A high performance SGTMOSFET power device

By optimizing the double-layer epitaxial structure, SiO2/Al2O3 composite insulating layer, and doping distribution, the problems of poor trade-off between breakdown voltage and on-resistance, low gate insulation reliability, and poor high-temperature stability of medium and low voltage SGTMOSFET devices have been solved, achieving a high-performance improvement in the device, which is suitable for power management of new energy vehicles and fast charging of consumer electronics.

CN224959958UActive Publication Date: 2026-11-10SHANGHAI YIDU POWER SEMICON TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202522512934.X
Authority / Receiving Office
CN · China
Patent Type
Utility models(China)
Current Assignee / Owner
Filing Date
2025-11-26
Publication Date
2026-11-10
Estimated Expiration
2035-11-26

AI Technical Summary

Technical Problem

Existing low- and medium-voltage SGTMOSFET devices have poor trade-offs between breakdown voltage and on-resistance, low gate insulation reliability, poor high-temperature stability, and uneven electric field distribution, which limits their performance improvement in the field of low- and medium-voltage power electronics.

Method used

By employing a double-layer epitaxial structure, a SiO2/Al2O3 composite insulating layer, and optimized doping distribution, combined with a shielded gate design, the electric field distribution and insulation performance of the device are optimized. By rationally designing the epitaxial layer thickness ratio and resistivity ratio, the lateral electric field distribution is optimized. Multilayer metal ohmic contact electrodes are used to improve the high-temperature stability and reliability of the device.

Benefits of technology

It significantly improves the trade-off between breakdown voltage and on-resistance, reduces Miller capacitance and gate leakage current, enhances high-temperature stability and switching losses, meets the requirements of high-frequency applications, and is suitable for power management of new energy vehicles and fast charging of consumer electronics.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN224959958U_ABST
    Figure CN224959958U_ABST
Patent Text Reader

Abstract

This utility model discloses a high-performance SGT MOSFET power device, comprising an N-type heavily doped silicon substrate, an N-type doped silicon buffer layer, and a double-layer epitaxial structure arranged sequentially from bottom to top. The epitaxial structure has a trench that extends through its thickness. The trench contains a gate assembly containing a gate polysilicon, a shielding gate polysilicon, and an isolation dielectric layer. A SiO2 / Al2O3 composite insulating layer is provided between the gate assembly and the trench sidewall. A P-type source region, an N-type drain region, and a multilayer metal ohmic contact electrode are respectively provided on both sides of the trench on the epitaxial structure. A Si3N4 passivation layer is provided on the surface of the epitaxial structure. The resistivity and thickness ratio of the double-layer epitaxial layer have been optimized. The composite insulating layer combines the advantages of thermally oxidized SiO2 and ALD-deposited Al2O3, which improves the breakdown voltage of the device by 8%-12%, reduces the on-resistance by 10%-15%, and reduces the Miller capacitance by 12-15 times. At a high temperature of 175℃, the on-resistance multiplication factor is only 1.8-1.9. It has excellent switching characteristics, high temperature stability and reliability, and is suitable for medium and low voltage power conversion scenarios such as power management of new energy vehicles.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This utility model relates to the field of power semiconductor device technology, and in particular to a high-performance SGT MOSFET power device, which is especially suitable for medium and low voltage power conversion scenarios with a rated voltage range of 50-100V, such as power management of new energy vehicles, fast charging of consumer electronics, and industrial frequency converters. Background Technology

[0002] Metal-oxide-semiconductor field-effect transistors (MOSFETs), as important power semiconductor devices, have been widely used in power electronics due to their high-frequency switching characteristics and good control performance. Traditional planar DMOSFETs, due to their relatively high on-resistance, are gradually being replaced by trench MOSFETs in low- and medium-voltage applications. Trench MOSFETs, by etching trenches in the semiconductor substrate and fabricating the gate structure within the trenches, effectively shorten the current path, reduce on-resistance, and simultaneously improve device integration.

[0003] However, traditional Trench MOSFETs have significant performance drawbacks, the biggest being their excessively large Miller capacitance (CGD), which leads to substantial switching losses during high-frequency switching and limits their use in high-frequency applications. To address this issue, Shielded Gate Trench MOSFETs (SGTMOSFETs) were developed. SGTMOSFETs innovate upon the Trench MOSFET design by separating the polysilicon within a deep trench into two dielectric-isolated parts: a gate polysilicon connected to the gate and a source polysilicon connected to the source (shielded gate). This structural design brings several performance improvements: First, the shielding gate can effectively shield the drain potential, significantly reducing Miller capacitance (CGD) and gate charge (Qg), thereby reducing switching losses, with Miller capacitance reduced by more than 10 times; Second, the charge coupling effect between the shielding gate and the drift region can induce a depletion region in the drift region, making the electric field distribution in the drift region quasi-rectangular, similar to a superjunction structure, which helps to increase the doping concentration of the epitaxial layer while maintaining the withstand voltage, further reducing the on-resistance. Compared with traditional Trench MOSFETs, the on-resistance under the same area can be reduced by more than 50%; In addition, SGTMOSFETs also have higher breakdown voltage (BVDSS), stronger avalanche capability (EAS), and better dv / dt resistance, and are less prone to false turn-on under high voltage change rate environments.

[0004] Although SGTMOSFETs represent a significant improvement over traditional structures, their performance still has room for optimization in low-to-medium voltage (50-100V) applications. In existing technologies, the trade-off between breakdown voltage and on-resistance in low-to-medium voltage SGTMOSFETs has not been fully optimized, limiting the overall performance improvement. Research indicates that a double-layer epitaxial design can further optimize this trade-off by improving the lateral electric field distribution near the trench. Sentaurus TCAD simulations verify that, compared to traditional 60V SGT devices, devices using a double-layer epitaxial structure exhibit a 6.9% increase in breakdown voltage, a 9.3% reduction in on-resistance (Ron,sp), and a 25.8% improvement in Baliga figure of merit (BFOM). However, existing double-layer epitaxial SGTMOSFETs often use a single SiO2 insulating layer for their gate insulation, and their dielectric properties and reliability still need improvement. Under high temperature and high voltage stress, gate leakage is prone to occur, affecting the device's lifespan and operational stability.

[0005] Meanwhile, existing SGTMOSFETs mostly employ uniform doping in their drift regions. This doping method cannot be specifically optimized based on the characteristics of the electric field distribution, resulting in localized concentration of the electric field in the drift region, which limits further improvements in breakdown voltage. Furthermore, the high-temperature performance of the device also needs further improvement. Traditional SGTMOSFETs typically exhibit an on-resistance multiplication factor above 2.0 at 175°C, making it difficult to guarantee stable output characteristics under harsh high-temperature conditions.

[0006] To address the shortcomings of the existing technologies, this application proposes a high-performance SGT MOSFET power device. By optimizing the epitaxial structure, gate insulation structure, and doping distribution, the breakdown voltage of the device is further improved, the on-resistance is reduced, and the high-temperature stability and reliability are improved, so as to meet the demand for high-performance devices in the field of medium and low voltage power electronics. Utility Model Content

[0007] The purpose of this invention is to overcome the technical problems existing in the current low- and medium-voltage SGT MOSFET power devices, such as poor trade-off between breakdown voltage and on-resistance, low gate insulation reliability, poor high-temperature stability, and uneven electric field distribution. The invention provides a high-performance SGT MOSFET power device that can significantly reduce on-resistance while ensuring high breakdown voltage, and has better switching characteristics, higher high-temperature stability, and longer service life.

[0008] The above-mentioned utility model objective is achieved through the following technical solution: A high-performance SGT MOSFET power device includes a substrate, a buffer layer, and an epitaxial structure arranged sequentially from bottom to top. A trench extending through its thickness direction is provided in the epitaxial structure, and a gate assembly is provided in the trench. A source region and a drain region are respectively provided on both sides of the trench on the epitaxial structure, and ohmic contact electrodes are provided on both the source region and the drain region. The epitaxial structure is a double-layer epitaxial structure, including a lower first epitaxial layer and an upper second epitaxial layer, wherein the resistivity of the first epitaxial layer is less than the resistivity of the second epitaxial layer; The gate assembly includes a gate polysilicon and a shielding gate polysilicon. The gate polysilicon is located in the upper part of the trench, and the shielding gate polysilicon is located in the lower part of the trench. An isolation dielectric layer is disposed between the gate polysilicon and the shielding gate polysilicon, and a composite insulating layer is disposed between the gate polysilicon, the shielding gate polysilicon and the trench sidewall.

[0009] As a further technical solution of this utility model: the substrate is an N-type heavily doped silicon substrate with a doping concentration of 1×10¹ 9 -5×10² 0 The silicon substrate has a thickness of 300-500 μm and a resistivity of ≤0.005 Ω·cm. Using a heavily doped N-type silicon substrate provides a good current path for the device, reduces additional losses due to substrate resistance, and ensures good mechanical strength, facilitating subsequent packaging processes.

[0010] As a further technical solution of this utility model: the buffer layer is an N-type doped silicon layer with a doping concentration of 5×10¹. 6 -2×10¹ 7 The buffer layer has a thickness of 5-10 μm and is filled with phosphorus or arsenic ions. The buffer layer is positioned between the substrate and the epitaxial structure, and its main function is to alleviate lattice mismatch and stress concentration between the substrate and the epitaxial layer. Simultaneously, it absorbs minority carriers conducted from the epitaxial layer, reducing leakage current and improving device reliability. Phosphorus or arsenic ions are chosen as dopant ions because these two ions have moderate diffusion coefficients in silicon, facilitating precise control of the doping concentration distribution.

[0011] As a further technical solution of this utility model: the total thickness of the double-layer epitaxial structure is 20-30 μm, wherein the thickness of the first epitaxial layer is 12-18 μm, the thickness of the second epitaxial layer is 8-12 μm, the thickness ratio of the first epitaxial layer to the second epitaxial layer is 1.2-1.8:1, and the resistivity ratio is 0.3-0.6:1. The design of the double-layer epitaxial structure is based on electric field optimization theory. The lower low-resistivity first epitaxial layer can provide a good current conduction path, while the upper high-resistivity second epitaxial layer can withstand higher voltage stress. By optimizing the thickness ratio and resistivity ratio of the two epitaxial layers, the lateral electric field distribution near the trench can be made more uniform, effectively suppressing the electric field concentration phenomenon, thereby increasing the breakdown voltage while reducing the on-resistance. Simulation verification shows that when the thickness ratio is 1.5:1 and the resistivity ratio is 0.4:1, the Baliga figure of merit of the device reaches its maximum value.

[0012] As a further technical solution of this utility model: the composite insulating layer consists of a bottom SiO2 layer and an upper Al2O3 layer. The thickness of the SiO2 layer is 50-100 nm, and the thickness of the Al2O3 layer is 20-50 nm. The SiO2 layer is prepared by thermal oxidation, and the Al2O3 layer is prepared by atomic layer deposition (ALD). Traditional single SiO2 insulating layers have the problems of low dielectric constant and large leakage current at high temperatures, while Al2O3 has a higher dielectric constant (approximately 8-10 times that of SiO2) and better thermal stability. This utility model adopts a SiO2 / Al2O3 composite insulating layer structure. The bottom SiO2 layer can form a good interface with the silicon substrate, reducing the interface state density, while the upper Al2O3 layer can improve the overall dielectric properties and breakdown field strength of the insulating layer, effectively reducing the gate leakage current. The SiO2 layer prepared by thermal oxidation has high purity and good interface quality, while the atomic layer deposition process can precisely control the thickness and uniformity of the Al2O3 layer, ensuring the consistency of the insulating layer.

[0013] As a further technical solution of this utility model: the trench has a depth of 15-25μm, a width of 0.8-1.2μm, a spacing of 2-3μm between adjacent trenches, a sidewall verticality ≥89.5°, and a bottom fillet radius of 0.1-0.2μm. The dimensional parameters of the trench are crucial to device performance. Deeper trenches can increase the control range of the gate and improve the switching speed of the device; appropriate width and spacing can reduce the parasitic capacitance of the device while ensuring current capacity. The high verticality of the trench sidewalls ensures the uniformity of the gate structure, while the rounded bottom design effectively alleviates the electric field concentration at the bottom of the trench, improving the breakdown voltage and reliability of the device.

[0014] As a further technical solution of this utility model: the source region is a heavily doped P-type region with a doping concentration of 1×10². 0 The dopant region has a density of -5×10²¹cm⁻³ and a depth of 0.5-1μm, and is doped with boron ions; the drain region is an N-type heavily doped region with a doping concentration of 5×10¹. 9 The phosphorus ion has a doping density of -2×10²¹cm⁻³ and a depth of 1-1.5μm. The source and drain regions are heavily doped to form good ohmic contacts and reduce contact resistance. The P-type source region forms a PN junction with the N-type epitaxial layer, constituting the main structure of the device. Boron ions, as P-type dopants, have high solubility and diffusion coefficients, facilitating the formation of shallow junctions; phosphorus ions, as N-type dopants, enable high-concentration doping, ensuring low resistance characteristics in the drain region.

[0015] As a further technical solution of this utility model: the isolation dielectric layer is a SiO2 dielectric layer with a thickness of 30-60nm, and the gate polysilicon is N-type doped polysilicon with a doping concentration of 1×10¹. 9 -5×10² 0 The shielding gate polysilicon is cm⁻³ and electrically connected to the source region. An isolation dielectric layer is used to achieve electrical isolation between the gate polysilicon and the shielding gate polysilicon, preventing leakage between them. The N-type doped gate polysilicon has good conductivity, enabling rapid transmission of gate control signals. After the shielding gate polysilicon is electrically connected to the source region, it effectively shields the drain potential, further reducing Miller capacitance and lowering switching losses.

[0016] As a further technical solution of this utility model: the ohmic contact electrode is a multilayer metal structure, consisting of a titanium layer, a nickel layer, and a silver layer from bottom to top. The titanium layer has a thickness of 20-50 nm, the nickel layer has a thickness of 50-100 nm, and the silver layer has a thickness of 100-200 nm. The titanium layer can form a good ohmic contact with the silicon substrate, reducing contact resistance; the nickel layer, as a transition layer, can improve the adhesion and reliability of the electrode; the silver layer has excellent electrical and thermal conductivity, effectively dissipating the heat generated during device operation and reducing the series resistance of the electrode. This multilayer metal structure balances contact performance, reliability, and heat dissipation performance, ensuring the stability of the device under high current operation.

[0017] As a further technical solution of this utility model: a passivation layer is further provided on the surface of the epitaxial structure. The passivation layer is a Si3N4 layer with a thickness of 100-200 nm, and an opening is provided on the passivation layer corresponding to the position of the ohmic contact electrode. The passivation layer can protect the device surface from external environmental pollution and corrosion, reduce surface leakage current, and improve the reliability and service life of the device. Si3N4 has good chemical stability and insulation properties, making it an ideal passivation layer material. The openings on the passivation layer are used to realize the electrical connection between the ohmic contact electrode and the source and drain regions.

[0018] In summary, compared with the prior art, the present invention has at least one of the following beneficial technical effects: Significant optimization of breakdown voltage and on-resistance: This invention employs a double-layer epitaxial structure. By rationally designing the thickness ratio and resistivity ratio of the two epitaxial layers, the lateral electric field distribution near the trench is made more uniform, effectively suppressing electric field concentration. Simulation tests show that compared with traditional single-layer epitaxial SGT devices, the breakdown voltage of the device in this application is increased by 8%-12%, the on-resistance is reduced by 10%-15%, and the Baliga figure of merit (BFOM=BVDSS² / Ron,sp) is increased by 25%-30%, significantly optimizing the trade-off between breakdown voltage and on-resistance.

[0019] Significantly reduced switching losses: Thanks to the synergistic effect of the shielded gate structure and composite insulating layer, the device’s Miller capacitance (CGD) is reduced by 12-15 times compared to traditional Trench MOSFETs, the gate charge (Qg) is reduced by 40%-50%, and the switching losses are reduced by 35%-45%, which can meet the needs of high-frequency switching applications and improve power conversion efficiency.

[0020] Improved gate insulation reliability: The use of the SiO2 / Al2O3 composite insulation layer combines the good interface characteristics of SiO2 with the excellent dielectric properties of Al2O3, which reduces the gate leakage current by more than an order of magnitude. Even at a high temperature of 120℃, the gate leakage current can still be kept below 1×10⁻¹¹A. ​​At the same time, the breakdown field strength of the insulation layer is increased to 8-10MV / cm, which significantly improves the reliability and service life of the gate structure.

[0021] Excellent high-temperature stability: By optimizing the doping concentration of the source and drain regions and the doping distribution of the epitaxial layer, the high-temperature performance of the device is significantly improved. At a high temperature of 175℃, the on-resistance multiplication factor is reduced to 1.8-1.9, which is 10%-15% lower than that of existing SGT devices, ensuring that the device can still operate stably under harsh high-temperature conditions.

[0022] Strong avalanche capability: The synergistic design of the deep trench structure and buffer layer can effectively absorb the energy generated during the avalanche process. The avalanche energy (EAS) of the device is increased by 20%-30% compared with the traditional structure. It has stronger impact resistance when surge voltage occurs, thus improving the reliability of the system.

[0023] Good process compatibility: The device structure of this utility model adopts mature semiconductor manufacturing processes, such as thermal oxidation, atomic layer deposition, ion implantation, photolithography, and etching. No special process equipment is required, which facilitates mass production and reduces manufacturing costs. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the overall structure of this utility model.

[0025] Figure 2 This is a cross-sectional view of the present invention.

[0026] Reference numerals: 1. Substrate; 2. Buffer layer; 3. Epitaxial structure; 31. First epitaxial layer; 32. Second epitaxial layer; 4. Trench; 5. Gate assembly; 51. Gate polysilicon; 52. Shielding gate polysilicon; 53. Isolation dielectric layer; 54. Composite insulating layer; 541. SiO2 layer; 542. Al2O3 layer; 6. Source region; 7. Drain region. Detailed Implementation

[0027] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0028] In the description of this application, it should be noted that the terms "upper," "lower," "inner," "outer," "top / bottom," etc., indicating the orientation or positional relationship are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0029] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installed," "equipped with," "sleeved / connected," "connected," etc., should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be a connection within two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0030] Example 1: Reference Figure 1 and Figure 2 This embodiment provides a high-performance SGT MOSFET power device, the structure of which is as follows: Figure 1 As shown, the structure includes a substrate 1, a buffer layer 2, and an epitaxial structure 3 arranged sequentially from bottom to top. A trench 4 extending through the thickness direction is provided in the epitaxial structure 3. A gate assembly 5 is provided in the trench 4. A source region 6 and a drain region 7 are respectively provided on both sides of the trench 4 on the epitaxial structure 3. Ohmic contact electrodes are provided on both the source region 6 and the drain region 7. A passivation layer is also provided on the surface of the epitaxial structure 3.

[0031] The specific parameters for each part are as follows: Substrate 1: N-type heavily doped silicon substrate with a doping concentration of 3 × 10¹ 9 The crystal structure, with a thickness of 400 μm and a resistivity of 0.003 Ω·cm, was prepared using the Czochralski method and has the following crystal orientation. <100> .

[0032] Buffer layer 2: N-type doped silicon layer, with phosphorus ions as the dopant, and a doping concentration of 1×10¹. 7 The material has a thickness of 8 μm and is grown on the surface of substrate 1 by vapor phase epitaxy at a growth temperature of 1100℃ and a growth rate of 1 μm / min.

[0033] Epitaxial structure 3: A double-layer epitaxial structure with a total thickness of 25 μm, wherein the first epitaxial layer 31 (lower layer) has a thickness of 15 μm, a resistivity of 0.5 Ω·cm, and a doping concentration of 8 × 10¹. 5 cm⁻³; the second epitaxial layer 32 (upper layer) has a thickness of 10 μm, a resistivity of 1.2 Ω·cm, and a doping concentration of 3 × 10¹. 5 cm⁻³. The first epitaxial layer 31 and the second epitaxial layer 32 are both prepared by vapor phase epitaxy, with growth temperatures of 1050℃ and 1000℃, respectively. During the growth process, the resistivity and thickness of the epitaxial layer are precisely controlled by controlling the flow rates of the silicon source and the dopant source.

[0034] Trench 4: 20 μm deep, 1 μm wide, with a spacing of 2.5 μm between adjacent trenches. It was prepared using a dry etching process with a mixture of SF6 and O2 as the etching gas at a rate of 0.5 μm / min. After etching, the bottom of the trench was rounded using a wet etching process to form a rounded corner with a radius of 0.15 μm. The verticality of the trench sidewalls was 89.8°.

[0035] Gate component 5 includes a gate polysilicon 51, a shielding gate polysilicon 52, an isolation dielectric layer 53, and a composite insulating layer 54. The composite insulating layer 54 consists of a bottom SiO2 layer 541 and an upper Al2O3 layer 542. The SiO2 layer 541 is 80 nm thick and is prepared by thermal oxidation at 950 °C for 60 min. The Al2O3 layer 542 is 30 nm thick and is prepared by atomic layer deposition at 300 °C using trimethylaluminum and water as precursors. The isolation dielectric layer 53 is a 45 nm thick SiO2 dielectric layer prepared by plasma-enhanced chemical vapor deposition (PECVD). The gate polysilicon 51 is N-type doped polysilicon with a doping concentration of 3 × 10¹. 9 The polysilicon shielding gate is 5μm thick and has a diameter of cm⁻³. The shielding gate polysilicon 52 is N-type doped polysilicon with a doping concentration of 2×10¹. 9 The membrane is 1 cm⁻³ thick and 10 μm in diameter, and is electrically connected to source region 6 via ion implantation.

[0036] Source region 6: P-type heavily doped region, doped with boron ions, with a doping concentration of 3 × 10². 0 The atom diameter was cm⁻³, and the depth was 0.8 μm. It was prepared using ion implantation at an implantation energy of 50 keV and an implantation dose of 5 × 10¹⁸. 5 cm⁻², after implantation, is annealed at 850℃ for 30 min to activate dopant ions and repair lattice damage.

[0037] Drain region 7: N-type heavily doped region, doped with phosphorus ions, with a doping concentration of 1×10². 0 The atom diameter was cm⁻³, and the depth was 1.2 μm. It was prepared using ion implantation at an implantation energy of 80 keV and an implantation dose of 8 × 10¹. 5 cm⁻², annealing conditions are 900℃ for 40 min.

[0038] Ohmic contact electrode: multilayer metal structure, consisting of a titanium layer (30 nm thick), a nickel layer (80 nm thick), and a silver layer (150 nm thick) from bottom to top. The titanium and nickel layers were prepared by electron beam evaporation, and the silver layer was prepared by sputtering. After preparation, alloying treatment was carried out at 450°C for 20 minutes to form a good ohmic contact.

[0039] Passivation layer: Si3N4 layer with a thickness of 150nm, prepared by PECVD process at a deposition temperature of 350℃. After deposition, openings are made at the corresponding source region 6 and drain region 7 using photolithography and etching processes. The opening size matches the size of the ohmic contact electrode.

[0040] Preparation process steps: The fabrication process steps of the high-performance SGTMOSFET power device in this embodiment are as follows: 1. Substrate cleaning: Select N-type heavily doped silicon substrate 1, and perform ultrasonic cleaning with acetone, ethanol and deionized water in sequence to remove organic contaminants and impurities on the surface. Then, treat with BOE (buffered oxide etchant) for 5 minutes to remove the natural oxide layer on the surface. Finally, rinse with deionized water and dry.

[0041] 2. Buffer layer growth: The cleaned substrate 1 was placed in a vapor phase epitaxy furnace, and a silicon source (silicon tetrachloride) and a dopant source (phosphine) were introduced. An N-type buffer layer 2 was grown at 1100℃ for 8 minutes, with a thickness of 8 μm and a doping concentration of 1 × 10¹. 7 cm⁻³.

[0042] 3. Growth of double-layer epitaxial layers: After the buffer layer is grown, the temperature of the epitaxial furnace is adjusted to 1050℃, and the flow rates of the silicon source and dopant source are changed to grow the first epitaxial layer 31. The growth time is 15 min, the thickness is 15 μm, and the resistivity is 0.5 Ω·cm. Then the temperature is lowered to 1000℃, and the second epitaxial layer 32 is grown for 10 min, the thickness is 10 μm, and the resistivity is 1.2 Ω·cm.

[0043] 4. Trench Etching: Photoresist is coated on the surface of the second epitaxial layer 32, and the trench pattern is defined by photolithography. Then, the trench 4 is etched by dry etching process with an etching depth of 20μm. After etching, the photoresist is removed, and wet etching is performed by a mixed solution of HF and NH4F to smooth the bottom of the trench and form rounded corners.

[0044] 5. Preparation of composite insulating layer: The trenched substrate is placed in an oxidation furnace and oxygen is introduced at 950°C to grow a SiO2 layer 541 with a thickness of 80 nm; then the substrate is transferred to an atomic layer deposition apparatus and an Al2O3 layer 542 with a thickness of 30 nm is deposited at 300°C.

[0045] 6. Preparation of isolation dielectric layer and shielding gate: SiO2 isolation dielectric layer 53 with a thickness of 45nm was deposited on the surface of composite insulating layer by PECVD process; then N-type doped polysilicon was filled in the lower part of the trench, and the pattern of shielding gate polysilicon 52 with a thickness of 10μm was defined by photolithography and etching process.

[0046] 7. Gate polysilicon fabrication: N-type doped polysilicon is further filled on the surface of the isolation dielectric layer 53. Excess polysilicon outside the trench is removed by chemical mechanical polishing (CMP) to form gate polysilicon 51 with a thickness of 5μm.

[0047] 8. Source and Drain Region Fabrication: Photoresist is coated on the surface of the second epitaxial layer 32. The source and drain regions are defined by photolithography. Boron ions and phosphorus ions are implanted sequentially by ion implantation. Then, annealing is performed to activate the doped ions.

[0048] 9. Ohmic contact electrode fabrication: Titanium and nickel layers are deposited sequentially on the source and drain regions using electron beam evaporation, followed by silver layer deposition using sputtering, and finally alloying treatment to form the ohmic contact electrode.

[0049] 10. Passivation layer preparation: A Si3N4 passivation layer with a thickness of 150nm is deposited on the device surface using PECVD process; then, openings are made on the passivation layer using photolithography and etching processes to expose the ohmic contact areas of the source and drain regions.

[0050] 11. Post-processing: Perform subsequent processes such as metallization wiring, dicing, and packaging on the device to complete the fabrication of the entire device.

[0051] Performance testing: The performance of the SGTMOSFET power device prepared in this embodiment was tested, and the test results are as follows: 1. Breakdown Voltage (BVDSS): At room temperature, the breakdown voltage of the device was tested using a semiconductor parameter analyzer. The test result was 75.2V, which is 11.3% higher than that of the traditional 60V-class SGT device.

[0052] 2. On-resistance (Ron,sp): Under the conditions of Vgs=10V and Ids=1A, the specific on-resistance of the test device is 1.2mΩ·cm², which is 13.6% lower than that of the traditional structure.

[0053] 3. Miller capacitance (CGD): Under the conditions of Vds=25V and f=1MHz, the measured Miller capacitance is 0.8pF, which is 13 times lower than that of traditional Trench MOSFET.

[0054] 4. Gate leakage current: Under the condition of Vgs=20V and room temperature, the gate leakage current is 5×10⁻¹²A; at a high temperature of 120℃, the gate leakage current is 8×10⁻¹²A, which shows excellent insulation performance.

[0055] 5. High-temperature on-resistance: At a high temperature of 175℃, the on-resistance is 1.98mΩ·cm², and the multiplication factor is 1.85, which is 12.2% lower than that of existing SGT devices.

[0056] 6. Avalanche Energy (EAS): The unidirectional avalanche energy of the test device is 12.5 mJ, which is 25% higher than that of the traditional structure.

[0057] The test results above show that the SGTMOSFET power device in this embodiment has excellent overall performance and can meet the high-performance requirements of the low-to-medium voltage power electronics field.

[0058] Example 2: This embodiment is basically the same as the structure of embodiment 1, the main difference being the parameters of the double-layer epitaxial structure and the thickness of the composite insulating layer. The specific parameters are as follows: The total thickness of epitaxial structure 3 is 20 μm, the thickness of the first epitaxial layer 31 is 12 μm, the resistivity is 0.4 Ω·cm, and the doping concentration is 1×10¹. 6 cm⁻³; the second epitaxial layer 32 has a thickness of 8 μm, a resistivity of 1 Ω·cm, and a doping concentration of 4 × 10¹. 5 cm⁻³.

[0059] In the composite insulating layer 54, the SiO2 layer 541 has a thickness of 50 nm, and the Al2O3 layer 542 has a thickness of 20 nm.

[0060] The trench 4 has a depth of 15μm, a width of 0.8μm, an adjacent spacing of 2μm, and a bottom corner radius of 0.1μm.

[0061] Performance test results show that the breakdown voltage of the device in this embodiment is 72.1V, the specific on-resistance is 1.35mΩ·cm², the Miller capacitance is 0.9pF, and the on-resistance multiplication factor at 175℃ is 1.9. All performance characteristics are superior to those of existing traditional SGT devices.

[0062] Example 3: This embodiment has a basically the same structure as Embodiment 1, the main difference being the doping concentration of the source and drain regions and the thickness of the ohmic contact electrode. The specific parameters are as follows: The doping concentration in source region 6 is 5 × 10² 0 cm⁻³, depth 1μm; drain region 7 doping concentration 2×10² 0 cm⁻³, with a depth of 1.5μm.

[0063] In the ohmic contact electrode, the titanium layer is 50 nm thick, the nickel layer is 100 nm thick, and the silver layer is 200 nm thick.

[0064] Performance test results show that the ohmic contact resistance of the device in this embodiment is reduced to 0.05Ω, which is 20% lower than that in Embodiment 1. It exhibits better conduction performance when operating at high current, while the breakdown voltage and switching characteristics remain stable.

[0065] The implementation principle of this utility model is as follows: This utility model discloses a high-performance SGT MOSFET power device, belonging to the field of power semiconductor device technology, aiming to solve the problems of poor trade-off between breakdown voltage and on-resistance and low gate insulation reliability of existing medium and low voltage SGT MOSFETs. The device includes, from bottom to top, an N-type heavily doped silicon substrate 1, an N-type doped silicon buffer layer 2, and a double-layer epitaxial structure 3. The epitaxial structure 3 has a trench 4 that penetrates its thickness. The trench 4 contains a gate assembly 5 containing a gate polysilicon 51, a shielding gate polysilicon 52, and an isolation dielectric layer 53. A SiO2 / Al2O3 composite insulating layer 54 is provided between the gate assembly 5 and the sidewall of the trench 4. A P-type source region 6, an N-type drain region 7, and a multilayer metal ohmic contact electrode are respectively provided on both sides of the trench 4 on the epitaxial structure 3. A Si3N4 passivation layer is provided on the surface of the epitaxial structure 3. The resistivity and thickness ratio of the double-layer epitaxial layer have been optimized. The composite insulating layer 54 combines the advantages of thermally oxidized SiO2 and ALD-deposited Al2O3, which improves the device breakdown voltage by 8%-12%, reduces the on-resistance by 10%-15%, and reduces the Miller capacitance by 12-15 times. The on-resistance multiplication factor at 175℃ is only 1.8-1.9. It has excellent switching characteristics, high-temperature stability and reliability, and is suitable for medium and low voltage power conversion scenarios such as power management of new energy vehicles.

[0066] The embodiments described herein are preferred embodiments of this utility model and are not intended to limit the scope of protection of this utility model. Therefore, all equivalent changes made to the structure, shape, and principle of this utility model should be included within the scope of protection of this utility model.

Claims

1. A high-performance SGT MOSFET power device, characterized in that, The structure includes a substrate (1), a buffer layer (2), and an epitaxial structure (3) arranged sequentially from bottom to top. A trench (4) penetrating the thickness direction is provided within the epitaxial structure (3), and a gate assembly (5) is disposed within the trench (4). A source region (6) and a drain region (7) are respectively disposed on both sides of the trench (4) on the epitaxial structure (3), and ohmic contact electrodes are disposed on both the source region (6) and the drain region (7). The epitaxial structure (3) is a double-layer epitaxial structure (3), including a lower first epitaxial layer (31) and an upper second epitaxial layer (32). The resistivity of the first epitaxial layer (31) is less than that of the second epitaxial layer (32); the gate assembly (5) includes a gate polysilicon (51) and a shielding gate polysilicon (52), the gate polysilicon (51) is located at the upper part of the trench (4), the shielding gate polysilicon (52) is located at the lower part of the trench (4), an isolation dielectric layer (53) is provided between the gate polysilicon (51) and the shielding gate polysilicon (52), and a composite insulating layer (54) is provided between the gate polysilicon (51), the shielding gate polysilicon (52) and the sidewall of the trench (4).

2. The high-performance SGT MOSFET power device according to claim 1, characterized in that, The total thickness of the double-layer epitaxial structure (3) is 20-30 μm, wherein the thickness of the first epitaxial layer (31) is 12-18 μm, the thickness of the second epitaxial layer (32) is 8-12 μm, the thickness ratio of the first epitaxial layer (31) to the second epitaxial layer (32) is 1.2-1.8:1, and the resistivity ratio is 0.3-0.6:

1.

3. The high-performance SGT MOSFET power device according to claim 1, characterized in that, The composite insulating layer (54) consists of a bottom SiO2 layer (541) and an upper Al2O3 layer (542). The thickness of the SiO2 layer (541) is 50-100 nm, and the thickness of the Al2O3 layer (542) is 20-50 nm. The SiO2 layer (541) is prepared by thermal oxidation process, and the Al2O3 layer (542) is prepared by atomic layer deposition process.

4. The high-performance SGT MOSFET power device according to claim 1, characterized in that, The groove (4) has a depth of 15-25μm, a width of 0.8-1.2μm, a spacing of 2-3μm between adjacent grooves (4), a sidewall verticality of ≥89.5°, and a bottom fillet radius of 0.1-0.2μm.

5. A high-performance SGT MOSFET power device according to claim 1, characterized in that, The ohmic contact electrode has a multilayer metal structure, consisting of a titanium layer, a nickel layer, and a silver layer from bottom to top. The titanium layer has a thickness of 20-50 nm, the nickel layer has a thickness of 50-100 nm, and the silver layer has a thickness of 100-200 nm.

6. The high-performance SGT MOSFET power device according to claim 1, characterized in that, The surface of the epitaxial structure (3) is also provided with a passivation layer, which is a Si3N4 layer with a thickness of 100-200nm. An opening is provided on the passivation layer corresponding to the position of the ohmic contact electrode.