Simulation method for abrasive particle vibration processing based on reaction molecular dynamics

Abstract

The invention discloses a simulation method for abrasive particle vibration machining based on reaction molecular dynamics, and relates to the technical field of reaction molecular dynamics, a vibration auxiliary model of a Si surface of 4H-SiC is established, the process of vibration auxiliary chemical mechanical polishing of diamond abrasive particles in the Z-axis direction is simulated, the Si surface model of the 4H-SiC is consistent with an actual machining surface of silicon carbide, and the abrasive particle vibration machining simulation method is suitable for the actual machining surface of silicon carbide. The vibration direction of the vibration auxiliary simulation is consistent with the motion direction of the abrasive particles in the actual chemical mechanical polishing process, so that the acting force and motion trail simulation of the diamond abrasive particle model in the polishing process are more real, and the simulation process is closer to the actual machining process; the atom removal mechanism of the Si surface of 4H-SiC under the conditions of different vibration amplitudes and frequencies and different scratching depths can be researched, so that a chemical mechanical polishing technology is conveniently optimized by ultrasonic vibration, and the polishing quality of silicon carbide is further improved; the technical problem that an existing vibration auxiliary model based on reaction molecular dynamics simulation cannot effectively optimize the chemical mechanical polishing process of silicon carbide is solved.

Description

Technical Field

[0001] The invention relates to the technical field of reactive molecular dynamics, in particular to a simulation method for abrasive vibration processing based on reactive molecular dynamics. Background Art

[0002] Silicon carbide is widely used in the electronic field of high voltage, high temperature, high power, high frequency, etc. With the continuous improvement of its application requirements, silicon carbide is developing towards ultra-precision and low damage of machining surface quality. A single physical or chemical action has been difficult to meet the industrial requirements for the surface of silicon carbide workpieces. Chemical mechanical polishing combines the effects of chemical corrosion and mechanical grinding. It is a very effective method for processing silicon carbide surface flattening and is currently the only way to achieve global polishing. After polishing with chemical mechanical polishing technology, the surface quality of silicon carbide is good and there is almost no damage layer, but the material removal efficiency is low, so the research on chemical mechanical polishing technology mainly focuses on material removal. Traditional experimental research has limited monitoring methods for material removal mechanisms, so molecular dynamics simulation is introduced to study the deformation and removal mechanism of nanoscale materials.

[0003] For example, the method for simulating the friction and wear process of nanoscale diamond based on molecular dynamics disclosed in the Chinese invention patent with the authorization announcement date of 2022.08.12 and the authorization announcement number of CN 113012765 B can analyze the influence of temperature change on the wear process of nanoscale diamond, among which the method for simulating the friction and wear process of nanoscale diamond is the key protected property right of the patent, involving the process of constructing a molecular dynamics model, determining the simulation parameters of the molecular dynamics model, and using diamond abrasives to rub and wear the diamond substrate. For example, the method and device for simulating the nano-grinding of silicon carbide with defects disclosed in the Chinese invention patent with the application publication date of 2022.09.13 and the application publication number of CN115048849 A are used to study the changes of defects in the processing process and the influence of defects on the processing results and parameters, among which the method and device for simulating the nano-grinding of silicon carbide with defects is the key protected property right of the patent, involving the process of constructing a molecular dynamics model, determining the simulation parameters of the molecular dynamics model, and using a tool model to simulate the grinding of a silicon carbide workpiece model with defects. For example, the Chinese invention patent with an application publication date of 2024.06.21 and application publication number CN 118230874 A discloses an abrasive processing simulation method based on reaction molecular dynamics, which observes the changes in the microscopic molecular structure during chemical corrosion and abrasive scratching at the atomic level. Among them, the reaction process of silicon carbide and the reaction layer is the property rights content that the patent focuses on protecting.

[0004] From the perspective of molecular dynamics simulation, existing research mainly focuses on the single vibration behavior of abrasive particles and their effects on the removal and damage of silicon carbide substrate atoms. The vibration mode of abrasive particles and their effects on the removal, oxidation and damage of silicon carbide atoms have not been studied. The mechanism of interaction between silicon carbide substrate, hydrogen peroxide solution and abrasive atoms has not been studied in depth.

[0005] In order to analyze the atomic-level material removal mechanism during the abrasive vibration-assisted polishing of silicon carbide in hydrogen peroxide solution, there has been research in the prior art that adds the single vibration behavior of the abrasive to the ultrasonic vibration assistance, and uses the reaction molecular pair dynamics to simulate the polishing process. By establishing a vibration-assisted model of the C surface of the silicon carbide substrate in hydrogen peroxide solution, the abrasive is scratched along the tangential direction of the silicon carbide substrate, and ultrasonic vibration is performed along the lateral direction of the silicon carbide substrate. The simulation results show that the lateral ultrasonic vibration of the silicon carbide substrate can effectively improve the quality and efficiency of nano-polishing.

[0006] However, in actual silicon carbide processing, the C-surface of the silicon carbide substrate is rarely used for chemical mechanical polishing, and the vibration direction of the abrasive is usually consistent with the processing direction to achieve more effective material removal and surface polishing. Therefore, the vibration-assisted model of the C-surface of the silicon carbide substrate cannot fully reveal the atomic-level removal mechanism in the actual silicon carbide chemical mechanical polishing process. Therefore, the simulation results cannot effectively guide the ultrasonic vibration-assisted optimization of the chemical mechanical polishing process of silicon carbide, which also limits the universality and application of the research results. Summary of the invention

[0007] In view of the deficiencies in the above-mentioned background technology, the present invention proposes a simulation method of abrasive vibration processing based on reaction molecular dynamics, which solves the technical problem that the existing vibration-assisted model based on reaction molecular dynamics simulation cannot effectively optimize the chemical mechanical polishing process of silicon carbide.

[0008] The technical solution of this application is: The simulation method of abrasive vibration machining based on reactive molecular dynamics includes the following steps: Step 1: establishing a simulation system with a coordinate system consisting of a hydrogen peroxide solution model and a silicon carbide substrate model, and setting boundary conditions for the simulation system; Step 2: Selecting a reaction force field potential function that describes the bonding, bond breaking and interaction between atoms in the simulation system; Step 3: establishing a diamond abrasive grain model in the simulation system, wherein the diamond abrasive grain model is arranged above the silicon carbide substrate model; Step 4: Divide the silicon carbide substrate model into three regions, namely, a fixed layer, a constant temperature layer and a Newtonian layer, from bottom to top along the Z-axis direction of the coordinate system, keep the atoms in the fixed layer stationary, and relax the simulation system; Step 5: Control the diamond abrasive model to press down to a set depth with a set pressing parameter, and then scratch at a set moving speed in the X-axis direction of the coordinate system, and at the same time, mechanically vibrate along a sinusoidal function trajectory in the Z-axis direction of the coordinate system to polish the silicon carbide substrate model; Step 6: Output the positional mechanical parameters of the diamond abrasive model vibrating and scratching the silicon carbide substrate model, and perform visual analysis by importing the positional mechanical parameters into the OVITO visualization software; Step 7: respectively change the pressing parameters of the diamond abrasive model, the amplitude and frequency parameters of the mechanical vibration of the diamond abrasive model, and repeat steps 5 and 6.

[0009] Preferably, the boundary conditions in step one are: setting periodic boundary conditions in the X-axis direction and the Y-axis direction of the coordinate system, and setting fixed boundary conditions in the Z-axis direction of the coordinate system.

[0010] Preferably, the shape of the diamond abrasive grain model in step 1 is spherical, and the lattice constant is 3.567.

[0011] Preferably, the silicon carbide substrate model in step one is a Si surface model of 4H-SiC.

[0012] Preferably, the reaction force field potential function in step 2 is SiCHO.ffield.reax.

[0013] Preferably, in step 2, the safety region and the minimum coordination number of the molecular dynamics simulation are first set; and then the bonding, bond breaking and interaction between the atoms in the simulation system are defined by setting the parameters of the SiCHO.ffield.reax.

[0014] Preferably, the simulation system in step 4 is relaxed under an NVT ensemble.

[0015] Preferably, the relaxation process output parameters in step 4 include: thermodynamic parameters, calculation of the number of bonds between atoms, the number of different species, and positional mechanical parameters, and the progress of relaxation is determined based on the relaxation process output parameters.

[0016] Preferably, the simulation system in step five is polished under the NVE ensemble.

[0017] Preferably, the scratching trajectory of the diamond abrasive model in the X-axis direction in step five is X = vt, and the sinusoidal function trajectory of the diamond abrasive model in the Z-axis direction is Z = Asin(2πft), and v, t, A, and f are the moving speed, nanopolishing time, amplitude, and frequency of the diamond abrasive model, respectively.

[0018] Compared with the prior art, the technical solution disclosed in the present invention has the following beneficial effects: 1. In actual processing, the silicon surface of silicon carbide is more widely used. The present invention establishes a Si surface model of 4H-SiC, which is more in line with the actual processing surface of silicon carbide. By studying the chemical mechanical polishing process of the Si surface in an aqueous hydrogen peroxide solution, the response and removal mechanism of different crystal surfaces of silicon carbide to the chemical mechanical polishing process can be more comprehensively understood, and the atomic-level removal mechanism on the actual processing surface of silicon carbide can be more comprehensively understood, which is helpful to use simulation methods to provide optimization guidance for the actual processing of silicon carbide and improve the practicality of the research results.

[0019] 2. The present invention realizes the vibration-assisted model research of the Si surface model of 4H-SiC in the hydrogen peroxide solution model by adding a diamond abrasive model that performs mechanical vibration along the Z-axis direction above the Si surface model of 4H-SiC. The vibration direction of the diamond abrasive model is more consistent with the movement direction of the abrasive in the actual chemical mechanical polishing process. The vibration-assisted model can more realistically simulate the force and movement trajectory of the abrasive in the polishing process, and make the simulation process closer to the actual processing technology.

[0020] 3. The simulation method can clearly output the surface morphology of the Si surface model of 4H-SiC scratched by the diamond abrasive model, and then can clearly and intuitively describe the oxidation, removal and damage mechanism of the Si surface of 4H-SiC. The vibration-assisted model established by the present invention is convenient for changing the vibration parameters and scratching parameters of the diamond abrasive model, which is helpful for studying the oxidation, removal and damage mechanism of the Si surface of 4H-SiC under the conditions of ultrasonic vibration of different vibration amplitudes and frequencies along the Z-axis direction and different scratching depths along the Z-axis direction of the diamond abrasive model, and is convenient for combining ultrasonic vibration with chemical mechanical polishing technology to improve the polishing quality of silicon carbide. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying any creative work.

[0022] Figure 1 is a flow chart of the simulation method of the present invention; Figure 2 It is a model diagram of abrasive vibration processing based on reaction molecular dynamics of the present invention; Figure 3 It is a surface morphology diagram of the silicon carbide substrate model polishing process of the present invention.

[0023] Description of Figure Numbers: 1 Diamond abrasive model, 2 Hydrogen peroxide solution model, 3 Silicon carbide substrate model. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the core concept of the present invention and the following embodiments, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] For ease of understanding, the technical solution of the present invention and the terms and / or custom words involved in the embodiments of the present invention are first explained: ASE: Atomic Simulation Environment, is a Python library for atomic-scale simulation and calculation of material properties. It provides a range of tools and interfaces for building, manipulating and analyzing atomic structures, and performing various simulation methods, such as molecular dynamics, quantum mechanics, and molecular mechanics. ASE supports a variety of atomic-scale simulation methods, including molecular dynamics, quantum mechanics, molecular mechanics, Monte Carlo, etc., enabling users to study the properties and behaviors of materials from multiple perspectives. ASE provides classes to represent atomic structures, and users can create, manipulate and modify information such as the position and type of atoms, while supporting reading and writing structures from external files. ASE provides interfaces with multiple third-party computing software (such as VASP, Quantum ESPRESSO, LAMMPS, etc.), allowing users to easily integrate these software into their simulation workflows. ASE provides a variety of tools for analyzing simulation results, such as calculating energy, force, stress, vibration frequency, electronic structure, etc. ASE has a modular design, and users can write custom plug-ins and extensions as needed to meet specific research needs. ASE is a library written in Python, making it easy to integrate with other scientific computing libraries and tools, such as NumPy, SciPy, and matplotlib.

[0026] LAMMPS: Large-scale Atomic / Molecular Massively Parallel Simulator, a large-scale atomic and molecular parallel simulator, is mainly used for some calculations and simulations related to molecular dynamics.

[0027] Example 1, a simulation method based on reactive molecular dynamics abrasive vibration machining, such as Figure 1 As shown, the following steps are included: Step 1: establish a simulation system with a coordinate system consisting of a hydrogen peroxide solution model 2 and a silicon carbide substrate model 3, and set boundary conditions for the simulation system.

[0028] In the specific implementation, the ASE software package in Python is used to establish the hydrogen peroxide solution model 2 and the 4H-SiC substrate model 3, and the output is a data file recognizable by the LAMMPS software; the LAMMPS software is used to build a simulation box, read the data file, and form the 4H-SiC substrate model 3 and the hydrogen peroxide solution model 2 in the simulation box; the atoms of the simulation system are all set to real units, so that the simulation system has periodic boundary conditions p in the X and Y directions and fixed boundary conditions f in the Z direction; the mass command is used to set the relative atomic masses of the four atoms of C, H, O and Si in the simulation system.

[0029] Furthermore, the silicon carbide substrate model 3 is a Si surface model of 4H-SiC, and the hydrogen peroxide solution model 2 is arranged on the surface of the Si surface model of 4H-SiC. In actual processing, the silicon surface of silicon carbide is more widely used. The present invention establishes a Si surface model of 4H-SiC, which is more in line with the actual processing surface of silicon carbide. By studying the chemical mechanical polishing process of the Si surface model in the hydrogen peroxide aqueous solution model 2, the response and removal mechanism of different crystal surfaces of silicon carbide to the chemical mechanical polishing process can be more comprehensively understood, and the atomic-level removal mechanism on the actual processing surface of silicon carbide can be more comprehensively understood, which is helpful to provide optimization guidance for the actual processing of silicon carbide and helpful for the practical application of the simulation method.

[0030] Furthermore, the data file is a model file with a coordinate system, so the simulation system composed of the hydrogen peroxide solution model 2 and the silicon carbide substrate model 3 also has a coordinate system, which is convenient for analyzing the position change of atoms after the oxidation reaction between the hydrogen peroxide solution model 2 and the silicon carbide substrate model 3, and the displacement change of atoms in the silicon carbide substrate model 3 during the polishing process. After the simulation system is established, the corresponding parameters of the simulation system are first set according to the simulation needs. In this embodiment, the real unit, boundary conditions and relative atomic mass of the atoms are set for the constructed simulation system. It should be noted that the set parameters are applicable to any model in the simulation system.

[0031] Step 2, selecting a reaction force field potential function that describes the bonding, bond breaking and interaction between atoms in the simulation system.

[0032] In this embodiment, the reaction force field potential function is SiCHO.ffield.reax, and each atom is C, H, O and Si. In specific implementation, the reaction force field potential function ReaxFF is first called using the pair_style command, and the safety area and the minimum coordination number are set. The energy formula of the reaction force field potential function ReaxFF is: Esys=Ebond+Eval+Etor+Eover+Eunder+Elp+EvdWaals+Ecoulomd, where Esys, Ebond, Eval, Etor, Eover, Eunder, Elp, EvdWaals and Ecoulomb are the total energy, bond energy, bond angle energy, torsion energy, overcoordination energy correction value, undercoordination stabilization energy, lone pair electron system energy, van der Waals energy term and Coulomb energy term of the entire simulation system respectively; then use the pair_coeff command to call and set the parameters required for the potential function SiCHO.ffield.reax to define the interactions between atoms.

[0033] Furthermore, the safe area and the minimum coordination number are parameters of the potential function SiCHO.ffield.reax. Limiting the safe area and the minimum coordination number helps the simulation proceed smoothly. The parameters set by the pair_coeff command are used to define the interaction between different types of atoms in the simulation system. In this embodiment, it is specifically the interaction between the four types of atoms: C, H, O and Si.

[0034] In addition, in order to ensure the stability and rationality of the simulation, it is also necessary to use the fix command to set the dynamic calculation of the charge balance and handle overlapping atoms; use the group command to group different atom types; use the fix zwalls command to set reflection boundaries for water molecules and hydrogen peroxide molecules. Specifically, overlapping atoms will affect the stability of the simulation. Handling overlapping atoms can better simulate the actual processing environment and make the simulation results closer to the actual processing process; different atom types are grouped to form different atomic groups, and classification by atomic group can facilitate subsequent calculations and file output about atomic groups; in this embodiment, water molecules and hydrogen peroxide molecules are divided into one type of atomic group, and setting reflection boundaries for water molecules and hydrogen peroxide molecules can allow water molecules and hydrogen peroxide molecules to fully react with the silicon carbide substrate model 3, making the simulation effect closer to the experimental state.

[0035] Step 3: establishing a diamond abrasive grain model 1 in the simulation system, wherein the diamond abrasive grain model 1 is arranged above the silicon carbide substrate model 3 .

[0036] In the specific implementation, the region command is used to define the shape of the diamond abrasive model 1 as a sphere with a lattice constant of 3.567, and the diamond abrasive model 1 is placed above the silicon carbide substrate model 3. Further, the diamond abrasive model 1 is established using LAMMPS software, which is different from the modeling stage and modeling software used in the hydrogen peroxide solution model 2 and the silicon carbide substrate model 3. The advantage of this is that the shape, size and placement of the diamond abrasive model 1 can be directly changed in the LAMMPS software, which is more convenient to operate and can better simulate the actual processing environment, so that the simulation results are closer to the actual processing process.

[0037] Step 4, dividing the silicon carbide substrate model 3 into three regions, namely, a fixed layer, a constant temperature layer and a Newtonian layer, from bottom to top along the Z-axis direction of the coordinate system, so that the atoms in the fixed layer remain stationary and the simulation system is relaxed.

[0038] In the specific implementation, according to the divided areas of the silicon carbide substrate model 3, the region and group commands are combined to create atomic groups in the three areas respectively; the setforce command of the fix statement is used to always set the atomic force of the fixed layer to 0, so that the atoms in the fixed layer remain stationary during the simulation, ensuring that the atoms in the fixed layer will not move or be affected by external forces; the minimize command is used to perform energy minimization; and the simulation system is relaxed under the NVT ensemble.

[0039] Furthermore, creating atomic groups in three regions can facilitate the setting of atoms according to the three regions. In this embodiment, it is mainly convenient to set the atoms of the fixed layer so that the atoms of the fixed layer remain stationary during the simulation process; energy minimization is performed before relaxation to ensure the accuracy of the simulation. The NVT ensemble facilitates keeping the temperature of the simulation system constant, so that the temperature of the simulation system is set at 298K, and relaxation is performed under the NVT ensemble to simulate the actual processing environment of oxidation of silicon carbide by hydrogen peroxide solution at room temperature, so that the simulation results are closer to the actual processing process.

[0040] In order to monitor the relaxation process and provide a basis for analyzing the atomic removal mechanism in the chemical corrosion stage during chemical mechanical polishing, the following operations are also performed: use the thermo command to output thermodynamic parameters, and output them every 100 steps; use the thermo_style command to customize the output method, and the output thermodynamic parameters are step, temp, press, lx, ly, and lz, which represent the number of simulation steps, temperature, stress, length in the X direction, length in the Y direction, and length in the Z direction, respectively; use the fix command to calculate the number of bonds between C, H, O, and Si atoms, and output the results to the file bonds.out; use the fix command to calculate the number of different species, and output the results to the file species.out; use the dump command to output the EQ.lammpstrj file to obtain the positional mechanical parameter information during the relaxation process; import the EQ.lammpstrj file into the software The Open Visualization Tool, i.e., the visualization software OVITO, to view the simulation results.

[0041] Among them, the simulation step number can be used to monitor the progress of relaxation, the temperature is used to monitor the temperature of the simulation system, the stress is used to monitor the force of the silicon carbide substrate model 3, and the length in the X direction, the length in the Y direction, and the length in the Z direction are used to monitor the size of the simulation system; in this embodiment, the number of oxidation bonds between atoms is mainly monitored to determine the degree of oxidation of the silicon carbide substrate model 3 by the hydrogen peroxide solution model 2, and different species refer to various types of atomic composition composed of different atoms during the relaxation process. In this embodiment, the number of water and hydrogen peroxide is mainly monitored to determine the degree of oxidation of the silicon carbide substrate model 3 by the hydrogen peroxide solution model 2; the positional mechanical parameter information is used to determine the atomic state change of the silicon carbide substrate model 3 during the relaxation process. The position information in the positional mechanical parameter information is based on the data file with the coordinate system established in step 1, and the mechanical information in the positional mechanical parameter information is based on the reaction force field potential function SiCHO.ffield.reax selected in step 2. The EQ.lammpstrj file is a data file. In order to more intuitively describe the atomic state of the silicon carbide substrate model 3 during the relaxation process, the data file is visualized and analyzed using visualization software to facilitate the analysis of the atomic-level removal mechanism of the silicon carbide substrate model 3.

[0042] Step 5, control the diamond abrasive model 1 to press down to a set depth with a set pressing parameter, then scratch at a set moving speed in the X-axis direction of the coordinate system, and at the same time perform mechanical vibration along a sinusoidal function trajectory in the Z-axis direction of the coordinate system to polish the silicon carbide substrate model 3.

[0043] In the specific implementation, the fix move command is used to apply a constant horizontal speed to the diamond abrasive model 1 in the X-axis direction, and mechanical vibration is applied in the Z-axis direction, so that the diamond abrasive model 1 is scratched at a constant speed in the X-direction of the surface of the silicon carbide substrate model 3, and an output file containing atomic coordinate information is obtained; the moving speed, nano-polishing time, amplitude and frequency of the diamond abrasive model 1 are defined by the variable command; the moving speed of the diamond abrasive model 1 in the X-axis direction is set to 1Å / ps by the variable command, and the scratching distance of the diamond abrasive model 1 in the X-axis direction is set to 120Å by the variable command; the system is simulated under the NVE ensemble. Among them, the displacement of the diamond abrasive model 1 in the X-axis direction and the Z-axis direction is represented by X= vt, Z = Asin(2πft), where v, t, A, and f are the moving speed, nano-polishing time, amplitude and frequency of the diamond abrasive model 1, respectively.

[0044] Furthermore, the diamond abrasive model 1 vibrates and scratches the silicon carbide substrate model 3 along the X-axis at a moving speed of 1Å / ps within a range of 0-120Å. The output file of the atomic coordinate information includes the positional mechanical parameter information when the hydrogen peroxide solution model 2 oxidizes the silicon carbide substrate model 3, and also includes the positional mechanical parameter information when the diamond abrasive model 1 scratches in the X-axis direction and mechanically vibrates along the sine function trajectory in the Z-axis direction.

[0045] Furthermore, by adding a diamond abrasive model 1 that vibrates in a sinusoidal trajectory along the Z-axis direction above the Si surface model of 4H-SiC, the vibration-assisted model research of the Si surface model of 4H-SiC in the hydrogen peroxide solution model 2 is realized. The vibration direction of the diamond abrasive model 1 is more consistent with the movement direction of the abrasive in the actual chemical mechanical polishing process. The vibration-assisted model can more realistically simulate the force and movement trajectory of the abrasive in the polishing process, and make the simulation process closer to the actual processing technology, which is convenient for providing guidance for optimizing the ultrasonic-assisted silicon carbide chemical mechanical polishing process.

[0046] Step 6, outputting the positional mechanical parameters of the diamond abrasive model 1 vibrating and scratching the silicon carbide substrate model 3, and importing the positional mechanical parameters into OVITO visualization software for visualization analysis.

[0047] In the specific implementation, the dump command is used to output the Move.lammpstrj file to obtain the positional mechanical parameter information of the diamond abrasive model 1 vibrating and scratching the silicon carbide substrate model 3; the Move.lammpstrj file is imported into the software The OpenVisualization Tool, i.e., the visualization software OVITO, to view the simulation results.

[0048] Furthermore, the position information in the position mechanical parameter information is based on the data file with the coordinate system established in step 1, and the mechanical information in the position mechanical parameter information is based on the reaction force field potential function SiCHO.ffield.reax selected in step 2. The Move.lammpstrj file is a data file. In order to more intuitively describe the atomic state of the silicon carbide substrate model 3 during the process of the diamond abrasive model 1 vibrating and scratching the silicon carbide substrate model 3, the data file is visualized and analyzed using visualization software, which is convenient for analyzing the atomic-level removal mechanism of the silicon carbide substrate model 3.

[0049] Furthermore, by outputting the surface morphology of the Si surface model of 4H-SiC scratched by the diamond abrasive model 1, the mechanism of oxidation, removal and damage of the Si surface of 4H-SiC can be clearly and intuitively described. The vibration-assisted model established by the present invention is convenient for changing the vibration parameters and scratching parameters of the diamond abrasive model 1, which is helpful for studying the mechanism of oxidation, removal and damage of the Si surface of 4H-SiC under the conditions of ultrasonic vibration of different vibration amplitudes and frequencies along the Z-axis direction and different scratching depths along the Z-axis direction of the diamond abrasive model 1, and is convenient for combining ultrasonic vibration with chemical mechanical polishing technology to improve the polishing quality of silicon carbide.

[0050] In order to obtain more basis for analyzing and studying atomic-level removal mechanism, the following parameters are outputted together with positional mechanical parameter information. The specific operations are as follows: use compute, fix, and variable commands in the in file to calculate the effective stress of atoms; use compute and rdf commands in the in file to calculate the radial distribution function and output the result to the tmp.rdf file; use compute command to calculate the forces of diamond abrasive model 1 in the X-axis direction, Y-axis direction, and Z-axis direction, use variable command to define the forces of diamond abrasive model 1 in the X-axis direction, Y-axis direction, and Z-axis direction, and use fixprint command to output and save diamond abrasive model 1. Stress state in three directions; use the fix command to calculate the number of bonds between C, H, O and Si atoms, and output the results to the file bonds.out; use the fix command to calculate the number of different species, and output the results to the file species.out; use the thermo command to output thermodynamic parameters, output once every 500 steps; use the thermo_style command to customize the output method, the output thermodynamic parameters are step, temp, press, lx, ly, lz, epair, etotal, which represent the number of simulation steps, temperature of the simulation system, stress, length in the X direction, length in the Y direction, length in the Z direction, potential energy, and total energy of the simulation system, respectively.

[0051] In this embodiment, the effective stress of atoms in the silicon carbide substrate model 3, the radial distribution function, the stress state of the diamond abrasive model 1 in the X-axis direction, the stress state of the diamond abrasive model 1 in the Y-axis direction, the stress state of the diamond abrasive model 1 in the Z-axis direction, the number of oxidation bonds, the amount of water and hydrogen peroxide, and thermodynamic parameters are mainly analyzed.

[0052] Step 7, respectively changing the pressing parameters of the diamond abrasive model 1, the amplitude and frequency parameters of the mechanical vibration of the diamond abrasive model 1, and repeating steps 5 and 6.

[0053] In the specific implementation, by changing the downward pressure parameters of the diamond abrasive model 1, changing the initial position coordinate parameters of the diamond abrasive model 1, repeating steps 5 and 6, and performing repeated calculations, the position mechanical parameter information at different scratching depths is obtained; by changing the amplitude and frequency parameters of the mechanical vibration of the diamond abrasive model 1, repeating steps 5 and 6, and performing repeated calculations, the position mechanical parameter information at different amplitudes and frequencies is obtained.

[0054] The surface morphology of diamond abrasive model 1 scratching silicon carbide substrate model 3 determines the surface roughness after polishing. The scratch morphology results are obtained by visual analysis using OVITO visualization software. Figure 3 As shown, as the frequency of mechanical vibration of diamond abrasive model 1 increases, the scratch surface becomes flatter, which is conducive to obtaining a better surface morphology.

[0055] The present invention simulates the process of introducing the vibration behavior of the diamond abrasive model 1 along the Z-axis direction into the chemical mechanical polishing, which is helpful to study the oxidation, removal and damage mechanism of silicon carbide under the action of chemical solution and the vibration of the diamond abrasive model 1 along the Z-axis direction.

[0056] The details not described in detail in the present invention are all conventional technical means known to those skilled in the art.

[0057] The above content shows and describes the basic principles, main features and beneficial effects of the present invention. The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A simulation method for abrasive vibration machining based on reactive molecular dynamics, characterized in that: The following steps are involved: Step 1: establishing a simulation system with a coordinate system consisting of a hydrogen peroxide solution model (2) and a silicon carbide substrate model (3), and setting boundary conditions for the simulation system; Step 2: Selecting a reaction force field potential function that describes the bonding, bond breaking and interaction between atoms in the simulation system; Step 3: establishing a diamond abrasive grain model (1) in the simulation system, wherein the diamond abrasive grain model (1) is arranged above the silicon carbide substrate model (3); Step 4: Divide the silicon carbide substrate model (3) into three regions, namely, a fixed layer, a constant temperature layer and a Newtonian layer, from bottom to top along the Z-axis direction of the coordinate system, so that the atoms in the fixed layer remain stationary and the simulation system is relaxed; Step 5: Controlling the diamond abrasive model (1) to press down to a set depth with a set pressing parameter, and then rubbing at a set moving speed in the X-axis direction of the coordinate system, while mechanically vibrating along a sinusoidal function trajectory in the Z-axis direction of the coordinate system to polish the silicon carbide substrate model (3); Step 6: Output the positional mechanical parameters of the diamond abrasive model (1) and the vibration scratching silicon carbide substrate model (3), and perform visual analysis by importing the positional mechanical parameters into OVITO visualization software; Step 7: respectively change the pressing parameters of the diamond abrasive model (1), the amplitude and frequency parameters of the mechanical vibration of the diamond abrasive model (1), and repeat steps 5 and 6.

2. The simulation method of abrasive vibration machining based on reactive molecular dynamics according to claim 1, characterized in that: The boundary conditions in step 1 are: setting periodic boundary conditions in the X-axis direction and the Y-axis direction of the coordinate system, and setting fixed boundary conditions in the Z-axis direction of the coordinate system.

3. The simulation method of abrasive vibration machining based on reactive molecular dynamics according to claim 1, characterized in that: The shape of the diamond abrasive grain model (1) in step 1 is spherical, and the lattice constant is 3.

567.

4. The simulation method of abrasive vibration machining based on reactive molecular dynamics according to claim 1, characterized in that: In the step 1, the silicon carbide substrate model (3) is a Si surface model of 4H-SiC.

5. The simulation method of abrasive vibration machining based on reactive molecular dynamics according to claim 1, characterized in that: The reaction force field potential function of step 2 is SiCHO.ffield.reax.

6. The simulation method of abrasive vibration machining based on reactive molecular dynamics according to claim 5, characterized in that: The step 2 first sets the safe region and the minimum coordination number of the molecular dynamics simulation; and then defines the bonding, bond breaking and interaction between the atoms in the simulation system by setting the parameters of the SiCHO.ffield.reax.

7. The simulation method of abrasive vibration machining based on reactive molecular dynamics according to claim 1, characterized in that: The simulation system described in step 4 is relaxed under the NVT ensemble.

8. The simulation method of abrasive vibration machining based on reactive molecular dynamics according to claim 7, characterized in that: The output parameters of the relaxation process in step 4 include: thermodynamic parameters, calculation of the number of bonds between atoms, the number of different species, and positional mechanical parameters. The progress of relaxation is determined based on the output parameters of the relaxation process.

9. The simulation method of abrasive vibration machining based on reactive molecular dynamics according to claim 1, characterized in that: The simulation system described in step 5 is polished under the NVE ensemble.

10. The simulation method of abrasive vibration machining based on reactive molecular dynamics according to claim 9, characterized in that: The scratching trajectory of the diamond abrasive model (1) in the X-axis direction in step 5 is X = vt, and the sinusoidal function trajectory of the diamond abrasive model (1) in the Z-axis direction is Z = Asin(2πft), and v, t, A, and f are the moving speed, nanopolishing time, amplitude, and frequency of the diamond abrasive model (1), respectively.

Images