A Ga2O3 power device heat dissipation optimization method based on a heterogeneous substrate structure

Abstract

The present application relates to the technical field of wide band gap semiconductor devices, and particularly relates to a Ga2O3 power device heat dissipation optimization method based on a hetero-substrate structure, which comprises the following steps: combining a first principle to construct an interface atomic model, adopting the first principle to screen a high-thermal-conductivity substrate material, constructing a hetero-interface model to analyze key physical parameters, calculating an interface thermal resistance based on a machine learning potential function and a molecular dynamics method, and combining finite element thermal simulation to carry out macro-scale thermal analysis and structure optimization. The present application covers the whole process from material screening to interface optimization and then to device-level thermal simulation, systematically improves the thermal management performance of the Ga2O3 power device, and simultaneously improves the efficiency and reduces the trial and error cost, thereby providing differentiated strategies for adapting to various complex application scenarios.

Description

Technical Field

[0001] This invention relates to the technical field of wide bandgap semiconductor devices, and more specifically to a heat dissipation optimization method for Ga2O3 power devices based on heterogeneous substrate structures. Background Technology

[0002] Gallium oxide (GaO), as a novel ultrawide bandgap semiconductor material, is considered a candidate material for next-generation high-voltage power devices due to its ultrawide bandgap characteristics (approximately 4.8 eV), high critical breakdown electric field strength (>8 MV / cm), and high Baliga figure of merit (approximately 4 times that of SiC and 2 times that of GaN), showing great application potential in power electronics and radio frequency fields. However, with the rapid development of electronic devices towards miniaturization, high integration, and high power density, the heat dissipation problem of power devices has become increasingly prominent. Due to bottlenecks such as the low thermal conductivity of Ga2O3 (approximately 10–27 W / mK), brittleness, and high defect density, intrinsic materials alone are insufficient to meet the heat dissipation requirements of high-power devices. When Ga2O3 is used in heterostructures or encapsulated, it generates a large amount of heat during operation, increasing the overall thermal resistance of the device and making heat dissipation difficult. If heat dissipation cannot be carried out in a timely and effective manner, it will not only reduce device performance but may also lead to failure or even thermal breakdown, thus seriously threatening the stability and lifespan of the device and severely restricting its application and promotion in high-power scenarios.

[0003] The current mainstream heat dissipation technology for power devices is active cooling, including natural convection air cooling, forced convection liquid cooling, and heat pipes and microchannels. However, these methods generally suffer from the following problems: system complexity, high cost, and difficulty in integration into compact packages; cooling efficiency is easily affected by external conditions, resulting in insufficient stability; and limited near-field thermal management capabilities for internal heat sources, failing to fundamentally solve the problem of high substrate thermal resistance. Therefore, relying solely on traditional cooling methods is insufficient to meet the development needs of high-power, highly integrated power devices. A new method for optimizing heat dissipation in power devices is urgently needed to overcome the technical bottlenecks of low thermal conductivity semiconductor materials such as Ga2O3 in power applications. Summary of the Invention

[0004] The purpose of this invention is to provide a heat dissipation optimization method for Ga2O3 power devices based on heterogeneous substrate structure, which can systematically and efficiently improve the thermal management performance of Ga2O3 power devices and provide differentiated strategies for various complex application scenarios.

[0005] The solution adopted to achieve the objective of this invention is as follows: a heat dissipation optimization method for Ga2O3 power devices based on heterogeneous substrate structure, comprising the following steps: (1) Screening of candidate substrate materials with a lattice mismatch rate of less than 5% with Ga2O3, and further simulation analysis of the properties of the candidate substrate materials to screen out substrate materials that match Ga2O3; (2) Predict the thermal conductivity of the substrate material selected in step (1) and select a high thermal conductivity substrate material with suitable thermal conductivity; (3) Based on the high thermal conductivity substrate material selected in step (2), construct its heterogeneous interface model with Ga2O3, and select a stable interface structure model. (4) The interface structure model selected in step (3) is used to perform micro-nano scale thermal transport simulation based on the potential function. The interface thermal conductivity of the interface structure model composed of different heterogeneous substrate materials under different sizes is analyzed, and the interface structure model with suitable interface thermal conductivity is selected. (5) Construct a device including the interface structure model selected in step (4), model and simulate the thermal field of the device from a macroscopic scale, simulate and evaluate the junction temperature change law under different power densities and packaging conditions, and select the best substrate material for heat dissipation of Ga2O3 power devices in practical applications.

[0006] Preferably, in step (1), the candidate substrate material includes at least one of diamond, boron arsenide, boron nitride, silicon carbide, gallium nitride and silicon.

[0007] Preferably, in step (1), the properties of the candidate substrate material are analyzed by first-principles calculation and simulation based on density functional theory, including lattice constant, formation energy, phonon spectrum and band structure.

[0008] Preferably, in step (2), the thermal conductivity of the material is predicted by first-principles calculation combined with ShengBTE, and the thermal conductivity of the selected high thermal conductivity substrate material is greater than 100 W / mK.

[0009] Preferably, in step (3), the binding energy of the interface structure model is less than 0 and there are no defective states.

[0010] Preferably, step (3) further includes obtaining key physical parameters for the interface structure model and analyzing the interface band continuity and carrier transport performance. The key physical parameters include charge density, differential charge density, electrostatic potential, density of states, and band alignment.

[0011] Preferably, in step (4), the potential function is a machine learning potential function, including a dp potential function or a nep potential function constructed by DeepMD or GPUMD methods.

[0012] Preferably, in step (4), the non-equilibrium molecular dynamics method is used to simulate the interfacial thermal transport behavior of the interfacial structure model and obtain the interfacial thermal conductivity.

[0013] Preferably, in step (5), the device including the interface structure model selected in step (4) is constructed using the finite element method, and the thermal field of the device is modeled and simulated from a macroscopic scale to analyze the thermal field distribution of the device.

[0014] Preferably, the analysis of the thermal field distribution of the device includes steady-state thermal analysis and transient thermal analysis.

[0015] Compared with the prior art, the present invention has the following advantages: This invention utilizes a multi-scale modeling approach, combining simulation analysis of substrate material properties, construction of a heterogeneous interface model between substrate material and Ga2O3, potential function-based thermal transport simulation, and macroscopic-scale device thermal field modeling and simulation. This approach achieves full-process coverage from material selection to interface optimization and device-level thermal simulation, ensuring that optimal solutions are found and theoretically supported from the atomic level to the device level and from the microscopic to the macroscopic.

[0016] This invention reduces the number of material selection and structural iterations required in traditional physics experiments by conducting multiple material screenings and modeling simulations, thereby shortening the R&D cycle, improving design efficiency, and reducing trial-and-error costs.

[0017] This invention effectively reduces the thermal resistance of Ga2O3 power devices and improves the heat flux density conduction capability through heterogeneous substrate structure design, thereby suppressing junction temperature rise, enhancing device stability and reliability, and improving device thermal management performance.

[0018] This invention combines simulation of external field control and interface morphology design to provide differentiated thermal management strategies for complex application scenarios, adapting to a variety of working conditions.

[0019] This invention uses real material parameters and boundary conditions to model and simulate the device-level structure on a macroscopic scale, evaluating the junction temperature variation of Ga2O3 power devices under different power densities and packaging conditions. This improves the reliability of Ga2O3 power devices in high-density packaging and high-power conversion applications, thereby expanding their applications in electric vehicles, 5G communications, satellite radar, and other fields. Attached Figure Description

[0020] Figure 1 A flowchart of the heat dissipation optimization method for Ga2O3 power devices based on heterogeneous substrate structure provided by the present invention; Figure 2 This is a flowchart of the heat dissipation optimization method for Ga2O3 power devices based on heterogeneous substrate structure provided in Embodiment 1 of the present invention; Figure 3 The supercell model for the first-principles analysis of the Ga2O3 / Si heterointerface in Embodiment 2 of the present invention; Figure 4The supercell model for the first-principles analysis of the Ga2O3 / Diamond heterostructure interface in Embodiment 2 of the present invention; Figure 5 This is the interface model for molecular dynamics analysis of the Ga2O3 / Si heterostructure in Example 2 of the present invention; Figure 6 This is the interface model for molecular dynamics analysis of the Ga2O3 / Diamond heterostructure in Example 2 of the present invention; Figure 7 This is a finite element analysis flowchart of the multi-scale simulation method for the thermal field of Ga2O3 power devices provided in Embodiment 2 of the present invention. Detailed Implementation

[0021] To better understand the present invention, the following embodiments are further illustrations of the present invention, but the content of the present invention is not limited to the following embodiments.

[0022] Example 1 A method for optimizing heat dissipation of Ga2O3 power devices based on heterogeneous substrate structures, such as... Figure 1 As shown, it includes the following steps: (1) Screening of candidate substrate materials with a lattice mismatch rate of less than 5% with Ga2O3, and further simulation analysis of the properties of the candidate substrate materials to screen out substrate materials that match Ga2O3; (2) Predict the thermal conductivity of the substrate material selected in step (1) and select a high thermal conductivity substrate material with suitable thermal conductivity; (3) Based on the high thermal conductivity substrate material selected in step (2), construct its heterogeneous interface model with Ga2O3, and select a stable interface structure model. (4) The interface structure model selected in step (3) is used to perform micro-nano scale thermal transport simulation based on the potential function. The interface thermal conductivity of the interface structure model composed of different heterogeneous substrate materials under different sizes is analyzed, and the interface structure model with suitable interface thermal conductivity is selected. (5) Construct a device including the interface structure model selected in step (4), model and simulate the thermal field of the device from a macroscopic scale, simulate and evaluate the junction temperature change law under different power densities and packaging conditions, and select the best substrate material for heat dissipation of Ga2O3 power devices in practical applications.

[0023] like Figure 2 As shown, the detailed steps are as follows: 1. Preliminary Screening of Candidate Heterogeneous Substrate Materials: A candidate material library is established based on material databases (such as Materials Project, OQMD, AFLOW, etc.). Inorganic crystal materials with potentially high thermal conductivity are selected as the initial candidate set, prioritizing materials with high crystal symmetry and strong interatomic bonding (such as cubic diamond structure), high elastic modulus and chemical stability, and the ability to produce large-size single crystals or epitaxial layers. Using the lattice constants of Ga2O3 (monoclinic system, a≈12.49 Å, b≈3.09 Å, c≈5.90 Å) as a benchmark, the in-plane lattice mismatch between Ga2O3 and each candidate material is calculated. Only materials with a mismatch rate of less than 5% are retained for subsequent thermal conductivity analysis and interface structure design. For materials with a mismatch rate within 5%, first-principles calculations and simulations are used to analyze the properties of the candidate substrate materials, including lattice constant, formation energy, phonon spectrum, and band structure, to further screen for matching materials.

[0024] 2. Screening of high thermal conductivity substrate materials with suitable thermal conductivity: Using first-principles calculations combined with ShengBTE to predict the thermal conductivity of materials, high thermal conductivity substrate materials with a thermal conductivity greater than 100 W / mK were screened.

[0025] 3. Screening for stable interface structure models: Construct a heterostructure model of Ga2O3, a high thermal conductivity substrate material. Use first-principles calculations (such as VASP) to evaluate the energy relaxation and stability of the constructed interface structure, obtaining key physical parameters including formation energy, interface binding energy, charge density distribution, differential charge density, electrostatic potential distribution, density of states, and band alignment. Analyze the interface band continuity and carrier transport performance. Screen interface structure models with a formation energy less than 0 and no defect states to ensure that the structure has good physical reliability and interface thermal conductivity in practical devices.

[0026] 4. Micro / Nano-Scale Thermal Transport Simulation Based on Potential Functions: Select the optimal interface structure from step 3, generate a sufficient dataset through Articulated Molecular Dynamics (AIMD) simulations, and use software such as DeepMD and GPUMD to obtain high-precision machine learning dp or nep potential functions. Utilize these potential functions to simulate interface thermal conductivity using Non-Equilibrium Molecular Dynamics (NEMD) and explore interface properties using methods such as Spectral Energy Density (SED). Simulate the interface thermal conductivity of different heterogeneous substrates at different sizes, and compare the results to find the optimal substrate for enhancing heat dissipation in Ga2O3 materials.

[0027] 5. Conduct multi-scale simulations and evaluate the effectiveness of thermal management optimization: Introduce atomic-scale thermal conductivity simulation results into a multi-scale modeling framework to model and simulate the device thermal field at a macroscopic scale, establishing a device-level thermal model including a multilayer structure such as a Ga2O3 epitaxial layer, heterogeneous substrate, and metal electrodes. Use finite element simulation software (such as COMSOL Multiphysics, ANSYS, etc.) to model and simulate the device thermal field, considering factors such as material anisotropic thermal conductivity, interface thermal resistance, and heat dissipation boundary conditions. Perform steady-state and transient thermal analyses, and simulate and evaluate the junction temperature variation under different power densities and packaging conditions to verify the thermal management effectiveness of the optimization method.

[0028] Example 2 A heat dissipation optimization method for Ga2O3 power devices based on heterogeneous substrate structures, taking Ga2O3 / Si and Ga2O3 / Diamond devices as examples, includes the following steps: (1) Interface models of Ga2O3 / Si and Ga2O3 / Diamond were constructed using Materials Studio to verify that the mismatch was <5% and that electronic requirements were met. The supercell model of the Ga2O3 / Si heterointerface was analyzed using first-principles methods. Figure 3 As shown, the supercell model of the Ga2O3 / Diamond heterointerface for first-principles analysis is as follows: Figure 4 As shown; Starting from the crystal structure theory in solid-state physics and combining the density functional theory in the Schrödinger equation of quantum theory, a suitable Brillouin zone reciprocal space is constructed. To avoid underestimating semiconductor band structure in plane wave expansion (PBE) calculations, hybrid functional (HSE) calculations are used to perform band structure and density of states related calculations. The simulated supercell band alignment, carrier mobility, and formation energy increase the accuracy of the calculation of the microscopic electronic structure, thus providing a reasonable theoretical basis for subsequent simulation coupling steps and material growth monitoring.

[0029] (2) The constructed interface models of Ga2O3 / Si and Ga2O3 / Diamond are cuboid in shape, such as... Figure 5 and Figure 6 As shown; We constructed an interface structure model with suitable interfacial thermal conductivity, performed non-equilibrium molecular dynamics simulations (NEMD), calculated the interfacial thermal conductivity of the interface model, and studied the effects of different sizes and temperatures on the interfacial thermal conductivity.

[0030] (3) Construct Ga2O3 / Si and Ga2O3 / Diamond heterostructure model devices. The finite element analysis flowchart of the multi-scale simulation method for the thermal field of Ga2O3 / Si and Ga2O3 / Diamond power devices is shown below. Figure 7 As shown; The optimal substrate quality was determined by combining micron-scale simulation results to enhance the heat dissipation of Ga2O3 material, and a macro-scale model of Ga2O3 power device was established by combining finite element software such as COMSOL Multiphysics. Set appropriate mesh quality and boundary conditions for the finite element model; Then, multi-physics simulations, including electro-thermal-mechanical simulations, are performed to obtain the thermal field distribution of the device. The simulation parameters were optimized and adjusted multiple times, including mesh quality, solver settings, and external conditions, to comprehensively determine the heat dissipation performance of Ga2O3 power devices in practical applications. Ultimately, Diamond was determined to be the best substrate material for heat dissipation.

[0031] This invention employs screening and multi-scale simulation analysis to design a heat dissipation optimization method for Ga2O3 power devices based on heterogeneous substrate structures. This method can greatly reduce the trial-and-error costs of actual device-level manufacturing and can be applied to the design of heat dissipation substrates made of other similar materials.

[0032] The above description is merely a preferred embodiment of the present invention, and should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for optimizing heat dissipation of Ga2O3 power devices based on heterogeneous substrate structures, characterized in that, Includes the following steps: (1) Screening of candidate substrate materials with a lattice mismatch rate of less than 5% with Ga2O3, and further simulation analysis of the properties of the candidate substrate materials to screen out substrate materials that match Ga2O3; (2) Predict the thermal conductivity of the substrate material selected in step (1) and select a high thermal conductivity substrate material with suitable thermal conductivity; (3) Based on the high thermal conductivity substrate material selected in step (2), construct its heterogeneous interface model with Ga2O3, and select a stable interface structure model. (4) The interface structure model selected in step (3) is used to perform micro-nano scale thermal transport simulation based on the potential function. The interface thermal conductivity of the interface structure model composed of different heterogeneous substrate materials under different sizes is analyzed, and the interface structure model with suitable interface thermal conductivity is selected. (5) Construct a device including the interface structure model selected in step (4), model and simulate the thermal field of the device from a macroscopic scale, simulate and evaluate the junction temperature change law under different power densities and packaging conditions, and select the best substrate material for heat dissipation of Ga2O3 power devices in practical applications.

2. The heat dissipation optimization method for Ga2O3 power devices based on heterogeneous substrate structure according to claim 1, characterized in that: In step (1), the alternative substrate material includes at least one of diamond, boron arsenide, boron nitride, silicon carbide, gallium nitride, and silicon.

3. The heat dissipation optimization method for Ga2O3 power devices based on heterogeneous substrate structure according to claim 1, characterized in that: In step (1), the properties of the candidate substrate material are analyzed by first-principles calculation and simulation based on density functional theory. The properties include lattice constant, formation energy, phonon spectrum, and band structure.

4. The heat dissipation optimization method for Ga2O3 power devices based on heterogeneous substrate structure according to claim 1, characterized in that: In step (2), the thermal conductivity of the material is predicted by first-principles calculation combined with ShengBTE, and the thermal conductivity of the high thermal conductivity substrate material is greater than 100 W / mK.

5. The heat dissipation optimization method for Ga2O3 power devices based on heterogeneous substrate structure according to claim 1, characterized in that: In step (3), the binding energy of the interface structure model is less than 0 and there are no defect states.

6. The heat dissipation optimization method for Ga2O3 power devices based on heterogeneous substrate structure according to claim 1, characterized in that: Step (3) also includes obtaining key physical parameters for the interface structure model and analyzing the interface band continuity and carrier transport performance. The key physical parameters include charge density, differential charge density, electrostatic potential, density of states, and band alignment.

7. The heat dissipation optimization method for Ga2O3 power devices based on heterogeneous substrate structure according to claim 1, characterized in that: In step (4), the potential function is a machine learning potential function, including a dp potential function or a nep potential function constructed by DeepMD or GPUMD methods.

8. The heat dissipation optimization method for Ga2O3 power devices based on heterogeneous substrate structure according to claim 1, characterized in that: In step (4), the non-equilibrium molecular dynamics method is used to simulate the interfacial thermal transport behavior of the interface structure model and obtain the interfacial thermal conductivity.

9. The heat dissipation optimization method for Ga2O3 power devices based on heterogeneous substrate structure according to claim 1, characterized in that: In step (5), the device including the interface structure model selected in step (4) is constructed using the finite element method. The thermal field of the device is modeled and simulated from a macroscopic scale to analyze the thermal field distribution of the device.

10. The heat dissipation optimization method for Ga2O3 power devices based on heterogeneous substrate structure according to claim 9, characterized in that: The thermal field distribution of the analytical device includes steady-state thermal analysis and transient thermal analysis.

Images