of the multimode waveguide. The two bending units have identical structural dimensions, are composed of two identical and tangent arc segments, and meet single-mode transmission conditions; The two modulation arm units are two straight waveguides with exactly the same structural dimensions.

4. The lithium niobate thin film phase modulator heterogeneously integrated with silicon nitride according to claim 1, characterized in that: There are three modulation electrodes in total, which are located at the upper, middle and lower positions of the two modulation arm units, respectively, forming a push-pull electrode configuration to reduce the size of the device; and the modulation electrode material is gold or platinum, and the three modulation electrodes have exactly the same structural dimensions, and each has the same electrode-waveguide spacing between adjacent modulation arms.

5. The lithium niobate thin film phase modulator heterogeneously integrated with silicon nitride according to claim 1, characterized in that: The data parameters of the waveguide structure that meet the single-mode transmission conditions are selected as follows: When the thickness of the lithium niobate film layer is 300±50nm and the thickness of the silicon nitride carrier is 80±50nm, the length and width of the silicon nitride carrier can be simplified by curve fitting to satisfy the equation 60.65sin(0.6191L s +2.295)+58.07sin(0.6399L s -0.8487)-D s ≥0 Among them L s and D s Respectively represent the length and width of the silicon nitride carrier.

6. The lithium niobate thin film phase modulator heterogeneously integrated with silicon nitride according to claim 1, characterized in that: The specific working principle of the lithium niobate thin film phase modulator heterogeneously integrated with silicon nitride is as follows: The light emitted by the laser enters the multimode waveguide from the input end of the multimode interference coupler unit input waveguide. It then passes through two tapered gradient output waveguides and a bending unit to split the input light field into two, generating two identical beams of light that enter the two modulation arm units. Simultaneously, the three modulation electrodes generate an electric field. Due to the linear electro-optic effect of lithium niobate, the refractive index of lithium niobate changes under the action of the applied electric field, causing the phase of the light transmitted in the two modulation arms to change, thereby outputting two beams of light with a certain phase difference. The phase shift introduced by the applied electric field is expressed as: in, represents the phase shift, n e represents the refractive index of lithium niobate e-light, γ 33 =30.9×10 -12 m / V represents the electro-optic tensor, L z Represents the electro-optical modulation length, E op represents the transmission light field in the waveguide, E ele is the modulated external electric field generated by the electrode, and the light transmission direction is defined as the x direction; When the phase shift is π, the corresponding voltage is the half-wave voltage V π , then the modulation efficiency is expressed as: Where V is the applied voltage; Numerical calculations show that the decisive factor affecting the modulation efficiency is the distance between the adjacent surfaces of the electrode and the waveguide. The smaller the distance, the higher the modulation efficiency. However, if the distance is too small, the electrode will introduce absorption loss of the light field in the waveguide. Therefore, the modulation efficiency should be improved as much as possible while avoiding metal electrode absorption.

Images