Electro-optic modulation chip, electro-optic modulator and forming method thereof

Abstract

The application relates to the technical field of semiconductors, in particular to an electro-optical modulation chip, an electro-optical modulator and a forming method thereof; the Z-cut thin film lithium niobate chip is used to realize planar isotropy and vertical direction modulation electric field, the defects of crystal axis inversion before and after the bending of a waveguide in the prior art are overcome, a flexible and compact layout is realized, meanwhile, the characteristics that the vertical electrode structure can also be modulated when the waveguide is bent and the structure is changed are utilized, the length of a modulation area can be increased without increasing the size of a device, the modulation efficiency is improved, the defects of phase distortion, amplitude attenuation and the like in the prior art modulator are improved, and a higher rate and a wider bandwidth are achieved.

Description

Technical Field

[0001] This invention relates to the field of semiconductor packaging materials technology, and in particular to an electro-optic modulation chip, an electro-optic modulator, and a method for forming the same. Background Technology

[0002] Lithium niobate on insulator (LNOI) single-crystal thin films possess excellent physical properties, such as wide transparency bandwidth, strong electro-optic coefficient, and good thermal stability, making them ideal materials for fabricating high-performance electro-optic modulators. However, current technologies present challenges: modulating the absolute refractive index of LNOI using voltage remains difficult; to accumulate the π phase shift and provide a reasonable driving voltage, electro-optic modulators require lengths of millimeters or even centimeters; and the device size is insufficient to meet the compactness requirements of photonic integrated circuits, hindering the improvement of the integration density of photonic integrated circuits.

[0003] To address the above technical issues, existing technologies have fabricated X-cut LNOI modulators, but these have the following drawbacks: Firstly, to utilize the maximum linear electro-optic coefficient of LN material in the Z direction, horizontally arranged GSG electrodes are required as modulation electrodes, but horizontally arranged GSG electrodes are not conducive to the horizontal folding of the LNOI waveguide; secondly, the refractive indices of LN material in the X-cut and Y-cut are anisotropic in the plane, which requires X-cut and Y-cut (such as...) Figure 2 The optical waveguides made of LN material (as shown) can only operate in a straight line to avoid anisotropy and utilize the electro-optic effect. This limits the further reduction in size of X-cut or Y-cut LNOI modulators, making it difficult to meet the high integration requirements of photonic integrated circuits. Currently, there is an urgent need in the market for an electro-optic modulator that is small in size, highly integrated, and meets the requirements of photonic integrated circuits. Summary of the Invention

[0004] In view of the above analysis, the present invention aims to provide an electro-optic modulation chip, an electro-optic modulator and a molding method thereof, so as to solve at least one of the problems of large length, large volume and low integration of electro-optic modulators in the prior art.

[0005] An electro-optic modulation chip, comprising:

[0006] Substrate layer;

[0007] A back electrode is formed on top of the substrate layer;

[0008] A bottom dielectric layer is formed on top of the back electrode;

[0009] A bottom modulation layer is formed on top of the bottom dielectric layer;

[0010] Top waveguide layer: formed on top of bottom modulation layer; the top waveguide layer has multiple waveguide units connected end to end, and the input optical signal propagates in the waveguide region formed by the waveguide units and bottom modulation layer; the waveguide units and the bottom modulation layer below form a continuous and folded modulation region connected end to end.

[0011] The top dielectric layer is formed in the free region of the bottom modulation layer and on the top and sides of the waveguide unit;

[0012] The top electrode is formed on top of the top dielectric layer.

[0013] Preferably, the bottom modulation layer is a Z-cut thin-film lithium niobate, and the top waveguide layer is silicon nitride.

[0014] Preferably, the back electrode and the top electrode are arranged horizontally to provide a vertically modulated electric field.

[0015] Preferably, the top dielectric layer is photoresist.

[0016] Preferably, the photoresist is SU8 photoresist.

[0017] Preferably, the top waveguide layer has a thickness of 200nm to 300nm, a width of 1µm to 1.5µm, and a spacing of 8µm to 10µm between adjacent waveguide units.

[0018] A method for fabricating an electro-optic modulation chip, comprising the following steps:

[0019] Step 1: Grow a top waveguide layer on the bottom modulation layer of a thin-film lithium niobate wafer;

[0020] Step 2: Etch the top waveguide layer to the target shape to fabricate waveguide units connected end to end;

[0021] Step 3: In the blank area where no waveguide unit is set, etch the bottom modulation layer and bottom dielectric layer of the thin film lithium niobate wafer down to the back electrode to obtain the exposed back electrode in the etched area.

[0022] Step 4: Spin-coat photoresist onto the surface of the preformed chip prepared in step 3 and develop it. Form a top dielectric layer on top of the bottom modulation layer and the top waveguide layer, in the etched area of ​​step 3. Further etch the top dielectric layer in the etched area of ​​step 3 to obtain the exposed back electrode.

[0023] Step 5: Form the original electrode layer in the etched area of ​​Step 4 and on top of the top dielectric layer;

[0024] Step 6: Etch the original electrode layer to the target shape, form a top electrode on top of the top dielectric layer, and form an external electrode connected to the back electrode in the area where the back electrode is exposed.

[0025] Preferably, after growing the top waveguide in step 1, a chromium layer is grown as a protective layer.

[0026] An electro-optic modulator includes an input coupler, an output coupler, a first modulation arm, and a second modulation arm; each of the first and second modulation arms includes one of the aforementioned electro-optic modulation chips.

[0027] Preferably, the half-wave voltage of the electro-optic modulator is 4V to 7V, the modulation bandwidth is 19GHz to 28GHz, and the insertion loss is 8dB to 11dB.

[0028] Compared with the prior art, the present invention can achieve at least one of the following beneficial effects:

[0029] (1) The present invention uses Z-cut thin film lithium niobate chip to prepare modulator. By utilizing the planar isotropic nature of Z-cut thin film lithium niobate chip, the defect of crystal axis reversal before and after waveguide bending in the prior art is overcome, and a flexible and compact layout is achieved. At the same time, the planar isotropic nature of Z-cut thin film lithium niobate chip allows it to match vertical electrode structure and has the characteristic of being able to modulate when waveguide is bent and structure is changed. It can increase the length of modulation region and improve modulation efficiency without increasing device size.

[0030] (2) The present invention uses a Z-cut thin film lithium niobate chip to prepare a modulator. The Z-cut thin film lithium niobate crystal has an electrode direction perpendicular to the light wave propagation direction. The electrode has little influence on the light wave, which improves the defects such as phase distortion and amplitude attenuation in X-cut and Y-cut thin film lithium niobate modulators.

[0031] (3) The present invention uses Z-cut thin film lithium niobate chip to prepare modulator. The Z-cut thin film lithium niobate crystal has a high nonlinear coefficient and small dispersion, so as to obtain higher speed and wider bandwidth. The half-wave voltage of the electro-optic modulator prepared by Z-cut thin film lithium niobate chip is 4V to 7V, the modulation bandwidth is 19GHz to 28GHz, and the insertion loss is 8dB to 11dB.

[0032] (4) The present invention uses Z-cut thin film lithium niobate chip to prepare modulator. The Z-cut thin film lithium niobate crystal has high thermal stability and small thermal drift, which makes the Z-cut thin film lithium niobate modulator work in a wider temperature range and has better long-term stability.

[0033] (5) In the preparation of Z-cut thin film lithium niobate chip, the present invention, on the one hand, sets a metal chromium layer before dry etching to reduce the damage of dry etching to each layer; on the other hand, the metal chromium layer acts as a conductive layer to reduce the charging phenomenon during photolithography, and overcomes the defects such as reduced modulation bandwidth, increased insertion loss and decreased electro-optic modulation performance caused by damage.

[0034] (6) In the preparation of Z-cut thin film lithium niobate chip, the lithium niobate layer is dry etched using argon-based plasma gas containing trifluoromethane. This mixed gas has high selectivity, which helps to obtain accurate dimensions and smooth surface and better etching feature quality. It can improve modulation bandwidth, reduce insertion loss and improve the performance of electro-optic modulator.

[0035] (7) In this invention, SU8 photoresist is selected for the top dielectric layer. On the one hand, SU8 photoresist has high resolution and small lateral drift, and a large refractive index. On the other hand, SU8 photoresist can achieve higher processing accuracy through multiple exposures and development, and can be processed using ordinary ultraviolet exposure machines, which helps to further reduce costs.

[0036] In this invention, the above-described technical solutions can be combined with each other to achieve more preferred combinations. Other features and advantages of this invention will be set forth in the following description, and some advantages may become apparent from the description or be learned by practicing the invention. The objects and other advantages of this invention can be realized and obtained through the embodiments described and the accompanying drawings, which are particularly pointed out. Attached Figure Description

[0037] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts.

[0038] Figure 1 This is a schematic diagram of the refractive index of an anisotropic medium.

[0039] Figure 2 Schematic diagram of TE mode and modulation electric field direction when X-cut or Y-cut;

[0040] Figure 3 When Z-cut, the TM mode and the direction of the modulation electric field;

[0041] Figure 4 This is a schematic diagram of a Z-cut thin-film lithium niobate chip according to one embodiment of the present invention;

[0042] Figure 5 This is a schematic diagram of the waveguide unit arrangement in the top waveguide layer of a Z-cut thin-film lithium niobate chip according to one embodiment of the present invention;

[0043] Figure 6 This is a schematic diagram of the structure of a thin-film lithium niobate wafer in one embodiment of the present invention;

[0044] Figure 7 This is a schematic diagram of the structure of a pre-formed Z-cut thin-film lithium niobate chip after growing a silicon nitride layer in one embodiment of the present invention;

[0045] Figure 8 This is a schematic diagram of the structure of a pre-formed Z-cut thin-film lithium niobate chip after waveguide unit forming in one embodiment of the present invention;

[0046] Figure 9 This is a schematic diagram of the structure of a pre-formed Z-cut thin-film lithium niobate chip after forming the top dielectric layer in one embodiment of the present invention;

[0047] Figure 10 This is a schematic diagram of the structure of a pre-formed Z-cut thin-film lithium niobate chip after etching the top dielectric layer in one embodiment of the present invention;

[0048] Figure 11 This is a schematic diagram of the structure of a pre-formed Z-cut thin-film lithium niobate chip after forming the original electrode layer in one embodiment of the present invention;

[0049] Figure 12 This is a schematic diagram of the structure of a pre-formed Z-cut thin-film lithium niobate chip after forming a top electrode photoresist protective layer in one embodiment of the present invention;

[0050] Figure 13 This is a schematic diagram of the structure of the finished Z-cut thin-film lithium niobate chip after forming the top electrode and the external electrode in one embodiment of the present invention;

[0051] Figure 14 This is a schematic diagram of a Z-cut thin-film lithium niobate electro-optic modulator in one embodiment of the present invention.

[0052] Figure Labels

[0053] Input coupler 01; Output coupler 02; First modulation arm 03; Second modulation arm 04;

[0054] Substrate 1; Back electrode 2; Bottom dielectric layer 3; Bottom modulation layer 4; Top waveguide layer 5; Waveguide unit 501; Top dielectric layer 6; Top electrode 7; Original electrode layer 8; Top electrode photoresist 9; External electrode 10. Detailed Implementation

[0055] The preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which constitute a part of the present invention and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.

[0056] To better illustrate the technical solution of the present invention, the following terms are explained:

[0057] Electro-optic effect: When a voltage is applied to an electro-optic crystal, the refractive index of the electro-optic crystal will change, resulting in a change in the optical wave characteristics passing through the crystal.

[0058] Electro-optic modulation: Based on the electro-optic effect, it realizes the modulation of the phase, amplitude, intensity and polarization state of optical signals.

[0059] Half-wave voltage: When a light wave propagates in an optical crystal, the voltage required to apply when the optical path difference between the two perpendicular components Ex' and Ey' of the light wave is half a wavelength (corresponding to a phase difference of 180 degrees) is called the half-wave voltage; it is related to the electro-optic properties and geometric dimensions of the electro-optic crystal.

[0060] CPW electrode structure:

[0061] The CPW electrode structure belongs to the coplanar waveguide electrode structure, which is a type of two-conductor transmission line that can transmit transverse electromagnetic waves (TEM waves).

[0062] TE mode: Transverse electric mode, refers to a mode where the electric field direction is perpendicular to the propagation direction; it has a magnetic field component but no electric field component in the propagation direction, and is called a transverse electric wave; in a waveguide (closed cavity structure), the electromagnetic field components have E y H x H z The propagation direction is the z-axis. Its propagation mode is TE mode.

[0063] TM mode: Transverse magnetic mode, referring to a magnetic field direction perpendicular to the propagation direction. It has an electric field component but no magnetic field component in the propagation direction, and is called a transverse magnetic wave; in a waveguide (closed cavity structure), the electromagnetic field components have H... y E x E z The propagation direction is the z-direction, and its propagation mode is the TM mode.

[0064] Fundamental mode and higher-order modes: The wavelength of light in an optical fiber has first-order, second-order, and higher-order modes. Different modes of light propagate in different ways, have different transmission speeds, and different intrinsic equations of the optical fiber. According to the optical fiber equation, the transmitted light can only be transmitted in certain modes.

[0065] This invention discloses an electro-optic modulation chip, such as... Figure 4 , Figure 5 As shown, it includes:

[0066] Substrate 1;

[0067] Back electrode 2 is formed on top of the substrate layer 1;

[0068] Bottom dielectric layer 3 is formed on top of back electrode 2;

[0069] Bottom modulation layer 4 is formed on top of bottom dielectric layer 3;

[0070] Top waveguide layer 5: formed on top of bottom modulation layer 4; top waveguide layer 5 is provided with multiple waveguide units 501 connected end to end, and the input optical signal propagates in the waveguide region formed by waveguide unit 501 and bottom modulation layer 4; waveguide unit 501 and bottom modulation layer 4 below form a continuous and folded modulation region connected end to end.

[0071] The top dielectric layer 6 is formed in the free region of the bottom modulation layer 4 and on the top and sides of the waveguide unit 501;

[0072] Top electrode 7 is formed on top of top dielectric layer 6.

[0073] Regarding the modulation principle of the electro-optic modulator, it should be noted that: a beam of light of a specific wavelength is transmitted to the top waveguide layer through an external coupler. The top waveguide layer and the bottom modulation layer are an upper and lower structure. Through evanescent coupling, the light in the top waveguide layer can be coupled to the bottom modulation layer, and then modulated in the bottom modulation layer. After modulation, the light oscillates and propagates between the bottom modulation layer and the top waveguide layer along the extension direction of the waveguide unit.

[0074] Specifically, modulation in the bottom modulation layer includes: when the top electrode and the back electrode are loaded with a reverse modulation electrical signal, the propagation speed, refractive index and propagation phase of the light beam in the bottom modulation layer change with the change of the modulation electrical signal, and the propagation phase of the light beam after passing through the modulation region also changes accordingly, thereby realizing electro-optic modulation of the light beam.

[0075] It should be noted that the top waveguide layer is configured as multiple waveguide units connected end to end. Preferably, each waveguide unit is connected end to end and arranged in parallel, forming a continuous and folded modulation region connected end to end with the bottom modulation layer below, thereby achieving high-integration modulation within a limited space.

[0076] Preferably, the bottom modulation layer may be lithium niobate.

[0077] It can be understood that thin-film lithium niobate has excellent physical properties, such as wide transparency bandwidth, strong electro-optic coefficient, and good thermal stability. Moreover, the electro-optic response time of lithium niobate material is on the femtosecond scale, indicating that its intrinsic bandwidth is on the terahertz scale, which is much higher than the intrinsic bandwidth of silicon-based modulators based on plasma dispersion effects in existing technologies.

[0078] Preferably, the top waveguide layer is silicon nitride.

[0079] It is understandable that silicon nitride materials have a similar refractive index to lithium niobate materials, and at the same time have ultra-low propagation loss, low second-order nonlinearity, small thermo-optic coefficient, and extremely high power handling capability.

[0080] More preferably, to adapt to the high integration and folding requirements of the modulation region, a vertical modulation electric field is selected, and the back electrode and top electrode are set horizontally.

[0081] More preferably, to adapt to the high integration of the modulation region, folding requirements, and vertical modulation electrodes, Z-cut thin-film lithium niobate is selected. Its refractive index is the same in all directions on the xy plane, and it is isotropic. There is no problem of crystal axis reversal before and after waveguide bending, which allows for flexible and compact layout. At the same time, by utilizing the characteristic that the vertical electrode structure can also be modulated when the waveguide is bent and the structure is changed, the length of the modulation region can be increased without increasing the device size, thereby improving the modulation efficiency.

[0082] To better illustrate the contribution of this invention compared to the prior art, the following section further elaborates on the electro-optic modulation of thin-film lithium niobate and the influence of different tangential orientations of thin-film lithium niobate on the electro-optic modulation performance:

[0083] The electro-optic effect refers to the phenomenon where the dielectric constant, i.e., the refractive index, of a crystal changes under the influence of an applied electric field; lithium niobate materials utilize the linear electro-optic effect to modulate the optical field. The electro-optic effect analysis of lithium niobate materials can be implemented based on a non-simple harmonic oscillator model. Under the influence of an applied electric field E, the refractive index n(E) of the medium satisfies:

[0084] n(E) = n + aE + bE 2

[0085] Where a represents the primary electro-optic effect, also known as the Pockels linear electro-optic effect; b represents the secondary electro-optic effect, also known as the Kerr effect; E represents the applied electric field, and when there is no applied electric field E, n is the refractive index in the general sense.

[0086] It should be noted that lithium niobate crystal is an anisotropic material with no reverse symmetry and exhibits the Pockels linear electro-optic effect. In lithium niobate crystal, the linear Pockels electro-optic effect is dominant, so the Kerr effect can be neglected. When there is no external electric field applied to the lithium niobate crystal, its standard refractive index ellipsoid equation in the principal axis coordinate system satisfies:

[0087]

[0088] The above formula is based on Figure 1 The refractive index ellipsoid shown represents the refractive index of the anisotropic medium, where n x n y n z These are the lengths of the three axes of the ellipsoid, called the principal axis refractive indices. x, y, and z are the principal axis directions of the medium, meaning that the electric displacement vector D and electric field intensity E are parallel along these directions within the crystal. After determining the direction of light propagation, a plane perpendicular to the propagation direction and passing through the origin of the refractive index ellipsoid is chosen. The semi-major axis and semi-minor axis in this plane are the two refractive indices corresponding to the selected propagation direction.

[0089] Specifically, for lithium niobate crystals, the following condition must be met: n x =n y =n o n z =n e n o n is the ordinary refractive index. e Where n is a very high refractive index, and n is a very high refractive index. o >n e Therefore, lithium niobate crystal is a negative uniaxial crystal, and its refractive index ellipsoid equation satisfies:

[0090]

[0091] The maximum value of the electro-optic coefficient determines the change in refractive index after applying an external electric field, playing the largest role in the refractive index change. Furthermore, when an arbitrary electric field E is applied, the refractive index ellipsoid will be disturbed, causing an angular shift in the principal axis. Therefore, to avoid birefringence effects during design, the polarization direction of the optical carrier is usually aligned with the Z-axis of the crystal, and the applied electric field is also parallel to the Z-axis. Measurements show that when the communication light wavelength is 1550 nm, n... o and n e The values ​​are 2.214 and 2.138, respectively. When the polarization direction is along the Z-axis of the crystal and the applied electric field is parallel to the Z-axis, the maximum measured electro-optic coefficient is 31 pm / V.

[0092] In the design of lithium niobate waveguide-optic modulators, correctly selecting the orientation of the lithium niobate wafer is crucial to maximizing the electro-optic coefficient. Wafer orientation is typically represented by a "cut" in a certain direction, such as a Z-cut where the Z-axis of the crystal is perpendicular to the smooth surface of the crystal; X-cut and Y-cut are similar. The smooth surface is the plane on which the optical waveguide and electrodes are fabricated. The relationship between the crystal tangent and the applied electric field in a lithium niobate electro-optic modulator is as follows: Figure 2 and Figure 3 As shown:

[0093] Figure 2 Indicates: When X-cut or Y-cut, the TE mode and the modulation electric field (E) ‖ ) Direction diagram; Figure 3 Indicates: During Z-cut, the TM mode and the modulation electric field (E) ⊥ The crystal thickness is along the X-axis, the width along the Z-axis, and the field is assumed to remain constant in the Z-axis. The propagation direction is the Y-axis. Analysis shows that the TE mode in the crystal has only three field components E. z H x and H y And satisfy:

[0094]

[0095]

[0096] The field components of the TM mode are Hz, Ex, and Ey, and satisfy:

[0097]

[0098]

[0099] Among them, H x H y Hz and Hz represent the components of the magnetic field at x and y, respectively, and E x E y E z Let x, y, and z represent the components of the electric field at x, y, and z, respectively; β represent the propagation constant; ω represent the angular frequency of the signal; ε0 represent the dielectric constant in vacuum; μ0 represent the permeability in vacuum; and n represent the refractive index of the dielectric material.

[0100] As can be seen from the above formula, in order to utilize the maximum electro-optic coefficient of the crystal, the transmission mode of the X-cut crystal waveguide should be the TE mode, with the electric field direction parallel to the crystal surface; the transmission mode of the Z-cut crystal waveguide should be the TM mode, with the electric field direction perpendicular to the crystal surface.

[0101] In the prior art, when using a CPW electrode structure with a horizontal electric field and X-cut thin-film lithium niobate to fabricate a compact modulator, in order to solve the problem of crystal axis reversal before and after waveguide bending, a polarization process or a cross waveguide structure needs to be used, which undoubtedly increases the complexity of the modulator fabrication process.

[0102] Compared with existing technologies, the modulator fabricated using Z-cut thin-film lithium niobate chips overcomes the defect of crystal axis reversal before and after waveguide bending by utilizing the planar isotropic nature of Z-cut thin-film lithium niobate chips, thus achieving a flexible and compact layout. At the same time, the planar isotropic nature of Z-cut thin-film lithium niobate chips allows them to be matched with vertical electrode structures and has the characteristic of being able to modulate even when the waveguide is bent or the structure is changed. This can increase the length of the modulation region and improve the modulation efficiency without increasing the device size.

[0103] In addition, in X-cut and Y-cut thin-film lithium niobate modulators, the electrode direction is parallel or perpendicular to the light wave propagation direction. Due to design defects, the electrode will affect the light wave, thereby reducing the performance of the modulator and causing problems such as phase distortion and amplitude attenuation.

[0104] Compared with the prior art, the present invention uses a Z-cut thin-film lithium niobate chip to fabricate a modulator. In the Z-cut thin-film lithium niobate crystal, the electrode direction is matched with the light wave propagation direction and the isotropic plane of the chip. The influence of the electrode on the light wave is very small, which improves the defects such as phase distortion and amplitude attenuation in X-cut and Y-cut thin-film lithium niobate modulators.

[0105] In addition, Z-cut thin-film lithium niobate crystals have high nonlinear coefficients and low dispersion, resulting in higher speeds and wider bandwidths.

[0106] In addition, Z-cut thin-film lithium niobate crystals have high thermal stability and low thermal drift, which allows Z-cut thin-film lithium niobate modulators to operate over a wider temperature range and have better long-term stability.

[0107] Specifically, a silicon substrate is selected as the substrate layer.

[0108] Specifically, the external electrode of the back electrode is a gold electrode.

[0109] Understandably, gold's excellent electrical conductivity and chemical stability can improve the performance and stability of modulators.

[0110] Specifically, the bottom dielectric layer can be silicon dioxide.

[0111] Understandably, silicon dioxide possesses excellent electrical properties and chemical stability, which can improve the quality and stability of devices. In addition, silicon dioxide has a high refractive index, which can form high refractive index difference waveguides with lithium niobate, which is beneficial for achieving compact designs. At the same time, silicon dioxide has good optical transparency, which can be used as an optical waveguide material.

[0112] Specifically, the thickness of the bottom modulation layer is 400nm to 600nm.

[0113] Understandably, the bottom modulation layer and the top waveguide layer form a dielectric-loaded hybrid waveguide, with some optical modes distributed in the top waveguide layer and some in the bottom modulation layer. The dimensions of the hybrid waveguide should meet the requirements of a single-mode waveguide, ensuring that most of the power remains in the fundamental mode during transmission, reducing power transfer from the fundamental mode to higher-order modes, and thus helping to reduce optical transmission loss.

[0114] It should be noted that, as an example, when lithium niobate is selected as the bottom modulation layer, the thickness of the thin-film lithium niobate wafer used as raw material is between 400nm and 600nm. Thicker lithium niobate films require higher ion implantation capabilities during intelligent ion cutting of lithium niobate materials, resulting in more surface defects and poorer quality of the lithium niobate film after annealing and stripping. At the same time, when the thickness of the bottom modulation layer is less than 400nm, the power distribution of light in the lithium niobate material decreases, leading to a reduction in device modulation efficiency.

[0115] Specifically, the top waveguide layer has a thickness of 200nm to 300nm; a width of 1µm to 1.5µm; and a spacing of 8µm to 10µm between adjacent waveguide units.

[0116] It should be noted that the thickness of the top waveguide layer is 200nm to 300nm. When the thickness of the top waveguide layer is greater than this range, in order to maintain the single-mode condition of the hybrid waveguide, the width of the top waveguide layer needs to be smaller, making device fabrication extremely difficult. At the same time, as an example, when silicon nitride is chosen as the top waveguide layer and lithium niobate is chosen as the bottom modulation layer, the thickness of the thin-film lithium niobate wafer used as raw material is between 400nm and 600nm. When silicon nitride and lithium niobate materials form a hybrid waveguide, the thickness and width of silicon nitride need to meet the single-mode condition of the hybrid waveguide, so that most of the power of light remains in the fundamental mode during transmission, reducing the transfer of power from the fundamental mode to higher-order modes and reducing light transmission loss.

[0117] If the thickness of the top waveguide layer is greater than this range, the proportion of optical power in the lithium niobate material will be reduced, thus reducing device performance; if the thickness of the top waveguide layer is less than this range, the waveguide's confinement effect on light will be weakened, causing the optical modes distributed in the lithium niobate to evolve into planar modes, increasing light transmission loss.

[0118] Similarly, if the width of the top waveguide layer is greater than 1.5 μm, the hybrid waveguide size will not meet the single-mode condition, and the power will be transferred from the fundamental mode to higher-order modes during optical transmission, increasing the transmission loss of light. If the width of the top waveguide layer is less than 1.0 μm, the ridge waveguide's confinement effect on light will be reduced, causing the distribution mode in the hybrid waveguide to evolve into a planar mode.

[0119] Similarly, if the spacing between adjacent waveguide units is greater than 10µm, the overall device size will be larger. The larger device size cannot meet the compactness requirements of photonic integrated circuits and is not conducive to improving the integration density of the device. If the spacing between adjacent waveguide units is less than 8µm, evanescent wave coupling will occur between two adjacent waveguides, which will have an adverse effect on the device performance. Moreover, it is more difficult to manufacture waveguides with too small a spacing.

[0120] Specifically, the top dielectric layer can be photoresist.

[0121] Preferably, the top dielectric layer is SU8 photoresist.

[0122] It should be noted that, on the one hand, SU8 photoresist can be spin-coated onto the top of the bottom modulation layer to create a highly uniform surface with a relatively large thickness, and this thickness can be easily controlled by the spin-coating speed; on the other hand, SU8 photoresist has high processing precision and can be used to create more complex structures.

[0123] Compared with existing technologies, SU8 photoresist has higher resolution and smaller lateral drift; SU8 photoresist has good transparency, with a refractive index of 1.57 at 1550nm, which is greater than that of silicon dioxide (1.46); SU8 photoresist can achieve higher processing precision through multiple exposures and development, and can be processed using ordinary UV exposure machines. Existing technologies often use silicon dioxide as the top dielectric layer, but silicon dioxide has high processing costs and requires expensive equipment, such as exposure machines and deep etching equipment, which not only makes it difficult to control processing precision but also incurs high costs.

[0124] Preferably, the top waveguide layer has a rectangular cross-section.

[0125] It should be noted that the waveguide layer has a rectangular cross-section, which can reduce the leakage of light in all directions of the cross-section, better confine the light within the waveguide layer, and reduce the loss of light during light propagation.

[0126] Preferably, the thin-film lithium niobate chip has a groove for external connection of the back electrode. The groove can be prepared by etching, and an exposed back electrode is provided therein for external connection of the back electrode.

[0127] In practice, the two outer electrodes are connected to the top electrode and the back electrode in the groove, respectively, and an external voltage is applied to the top electrode and the back electrode for chip modulation.

[0128] On the other hand, this invention discloses a method for molding an electro-optic modulation chip, such as... Figures 6-13 As shown, it includes the following steps:

[0129] Step 1: Grow the top waveguide layer 5 on the bottom modulation layer 4 of the thin-film lithium niobate wafer;

[0130] Step 2: Etch the top waveguide layer 5 to the target shape to fabricate waveguide units 501 connected end to end;

[0131] Step 3: In the blank area where no waveguide unit is set, etch the bottom modulation layer 4, the bottom dielectric layer 3 to the back electrode 2 of the thin film lithium niobate wafer to obtain the bare back electrode 2 in the etched area.

[0132] Step 4: Spin-coat photoresist onto the surface of the preformed chip prepared in step 3 and develop it. Form a top dielectric layer 6 on top of the bottom modulation layer 4 and the top waveguide layer 5, in the etched area of ​​step 3. Further etch the top dielectric layer 6 in the etched area of ​​step 3 to obtain the exposed back electrode.

[0133] Step 5: Form the original electrode layer 8 on top of the etched area in step 4 and the top dielectric layer 6;

[0134] Step 6: Etch the original electrode layer 8 to the target shape, form a top electrode 7 on the top of the top dielectric layer 6, and form an external electrode 10 connected to the back electrode 2 in the exposed area of ​​the back electrode 2.

[0135] It should be noted that the target shape of the silicon nitride layer etching is multiple silicon nitride waveguide units connected end to end, with each waveguide unit arranged in parallel and connected end to end.

[0136] It should be noted that, as an example, a typical commercially available thin-film lithium niobate wafer structure, such as... Figure 6 As shown, it includes:

[0137] Substrate 1;

[0138] Back electrode 2 is formed on top of the substrate layer 1;

[0139] Bottom dielectric layer 3 is formed on top of back electrode 2;

[0140] Bottom modulation layer 4 is formed on top of bottom dielectric layer 3.

[0141] As an example, a typical lithium niobate wafer has the following characteristics: a silicon substrate, a gold electrode, a silicon dioxide bottom dielectric layer, and a thin-film lithium niobate bottom modulation layer.

[0142] Preferably, the bottom modulation layer is a Z-cut thin film lithium niobate.

[0143] Specifically, the top waveguide layer is silicon nitride.

[0144] Specifically, in step 1, after growing the top waveguide layer, a chromium layer is grown as a photolithography protective layer.

[0145] Step 1 involves growing the top waveguide layer, including:

[0146] S101: Deposit the top waveguide layer using depressurized chemical vapor deposition (DCF).

[0147] S102: A first chromium layer is formed on the top waveguide layer;

[0148] S103: Spin-coating negative photoresist, followed by exposure and development to obtain a soft mask of the target shape.

[0149] Specifically, the feed gas and its inlet volume for depressurized chemical vapor deposition in S101 satisfy the following: SiH4 (8sccm~12sccm): NH3 (4sccm~8sccm): N2 (250sccm~350sccm).

[0150] For example, the SiH4 flow rate can be: 8 sccm, 8.5 sccm, 9 sccm, 9.5 sccm, 10 sccm, 10.5 sccm, 11 sccm, 11.5 sccm and 12 sccm.

[0151] For example, the NH3 flow rate can be: 4 sccm, 4.5 sccm, 5 sccm, 5.5 sccm, 6 sccm, 6.5 sccm, 7 sccm, 7.5 sccm and 8 sccm.

[0152] For example, the N2 flow rate can be: 250 sccm, 260 sccm, 270 sccm, 280 sccm, 290 sccm, 300 sccm, 310 sccm, 320 sccm, 330 sccm, 340 sccm, and 350 sccm.

[0153] Specifically, the chamber pressure for depressurized chemical vapor deposition in S101 is 50 Pa to 150 Pa, and the temperature is 280 °C to 320 °C.

[0154] For example, the chamber pressure for depressurized chemical vapor deposition is 50 Pa, 60 Pa, 70 Pa, 80 Pa, 90 Pa, 100 Pa, 110 Pa, 120 Pa, 130 Pa, 140 Pa, or 150 Pa.

[0155] For example, the temperature for reduced pressure chemical vapor deposition is 280°C, 290°C, 300°C, 310°C, or 320°C.

[0156] The reason why the chamber pressure for depressurized chemical vapor deposition (DCV) is chosen to be between 50 Pa and 150 Pa is understandable: If the chamber pressure is less than 50 Pa, insufficient pressure will lead to poor film growth. This may result in reduced reactant concentration near the substrate surface, causing slow or incomplete chemical reactions and film deposition. Furthermore, it will cause uneven reactant distribution, leading to uneven film thickness across the entire substrate surface, negatively impacting the desired properties and functions of the deposited film. Simultaneously, low chamber pressure may allow contaminants, such as air or moisture, to penetrate into the deposition chamber. These impurities will adversely affect film quality, introducing defects or altering the film composition. If the chamber pressure is greater than 150 Pa, the higher gas velocity and turbulence can disrupt the deposition process, resulting in uneven film thickness, surface roughness, or the formation of unwanted byproducts. Additionally, high pressure conditions may inhibit effective intermolecular interactions and reduce the adhesion between the film and the substrate surface, hindering proper adhesion. Excessive pressure can also exert mechanical stress on the DCV equipment and chamber, potentially leading to equipment failure, leaks, or decreased process stability.

[0157] Understandably, the temperature range of 280℃ to 320℃ is chosen for vacuum chemical vapor deposition (CVD) because: If the temperature is below 280℃, firstly, insufficient temperature makes it difficult to initiate and maintain the CVD reaction process, leading to slow or incomplete chemical reactions between the precursor gas and the substrate surface, resulting in poor film growth rate and insufficient film coverage; secondly, insufficient temperature reduces the desorption of volatile byproducts and impurities, causing them to incorporate into the deposited film, negatively impacting film quality, purity, and desired performance. Conversely, if the temperature is above 320℃, excessively high temperatures can lead to problems with the CVD reaction... Thermal decomposition of precursor gases before reaching the substrate surface leads to the formation of unwanted byproducts or the loss of required precursor materials, affecting the composition and quality of the film. On the other hand, high temperatures can increase the surface fluidity of atoms or molecules, leading to excessive surface diffusion, resulting in uneven film growth, surface roughness, and difficulty in controlling film thickness and morphology. In addition, high temperatures can damage the substrate material or the previously deposited layer structure, causing substrate melting, interpenetration, or chemical reactions between the substrate and the deposited film, thereby impairing the adhesion and integrity of the film. At the same time, operating at excessively high temperatures can impose enormous thermal stress on the vapor deposition equipment, leading to equipment degradation, decreased reliability, and potential safety hazards.

[0158] Specifically, in S102, the thickness of the chromium layer is 60nm to 100nm. The chromium layer is placed on the silicon nitride layer to act as a conductive layer to reduce the charging phenomenon during photolithography.

[0159] Specifically, the negative photoresist in S103 is AR-N 7520.18 negative photoresist, and it is exposed using electron beam exposure technology.

[0160] Compared with the prior art, the present invention can reduce charge accumulation in silicon nitride layers and other materials during dry etching by setting a chromium layer, thereby reducing etching damage.

[0161] Specifically, in step 2, the top waveguide layer is etched to the target shape. Taking a silicon nitride layer as the top waveguide layer and a thin-film lithium niobate layer as the bottom modulation layer as an example, the process includes:

[0162] S201: Based on the first soft mask corresponding to the target shape after etching of the silicon nitride layer, the first metal chromium layer is subjected to a first dry etching to obtain a first metal chromium mask corresponding to the development shape of the soft mask, wherein the first dry etching satisfies the etching depth to the silicon nitride layer, and the etched area forms an exposed silicon nitride layer.

[0163] S202: Based on the metal chromium mask, the exposed silicon nitride layer in the etched area in S201 is subjected to a second dry etching to obtain a silicon nitride layer corresponding to the development shape of the first metal chromium mask. The second dry etching satisfies the etching depth to the lithium niobate layer, and the etched area forms an exposed lithium niobate layer.

[0164] S203: Perform a third dry etching on the pre-formed chip etched in S202 to remove the remaining photoresist; remove the first chromium layer by selective wet etching to obtain waveguide units 501 connected end to end (e.g., Figure 5 (As shown).

[0165] Specifically, in step S201, the first dry etching method is chlorine-based inductively coupled plasma dry etching, which can etch the first chromium layer according to the soft mask to form the first chromium mask without damaging the silicon nitride layer under the chromium layer.

[0166] Specifically, in step S202, the second dry etching uses fluorine-based inductively coupled plasma etching technology to transfer the target pattern of the metal chromium mask onto the silicon nitride thin film without damaging the lithium niobate layer.

[0167] Specifically, in step S202, the second dry etching gas raw materials and their inlet volume satisfy the following: SF6 (2sccm~7sccm): C4F8 (2sccc~7sccc): Ar (70sccm~110sccm).

[0168] Preferably, the second dry etching in step S202 is a depressurized etching process, with a chamber pressure of 0.5 Pa to 2.0 Pa.

[0169] Preferably, the etching rate of the second dry etching in step S202 is 2.0 nm / sec to 3.2 nm / sec.

[0170] Specifically, in step S203, the third dry etching process uses oxygen plasma technology to remove residual resist while having minimal impact on the silicon nitride layer and lithium niobate layer.

[0171] Preferably, the third dry etching in step S203 includes cleaning in a solution of sulfuric acid and hydrogen peroxide at a mass ratio of 2 to 4:1 at 60°C to 80°C for 1 to 10 minutes to further remove residual photoresist.

[0172] Specifically, step S203, after the third dry etching, also includes removing the remaining chromium layer and performing an oxygen plasma cleaning step.

[0173] Preferably, the remaining chromium layer is removed by selective wet etching.

[0174] Preferably, lithium niobate thin film etching is performed using inductively coupled reactive ion etching (ICP-RIE).

[0175] Existing technologies for etching lithium niobate thin films include wet etching, reactive ion etching, ion beam etching, plasma etching, and laser ablation. Among these technologies, wet etching is slow, requires longer processing time, and has limited control over the etched profile and sidewall roughness. Ion beam etching (IBE) requires expensive equipment and has limited wafer yield. Laser ablation technology is limited in large-area processing, the deep etching process is relatively slow, the cost is high, and the etching efficiency is heavily dependent on the laser wavelength and energy.

[0176] Compared with other etching methods in the prior art, ICP-RIE can better control the etching depth and contour uniformity. The ICP-RIE used in this invention has the following advantages:

[0177] 1. High etching rate:

[0178] ICP-RIE can achieve high etching rates and can achieve faster processing compared to wet etching or other plasma etching technologies, making it suitable for processing large-area substrates or time-sensitive manufacturing processes.

[0179] 2. Anisotropic etching:

[0180] ICP-RIE offers excellent anisotropic etching capabilities, resulting in well-defined vertical sidewalls that facilitate precise pattern transfer and the fabrication of high aspect ratio structures with sharp features.

[0181] 3. Selectivity and control:

[0182] ICP-RIE offers high selectivity, allowing the selective removal of specific materials while preserving others, making it ideal for etching complex multilayer structures or when specific layers need to be etched without damaging the underlying or adjacent materials.

[0183] Specifically, step 3 involves obtaining the exposed back electrode, including:

[0184] S301: In step 2, the lithium niobate layer of the waveguide unit chip is formed into a second chromium layer;

[0185] S302: Spin-coating negative photoresist, followed by exposure and development to obtain a soft mask of the target shape;

[0186] S303: Based on a soft mask, a second chromium metal layer is dry etched to obtain a second chromium metal mask corresponding to the development shape of the soft mask, wherein the dry etching satisfies the etching depth to the lithium niobate layer, and the etched area forms an exposed lithium niobate layer.

[0187] S304: Based on the second chromium metal mask, the exposed lithium niobate layer in the etching area of ​​S303 is dry etched to obtain a bottom dielectric layer corresponding to the development shape of the second chromium metal mask. The dry etching satisfies the etching depth to the bottom dielectric layer, and the etched area forms an exposed bottom dielectric layer.

[0188] S305: Dry etching of the exposed bottom dielectric layer in the S304 etching area down to the back electrode to obtain the exposed back electrode.

[0189] Specifically, S304 uses argon-based plasma gas to dry etch the lithium niobate layer.

[0190] Preferably, the argon-based plasma gas includes a mixture of trifluoromethane and argon, with an inlet molar ratio of 1:5 to 10; the etching rate is between 20 nm / min and 30 nm / min, and the etching selectivity for Cr is approximately 1:10.

[0191] It should be noted that the present invention uses a mixture of trifluoromethane and argon gas to etch lithium niobate, which has the following advantages:

[0192] 1. Chemicals containing trifluoromethane, such as CHF3, exhibit high selectivity when etching LiNbO3, meaning that the etching process primarily removes LiNbO3 material while having little reaction with other materials in the substrate or underlayer; this selectivity enables precise pattern transfer and helps maintain the integrity of the surrounding structure.

[0193] 2. When a mixture of CHF3 and Ar is used in an inductively coupled plasma active ion etching (ICP-RIE) system, it can perform anisotropic etching of LiNbO3, i.e. preferential removal of material in the vertical direction, thereby producing well-defined vertical sidewalls, which helps to prepare precise pattern transfer and high aspect ratio structures.

[0194] 3. In ICP-RIE etching, the combination of CHF3 and Ar mixed gases often results in a smoother etching profile and minimal sidewall roughness, which helps to obtain accurate dimensions and a smooth surface, thus contributing to improved modulator performance.

[0195] 4. In ICP-RIE etching, the etching rate of LiNbO3 can be controlled by adjusting the flow rate of CHF3 and its ratio with Ar to achieve the desired etching depth while maintaining high selectivity and quality of the etching features.

[0196] Specifically, in S305, CF4 gas is used to etch the bottom dielectric layer down to the back electrode.

[0197] Specifically, the CF4 gas intake in S305 is 10 sccm to 100 sccm.

[0198] It should be noted that the advantage of using CF4 gas is that CF4 gas exhibits high selectivity when etching SiO2: CF4 is a chemically stable gas, which means that it is inert and does not easily react with other materials under normal etching conditions. It preferentially reacts with and removes silicon dioxide, while having minimal reactivity with other materials. This selectivity allows for precise pattern transfer and preserves the integrity of the underlying or adjacent structures, thus ensuring controllable and predictable etching and reducing the possibility of unnecessary side reactions or contamination.

[0199] Specifically, the dry etching process in S305 uses oxygen plasma technology to remove residual resist while having minimal impact on the silicon nitride and lithium niobate layers.

[0200] Preferably, the dry etching in S305 includes cleaning for 1 to 10 minutes in a solution of sulfuric acid and hydrogen peroxide at a mass ratio of 2 to 4:1 at 60°C to 80°C to further remove residual photoresist.

[0201] Specifically, in step 4, photoresist is spin-coated onto the surface of the preformed chip prepared in step 3 and developed. The top dielectric layer of the etched area in step 3 is further etched to obtain the exposed back electrode, including:

[0202] S401: Spin-coating photoresist using an automatic spin coater.

[0203] S402: Expose and develop the photoresist to form a top dielectric layer on top of the lithium niobate layer and silicon nitride layer, in the etched area of ​​step 3;

[0204] S403: Etch the top dielectric layer of the etching area in step 3 down to the back electrode layer to expose the back electrode.

[0205] Specifically, the photoresist in S401 is SU8 photoresist with a thickness of 0.5μm to 1.5μm and a spin coating rate of 1200rpm / s to 1800rpm / s.

[0206] In S403, the etching of the top dielectric layer of the etching area in step 3 is performed using chlorine-based inductively coupled plasma dry etching.

[0207] Specifically, step 5 involves selecting electron beam evaporation deposition to form the original electrode layer.

[0208] Specifically, step 6, etching the original electrode layer to the target shape, includes:

[0209] S601: Spin-coat the top electrode photoresist 9 onto the original electrode layer, and obtain a soft mask of the target shape after exposure and development; the top electrode photoresist 9 is a negative photoresist;

[0210] S602: Based on a soft mask, the original electrode layer 8 is dry-etched to form a top electrode 7 on top of the top dielectric layer 6, and an external electrode 10 connected to the back electrode 2 is formed in the exposed area of ​​the back electrode 2; wherein, the dry etching etches the original electrode layer to the depth of the back electrode 2, and the external electrode 10 is connected to the substrate layer 1 and the bottom dielectric layer 3 through the back electrode 2.

[0211] Specifically, the negative photoresist in S601 is AR-N 7520.18 negative photoresist, and it is exposed using electron beam exposure technology.

[0212] This invention discloses an electro-optic modulator, such as Figure 14 As shown, it includes an input coupler 01, an output coupler 02, a first modulation arm 03, and a second modulation arm 04; each of the first and second modulation arms contains one of the aforementioned thin-film lithium niobate chips.

[0213] The input coupler 01 is coupled to one end of the top waveguide layer of the thin-film lithium niobate chip in the first modulation arm 03 and the second modulation arm 04.

[0214] Optionally, the output coupler 02 is coupled to the other end of the top waveguide layer of the thin-film lithium niobate chip in the first modulation arm 03 and the second modulation arm 04.

[0215] The optical signal to be modulated is split into two optical signals at the input coupler 01 and enters the first modulation arm 03 and the second modulation arm 04 for modulation respectively.

[0216] In one feasible implementation, the wavelength and / or phase of the modulated optical signal change compared to before modulation.

[0217] In one feasible implementation, the first modulation arm 03 and the second modulation arm 04 output optical signals with different wavelengths and / or phases.

[0218] In one feasible implementation, the modulated optical signals from the first modulation arm 03 and the second modulation arm 04 are combined at the output coupler 02 to obtain a mixed optical signal composed of optical signals with different wavelengths and / or phases.

[0219] To further illustrate the advancements of this invention, the following embodiments and comparative examples are provided:

[0220] Example 1

[0221] This embodiment discloses a typical commercially available thin-film lithium niobate wafer structure, such as Figure 6 As shown, it includes:

[0222] Substrate 1;

[0223] Back electrode 2 is formed on top of the substrate layer 1;

[0224] Bottom dielectric layer 3 is formed on top of back electrode 2;

[0225] Bottom modulation layer 4 is formed on top of bottom dielectric layer 3.

[0226] As an example, a typical lithium niobate wafer has the following characteristics: a silicon substrate, a gold electrode as the back electrode, a silicon dioxide as the bottom dielectric layer, and a Z-cut thin-film lithium niobate as the bottom modulation layer.

[0227] This embodiment discloses a method for molding an electro-optic modulation chip, such as... Figures 6-13 As shown, it includes the following steps:

[0228] Step 1:

[0229] S101: When the chamber pressure is 50Pa, the temperature is 280℃, and the raw material gas and its inlet flow rate are SiH4 (8sccm):NH3 (4sccm):N2 (250sccm), a silicon nitride layer is deposited on the lithium niobate layer of the lithium niobate wafer using reduced pressure chemical vapor deposition technology.

[0230] S102: A first chromium layer is formed on the silicon nitride layer, with a thickness of 60 nm.

[0231] S103: Spin-coating negative photoresist, followed by exposure and development to obtain a soft mask of the target shape.

[0232] Step 2:

[0233] S201: Based on the first soft mask, the first metal chromium layer is dry etched for the first time to obtain a first metal chromium mask corresponding to the development shape of the soft mask, wherein the first dry etching satisfies the etching depth to the silicon nitride layer, and the etched area forms an exposed silicon nitride layer.

[0234] S202: Based on the first chromium mask, the silicon nitride layer is dry-etched when the gas raw material and its inlet volume meet the following conditions: SF6 (2 sccm): C4F8 (2 sccc): Ar (70 sccm); the chamber pressure is 0.5 Pa and the etching rate is 2.0 nm / sec.

[0235] S203: The preformed chip etched in S202 is subjected to a third dry etching process to remove the remaining photoresist; the first chromium layer is removed by selective wet etching to obtain waveguide units connected end to end.

[0236] Step 3: In the blank area where no waveguide unit is set, etch the lithium niobate layer, bottom dielectric layer to the back electrode of the thin film lithium niobate wafer to obtain the exposed back electrode in the etched area;

[0237] S301: Formation of a second metallic chromium layer on lithium niobate;

[0238] S302: Spin-coat AR-N 7520.18 negative photoresist, and obtain a soft mask of the target shape after exposure and development;

[0239] S303: Based on a soft mask, a second chromium metal layer is dry-etched for the fourth time to obtain a second chromium metal mask corresponding to the development shape of the soft mask. The fourth dry etching satisfies the etching depth to the lithium niobate layer, and the etched area forms an exposed lithium niobate layer.

[0240] S304: Based on the second chromium mask, the lithium niobate layer is dry etched using argon-based plasma gas. The argon-based plasma gas includes a mixture of trifluoromethane and argon, with an inlet molar ratio of 1:5. The etching rate is between 20 nm / min, and the etch selectivity for Cr is approximately 1:8.

[0241] S305: The exposed bottom dielectric layer of the S304 etching area is etched to the back electrode using oxygen plasma technology with CF4 gas to obtain the exposed back electrode; the CF4 gas inlet rate is 10 sccm; and the sample is cleaned for 1 to 10 minutes in a solution of sulfuric acid and hydrogen peroxide at a mass ratio of 2 to 4:1 at 60℃ to 80℃ to further remove residual photoresist.

[0242] Step 4: Spin-coat photoresist onto the surface of the preformed chip prepared in Step 3 and develop it to form a top dielectric layer on top of the top waveguide layer and the bottom modulation layer, in the etched area of ​​Step 3. Further etch the top dielectric layer in the etched area of ​​Step 3 to obtain the exposed back electrode.

[0243] S401: Spin-coating photoresist using an automatic spin coater, using SU8 photoresist with a thickness of 0.5μm to 1.5μm and a spin coating rate of 1200rpm / s to 1800rpm / s;

[0244] S402: Expose and develop the photoresist to expose the back electrode.

[0245] Step 5: An aluminum primary electrode layer is formed in the etched area of ​​Step 4 and on top of the top dielectric layer.

[0246] Step 6: Etch the original electrode layer to the target shape, form a top electrode on top of the top dielectric layer, and form an external electrode connected to the back electrode in the area where the back electrode is exposed.

[0247] S601: Spin-coat AR-N 7520.18 negative photoresist onto the original electrode layer, expose it using electron beam lithography, and obtain a soft mask of the target shape after exposure and development;

[0248] S602: Based on a soft mask, the aluminum original electrode layer is etched for the seventh time using dry etching, with the etching depth reaching the original electrode layer to form a top electrode on top of the top dielectric layer, and an external electrode connected to the back electrode is formed in the exposed area of ​​the back electrode; wherein, the seventh dry etching etches the original electrode layer to the back electrode layer, and the external electrode is connected to the substrate and the bottom dielectric layer through the back electrode.

[0249] This embodiment discloses an electro-optic modulation chip, which is prepared by the above method and includes: a thin-film lithium niobate wafer with a lithium niobate layer on top, a silicon nitride top waveguide layer formed on the lithium niobate layer, an SU8 photoresist top dielectric layer, and a top electrode on top of the top dielectric layer.

[0250] The top waveguide layer of silicon nitride has waveguide units connected end to end, and the input optical signal propagates in the waveguide region formed by the waveguide units and the bottom modulation layer of lithium niobate.

[0251] The SU8 top dielectric layer is formed in the free region of the bottom modulation layer and on the top and sides of the waveguide unit;

[0252] The top electrode is formed on top of the top dielectric layer.

[0253] The bottom modulation layer is 400 nm thick; the top waveguide layer is 200 nm thick; the top waveguide layer is 1 μm wide; and the spacing between adjacent waveguide units is 8 μm.

[0254] This embodiment discloses an electro-optic modulator, such as Figure 14 As shown, it includes an input coupler 01, an output coupler 02, a first modulation arm 03, and a second modulation arm 04; each of the first and second modulation arms contains an electro-optic modulation chip prepared in this embodiment.

[0255] The input coupler 01 is coupled to one end of the top waveguide layer of the thin-film lithium niobate chip in the first modulation arm 03 and the second modulation arm 04, and the output coupler 02 is coupled to the other end of the top waveguide layer of the thin-film lithium niobate chip in the first modulation arm 03 and the second modulation arm 04.

[0256] Example 2

[0257] This embodiment discloses a typical commercially available thin-film lithium niobate wafer structure, such as Figure 6 As shown, it includes:

[0258] Substrate 1;

[0259] Back electrode 2 is formed on top of the substrate layer 1;

[0260] Bottom dielectric layer 3 is formed on top of back electrode 2;

[0261] Bottom modulation layer 4 is formed on top of bottom dielectric layer 3.

[0262] As an example, a typical lithium niobate wafer has the following characteristics: a silicon substrate, a gold electrode as the back electrode, a silicon dioxide as the bottom dielectric layer, and a Z-cut thin-film lithium niobate as the bottom modulation layer.

[0263] This embodiment discloses a method for molding an electro-optic modulation chip, such as... Figures 6-13 As shown, it includes the following steps:

[0264] Step 1:

[0265] S101: When the chamber pressure is 150Pa, the temperature is 320℃, and the raw material gas and its inlet flow rate are SiH4 (12sccm):NH3 (8sccm):N2 (350sccm), a silicon nitride layer is deposited on the lithium niobate layer of the lithium niobate wafer using reduced pressure chemical vapor deposition technology.

[0266] S102: A first chromium layer is formed on the silicon nitride layer, and the thickness of the chromium layer is 100 nm.

[0267] S103: Spin-coating negative photoresist, followed by exposure and development to obtain a soft mask of the target shape.

[0268] Step 2:

[0269] S201: Based on a soft mask, a first metal chromium layer is dry-etched for the first time to obtain a first metal chromium mask corresponding to the development shape of the soft mask, wherein the first dry etching satisfies the etching depth to the silicon nitride layer, and the etched area forms an exposed silicon nitride layer.

[0270] S202: Based on the first chromium mask, the silicon nitride layer is dry-etched when the gas raw material and its inlet volume meet the following conditions: SF6 (7 sccm): C4F8 (7 sccc): Ar (110 sccm); the chamber pressure is 2.0 Pa and the etching rate is 3.2 nm / sec.

[0271] S203: The preformed chip etched in S202 is subjected to a third dry etching process to remove the remaining photoresist; the first chromium layer is removed by selective wet etching to obtain waveguide units connected end to end.

[0272] Step 3: In the blank area where no waveguide unit is set, etch the lithium niobate layer, bottom dielectric layer to the back electrode of the thin film lithium niobate wafer to obtain the exposed back electrode in the etched area;

[0273] S301: Formation of a second metallic chromium layer on lithium niobate;

[0274] S302: Spin-coat AR-N 7520.18 negative photoresist, and obtain a soft mask of the target shape after exposure and development;

[0275] S303: Based on a soft mask, a second chromium metal layer is dry-etched for the fourth time to obtain a second chromium metal mask corresponding to the development shape of the soft mask. The fourth dry etching satisfies the etching depth to the lithium niobate layer, and the etched area forms an exposed lithium niobate layer.

[0276] S304: Based on the second chromium mask, the lithium niobate layer is dry etched using argon-based plasma gas. The argon-based plasma gas includes a mixture of trifluoromethane and argon, with an inlet molar ratio of 1:8. The etching rate is between 30 nm / min, and the etch selectivity for Cr is approximately 1:10.

[0277] S305: The exposed bottom dielectric layer of the S304 etching area is etched to the back electrode using oxygen plasma technology with CF4 gas, and the CF4 gas inlet is 50 sccm. The residual photoresist is further removed by cleaning in a solution of sulfuric acid and hydrogen peroxide at 80°C with a mass ratio of 4:1.

[0278] Step 4: Spin-coat photoresist onto the surface of the preformed chip prepared in Step 3 and develop it to form a top dielectric layer on top of the top waveguide layer and the bottom modulation layer, in the etched area of ​​Step 3. Further etch the top dielectric layer in the etched area of ​​Step 3 to obtain the exposed back electrode.

[0279] S401: Photoresist, SU8 photoresist, 1.5μm thickness, spin coating speed of 1800rpm / s is used for spin coating using an automatic spin coater;

[0280] S402: Expose and develop the photoresist to expose the back electrode.

[0281] Step 5: Form an aluminum primary electrode layer in the etched area of ​​Step 4 and on top of the top dielectric layer;

[0282] Step 6: Etch the original electrode layer to the target shape, form a top electrode on top of the top dielectric layer, and form an external electrode connected to the back electrode in the area where the back electrode is exposed.

[0283] S601: Spin-coat AR-N 7520.18 negative photoresist onto the original electrode layer, expose it using electron beam lithography, and obtain a soft mask of the target shape after exposure and development;

[0284] S602: Based on a soft mask, the aluminum original electrode layer is dry etched to a depth that forms a top electrode on top of the top dielectric layer, and an external electrode connected to the back electrode is formed in the exposed area of ​​the back electrode; wherein, the dry etching etches the original electrode layer to a depth that reaches the back electrode layer, and the external electrode is connected to the substrate and the bottom dielectric layer through the back electrode.

[0285] This embodiment discloses an electro-optic modulation chip, which is prepared by the above method and includes: a thin-film lithium niobate wafer with a lithium niobate layer on top, a silicon nitride top waveguide layer formed on the lithium niobate layer, an SU8 photoresist top dielectric layer, and a top electrode on top of the top dielectric layer.

[0286] The top waveguide layer of silicon nitride has waveguide units connected end to end, and the input optical signal propagates in the waveguide region formed by the waveguide units and the bottom modulation layer of lithium niobate.

[0287] The SU8 top dielectric layer is formed in the free region of the bottom modulation layer and on the top and sides of the waveguide unit;

[0288] The top electrode is formed on top of the top dielectric layer.

[0289] The bottom modulation layer is 600 nm thick; the top waveguide layer is 300 nm thick; the top waveguide layer is 1.5 μm wide; and the spacing between adjacent waveguide units is 10 μm.

[0290] This embodiment discloses an electro-optic modulator, such as Figure 14 As shown, it includes an input coupler 01, an output coupler 02, a first modulation arm 03, and a second modulation arm 04; each of the first and second modulation arms contains an electro-optic modulation chip prepared in this embodiment.

[0291] The input coupler 01 is coupled to one end of the top waveguide layer of the thin-film lithium niobate chip in the first modulation arm 03 and the second modulation arm 04, and the output coupler 02 is coupled to the other end of the top waveguide layer of the thin-film lithium niobate chip in the first modulation arm 03 and the second modulation arm 04.

[0292] Example 3

[0293] This embodiment discloses a typical commercially available thin-film lithium niobate wafer structure, such as Figure 6 As shown, it includes:

[0294] Substrate 1;

[0295] Back electrode 2 is formed on top of the substrate layer 1;

[0296] Bottom dielectric layer 3 is formed on top of back electrode 2;

[0297] Bottom modulation layer 4 is formed on top of bottom dielectric layer 3.

[0298] As an example, a typical lithium niobate wafer has the following characteristics: a silicon substrate, a gold electrode as the back electrode, a silicon dioxide as the bottom dielectric layer, and a Z-cut thin-film lithium niobate as the bottom modulation layer.

[0299] This embodiment discloses a method for molding an electro-optic modulation chip, such as... Figures 6-13 As shown, it includes the following steps:

[0300] Step 1:

[0301] S101: When the chamber pressure is 100Pa, the temperature is 300℃, and the raw material gas and its inlet flow rate are SiH4 (10sccm):NH3 (6sccm):N2 (300sccm), a silicon nitride layer is deposited on the lithium niobate layer of the lithium niobate wafer using reduced pressure chemical vapor deposition technology.

[0302] S102: A first chromium layer is formed on the silicon nitride layer, with a thickness of 80 nm.

[0303] S103: Spin-coating negative photoresist, followed by exposure and development to obtain a soft mask of the target shape.

[0304] Step 2:

[0305] S201: Based on a soft mask, a first metal chromium layer is dry-etched for the first time to obtain a first metal chromium mask corresponding to the development shape of the soft mask, wherein the first dry etching satisfies the etching depth to the silicon nitride layer, and the etched area forms an exposed silicon nitride layer.

[0306] S202: Based on the first chromium mask, the silicon nitride layer is dry-etched when the gas raw material and its inlet volume meet the following conditions: SF6 (5 sccm): C4F8 (5 sccc): Ar (90 sccm); the chamber pressure is 1.0 Pa and the etching rate is 2.6 nm / sec.

[0307] S203: The preformed chip etched in S202 is subjected to a third dry etching process to remove the remaining photoresist; the first chromium layer is removed by selective wet etching to obtain waveguide units connected end to end.

[0308] Step 3: In the blank area where no waveguide unit is set, etch the lithium niobate layer, bottom dielectric layer to the back electrode of the thin film lithium niobate wafer to obtain the exposed back electrode in the etched area;

[0309] S301: Formation of a second metallic chromium layer on lithium niobate;

[0310] S302: Spin-coat AR-N 7520.18 negative photoresist, and obtain a soft mask of the target shape after exposure and development;

[0311] S303: Based on a soft mask, a second chromium metal layer is dry-etched for the fourth time to obtain a second chromium metal mask corresponding to the development shape of the soft mask. The fourth dry etching satisfies the etching depth to the lithium niobate layer, and the etched area forms an exposed lithium niobate layer.

[0312] S304: Based on the second chromium metal mask, the lithium niobate layer is dry etched using argon-based plasma gas. The argon-based plasma gas includes a mixture of trifluoromethane and argon, with an inlet molar ratio of 1:10. The etching rate is between 20 nm / min and 30 nm / min, and the etch selectivity for Cr is approximately 1:12.

[0313] S305: The exposed bottom dielectric layer of the S304 etching area is etched to the back electrode using oxygen plasma technology with CF4 gas, and the CF4 gas inlet is 100 sccm. The remaining photoresist is further removed by cleaning in a solution of sulfuric acid and hydrogen peroxide at 70°C with a mass ratio of 3:1.

[0314] Step 4: Spin-coat photoresist onto the surface of the preformed chip prepared in Step 3 and develop it to form a top dielectric layer on top of the top waveguide layer and the bottom modulation layer, in the etched area of ​​Step 3. Further etch the top dielectric layer in the etched area of ​​Step 3 to obtain the exposed back electrode.

[0315] S401: Photoresist, SU8 photoresist, 1.0μm thickness, spin coating speed 1500rpm / s is used for spin coating using an automatic spin coater;

[0316] S402: Expose and develop the photoresist to expose the back electrode.

[0317] Step 5: Form an aluminum primary electrode layer in the etched area of ​​Step 4 and on top of the top dielectric layer;

[0318] Step 6: Etch the original electrode layer to the target shape, form a top electrode on top of the top dielectric layer, and form an external electrode connected to the back electrode in the area where the back electrode is exposed.

[0319] S601: Spin-coat AR-N 7520.18 negative photoresist onto the original electrode layer, expose it using electron beam lithography, and obtain a soft mask of the target shape after exposure and development;

[0320] S602: Based on a soft mask, the aluminum original electrode layer is dry etched to a depth that forms a top electrode on top of the top dielectric layer, and an external electrode connected to the back electrode is formed in the exposed area of ​​the back electrode; wherein, the dry etching etches the original electrode layer to a depth that reaches the back electrode layer, and the external electrode is connected to the substrate and the bottom dielectric layer through the back electrode.

[0321] This embodiment discloses a Z-cut thin-film lithium niobate chip, which is prepared by the above method and includes: a thin-film lithium niobate wafer with a lithium niobate layer on top, a silicon nitride top waveguide layer formed on the lithium niobate layer, an SU8 photoresist top dielectric layer, and a top electrode on top of the top dielectric layer.

[0322] The top waveguide layer of silicon nitride has waveguide units connected end to end, and the input optical signal propagates in the waveguide region formed by the waveguide units and the bottom modulation layer of lithium niobate.

[0323] The SU8 top dielectric layer is formed in the free region of the bottom modulation layer and on the top and sides of the waveguide unit;

[0324] The top electrode is formed on top of the top dielectric layer.

[0325] The bottom modulation layer is 500 nm thick; the top waveguide layer is 250 nm thick; the top waveguide layer is 1.2 μm wide; and the spacing between adjacent waveguide units is 9 μm.

[0326] This embodiment discloses an electro-optic modulator, such as Figure 14 As shown, it includes an input coupler 01, an output coupler 02, a first modulation arm 03, and a second modulation arm 04; each of the first and second modulation arms contains an electro-optic modulation chip prepared in this embodiment.

[0327] The input coupler 01 is coupled to one end of the top waveguide layer of the thin-film lithium niobate chip in the first modulation arm 03 and the second modulation arm 04, and the output coupler 02 is coupled to the other end of the top waveguide layer of the thin-film lithium niobate chip in the first modulation arm 03 and the second modulation arm 04.

[0328] This embodiment discloses an electro-optic modulator, such as Figure 14 As shown, it includes an input coupler 01, an output coupler 02, a first modulation arm 03, and a second modulation arm 04; each of the first and second modulation arms contains an electro-optic modulation chip prepared in this embodiment.

[0329] The input coupler 01 is coupled to one end of the top waveguide layer of the thin-film lithium niobate chip in the first modulation arm 03 and the second modulation arm 04, and the output coupler 02 is coupled to the other end of the top waveguide layer of the thin-film lithium niobate chip in the first modulation arm 03 and the second modulation arm 04.

[0330] Example 4

[0331] The only difference from Example 3 is that the top waveguide layer of the thin-film lithium niobate chip in the first modulation arm 03 and the second modulation arm 04 outputs their respective modulated optical signals to obtain optical signals with different frequencies and phases than the input optical signals.

[0332] Comparative Example 1

[0333] The only difference from Example 3 is that the thickness of the top waveguide layer in the electro-optic modulation chip is set to 150nm.

[0334] Comparative Example 2

[0335] The only difference from Example 3 is that the spacing between adjacent waveguide units in the electro-optic modulation chip is set to 6µm.

[0336] Comparative Example 3

[0337] The only difference from Example 3 is that the width of the top waveguide layer in the electro-optic modulation chip is set to 0.5 μm.

[0338] Comparative Example 4

[0339] The only difference from Example 3 is that the width of the top waveguide layer in the electro-optic modulation chip is set to 3.0 μm.

[0340] Comparative Example 5

[0341] The only difference from Example 3 is that, in the preparation of the electro-optic modulation chip, the first chromium layer was not set in step S102 and the second chromium layer was not set in step S301.

[0342] Comparative Example 6

[0343] The only difference from Example 3 is that the argon-based plasma gas in step S304 does not contain trifluoromethane when preparing the electro-optic modulation chip.

[0344] Experimental Example

[0345] The electro-optic modulators designed in the examples and comparative simulations were subjected to photoelectric modulation performance simulation tests. The test results are as follows:

[0346]

[0347]

[0348] Experimental conclusion:

[0349] It can be seen from the above table:

[0350] As can be seen from Examples 1-3, the half-wave voltage of the electro-optic modulator prepared with Z-cut thin-film lithium niobate chip is 4V to 7V, the modulation bandwidth is 19GHz to 28GHz, and the insertion loss is 8dB to 11dB.

[0351] Comparing Example 3 and Comparative Example 1, it can be seen that when the thickness of the top waveguide layer is less than 200 nm, the modulation bandwidth decreases, the insertion loss increases, and the performance of the electro-optic modulator deteriorates significantly.

[0352] Comparing Example 3 and Comparative Example 2, it can be seen that if the spacing between adjacent waveguide units is set too small, the modulation bandwidth decreases, the insertion loss increases, and the performance of the electro-optic modulator deteriorates significantly.

[0353] Comparing Example 3 with Comparative Examples 3 and 4, it can be seen that if the width of the top waveguide layer is set too small, the modulation bandwidth decreases, the insertion loss increases, and the performance of the electro-optic modulator decreases significantly; if the width of the top waveguide layer is set too large, the modulation bandwidth remains basically unchanged, the insertion loss decreases slightly, and the modulator volume increases.

[0354] Comparing Example 3 and Comparative Example 5, it can be seen that when preparing the Z-cut thin-film lithium niobate chip, the absence of a chromium layer during the dry etching process increases the damage, resulting in a decrease in modulation bandwidth, an increase in insertion loss, and a significant decline in the performance of the electro-optic modulator.

[0355] Comparing Example 3 and Comparative Example 6, it can be seen that when preparing the Z-cut thin-film lithium niobate chip, the argon-based plasma gas in step S304 does not contain trifluoromethane, resulting in a decrease in modulation bandwidth, an increase in insertion loss, and a significant decrease in the performance of the electro-optic modulator.

[0356] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. An electro-optic modulation chip, characterized in that, include: Substrate layer; A back electrode is formed on top of the substrate layer; A bottom dielectric layer is formed on top of the back electrode; A bottom modulation layer is formed on top of the bottom dielectric layer, wherein the bottom modulation layer is a Z-cut thin film lithium niobate; Top waveguide layer: formed on top of bottom modulation layer; the top waveguide layer is silicon nitride, and the top waveguide layer has multiple waveguide units connected end to end. The input optical signal propagates in the waveguide region formed by the waveguide units and the bottom modulation layer; the waveguide units and the bottom modulation layer below form a continuous and folded modulation region connected end to end. The folded modulation region achieves multiple modulations of the optical signal through multiple reflections and refractions of the waveguide units. A top dielectric layer is formed in the free region of the bottom modulation layer and on the top and sides of the waveguide unit, and the top dielectric layer is SU8 photoresist; The top electrode is formed on top of the top dielectric layer; The back electrode and the top electrode are horizontally arranged to provide a vertically modulated electric field. The electro-optic modulation chip also includes an external electrode formed in the exposed area of ​​the back electrode. The external electrode is connected to the back electrode and is used to supply power to the electro-optic modulation chip.

2. The electro-optic modulation chip according to claim 1, characterized in that, The top waveguide layer has a thickness of 200nm to 300nm; a width of 1µm to 1.5µm; and a spacing of 8µm to 10µm between adjacent waveguide units.

3. A method for forming an electro-optic modulation chip, characterized in that, The method for preparing the electro-optic modulation chip according to claim 1 or 2 includes the following steps: Step 1: Grow a top waveguide layer on the bottom modulation layer of a thin-film lithium niobate wafer; Step 2: Etch the top waveguide layer to the target shape to fabricate waveguide units connected end to end; Step 3: In the blank area where no waveguide unit is set, etch the bottom modulation layer and bottom dielectric layer of the thin film lithium niobate wafer down to the back electrode to obtain the exposed back electrode in the etched area. Step 4: Spin-coat photoresist onto the surface of the preformed chip prepared in step 3 and develop it. Form a top dielectric layer on top of the bottom modulation layer and the top waveguide layer, in the etched area of ​​step 3. Further etch the top dielectric layer in the etched area of ​​step 3 to obtain the exposed back electrode. Step 5: Form the original electrode layer in the etched area of ​​Step 4 and on top of the top dielectric layer; Step 6: Etch the original electrode layer to the target shape, form a top electrode on top of the top dielectric layer, and form an external electrode connected to the back electrode in the area where the back electrode is exposed.

4. The method for forming an electro-optic modulation chip according to claim 3, characterized in that, In step 1, after growing the top waveguide, a chromium layer is grown as a protective layer.

5. An electro-optic modulator, characterized in that, It includes an input coupler, an output coupler, a first modulation arm, and a second modulation arm; each of the first and second modulation arms includes an electro-optic modulation chip as described in claim 1 or 2.

6. An electro-optic modulator according to claim 5, characterized in that, The electro-optic modulator has a half-wave voltage of 4V to 7V, a modulation bandwidth of 19GHz to 28GHz, and an insertion loss of 8dB to 11dB.

Images