A silicon carbide MOS device and its fabrication method

By employing a composite superjunction design with longitudinal gradient doping and lateral segmented doping in silicon carbide MOS devices, the problems of electric field concentration and charge unevenness are solved, improving the device's withstand voltage and conduction performance, extending the device's lifespan, and making it suitable for high-reliability applications.

CN123124377APending Publication Date: 2026-11-17HEILONGJIANG HUIXIN SEMICONDUCTOR CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202610836779.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-06-10
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing silicon carbide MOS devices suffer from local electric field concentration and uneven charge matching under high voltage operation, making it difficult to optimize the trade-off between the device's withstand voltage and on-resistance, which can easily lead to early breakdown failure.

Method used

A composite superjunction design combining vertical gradient doping and lateral segmented doping is adopted. Gradient-doped N-type pillars and segmented-doped P-type pillars are formed through multiple ion implantations. The charge distribution and electric field state are optimized by combining the trench gate structure.

Benefits of technology

It achieves balanced charge control across the entire domain, improves the problem of local electric field concentration, enhances the device's withstand voltage and conduction performance, and strengthens its stability and lifespan under high temperature and high pressure conditions, meeting the requirements of automotive and industrial-grade high reliability applications.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123124377A_ABST
    Figure CN123124377A_ABST
Patent Text Reader

Abstract

This application belongs to the field of semiconductor technology and discloses a silicon carbide MOS device and its fabrication method. The method first grows an N-type drift layer epitaxially on the front side of a silicon carbide substrate, defining a superjunction region and a gate region. N-type pillars are formed through multiple ion implantations with increasing energy and dosage, followed by segmented ion implantation to prepare P-type pillars composed of high- and low-doped sections. Subsequently, a gate trench is etched to form a gate trench, and a P-type layer is fabricated on the inner wall of the trench, followed by the sequential formation of a gate insulating layer and a gate. Finally, a PN junction, source, and drain are fabricated to obtain the silicon carbide MOS device. This application achieves balanced control of the global charge within the silicon carbide MOS device through a composite superjunction design combining vertical and horizontal gradient doping. This effectively improves the problem of local electric field concentration under high-voltage conditions, overcomes the technical bottleneck of the mutual constraint between voltage withstand capability and conduction loss in traditional devices, and significantly improves the overall electrical performance of the device.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application belongs to the field of semiconductor technology, and specifically relates to a silicon carbide MOS device and its fabrication method. Background Technology

[0002] As power electronics technology continues to develop towards high voltage, high frequency, high efficiency, and high reliability, third-generation semiconductor materials, with their superior physical properties, are gradually becoming the core material for power devices in high-voltage, high-power applications. Among them, silicon carbide (SiC) is widely used in many key fields such as high-voltage platforms for new energy vehicles, photovoltaic energy storage systems, rail transit traction equipment, and industrial frequency converters due to its outstanding advantages such as wide bandgap, high critical breakdown field strength, high thermal conductivity, and low switching losses. As the mainstream silicon carbide power switching device, the performance of silicon carbide MOS devices directly determines the conversion efficiency, operational stability, and overall cost of power electronic systems. To overcome the industry challenge of the mutual constraint between the withstand voltage and on-resistance of traditional power devices, superjunction structures are widely used in the design of silicon carbide MOS devices. By relying on the charge compensation effect to optimize the electric field distribution in the drift region, the on-resistance is reduced while ensuring high withstand voltage, making it the mainstream technical route for high-voltage silicon carbide power devices.

[0003] Most commercially available silicon carbide superjunction MOS devices employ uniformly doped pillar structures to build the superjunction units. Both N-type and P-type pillars are set to a single doping concentration. This structure can only achieve basic charge complementarity, with limited charge matching accuracy, making it difficult to balance the uniformity of the electric field in both the longitudinal and lateral directions. Under high-voltage operation, localized electric field concentration easily occurs in the superjunction region and around the trench gate. This not only limits further increases in breakdown voltage but also easily leads to excessive local stress, causing early breakdown failure. Even when some existing technologies attempt to optimize the superjunction structure by introducing gradient doping, they often only apply a gradual concentration change to one type of doped pillar, while the other type remains at a fixed doping concentration. This fails to achieve dynamic charge balance across the entire domain, and the core problem of uneven lateral electric field distribution remains unresolved. The trade-off between breakdown voltage and on-resistance remains difficult to further optimize. Summary of the Invention

[0004] This application aims to improve at least one technical problem in the background art.

[0005] The first aspect of this application provides a method for fabricating a silicon carbide MOS device, comprising the following steps: A silicon carbide substrate is obtained, with one side of the silicon carbide substrate as the front side and the side corresponding to the front side as the back side. An N-type drift layer is epitaxially grown on the front side. A superjunction region and a gate region are defined on the surface of the N-type drift layer. The superjunction region includes a plurality of alternately arranged N-type pillar injection regions and P-type pillar injection regions. In the N-type column implantation region, N-type columns are formed in the N-type drift layer by sequentially performing multiple N-type ion implantation treatments with increasing implantation energy and dosage. The P-type pillar implantation region consists of a low-doped region and a high-doped region. The low-doped region is located on the side away from the gate region, and the high-doped region is located on the side closer to the gate region. A low-doped P-type pillar is formed in the N-type drift layer by a first P-type ion implantation process in the low-doped region, and a high-doped P-type pillar is formed in the N-type drift layer by a second P-type ion implantation process in the high-doped region. The implantation dose of the first P-type ion implantation process is less than the implantation dose of the second P-type ion implantation process. The low-doped P-type pillar and the high-doped P-type pillar constitute a P-type pillar. Etching is performed in the gate region to form gate trenches; The bottom and sidewalls of the gate trench are subjected to a third P-type ion implantation process to form a P-type layer; A gate insulating layer is deposited on the surface of the P-type layer; A gate is formed on the gate insulating layer; A fourth P-type ion implantation treatment and a second N-type ion implantation treatment are sequentially performed on one side of the N-type drift layer to form a PN junction. A source electrode is deposited on the PN junction; A drain electrode is deposited on the back side to obtain the silicon carbide MOS device.

[0006] In some embodiments, the first N-type ion implantation treatment is performed four times, with the implantation energies of the four treatments being 40keV-60keV, 110keV-130keV, 180keV-220keV, and 280keV-320keV, respectively, and the implantation dose of the four treatments being 0.8E13cm. -2 -1.2E13cm -2 1.8E13cm -2 -2.2E13cm -2 2.8E13cm -2 -3.2E13cm -2 and 4.8E14cm -2 -5.2E14cm -2 .

[0007] In some embodiments, the implantation energy of the first P-type ion implantation treatment is 80 keV-120 keV, and the implantation dose of the first P-type ion implantation treatment is 5E13cm. -2 -1E14cm -2 The implantation energy for the second P-type ion implantation treatment is 180 keV-220 keV, and the implantation dose for the second P-type ion implantation treatment is 2E14cm. -2 -5E14cm -2 .

[0008] In some embodiments, the width of the low-doped region is 0.8 μm-1.2 μm, and the width of the high-doped region is 0.4 μm-0.6 μm.

[0009] In some embodiments, the thickness of the N-type drift layer is 20 μm-60 μm.

[0010] In some embodiments, the depth of the gate trench is 15μm-40μm.

[0011] In some embodiments, the implantation energy of the third P-type ion implantation treatment is 100 keV-200 keV, and the implantation dose of the third P-type ion implantation treatment is 1E14cm. -2 -3E14cm -2 .

[0012] In some embodiments, the thickness of the gate insulating layer is 0.3 μm to 0.5 μm.

[0013] In some embodiments, the implantation energy of the fourth P-type ion implantation treatment is 80 keV-100 keV, and the implantation dose of the fourth P-type ion implantation treatment is 1E14cm. -2 -2E14cm -2 The implantation energy for the second N-type ion implantation treatment is 50 keV-70 keV, and the implantation dose for the second N-type ion implantation treatment is 5E15cm. -2 -1E16cm -2 .

[0014] The second aspect of this application discloses a silicon carbide MOS device, which is prepared by the above-described preparation method.

[0015] The beneficial effects of this application are as follows: This application achieves balanced control of the entire charge domain within the device through a composite superjunction design combining vertical and horizontal gradient doping. This effectively improves the problem of local electric field concentration under high-voltage conditions, breaking through the technical bottleneck of the mutual constraint between the withstand voltage and conduction loss of traditional devices, and significantly improving the overall electrical performance of silicon carbide MOS devices. Furthermore, this application performs targeted structural optimization of the gate trench region, weakening the negative impact of trench defects and strong electric fields, and improving the stability of the gate insulation system. Overall, this application enhances the withstand capability of silicon carbide MOS devices under harsh conditions such as high temperature, high voltage, and sudden short circuits, extending the device's service life and meeting the high reliability requirements of automotive and industrial applications. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the structure of a silicon carbide MOS device according to one embodiment. Figure 2 This is a schematic diagram of an N-type column and a P-type column structure according to one embodiment. Detailed Implementation

[0017] The present application will be further described below with reference to specific embodiments. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the application. Furthermore, it should be understood that after reading the contents of this application, those skilled in the art can make various alterations or modifications to this application, and these equivalent forms also fall within the scope defined by the appended claims.

[0018] This application provides a method for fabricating a silicon carbide MOS device, comprising the following steps: A silicon carbide substrate is obtained, with one side of the silicon carbide substrate as the front side and the side corresponding to the front side as the back side. An N-type drift layer is epitaxially grown on the front side. A superjunction region and a gate region are defined on the surface of the N-type drift layer. The superjunction region includes a plurality of alternately arranged N-type pillar injection regions and P-type pillar injection regions. In the N-type column implantation region, N-type columns are formed in the N-type drift layer by sequentially performing multiple N-type ion implantation treatments with increasing implantation energy and dosage. The P-type pillar implantation region consists of a low-doped region and a high-doped region. The low-doped region is located on the side away from the gate region, and the high-doped region is located on the side closer to the gate region. A low-doped P-type pillar is formed in the N-type drift layer by a first P-type ion implantation process in the low-doped region, and a high-doped P-type pillar is formed in the N-type drift layer by a second P-type ion implantation process in the high-doped region. The implantation dose of the first P-type ion implantation process is less than the implantation dose of the second P-type ion implantation process. The low-doped P-type pillar and the high-doped P-type pillar constitute a P-type pillar. Etching is performed in the gate region to form gate trenches; The bottom and sidewalls of the gate trench are subjected to a third P-type ion implantation process to form a P-type layer; A gate insulating layer is deposited on the surface of the P-type layer; A gate is formed on the gate insulating layer; A fourth P-type ion implantation treatment and a second N-type ion implantation treatment are sequentially performed on one side of the N-type drift layer to form a PN junction. A source electrode is deposited on the PN junction; A drain electrode is deposited on the back side to obtain the silicon carbide MOS device (e.g., Figure 1 As shown in the attached figure: 100 - silicon carbide substrate; 200 - N-type drift layer; 310 - N-type pillar; 320 - P-type pillar; 321 - low-doped P-type pillar; 322 - high-doped P-type pillar; 400 - P-type layer; 500 - gate insulating layer; 600 - gate; 700 - PN junction; 800 - source; 900 - drain.

[0019] The silicon carbide MOS device fabrication method proposed in this application uses a composite superjunction scheme that combines gradient-doped N-type pillars with segmented-doped P-type pillars, and optimizes the internal charge distribution and electric field state of the device by combining a trench gate structure. This effectively improves the problems of uneven charge matching and local electric field concentration in traditional silicon carbide power devices, and enhances the device's voltage withstand capability and conduction performance.

[0020] This fabrication method first selects a silicon carbide substrate, defining its two surfaces as the front and back sides. The front side is used as the core processing surface, where an N-type drift layer is epitaxially grown. This layer is the core drift region responsible for high-voltage withstand and high-current conduction in the device, and also provides a complete semiconductor substrate for the subsequent fabrication of the superjunction structure, gate structure, and channel structure. After the epitaxial formation of the N-type drift layer is completed, a superjunction region and a gate region are defined on its surface. Within the superjunction region, multiple alternating N-type pillar injection regions and P-type pillar injection regions are planned. This staggered pillar layout is the basic form of the superjunction structure, and the electric field distribution under high-voltage operating conditions can be optimized from a structural level by relying on the charge complementarity effect between adjacent doped pillars. For the N-type pillar implantation region, this method uses multiple first N-type ion implantation processes for doping, and the energy and implantation dose of each ion implantation are set to increase progressively. Finally, an N-type pillar with longitudinal gradient doping characteristics is formed inside the N-type drift layer. The doping concentration forms a continuous transition distribution along the longitudinal direction of the device, which can smooth the longitudinal electric field in the drift region and avoid local electric field spikes under high voltage conditions.

[0021] The corresponding P-type pillar implantation region is further divided into a low-doped region and a high-doped region. Spatially, the low-doped region is located on the side furthest from the gate region, while the high-doped region is located on the side closest to the gate region. During fabrication, two sets of P-type ion implantation processes with different doses are used to process the two regions. The first P-type ion implantation forms a low-doped segment of the P-type pillar in the low-doped region, and the second P-type ion implantation forms a high-doped segment of the P-type pillar in the high-doped region. The dose of the first P-type ion implantation is controlled to be less than that of the second P-type ion implantation. The two pillars with different doping concentrations in the lateral direction combine to form a complete P-type pillar. This segmented doping structure, designed based on positional differences, can match the corresponding doping concentration to the electric field environment of different regions. Combined with the aforementioned gradient-doped N-type pillars, it achieves precise charge balance in the lateral and longitudinal directions of the superjunction region, avoiding the charge matching deviation problem caused by the overall single doping concentration in traditional superjunctions.

[0022] After setting up the superjunction region, the pre-defined gate region is etched to form a gate trench. The trench structure provides physical space for the gate system. For the etched gate trench, a third P-type ion implantation is performed on its bottom and sidewalls to form a P-type layer on the inner wall of the trench. This structure shields the electric field around the trench, reducing the electric field concentration that is prone to occur at the trench location, while also mitigating interface defects caused by the etching process and improving the operational stability of the gate region. Subsequently, a gate insulating layer is deposited on the surface of the P-type layer. This insulating layer provides electrical isolation between the gate and the underlying semiconductor substrate, eliminating the risk of leakage and ensuring stable gate control. The gate is then formed on the gate insulating layer. As the control electrode of the device, the gate allows for flexible control of the device's on / off state through dynamic changes in the applied voltage.

[0023] After the gate structure is fabricated, fourth P-type ion implantation and second N-type ion implantation are sequentially performed on the corresponding side of the N-type drift layer. These two ion implantation processes, acting in succession, form a PN junction inside the semiconductor substrate. This PN junction forms the conductive channel of the device, working in conjunction with the gate to regulate and transport charge carriers, and is the core functional structure for realizing the switching characteristics of the device. Finally, the main electrodes are fabricated. A source electrode is deposited on the surface of the formed PN junction, and a drain electrode is deposited on the back side of the silicon carbide substrate. The source and drain electrodes serve as the two main power electrodes of the device, respectively responsible for current introduction and extraction. After all structures and electrodes are fabricated, the silicon carbide MOS device designed in this application is obtained.

[0024] In this specific embodiment, the silicon carbide substrate is selected as a 4H-SiC N-type substrate with a crystal orientation of <0001> The deflection angle is 4°, the thickness is 10 μm (in other embodiments, the thickness can also be any value in the range of 3 μm to 10 μm), and the N doping concentration is 1E20cm. -3 The microtubule density does not exceed 0.5 cm³. -2 .

[0025] In this specific embodiment, the silicon carbide substrate is first ultrasonically cleaned in acetone solution for 15 minutes at a power of 100W and a frequency of 40kHz to remove organic contaminants; then ultrasonically cleaned with ethanol for 15 minutes, and rinsed three times with deionized water (5 minutes each time); finally, it is placed in a quartz tube furnace and annealed at 1200°C in a hydrogen atmosphere (flow rate 5L / min) for 30 minutes to thoroughly remove the natural oxide layer and lattice defects on the surface of the silicon carbide substrate.

[0026] In this specific embodiment, an N-type drift layer is epitaxially grown on the front side of a silicon carbide substrate using a hot-wall chemical vapor deposition (HWCVD) process. The epitaxial growth temperature is 1550°C (in some other embodiments, the epitaxial growth temperature can be any value within the range of 1530°C-1570°C), and the epitaxial growth pressure is 80 mbar (in some other embodiments, the epitaxial growth pressure can be any value within the range of 70 mbar-90 mbar). The silicon source is trichloromethylsilane (MTS), the carbon source is propane (C3H8), the silicon-to-carbon molar ratio is 2:1, the doping source is phosphine (PH3), and the doping concentration is 1E15cm⁻¹. -3 (In some other embodiments, the doping concentration can also be 1E15cm) -3 -1E17cm -3 (Any point value within the range); the growth rate is 1 μm / h (in some other embodiments, the growth rate can also be any point value within the range of 1 μm / h-2 μm / h).

[0027] In some embodiments, the thickness of the N-type drift layer is 20 μm-60 μm. It is understood that the thickness of the N-type drift layer can be any value within the range of 25 μm, 30 μm, 40 μm, 50 μm, and 20 μm-60 μm. In this specific embodiment, the thickness of the N-type drift layer is 40 μm.

[0028] In this specific embodiment, multiple N-type pillar implantation regions are first defined in the superjunction region on the N-type drift layer. Specifically, a 2μm thick SiO2 hard mask layer is deposited on the surface of the N-type drift layer (deposition temperature 400℃, RF power 300W). Multiple N-type pillar implantation regions are defined by electron beam lithography (resolution ≤50nm). Each N-type pillar implantation region is square with a width of 1.2μm (in some other embodiments, this width can be any point value in the range of 1.2μm-1.8μm), and the center-to-center distance between two adjacent N-type pillar implantation regions is 5μm. The SiO2 hard mask layer is etched away after the N-type pillars are fabricated.

[0029] In some embodiments, the first N-type ion implantation treatment is performed four times, with the implantation energies of the four treatments being 40keV-60keV, 110keV-130keV, 180keV-220keV, and 280keV-320keV, respectively, and the implantation dose of the four treatments being 0.8E13cm. -2 -1.2E13cm -2 1.8E13cm -2 -2.2E13cm -2 2.8E13cm-2 -3.2E13cm -2 and 4.8E14cm -2 -5.2E14cm -2 It is understood that the implantation energy and dosage mentioned above can be any value within their respective ranges. By setting the first N-type ion implantation to a process of four progressively increasing implantation energies, the dopant ions can be precisely applied to different vertical regions within the N-type drift layer from shallow to deep, achieving layered doping. Compared to gradient implantation schemes with fewer implantation steps, four-step subdivision implantation allows for a continuous and linear gradient change in the doping concentration of the N-type pillar, completely avoiding the abrupt changes in doping concentration caused by a single or a few high-energy implantations. This results in a smoother and more uniform vertical electric field distribution in the N-type pillar, effectively preventing local electric field spikes at the doping interface under high-voltage conditions, significantly reducing the risk of premature device breakdown, and ensuring long-term stability of the breakdown voltage.

[0030] In this specific embodiment, the first N-type ion implantation treatment is performed four times. The first N-type ion implantation treatment uses phosphorus. The implantation energies for the four treatments are 50 keV, 120 keV, 200 keV, and 300 keV, respectively. The implantation doses for the four treatments are 1E13cm⁻¹, respectively. -2 2E13cm -2 3E13cm -2 and 5E14cm -2 .

[0031] In this specific embodiment, after the N-type pillar is fabricated, a low-doped region of the P-type pillar implantation region is defined in the same manner as the N-type pillar implantation region. After the first P-type ion implantation process is completed to form a low-doped P-type pillar, a high-doped region of the P-type pillar implantation region is defined in the same manner. A second P-type ion implantation process is then performed to form a high-doped P-type pillar. (Reference) Figure 2 The P-type column injection region is square with a width A1 of 1.2 μm (in some other embodiments, the width A1 can be any point value in the range of 1.2 μm-1.8 μm, but it is the same as the width of the N-type column injection region).

[0032] refer to Figure 2In some embodiments, the width B1 of the low-doped region is 0.8 μm-1.2 μm, and the width B2 of the high-doped region is 0.4 μm-0.6 μm. It is understood that the widths of both the low-doped and high-doped regions can be any value within these ranges. The P-type pillar implantation region is square, consisting of a low-doped region and a high-doped region. The lengths of the low-doped and high-doped regions are the same as the width C1 of the P-type pillar implantation region. The sum of the widths of the low-doped and high-doped regions is the width of the P-type pillar implantation region, i.e., B1 + B2 = C1.

[0033] In this specific embodiment, the width of the low-doped region is 0.8 μm, and the width of the high-doped region is 0.4 μm.

[0034] In some embodiments, the implantation energy of the first P-type ion implantation treatment is 80 keV-120 keV, and the implantation dose of the first P-type ion implantation treatment is 5E13cm. -2 -1E14cm -2 The implantation energy for the second P-type ion implantation treatment is 180 keV-220 keV, and the implantation dose for the second P-type ion implantation treatment is 2E14cm. -2 -5E14cm -2 It is understood that the implantation energy and implantation dose mentioned above can be any value within their respective ranges. The two sets of differentiated implantation parameters work together to create a reasonable doping gradient along the lateral direction of the P-type pillar. Combined with the longitudinal gradient doping design of the N-type pillar, this achieves charge density matching in both the lateral and longitudinal dimensions of the entire composite superjunction structure, maximizing the charge compensation advantage of the superjunction structure. In this specific embodiment, the first P-type ion implantation treatment uses aluminum, the implantation energy of the first P-type ion implantation treatment is 100 keV, and the implantation dose of the first P-type ion implantation treatment is 5E13cm⁻¹. -2 The second P-type ion implantation treatment uses aluminum, has an implantation energy of 200 keV, and an implantation dose of 2E14 cm⁻¹. -2 .

[0035] In some embodiments, the depth of the gate trench is 15 μm-40 μm. It is understood that the depth of the gate trench can be any value within the range of 18 μm, 20 μm, 25 μm, 30 μm, 35 μm, and 15 μm-40 μm.

[0036] In this specific embodiment, the depth of the gate trench is 20 μm.

[0037] In this specific embodiment, the gate trench is etched using inductively coupled plasma etching (ICP-C), with the etching performed sequentially using first and second etching parameters. The first etching parameters are: a mixture of SF6 and O2 (SF6 to O2 volume ratio 5:1) as the etching gas, a radio frequency power of 600W, a bias voltage of 100V, an etching rate of 1μm / min, and an etching depth of 5μm. The second etching parameters are: a mixture of SF6 and Ar (SF6 to Ar volume ratio 3:2) as the etching gas, a radio frequency power of 800W, a bias voltage of 150V, an etching rate of 1.5μm / min, and an etching depth of 15μm.

[0038] In this specific embodiment, after the gate trench etching is completed, oxygen plasma treatment is used with a power of 200W for 5 minutes to remove polymer residues on the sidewalls of the gate trench; then, it is soaked in diluted hydrofluoric acid (volume ratio HF:H2O=1:10) for 30 seconds to clean the etching residues.

[0039] In some embodiments, the implantation energy of the third P-type ion implantation treatment is 100 keV-200 keV, and the implantation dose of the third P-type ion implantation treatment is 1E14cm. -2 -3E14cm -2 It is understood that the injected energy and injected dose mentioned above can be any point value within their range.

[0040] In this specific embodiment, the third P-type ion implantation treatment uses aluminum, the implantation energy is 150 keV, and the implantation dose is 1E14 cm⁻¹. -2 After implantation, the cells are annealed in a tube furnace at 1650°C under an argon atmosphere (argon flow rate of 8 L / min) for 30 min to activate the implanted third P-type ions and simultaneously repair the etching damage of the gate trench.

[0041] In this specific embodiment, the gate insulating layer is made of silicon nitride and is deposited using plasma-enhanced chemical vapor deposition (PECVD) at a deposition temperature of 500°C, an RF power of 250W, and a deposition rate of 0.1 μm / min. Before the rated deposition of the gate insulating layer, an argon plasma pretreatment is performed at a power of 100W for 3 minutes to enhance the adhesion between the gate insulating layer and the P-type layer.

[0042] In some embodiments, the thickness of the gate insulating layer is 0.3 μm-0.5 μm. It is understood that the thickness of the gate insulating layer can be any value within the range of 0.35 μm, 0.4 μm, 0.45 μm, and 0.3 μm-0.5 μm. In this specific embodiment, the thickness of the gate insulating layer is 0.3 μm.

[0043] In some embodiments, the implantation energy of the fourth P-type ion implantation treatment is 80 keV-100 keV, and the implantation dose of the fourth P-type ion implantation treatment is 1E14cm. -2 -2E14cm -2 The implantation energy for the second N-type ion implantation treatment is 50 keV-70 keV, and the implantation dose for the second N-type ion implantation treatment is 5E15cm. -2 -1E16cm -2 It is understood that the injected energy and injected dose mentioned above can be any point value within their range.

[0044] In this specific embodiment, the fourth P-type ion implantation treatment uses aluminum, the implantation energy is 80 keV, and the implantation dose is 1E14 cm⁻¹. -2 The second N-type ion implantation treatment uses phosphorus, the implantation energy is 60 keV, and the implantation dose is 5E15cm⁻¹. -2 .

[0045] In this specific embodiment, the gate forming, source and drain deposition (metallization preparation) all adopt conventional processes in the field of silicon carbide MOS device manufacturing. The relevant process flow, operating parameters and implementation methods will not be described in detail.

[0046] In the description of this specification, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature.

[0047] The above description is only a preferred embodiment of this application. It should be noted that those skilled in the art can make several improvements and additions without departing from the method of this application, and these improvements and additions should also be considered within the scope of protection of this application.

Claims

1. A method for fabricating a silicon carbide MOS device, characterized in that, Includes the following steps: A silicon carbide substrate is obtained, with one side of the silicon carbide substrate as the front side and the side corresponding to the front side as the back side. An N-type drift layer is epitaxially grown on the front side. A superjunction region and a gate region are defined on the surface of the N-type drift layer. The superjunction region includes a plurality of alternately arranged N-type pillar injection regions and P-type pillar injection regions. In the N-type column implantation region, N-type columns are formed in the N-type drift layer by sequentially performing multiple N-type ion implantation treatments with increasing implantation energy and dosage. The P-type pillar implantation region consists of a low-doped region and a high-doped region. The low-doped region is located on the side away from the gate region, and the high-doped region is located on the side closer to the gate region. A low-doped P-type pillar is formed in the N-type drift layer by a first P-type ion implantation process in the low-doped region, and a high-doped P-type pillar is formed in the N-type drift layer by a second P-type ion implantation process in the high-doped region. The implantation dose of the first P-type ion implantation process is less than the implantation dose of the second P-type ion implantation process. The low-doped P-type pillar and the high-doped P-type pillar constitute a P-type pillar. Etching is performed in the gate region to form gate trenches; The bottom and sidewalls of the gate trench are subjected to a third P-type ion implantation process to form a P-type layer; A gate insulating layer is deposited on the surface of the P-type layer; A gate is formed on the gate insulating layer; A fourth P-type ion implantation treatment and a second N-type ion implantation treatment are sequentially performed on one side of the N-type drift layer to form a PN junction. A source electrode is deposited on the PN junction; A drain electrode is deposited on the back side to obtain the silicon carbide MOS device.

2. The method for fabricating a silicon carbide MOS device according to claim 1, characterized in that, The first N-type ion implantation treatment was performed four times, with implantation energies of 40keV-60keV, 110keV-130keV, 180keV-220keV, and 280keV-320keV respectively. The implantation dose for the four treatments was 0.8E13cm. -2 -1.2E13cm -2 1.8E13cm -2 -2.2E13cm -2 2.8E13cm -2 -3.2E13cm -2 and 4.8E14cm -2 -5.2E14cm -2 .

3. The method for fabricating a silicon carbide MOS device according to claim 1, characterized in that, The implantation energy of the first P-type ion implantation treatment is 80keV-120keV, and the implantation dose of the first P-type ion implantation treatment is 5E13cm. -2 -1E14cm -2 The implantation energy for the second P-type ion implantation treatment is 180 keV-220 keV, and the implantation dose for the second P-type ion implantation treatment is 2E14cm. -2 -5E14cm -2 .

4. The method for fabricating a silicon carbide MOS device according to claim 1, characterized in that, The width of the low-doped region is 0.8μm-1.2μm, and the width of the high-doped region is 0.4μm-0.6μm.

5. The method for fabricating a silicon carbide MOS device according to claim 1, characterized in that, The thickness of the N-type drift layer is 20μm-60μm.

6. The method for fabricating a silicon carbide MOS device according to claim 1, characterized in that, The depth of the gate trench is 15μm-40μm.

7. The method for fabricating a silicon carbide MOS device according to claim 1, characterized in that, The implantation energy of the third P-type ion implantation treatment is 100 keV-200 keV, and the implantation dose of the third P-type ion implantation treatment is 1E14cm. -2 -3E14cm -2 .

8. The method for fabricating a silicon carbide MOS device according to claim 1, characterized in that, The thickness of the gate insulating layer is 0.3μm-0.5μm.

9. The method for fabricating a silicon carbide MOS device according to claim 1, characterized in that, The implantation energy for the fourth P-type ion implantation treatment is 80 keV-100 keV, and the implantation dose for the fourth P-type ion implantation treatment is 1E14cm. -2 -2E14cm -2 The implantation energy for the second N-type ion implantation treatment is 50 keV-70 keV, and the implantation dose for the second N-type ion implantation treatment is 5E15cm. -2 -1E16cm -2 .

10. A silicon carbide MOS device, characterized in that, The silicon carbide MOS device is prepared by the method according to any one of claims 1-9.