Electroabsorption modulated laser and method of manufacturing the same

Abstract

The application provides an electroabsorption modulated laser and a preparation method thereof, and relates to the technical field of semiconductor optoelectronic integrated devices. The preparation method comprises the following steps: S1, preparing a laser active layer and a modulator active layer on a substrate; S2, preparing a grating on the surface of the laser active layer; S3, sequentially preparing a cladding layer, an electrical contact layer, a ridge waveguide structure, an electrical isolation groove and a silicon dioxide film on the modulator active layer and the grating; S4, preparing a P electrode on the silicon dioxide film; S5, growing a composite film of silicon nitride and hafnium oxide on the surface of the P electrode and the exposed area of the electrical isolation groove; S6, removing the composite film of the P electrode wire bonding area to expose the P electrode and form a wire bonding area; and S7, preparing an N electrode on the back of the substrate to obtain the electroabsorption modulated laser. The preparation method deposits the composite film of silicon nitride and hafnium oxide through an atomic layer deposition process, which meets the non-hermetic packaging requirements of the light emitting chip.

Description

Technical Field

[0001] This application relates to the field of semiconductor optoelectronic integrated device technology, and in particular to electroabsorption modulated lasers and their fabrication methods. Background Technology

[0002] With the rapid development of artificial intelligence, the demand for optical transmitter chips in data centers and short-range data transmission has exploded. On the one hand, optical transmitter chips are required to provide higher bandwidth, with single-channel data transmission exceeding 100G; on the other hand, higher requirements are placed on the packaging of optical transmitter chips, shifting from the original high-cost hermetically sealed packaging to non-hermetically sealed packaging. Non-hermetically sealed packaging requires optical transmitter chips to have higher requirements for high temperature and humidity resistance. Hermetically sealed packaging can meet the requirements with a single layer of silicon dioxide insulating film, but silicon dioxide is prone to moisture absorption and cannot meet the requirements of non-hermetically sealed packaging. Summary of the Invention

[0003] In view of this, this application provides an electroabsorption modulated laser and a method for fabricating the same, in order to at least partially solve the aforementioned technical problems.

[0004] One aspect of this application provides a method for fabricating an electroabsorption modulated laser, comprising: step S1, fabricating a laser active layer and a modulator active layer on a substrate; step S2, fabricating a grating on the surface of the laser active layer; step S3, sequentially fabricating a cladding layer, an electrical contact layer, a ridge waveguide structure, an electrical isolation trench, and a silicon dioxide thin film on the modulator active layer and the grating; step S4, fabricating a P-electrode on the silicon dioxide thin film; step S5, growing a silicon nitride thin film on the surface of the P-electrode and in the exposed area of ​​the electrical isolation trench, and growing a hafnium oxide thin film on the silicon nitride thin film; step S6, removing the silicon nitride thin film and hafnium oxide thin film in the wire bonding area of ​​the P-electrode to expose the P-electrode and form a wire bonding area; and step S7, fabricating an N-electrode on the side of the substrate away from the laser active layer and the modulator active layer to fabricate an electroabsorption modulated laser.

[0005] According to an embodiment of this application, in step S5, based on the thickness of the silicon dioxide film, the thickness ratio of the silicon nitride film and the hafnium oxide film is adjusted so that the absolute value of the residual stress is lower than a preset threshold.

[0006] According to an embodiment of this application, step S5 includes: performing plasma-enhanced atomic layer deposition at a temperature of 300°C, wherein at least two precursors are alternately introduced in a deposition cycle, and plasma is excited when at least one precursor is introduced to grow silicon nitride thin films and hafnium oxide thin films.

[0007] According to an embodiment of this application, step S6 includes: step S61, forming a patterned photoresist mask on the surface of the hafnium oxide film, wherein the patterned photoresist mask has an opening at the position corresponding to the P electrode bonding area to expose the hafnium oxide film in that area; step S62, using the patterned photoresist mask as a barrier layer, using reactive ion etching process to sequentially remove the exposed hafnium oxide film and silicon nitride film to expose the underlying P electrode and form a bonding area; step S63, removing the remaining patterned photoresist mask.

[0008] According to an embodiment of this application, step S1 includes: step S11, fabricating a laser active layer on a substrate, the laser active layer including a laser lower waveguide, a laser multiple quantum well, and a laser upper waveguide stacked sequentially; step S12, covering a portion of the surface of the laser upper waveguide with a mask pattern, and removing the portion of the laser active layer not covered by the mask pattern to expose the substrate; step S13, fabricating a modulator active layer on the substrate, the modulator active layer including a modulator lower waveguide, a modulator multiple quantum well, and a modulator upper waveguide stacked sequentially.

[0009] According to an embodiment of this application, step S12 includes: growing a silicon dioxide thin film layer on the waveguide surface of the laser, etching a mask pattern using photolithography and reactive ion etching processes, and removing the portion of the active layer of the laser that is not covered by the mask pattern using inductively coupled plasma etching and wet etching processes.

[0010] According to an embodiment of this application, step S3 includes: step S31, fabricating a cladding layer on the modulator active layer and the grating, and fabricating an electrical contact layer on the cladding layer; step S32, fabricating a ridge waveguide structure on the cladding layer and the electrical contact layer, and etching an electrical isolation trench exposing the cladding layer in the middle of the electrical contact layer to achieve electrical isolation between the laser and the modulator; step S33, growing a silicon dioxide thin film on the electrical contact layer, and removing the silicon dioxide thin film above the ridge waveguide structure.

[0011] According to an embodiment of this application, in step S33, a silicon dioxide thin film is grown using a plasma-enhanced chemical vapor deposition process, and the silicon dioxide thin film above the ridge waveguide structure is removed using photolithography and reactive ion etching processes.

[0012] Another aspect of this application provides an electrically absorbed modulated laser, comprising: a substrate; a laser active layer and a modulator active layer disposed side-by-side on the substrate; a grating located on the surface of the laser active layer; a cladding covering the modulator active layer and the grating; an electrical contact layer covering the cladding, having ridge waveguide structures on the cladding and the electrical contact layer, and having an electrical isolation trench; a silicon dioxide thin film covering the electrical contact layer; a P-electrode located on the silicon dioxide thin film; a silicon nitride thin film covering the P-electrode and the exposed area of ​​the electrical isolation trench; a hafnium oxide thin film covering the silicon nitride thin film; a wire bonding region located in the silicon nitride thin film and the hafnium oxide thin film, exposing the P-electrode at the wire bonding region; and an N-electrode located on the side of the substrate away from the laser active layer and the modulator active layer.

[0013] According to embodiments of this application, the thickness of both the silicon nitride thin film and the hafnium oxide thin film is 10 nanometers to 100 nanometers.

[0014] The electroabsorption modulated laser and its fabrication method provided in this application have at least the following beneficial effects:

[0015] After fabricating the P-electrode in the laser and modulator integrated chip, growing a composite thin film layer of silicon nitride and hafnium oxide can meet the non-hermetic packaging requirements of the optical emission chip. This utilizes both the excellent moisture resistance of hafnium oxide and the strong adhesion between silicon nitride and the P-electrode to jointly protect the chip from moisture corrosion. This meets the high-temperature and high-humidity resistance requirements of non-hermetic packaging for optical emission chips, making it suitable for non-hermetic applications. Non-hermetic optical emission chips can reduce costs, expand application scenarios, and provide a high-performance optical emission chip solution for the development of next-generation artificial intelligence. Attached Figure Description

[0016] The above and other objects, features and advantages of this application will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0017] Figure 1 A flowchart illustrating a method for fabricating an electroabsorption modulated laser according to an embodiment of this application is shown schematically.

[0018] Figure 2 The flowchart of step S1 in the fabrication method of an electroabsorption modulated laser according to an embodiment of this application is illustrated schematically.

[0019] Figure 3 The flowchart of step S3 in the fabrication method of an electroabsorption modulated laser according to an embodiment of this application is illustrated schematically.

[0020] Figure 4 The flowchart of step S6 in the method for fabricating an electroabsorption modulated laser according to an embodiment of this application is illustrated schematically.

[0021] Figure 5 A schematic diagram of an electroabsorption modulated laser according to an embodiment of this application is shown.

[0022] [Explanation of Labels in the Attached Image]

[0023] 1-Substrate; 2-Laser lower waveguide; 3-Laser multiple quantum wells; 4-Laser upper waveguide; 5-Modulator lower waveguide; 6-Modulator multiple quantum wells; 7-Modulator upper waveguide; 8-Grate; 9-Cladding; 10-Electrical contact layer; 11-Electrically isolating trench; 12-Silicon dioxide thin film; 13-P electrode; 14-Silicon nitride thin film; 15-Hafnium oxide thin film; 16-Wire bonding area; 17-N electrode. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with specific embodiments and the accompanying drawings.

[0025] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of this application. The terms “comprising,” “including,” etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0026] All terms used herein, including technical and scientific terms, have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0027] When using expressions such as "at least one of A, B, and C," the expression should generally be interpreted in accordance with the meaning commonly understood by those skilled in the art. For example, "a system having at least one of A, B, and C" should include, but is not limited to, systems having A alone, having B alone, having C alone, having A and B, having A and C, having B and C, and / or having A, B, and C.

[0028] It should also be noted that the directional terms mentioned in the embodiments, such as "up," "down," "front," "back," "left," and "right," are only for reference to the directions in the accompanying drawings and are not intended to limit the scope of protection of this application. Throughout the drawings, the same elements are represented by the same or similar reference numerals. Conventional structures or constructions will be omitted where they may cause confusion in understanding this application.

[0029] Figure 1 A flowchart illustrating a method for fabricating an electroabsorption modulated laser according to an embodiment of this application is shown. Figure 5A schematic diagram of an electroabsorption modulated laser according to an embodiment of this application is shown.

[0030] like Figure 1 and Figure 5 As shown, an embodiment of this application provides a method for fabricating an electroabsorption modulated laser, including steps S1 to S7, which are described in detail below in sequence.

[0031] Step S1: Fabricate the laser active layer and the modulator active layer on substrate 1. According to embodiments of this application, such as... Figure 2 As shown, step S1 specifically includes steps S11 to S13.

[0032] Step S11: A laser active layer is fabricated on substrate 1. The laser active layer includes a laser lower waveguide 2, a laser multiple quantum well 3, and a laser upper waveguide 4 stacked sequentially.

[0033] For example, an N-type indium phosphide (InP) substrate 1 is selected as the substrate for epitaxial growth. After standard cleaning and surface treatment, the substrate is placed in the reaction chamber of a metal-organic chemical vapor deposition (MOCVD) apparatus. Using the MOCVD apparatus, the active layer structure of the laser is sequentially epitaxially grown on the N-type InP substrate 1. The growth process of the active layer specifically includes the following steps.

[0034] First, a laser lower waveguide 2 made of indium gallium arsenide phosphide (InGaAsP) material is grown on an N-type InP substrate 1. The material composition of the laser lower waveguide 2 is precisely designed, and its bandgap wavelength is tuned within the range of 1100 nm to 1200 nm to ensure good transparency to light in the laser's operating wavelength band and to provide effective optical confinement. The typical thickness of the laser lower waveguide 2 is 50 nm to 200 nm, and the specific thickness is determined according to the design requirements of the optical field confinement factor and mode distribution.

[0035] Next, a multiple quantum well (MQW) 3 is grown on top of the laser's lower waveguide 2, serving as the core light-emitting region of the laser. This MQW 3 is composed of multiple alternating stacks of InGaAsP quantum wells and InGaAsP barriers. By precisely controlling the material composition of the MQW 3, its bandgap wavelength is designed to be in the range of 1530 nm to 1550 nm, covering the conventional C-band of optical fiber communication to meet the gain requirements of the target operating wavelength. The structural parameters of the MQW 3, such as the number of cycles, well width, and barrier width, are optimized based on the laser's threshold current, differential gain, and modulation bandwidth, with a typical number of cycles ranging from 5 to 10.

[0036] Finally, another layer of InGaAsP material, the upper waveguide 4, is grown on top of the laser multiple quantum wells 3. The composition and bandgap wavelength of this upper waveguide 4 are the same as or similar to those of the lower waveguide 2, with its bandgap wavelength also controlled within the range of 1100 nm to 1200 nm, and its thickness matched that of the lower waveguide 2. The upper waveguide 4 and the lower waveguide 2 together constitute a waveguide structure, effectively confining the optical field generated in the laser multiple quantum wells 3 in the vertical direction near the active region, ensuring that the light waves can efficiently interact with the subsequently fabricated grating structure and obtain gain.

[0037] Through the above steps, the epitaxial growth of the laser active layer on the N-type InP substrate 1 was completed. The obtained epitaxial structure, from bottom to top, includes the InGaAsP laser lower waveguide 2, the laser multiple quantum well 3, and the InGaAsP laser upper waveguide 4. All layers are single-crystal thin films with high-quality interfaces and precisely controllable thickness and composition.

[0038] Step S12: Cover a portion of the surface of the waveguide 4 on the laser with a mask pattern, and remove the portion of the active layer of the laser that is not covered by the mask pattern to expose the substrate 1.

[0039] According to an embodiment of this application, step S12 includes: growing a silicon dioxide thin film layer on the surface of the waveguide 4 on the laser, etching a mask pattern using photolithography and reactive ion etching processes, and removing the portion of the active layer of the laser that is not covered by the mask pattern using inductively coupled plasma etching and wet etching processes.

[0040] For example, a 250 nm thick silicon dioxide (SiO2) film is deposited on the surface of a wafer after the active layer has been grown using plasma-enhanced chemical vapor deposition (PECVD). This SiO2 film serves as a hard mask for subsequent inductively coupled plasma (ICP) etching processes.

[0041] After deposition, a layer of photoresist is spin-coated onto the SiO2 film surface. Through exposure and development processes, an opening pattern corresponding to the laser area is formed on the photoresist, exposing the underlying SiO2 film. Next, using the patterned photoresist as a mask, reactive ion etching (RIE) is employed to selectively etch the exposed SiO2 film, precisely transferring the pattern from the photoresist to the SiO2 film, forming a hard SiO2 mask pattern that covers and protects the laser area. After etching, the remaining photoresist is removed.

[0042] Subsequently, the wafer undergoes standard cleaning to remove etching residues. After cleaning, using the formed SiO2 hard mask pattern as an etching barrier layer, anisotropic dry etching is first performed on the active layer material not covered by the SiO2 hard mask using ICP etching, etching vertically downwards to etch most of the material thickness. Next, a wet etching process is used to selectively chemically etch the surface after ICP etching to remove residual damage layers, smooth the etched sidewalls, and finally completely remove the active layer material outside the SiO2 hard mask pattern, forming the mesa structure of the laser region.

[0043] Step S13: A modulator active layer is fabricated on substrate 1. The modulator active layer includes a modulator lower waveguide 5, a modulator multiple quantum well 6, and a modulator upper waveguide 7 stacked sequentially.

[0044] For example, MOCVD equipment is used for the docking growth of modulator active layer materials. The specific growth process of the modulator active layer is as follows.

[0045] First, an InGaAsP modulator lower waveguide 5 is grown on the exposed substrate 1 surface and the side of the laser mesa. The material composition of the modulator lower waveguide 5 is designed so that its bandgap wavelength is tuned within the range of 1100 nm to 1200 nm to ensure good transparency to light in the modulator's operating wavelength band. The thickness of the modulator lower waveguide 5 is matched with that of the laser lower waveguide 2 in the laser region to achieve effective optical field coupling.

[0046] Next, a modulator multi-quantum well 6 is grown on the lower waveguide 5 of the modulator, serving as the core electro-absorption region of the modulator. This modulator multi-quantum well 6 is composed of multiple alternating stacks of InGaAsP quantum wells and InGaAsP barriers. By precisely controlling the material composition of the modulator multi-quantum well 6, its bandgap wavelength is positioned within the range of 1500 nm to 1530 nm. This bandgap wavelength is slightly smaller than the bandgap wavelength of the laser multi-quantum well 3 (1530 nm to 1550 nm), meaning that the absorption edge of the modulator's active region undergoes a blue shift relative to the laser's emission wavelength. This allows for efficient electric field-induced light absorption when a reverse bias voltage is applied, utilizing the quantum confinement Stark effect (QCSE), thus achieving modulation. The structural parameters of the modulator multi-quantum well 6, such as the number of periods, well width, and barrier width, are optimized based on the modulator's extinction ratio, modulation bandwidth, and insertion loss performance indicators.

[0047] Finally, another InGaAsP modulator upper waveguide 7 is grown on top of the modulator multi-quantum well 6. The composition and bandgap wavelength of this modulator upper waveguide 7 are the same as or similar to those of the modulator lower waveguide 5, and its bandgap wavelength is also controlled within the range of 1100 nm to 1200 nm. Its thickness matches that of the modulator lower waveguide 5. The modulator upper waveguide 7 and the modulator lower waveguide 5 together constitute the waveguide structure of the modulator region, effectively confining the optical field generated in the modulator multi-quantum well 6 in the vertical direction near the active region of the modulator.

[0048] Through the above-described docking growth process, the monolithic integration of the modulator active layer and the laser active layer is completed. The modulator lower waveguide 5, modulator multiple quantum well 6, and modulator upper waveguide 7 of the modulator active layer are aligned vertically with the corresponding layers of the laser active layer, and continuous optical path coupling is achieved in the horizontal direction through the docking interface, ensuring that the light generated by the laser can be transmitted to the modulator region with low loss.

[0049] Step S2: Prepare grating 8 on the surface of the active layer of the laser.

[0050] For example, a wet etching process is used to remove the remaining SiO2 hard mask layer on the wafer surface, exposing the laser waveguide 4 surface in the laser region, and then the wafer is cleaned according to standard procedures.

[0051] Subsequently, a quarter-wavelength phase-shift grating 8 is fabricated on the laser waveguide 4 in the laser region. The fabrication process of the grating 8 is as follows: First, a layer of photoresist is coated on the surface of the laser waveguide 4 as a mask layer; then, an electron beam lithography or holographic lithography process is used to form a grating pattern with a quarter-wavelength phase-shift structure on the photoresist. The grating period is determined to be 240 nm based on the target operating wavelength of the laser (e.g., 1550 nm) and the effective refractive index of the material; then, using the patterned photoresist as a barrier layer, an inductively coupled plasma (ICP) etching process is used to transfer the grating pattern to the surface of the laser waveguide 4, forming a periodic corrugated grating with a depth of 50 nm; finally, the remaining photoresist is removed. This quarter-wavelength phase-shift grating 8 introduces a π / 2 phase transition at the center of the grating structure, ensuring that the laser can achieve stable single-mode operation at the Bragg wavelength.

[0052] Step S3: Cladding 9, electrical contact layer 10, ridge waveguide structure, electrical isolation trench 11, and silicon dioxide thin film 12 are sequentially fabricated on the modulator active layer and grating 8. According to embodiments of this application, such as... Figure 3 As shown, step S3 specifically includes steps S31 to S33.

[0053] Step S31: A cladding layer 9 is prepared on the modulator active layer and the grating 8, and an electrical contact layer 10 is prepared on the cladding layer 9.

[0054] For example, firstly, using an MOCVD device, an InP cladding layer 9 is formed by epitaxial regeneration on the exposed surfaces of the laser waveguide 4, grating 8, and modulator waveguide 7 in the modulator region. This InP cladding layer 9 completely covers the periodic corrugated structure of the grating 8, embedding it within it, and simultaneously covers the upper surfaces and mating interfaces of the laser and modulator regions. The InP cladding layer 9 has a thickness of 1500 nm, its bandgap wavelength is smaller than the laser's operating wavelength, it is transparent to the optical field, and its refractive index is lower than that of the active layer material. Together with the lower laser waveguide 2 and lower modulator waveguide 5 below, it forms a vertical optical confinement structure. By controlling the growth conditions, the InP cladding layer 9 is uniformly filled in the trenches and mesa regions of the grating 8, ensuring the crystal quality and optical performance of the grating interface.

[0055] Next, a second epitaxial growth was performed on the InP cladding layer 9 using an MOCVD apparatus to form an indium gallium arsenide (InGaAs) electrical contact layer 10. This InGaAs electrical contact layer 10 has a thickness of 200 nm and a higher doping concentration than the InP cladding layer 9. It is used to form a low-resistance ohmic contact with the subsequently fabricated P-electrode, reducing the series resistance of the device. The InGaAs electrical contact layer 10 covers the entire upper surface of the laser and modulator regions, providing a conductive path for the subsequent fabrication of the ridge waveguide and current injection.

[0056] Step S32: A ridge waveguide structure is fabricated on the cladding 9 and the electrical contact layer 10, and an electrical isolation trench 11 exposing the cladding 9 is etched in the middle of the electrical contact layer 10 to achieve electrical isolation between the laser and the modulator.

[0057] For example, a ridge waveguide structure is fabricated on an InP cladding layer 9 and an InGaAs electrical contact layer 10.

[0058] First, a 300 nm thick SiO2 film is deposited on the surface of the InGaAs electrical contact layer 10 using PECVD as an etching hard mask. A photoresist layer is then spin-coated onto the SiO2 film surface. Through exposure and development processes, a strip-shaped opening pattern corresponding to the ridge waveguide is formed on the photoresist. The width of the strip-shaped opening is 2 μm, exposing the underlying SiO2 film. Next, using the patterned photoresist as a mask, selective etching is performed on the exposed SiO2 film using RIE (Reverse Etching) technology to transfer the pattern from the photoresist to the SiO2 film, forming the SiO2 hard mask pattern. After etching, the remaining photoresist is removed.

[0059] Subsequently, using the formed SiO2 hard mask pattern as an etching barrier layer, anisotropic dry etching was performed on the InGaAs electrical contact layer 10 and InP cladding layer 9 not covered by the SiO2 hard mask using an ICP etching process. The etching depth was controlled to reach a position 200 nm from the upper surface of the laser upper waveguide 4 and the modulator upper waveguide 7 inside the InP cladding layer 9, forming a ridge waveguide structure. This ridge waveguide structure consists of strip-shaped protrusions extending along the light transmission direction in both the laser and modulator regions, with a ridge width of 2 μm. The etched areas on both sides of the ridge form flat plate regions. After etching, the remaining SiO2 hard mask was removed using a wet etching process.

[0060] After the ridge waveguide structure is fabricated, an electrical isolation trench 11 between the laser and the modulator is etched on the electrical contact layer 10.

[0061] A layer of photoresist is spin-coated onto the wafer surface forming the ridge waveguide structure. Through exposure and development processes, an opening pattern corresponding to the electrical isolation trench is formed on the photoresist. This opening pattern is located in a predetermined isolation region between the laser region and the modulator region. Next, using the patterned photoresist as a mask, anisotropic dry etching is performed on the exposed InGaAs electrical contact layer 10 and InP cladding layer 9 using an ICP etching process. The etching depth is controlled to just penetrate the InGaAs electrical contact layer 10 and stop at the upper surface of the InP cladding layer 9, forming the electrical isolation trench 11. This electrical isolation trench 11 penetrates the InGaAs electrical contact layer 10 perpendicular to the light transmission direction, completely separating the electrical contact layers 10 of the laser region and the modulator region electrically, achieving electrical isolation between the laser and the modulator. After etching, the remaining photoresist is removed.

[0062] Step S33: A silicon dioxide thin film 12 is grown on the electrical contact layer 10, and the silicon dioxide thin film 12 above the ridge waveguide structure is removed.

[0063] According to an embodiment of this application, in step S33, a silicon dioxide thin film 12 is grown using a plasma-enhanced chemical vapor deposition process, and the silicon dioxide thin film 12 above the ridge waveguide structure is removed using photolithography and reactive ion etching processes.

[0064] For example, using a PECVD apparatus at a growth temperature of 300°C, silane (SiH4) and nitrous oxide (N2O) are introduced as reaction precursors. A chemical reaction occurs with the assistance of plasma, depositing a silicon dioxide (SiO2) thin film 12 on the entire surface of the wafer. The thickness of this silicon dioxide thin film 12 is 200 nm to 400 nm, covering the top, sidewalls, and both planar regions of the ridge waveguide structure. During deposition, the flow ratio of silane to nitrous oxide is set to 1:3, the reaction chamber pressure is controlled at 2 Torr, and the RF power is 50 W. This silicon dioxide thin film 12 serves as an electrically insulating dielectric layer, electrically isolating the semiconductor material in the planar regions on both sides of the ridge waveguide from the subsequently fabricated P-electrodes. Simultaneously, its lower refractive index compared to the InP cladding layer 9 enhances the lateral confinement of the optical field by the ridge waveguide.

[0065] After deposition, a layer of photoresist is spin-coated onto the surface of the silicon dioxide thin film 12. An opening pattern is formed on the photoresist through exposure and development processes. This opening pattern is located directly above the top of the ridge waveguide structure, exposing the silicon dioxide thin film 12 covering the top of the ridge waveguide structure. Next, using the patterned photoresist as a mask, an anisotropic dry etching process is performed on the exposed silicon dioxide thin film 12 using a RIE process. A mixed gas of CF4 and CHF3 is introduced as the etching gas, with a CF4 to CHF3 flow rate ratio of 1:1. This removes the silicon dioxide thin film 12 at the top of the ridge waveguide structure, exposing the underlying InGaAs electrical contact layer 10, forming an electrical injection window. After etching, the remaining photoresist is removed. Through this electrical injection window, the subsequently fabricated P-electrode can directly form an ohmic contact with the InGaAs electrical contact layer 10, achieving effective injection of charge carriers from the electrode into the active region of the ridge waveguide.

[0066] Step S4: Prepare P electrode 13 on silicon dioxide thin film 12.

[0067] For example, firstly, a layer of photoresist is spin-coated onto the wafer surface forming the electrical injection window. Through exposure and development processes, an opening pattern corresponding to the P-electrode 13 pattern is formed on the photoresist. This opening pattern covers the electrical injection window regions of both the laser and modulator regions, exposing the underlying InGaAs electrical contact layer 10. Next, using electron beam evaporation or magnetron sputtering, a titanium (Ti) layer and a gold (Au) layer are sequentially deposited on the entire wafer surface. The Ti layer is 50 nm thick, serving as an adhesion and diffusion barrier layer, and the Au layer is 200 nm thick, serving as a conductive and bonding layer. After deposition, the wafer is immersed in an organic solvent for a lift-off process to remove the photoresist and the overlying metal layers, retaining only the Ti and Au metal stack in the opening pattern region, forming the P-electrode 13 located in the laser region and the P-electrode 13 located in the modulator region. Subsequently, a rapid thermal annealing process was performed in a nitrogen atmosphere at 350°C for 60 seconds to form a low-resistance ohmic contact between the P electrode 13 and the InGaAs electrical contact layer 10.

[0068] The P-electrode 13 in the laser region covers the electrical injection window at the top of the ridge waveguide structure in the laser region and is in direct contact with the InGaAs electrical contact layer 10 in this region. The P-electrode 13 in the modulator region covers the electrical injection window at the top of the ridge waveguide structure in the modulator region and is in direct contact with the InGaAs electrical contact layer 10 in this region. The two P-electrodes 13 are electrically completely separated by an electrical isolation trench 11, and are used to independently inject drive current into the laser and modulator, respectively. Each P-electrode 13 has a pre-reserved flat bonding region with an area of ​​50 μm × 50 μm for subsequent electrical connection to an external drive circuit via wire bonding.

[0069] Step S5: A silicon nitride thin film 14 is grown on the surface of the P electrode 13 and in the exposed area of ​​the electrical isolation trench 11, and a hafnium oxide thin film 15 is grown on the silicon nitride thin film 14.

[0070] The embodiments of this application employ a silicon nitride thin film as the composite film substrate. This utilizes the excellent interfacial adhesion between silicon nitride and the P-electrode metal material to enhance the adhesion strength of the protective layer on the metal electrode surface, reducing the risk of the film peeling off from the electrode surface during subsequent processes and use. Simultaneously, the dense structure of the silicon nitride thin film effectively prevents the outward diffusion of metal atoms, maintaining the chemical stability of the electrode interface.

[0071] Hafnium oxide thin film is used as the top layer of the composite film. Hafnium oxide has high chemical stability, a dense structure, and excellent water resistance. Using hafnium oxide as the outermost layer of the composite film directly resists the erosion of external moisture and pollutants, providing long-term reliable protection for non-hermetic devices. Experimental data shows that devices using the composite film of this invention maintain good photoelectric performance after 1000 hours of aging tests at 85°C and 85% relative humidity, meeting the reliability requirements of non-hermetic devices in high-humidity environments.

[0072] According to an embodiment of this application, step S5 includes: performing plasma-enhanced atomic layer deposition at a temperature of 300°C, wherein at least two precursors are alternately introduced in a deposition cycle, and plasma is excited when at least one precursor is introduced, to grow a silicon nitride thin film 14 and a hafnium oxide thin film 15.

[0073] For example, firstly, the wafer is placed in the reaction chamber of an atomic layer deposition (ALD) apparatus and heated to 300°C. Using tetratetra(dimethylamino)silane (TDMA-Si) as the silicon source precursor and ammonia (NH3) as the nitrogen source precursor, plasma-enhanced atomic layer deposition (PEALD) is performed with the assistance of plasma. One deposition cycle includes: pulsed TDMA-Si precursor, causing it to chemically adsorb onto the wafer surface; a first purging with an inert gas to remove excess precursor and byproducts; pulsed NH3 and simultaneously excited plasma, causing the active free radicals in the NH3 plasma to react with the surface-adsorbed TDMA-Si to generate a silicon nitride monolayer; and a second purging with an inert gas. This deposition cycle is repeated until the thickness of the silicon nitride film 14 reaches a predetermined thickness. This silicon nitride film 14 covers the entire upper surface of the wafer, including the upper surface of the P-electrode 13, the sidewalls of the electrical isolation trench 11, and the bottom.

[0074] Subsequently, at a process temperature of 300°C, the precursors were switched, using tetra(ethylmethylamino)hafnium (TEMAHf) as the hafnium source precursor and oxygen (O2) as the oxygen source precursor for PEALD deposition. One deposition cycle included: pulsed TEMAHf precursor, causing it to chemically adsorb onto the surface of the silicon nitride film 14; a first purging with an inert gas; pulsed O2 and simultaneously excited plasma, causing the active free radicals in the O2 plasma to react with the surface-adsorbed TEMAHf to generate a hafnium oxide monolayer; and a second purging with an inert gas. This deposition cycle was repeated until the thickness of the hafnium oxide film 15 reached a preset thickness. This hafnium oxide film 15 covered the entire surface of the silicon nitride film 14.

[0075] The embodiments of this application select ALD (Alternating Deposition) technology as the deposition method for composite thin films, making full use of the excellent step coverage capability of ALD technology. Due to the presence of ridge waveguide structures, electrically isolated trenches, and P-electrode metal patterns on the surface of the electroabsorption modulated laser chip, the chip surface morphology is complex. Based on a self-limiting surface chemical reaction mechanism, the ALD process can uniformly grow thin films at the same rate on the vertical sidewalls and horizontal planes, ensuring that the film thickness deposited at various locations such as the ridge waveguide sidewalls, the inner sidewalls of the electrically isolated trenches, and the edges of the P-electrode is basically consistent, avoiding the defects of thinning or delamination at step points that occur in traditional PECVD or PVD processes. The uniform film thickness ensures a continuous and dense protective layer throughout the entire chip surface, eliminating weak points that may be susceptible to moisture intrusion.

[0076] A silicon nitride thin film 14 and a hafnium oxide thin film 15 were deposited at a substrate temperature of 300°C using plasma-enhanced atomic layer deposition (PEALD). Compared to traditional high-temperature thermal oxidation or high-temperature chemical vapor deposition processes (typically requiring 600°C to 800°C), this significantly reduces the process temperature. The deposition temperature of 300°C falls within the compatible temperature range for back-end processes in semiconductor optoelectronic devices, and is far lower than the thermal decomposition temperatures of III-V group semiconductor materials such as the InGaAs electrical contact layer 10 and the InP cladding layer 9. It is also lower than the critical temperature for component interdiffusion in the multi-quantum well structure in the active regions of lasers and modulators. Depositing the composite thin film at this temperature effectively avoids thermal damage to the existing device structure caused by high-temperature processes, protects the interface steepness and gain characteristics of the multi-quantum wells, and prevents excessive alloying reactions or metal diffusion between the P electrode 13 and the InGaAs electrical contact layer 10.

[0077] Meanwhile, the introduction of plasma provides additional energy to the reaction precursors, enabling them to fully decompose and undergo chemical reactions at a low temperature of 300℃ to generate high-quality thin films. This overcomes the limitation of traditional ALD, which requires higher temperatures to obtain the same film quality. Due to the lower deposition temperature, the thermal expansion and contraction of the wafer during heating and cooling are reduced, resulting in lower thermal stress accumulation and a reduced risk of wafer warpage and film cracking. This improves the process compatibility of the entire fabrication process and the yield of the final device.

[0078] According to embodiments of this application, the thickness of both the silicon nitride thin film 14 and the hafnium oxide thin film 15 is 10 nanometers to 100 nanometers.

[0079] The thicknesses of both the silicon nitride film 14 and the hafnium oxide film 15 are limited to 10 nm to 100 nm, balancing the dual requirements of waterproof reliability and film stress control. The hafnium oxide film itself possesses excellent resistance to water vapor penetration and chemical stability; a thickness of only 10 nm or more is sufficient to form a continuous and dense water vapor barrier layer, meeting the moisture-proof requirements of non-hermetically sealed packaging at 85°C and 85% relative humidity. The silicon nitride film, at a thickness of 10 nm or more, provides effective protection against metal diffusion and interfacial bonding. Simultaneously, controlling the upper limit of the thickness of both films to within 100 nm avoids stress accumulation caused by excessively thick films. Furthermore, this thickness range also considers the production efficiency of the ALD process, minimizing deposition time while ensuring film quality and protective performance, which is beneficial for mass production.

[0080] A thickness range of 10 to 100 nanometers also helps reduce the impact of composite films on the optical and electrical performance of devices. The surface of an electroabsorption modulated laser chip has a ridge waveguide structure. When the light field propagates in the waveguide, some evanescent field extends into the dielectric layer above the waveguide. By controlling the total thickness of the composite film to the subwavelength level (below 200 nanometers), the disturbance to the light field is minimal, and it will not disrupt the optical transmission mode of the laser or modulator. Simultaneously, the thinner dielectric layer covering the P-electrode and the ridge waveguide structure introduces less parasitic capacitance, which helps maintain the high-speed modulation performance of the device. Controlling the thickness of a single-layer film to below 100 nanometers also effectively reduces the risk of film cracking or peeling caused by defect proliferation and stress concentration in thick films, ensuring the composite film maintains excellent mechanical integrity and long-term reliability.

[0081] According to an embodiment of this application, in step S5, based on the thickness of the silicon dioxide film 12, the thickness ratio of the silicon nitride film 14 and the hafnium oxide film 15 is adjusted so that the absolute value of the residual stress is lower than a preset threshold.

[0082] After deposition, the silicon dioxide thin film 12 introduces a non-negligible compressive stress, forming the stress substrate of the entire dielectric film. On this substrate, the silicon nitride thin film 14 exhibits tensile stress, and the hafnium oxide thin film 15 exhibits compressive stress; these three stresses are superimposed. Utilizing the linear cumulative law that the stress properties of each thin film are positively correlated with its thickness, the tensile stress introduced by the silicon nitride thin film 14 can be effectively compensated for the compressive stress jointly generated by the silicon dioxide thin film 12 and the hafnium oxide thin film 15 by adjusting the film thickness. Ultimately, the absolute value of the residual stress in the entire dielectric film, including the silicon dioxide thin film 12, silicon nitride thin film 14, and hafnium oxide thin film 15, is controlled below a preset threshold.

[0083] For example, if the thickness of the silicon dioxide film 12 is 300 nm, and the thickness of the silicon nitride film 14 is set to 100 nm and the thickness of the hafnium oxide film 15 is set to 50 nm, the absolute value of the residual stress can be kept below 100 MPa. This design, which uses the fixed thickness of the bottom layer as a constraint and achieves stress self-compensation by adjusting the thickness ratio of the upper film, avoids mechanical failures such as wafer warping, film cracking, or peeling caused by excessive stress. It ensures the processing accuracy and yield of subsequent photolithography, dicing, and packaging processes, while also ensuring the structural stability and reliability of the device during long-term operation.

[0084] Step S6: Remove the silicon nitride film 14 and hafnium oxide film 15 from the wire bonding region of the P electrode 13 to expose the P electrode 13, forming the wire bonding region 16. According to an embodiment of this application, such as... Figure 4 As shown, step S6 specifically includes steps S61 to S63.

[0085] Step S61: A patterned photoresist mask is formed on the surface of the hafnium oxide thin film 15. The patterned photoresist mask has an opening at the position corresponding to the wire bonding area of ​​the P electrode 13, exposing the hafnium oxide thin film 15 in that area.

[0086] For example, a layer of photoresist is spin-coated onto the surface of the hafnium oxide thin film 15. Through exposure and development processes, an opening pattern is formed on the photoresist. This opening pattern is located directly above the bonding area of ​​the P electrode 13, exposing the hafnium oxide thin film 15 corresponding to the bonding area, thus forming a patterned photoresist mask. The opening size of the photoresist corresponds to the size of the bonding area of ​​the P electrode 13, ensuring that the subsequent etching process can accurately remove the composite film in this area.

[0087] In step S62, using a patterned photoresist mask as a barrier layer, reactive ion etching is employed to sequentially remove the exposed hafnium oxide film 15 and silicon nitride film 14 to expose the underlying P electrode 13, forming the wire bonding area 16.

[0088] For example, using a patterned photoresist mask as a barrier layer, reactive ion etching (RIE) is employed. First, a mixture of boron trichloride (BCl3) and chlorine (Cl2) is introduced as the etching gas, with a BCl3 to Cl2 flow rate ratio of 1:3, to perform anisotropic dry etching on the exposed hafnium oxide film 15. Hafnium oxide is chemically stable, requiring the use of a chlorine-containing gas to utilize the reaction of chlorine radicals with hafnium atoms to generate volatile hafnium chloride for effective etching. Once the hafnium oxide film 15 is completely removed, exposing the underlying silicon nitride film 14, etching is stopped.

[0089] Subsequently, the etching gas formulation was switched, and a mixture of carbon tetrafluoride (CF4) and oxygen (O2) was introduced as the etching gas, with a CF4:O2 flow rate ratio of 4:1. The reason for switching the etching gas is that silicon nitride and hafnium oxide have different material compositions and chemical bonding characteristics. Chlorine-containing gases, which are suitable for hafnium oxide, have a low etching rate for silicon nitride, while fluorine-containing gases utilize fluorine radicals to react with silicon atoms to generate volatile silicon tetrafluoride, which has a higher etching rate and better selectivity for silicon nitride, and can efficiently remove the silicon nitride film 14 without excessively damaging the underlying P electrode 13 metal layer. The exposed silicon nitride film 14 was then subjected to anisotropic dry etching using the RIE process to completely remove it, exposing the surface of the underlying P electrode 13.

[0090] Through the two-step RIE etching process described above, an opening is formed on the composite film. This opening area is the wire bonding region 16 of the P electrode 13. The size of the wire bonding region 16 is consistent with the size of the patterned photoresist mask opening. It has steep sidewalls, and the bottom surface of the P electrode 13 is clean with no residual film, which is used for subsequent wire bonding to realize the electrical connection between the chip and external circuits.

[0091] Step S63: Remove the remaining patterned photoresist mask.

[0092] Step S7: An N-electrode 17 is fabricated on the side of substrate 1 away from the laser active layer and the modulator active layer to form an electroabsorption modulated laser.

[0093] For example, firstly, the back side of the N-type InP substrate 1 is thinned. A mechanical polishing process is used to reduce the substrate 1 from its original thickness to 100 μm. Then, a chemical mechanical polishing (CMP) process is used to polish the polished back side, eliminating the surface damage layer introduced by the polishing process and obtaining a smooth and flat back side. The reduced substrate thickness helps to lower the series resistance and thermal resistance of the device, while also facilitating subsequent dicing and chip separation.

[0094] Subsequently, an N-electrode 17 was fabricated on the back side of the thinned and polished substrate 1. Using electron beam evaporation or magnetron sputtering, a titanium (Ti) layer and a gold (Au) layer were sequentially deposited on the entire back side of substrate 1. The Ti layer was 50 nm thick and served as an adhesion layer, while the Au layer was 200 nm thick and served as a conductive layer. After deposition, a rapid thermal annealing process was performed at 350°C in a nitrogen atmosphere for 60 seconds to establish a low-resistance ohmic contact between the N-electrode 17 and the N-type InP substrate 1, thus completing the fabrication of the N-electrode 17.

[0095] After completing all the above process steps, the wafer is diced along the cleavage plane to separate it into individual dies, thus obtaining the electroabsorption modulated laser chip.

[0096] like Figure 5As shown, embodiments of this application also provide an electro-absorption modulated laser, comprising: a substrate 1, a laser active layer, a modulator active layer, a grating 8, a cladding 9, an electrical contact layer 10, an electrical isolation trench 11, a silicon dioxide thin film 12, a P-electrode 13, a silicon nitride thin film 14, a hafnium oxide thin film 15, a wire bonding region 16, and an N-electrode 17.

[0097] The laser active layer and the modulator active layer are disposed side-by-side on the substrate 1. A grating 8 is located on the surface of the laser active layer. A cladding layer 9 covers the modulator active layer and the grating 8. An electrical contact layer 10 covers the cladding layer 9, and both the cladding layer 9 and the electrical contact layer 10 have ridge waveguide structures. The electrical contact layer 10 has an electrical isolation trench 11. A silicon dioxide thin film 12 covers the electrical contact layer 10. A P-electrode 13 is located on the silicon dioxide thin film 12. A silicon nitride thin film 14 covers the P-electrode 13 and the exposed area of ​​the electrical isolation trench 11. A hafnium oxide thin film 15 covers the silicon nitride thin film 14. A wire bonding region 16 is located between the silicon nitride thin film 14 and the hafnium oxide thin film 15, exposing the P-electrode 13 at the wire bonding region 16. An N-electrode 17 is located on the side of the substrate 1 away from the laser active layer and the modulator active layer.

[0098] In summary, the embodiments of this application provide an electroabsorption modulated laser and its fabrication method. After fabricating the P-electrode on the laser and modulator integrated chip, a composite thin film of silicon nitride and hafnium oxide is deposited using atomic layer deposition (ALD). This utilizes both the excellent moisture resistance of hafnium oxide and the strong adhesion between silicon nitride and the P-electrode to jointly protect the chip from moisture corrosion. This meets the high temperature and high humidity resistance requirements of hermetically sealed optical emitting chips, making it suitable for hermetically sealed applications. The hermetically sealed approach completely eliminates the hermetically sealed processes such as metal casing, inert gas filling, and sealing welding required in traditional solutions, simplifying the packaging process and reducing material and manufacturing costs. This expands application scenarios and provides a high-performance optical emitting chip solution for the development of next-generation artificial intelligence.

[0099] The embodiments of this application have been described above. However, these embodiments are merely illustrative and not intended to limit the scope of this application. Although various embodiments have been described above, this does not mean that the measures in the various embodiments cannot be used advantageously in combination. Without departing from the scope of this application, those skilled in the art can make various substitutions and modifications, all of which should fall within the scope of this application.

Claims

1. A method for fabricating an electroabsorption modulated laser, characterized in that, include: Step S1: Prepare a laser active layer and a modulator active layer on the substrate (1); Step S2, a grating (8) is fabricated on the surface of the active layer of the laser. Step S3: Cladding (9), electrical contact layer (10), ridge waveguide structure, electrical isolation trench (11) and silicon dioxide thin film (12) are sequentially prepared on the modulator active layer and the grating (8). Step S4: Prepare a P electrode (13) on the silicon dioxide thin film (12). Step S5: A silicon nitride thin film (14) is grown on the surface of the P electrode (13) and in the exposed area of ​​the electrical isolation trench (11), and a hafnium oxide thin film (15) is grown on the silicon nitride thin film (14). Step S6: Remove the silicon nitride film (14) and hafnium oxide film (15) in the wire bonding area of ​​the P electrode (13) to expose the P electrode (13) and form the wire bonding area (16). Step S7: An N-electrode (17) is prepared on the side of the substrate (1) away from the active layer of the laser and the active layer of the modulator to form the electro-absorption modulated laser.

2. The method for fabricating an electroabsorption modulated laser according to claim 1, characterized in that, In step S5, based on the thickness of the silicon dioxide film (12), the thickness ratio of the silicon nitride film (14) and the hafnium oxide film (15) is adjusted so that the absolute value of the residual stress is lower than a preset threshold.

3. The method for fabricating an electroabsorption modulated laser according to claim 1, characterized in that, Step S5 includes: Plasma-enhanced atomic layer deposition is performed at a temperature of 300°C, wherein at least two precursors are alternately introduced in one deposition cycle and plasma is excited when at least one precursor is introduced to grow the silicon nitride film (14) and hafnium oxide film (15).

4. The method for fabricating an electroabsorption modulated laser according to claim 1, characterized in that, Step S6 includes: Step S61: A patterned photoresist mask is formed on the surface of the hafnium oxide film (15). The patterned photoresist mask has an opening at the position corresponding to the wire bonding area of ​​the P electrode (13) to expose the hafnium oxide film (15) in that area. In step S62, using the patterned photoresist mask as a barrier layer, reactive ion etching is used to sequentially remove the exposed hafnium oxide film (15) and silicon nitride film (14) to expose the underlying P electrode (13) and form a wire bonding area (16). Step S63: Remove the remaining patterned photoresist mask.

5. The method for fabricating an electroabsorption modulated laser according to claim 1, characterized in that, Step S1 includes: Step S11: A laser active layer is fabricated on a substrate (1). The laser active layer includes a laser lower waveguide (2), a laser multiple quantum well (3), and a laser upper waveguide (4) stacked sequentially. Step S12: Cover a portion of the surface of the waveguide (4) of the laser with a mask pattern, and remove the portion of the active layer of the laser that is not covered by the mask pattern to expose the substrate (1). Step S13: A modulator active layer is prepared on the substrate (1). The modulator active layer includes a modulator lower waveguide (5), a modulator multiple quantum well (6), and a modulator upper waveguide (7) stacked sequentially.

6. The method for fabricating an electroabsorption modulated laser according to claim 5, characterized in that, Step S12 includes: A silicon dioxide thin film layer is grown on the surface of the waveguide (4) of the laser, and a mask pattern is etched by photolithography and reactive ion etching. The part of the active layer of the laser that is not covered by the mask pattern is removed by inductively coupled plasma etching and wet etching.

7. The method for fabricating an electroabsorption modulated laser according to claim 1, characterized in that, Step S3 includes: Step S31: A cladding layer (9) is prepared on the modulator active layer and the grating (8), and an electrical contact layer (10) is prepared on the cladding layer (9). Step S32: A ridge waveguide structure is fabricated on the cladding (9) and the electrical contact layer (10), and an electrical isolation trench (11) exposing the cladding (9) is etched in the middle of the electrical contact layer (10) to achieve electrical isolation between the laser and the modulator; Step S33: A silicon dioxide thin film (12) is grown on the electrical contact layer (10), and the silicon dioxide thin film (12) above the ridge waveguide structure is removed.

8. The method for fabricating an electroabsorption modulated laser according to claim 7, characterized in that, In step S33, the silicon dioxide thin film (12) is grown using plasma-enhanced chemical vapor deposition, and the silicon dioxide thin film (12) above the ridge waveguide structure is removed using photolithography and reactive ion etching.

9. An electroabsorption modulated laser, characterized in that, include: Substrate (1); The laser active layer and the modulator active layer are disposed side by side on the substrate (1); The grating (8) is located on the surface of the active layer of the laser; A cladding layer (9) covers the modulator active layer and the grating (8); An electrical contact layer (10) covers the cladding layer (9), and the cladding layer (9) and the electrical contact layer (10) have ridge waveguide structures. The electrical contact layer (10) has an electrical isolation trench (11). A silicon dioxide film (12) is applied to the electrical contact layer (10); The P electrode (13) is located on the silicon dioxide thin film (12); A silicon nitride thin film (14) covers the P electrode (13) and the exposed area of ​​the electrical isolation trench (11); A hafnium oxide thin film (15) is applied over the silicon nitride thin film (14); The wire bonding area (16) is located in the silicon nitride film (14) and the hafnium oxide film (15), and the P electrode (13) is exposed at the wire bonding area (16). The N electrode (17) is located on the side of the substrate (1) away from the laser active layer and the modulator active layer.

10. The electroabsorption modulated laser according to claim 9, characterized in that, The thickness of both the silicon nitride thin film (14) and the hafnium oxide thin film (15) is 10 nanometers to 100 nanometers.

Images