A numerical simulation method for neutral gas capacitively coupled plasma discharge

By using the capacitive coupling numerical simulation method of neutral gas during plasma discharge, combined with DSMC and PIC methods, the neutral particle motion and electric field distribution are simulated, and the problem of difficult to simulate the plasma discharge process in the prior art is solved, and the problem of achieving the High-precision particle attribute distribution simulation improves the accuracy and practical value of the process simulation model.

CN115017788BActive Publication Date: 2024-11-22SOUTHEAST UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210712488.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-06-22
Publication Date
2024-11-22
Estimated Expiration
2042-06-22

AI Technical Summary

Technical Problem

The prior art is difficult to effectively simulate the particle energy, velocity and angle distribution during low-temperature plasma discharge process. Especially in complex vacuum low-voltage environments, it is difficult to directly solve the Boltzmann equation, and the existing methods are accurate when expanding the chemical reaction module. and insufficient practical value.

Method used

A numerical simulation method for capacitively coupled plasma discharge of neutral gas is proposed. By constructing a corresponding simulation model for neutral gas in the chamber, combining DSMC and PIC methods, the movement of neutral particles is simulated and the electric field distribution is calculated to promote charged particles. until the steady state is reached, the attribute distribution of the particles is obtained.

Benefits of technology

The high-precision simulation of particle energy, velocity and angle distribution is achieved, and the simulation results are closer to the actual physical process, which improves the accuracy and practical value of the process simulation model, and is suitable for plasma discharge simulation of different neutral gases.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115017788B_ABST
    Figure CN115017788B_ABST
Patent Text Reader

Abstract

The invention discloses a numerical simulation method for neutral gas capacitive coupled discharge plasma. The method comprises: step 1, constructing a corresponding neutral particle simulation model according to the flow of neutral gas in a chamber, simulating the movement of neutral particles by using a direct simulation Monte Carlo method, and outputting flow field information; step 2, constructing a corresponding charged particle simulation model according to charged particles generated by the capacitive coupled discharge of neutral gas, and using a particle grid method to initialize the charged particles to be uniformly distributed in the simulation space; step 3, using the Poisson equation to calculate the potential and electric field distribution on the grid points; step 4, adding the electric field on the grid points to the particle position through a weight function to promote the movement of the particles; step 5, using the Monte Carlo method to process the collision between particles; step 6, repeating steps 2 to 5, updating the grid point electric field and particle properties in each time step, until a steady state is reached, and outputting the distribution of particle energy, velocity, angle, etc.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the field of micro-electromechanical system processing simulation, and in particular relates to a numerical simulation method for neutral gas capacitively coupled plasma discharge. Background Art

[0002] Gas discharge is the main way to generate plasma in laboratories and industries. According to the different gas discharge devices, plasma can be divided into capacitively coupled plasma, inductively coupled plasma, microwave plasma, etc. Since low-temperature plasma discharge is very complex and the reaction process includes a lot of physical and chemical reactions, it is difficult to conduct a comprehensive analysis of its reaction mechanism. By building a model of the cavity and performing numerical simulation on the computer, it is an important means to understand the plasma discharge process. By simulating the particle angle and energy distribution under different process conditions, the accuracy and practical value of the process simulation model can be further improved.

[0003] Since the working environment of complex processing gases is mostly vacuum and low pressure, the Knudsen number is greater than 0.1, and the fluid control equation becomes the Boltzmann equation. It is difficult to directly solve the Boltzmann equation containing collision terms. The indirect solution using the DSMC method can not only obtain extremely high accuracy, but also expand the chemical reaction module in the future. At the same time, it has the potential to couple with the existing PIC model at the microscopic level. Therefore, by first building a model for the neutral gas in the chamber and then coupling it with the plasma simulation, the accuracy and practical value of the process simulation model can be further improved. Summary of the invention

[0004] Purpose of the invention: In view of the above-mentioned prior art, a numerical simulation method of neutral capacitively coupled plasma discharge is proposed to simulate the distribution of energy, velocity, angle, etc. of charged particles in etching and deposition processes.

[0005] Technical solution: The present invention proposes a numerical simulation method for neutral gas capacitively coupled plasma discharge, which comprises the following steps:

[0006] Step 1: construct a corresponding neutral particle simulation model for the flowing neutral gas in the chamber, and establish a grid model for the chamber and perform grid division to simulate the movement of neutral particles and obtain flow field information;

[0007] Step 2: construct a charged particle simulation model for the charged particles generated by the capacitive coupled discharge of the neutral gas, re-grid the chamber, and initialize the charged particles with initial velocities to be evenly distributed in the grid space;

[0008] Step 3: At the current moment, the charge of the charged particles in each grid is distributed to the grid points around the grid through the weight function, and a driving voltage is applied to the bottom boundary of the chamber model to calculate the electric field distribution on the grid points;

[0009] Step 4: The electric field on the grid point is added to the charged particles in the corresponding grid through the weight function, so as to promote the movement of the charged particles in unit time and update the speed and position information of the charged particles;

[0010] Step 5: traverse all charged particles to determine whether charged particles collide with neutral particles. If a collision occurs, update the velocity information of the charged particles after the collision.

[0011] Step 6: Repeat steps 2 to 5, updating the electric field of the grid points and the velocity and position information of the charged particles in each unit time until a steady state is reached, and obtaining the property distribution of the particles.

[0012] Furthermore, the properties of the particles include one or more of particle energy, velocity, and an angle between the particle velocity direction and the vertical direction.

[0013] Furthermore, the meshing method in step 1 is the DSMC method and / or the meshing method in step 2 is the PIC method.

[0014] Furthermore, in step 1, the flow field information is obtained by statistically weighting the neutral particle information in the current grid:

[0015]

[0016]

[0017]

[0018] Among them, ρ is the neutral gas concentration, V is the volume of the grid where the neutral particles are located, m is the mass of the neutral particles, and n is the number of neutral particles in the grid; is the flow velocity, is the velocity of the pth particle; P is the pressure in the grid, and σ(V) represents the mean square deviation of the velocities of all neutral particles in the grid.

[0019] Furthermore, in step 1, when the DSMC method is used to establish a grid model for the chamber, the size of the grid is the mean free path of the neutral particles. And the grid considers different boundary conditions. For the solid boundary, the diffuse reflection model is used to describe the influence of the chamber wall on the neutral gas flow. The neutral particles moving to the solid boundary are assigned a new random velocity:

[0020]

[0021] V t =V n ·cosA (5)

[0022] V s =V n ·sinA (6)

[0023]

[0024] Among them, V n is the velocity component of the neutral particle in the normal direction of the boundary, V t , V s is the velocity component on the parallel boundary, k B is the Boltzmann constant, T wall is the set temperature of the solid boundary, R1 and R2 are random numbers ranging from 0 to 1;

[0025] For the inlet and outlet boundaries, the neutral particle inlet and outlet flow field is considered to satisfy the Maxwell rate distribution:

[0026]

[0027]

[0028] Among them, N in is the inlet and outlet neutral gas flow rate, n g is the neutral gas concentration at the inlet and outlet, A1 is the inlet and outlet area, erf is the error function, θ is the angle between the inlet and outlet normal vector and the velocity direction, is the velocity sampling amplitude of the flow field at the boundary, T in is the set temperature of the inlet and outlet boundaries.

[0029] Furthermore, in step 2, when the chamber is re-meshed, the length of each grid does not exceed the Debye length of the charged particle, and based on the axial symmetry of the cylindrical chamber, the coordinates in the grid are all in the form of cylindrical coordinates, and the entire simulation area is a two-dimensional axisymmetric RZ plane. At the beginning of the simulation, the charged particles are randomly and evenly distributed in the grid space:

[0030]

[0031] z=R4·LZ (11)

[0032] Among them, r and z are the random positions of the charged particles in the radial and axial directions, LR and LZ are the maximum lengths in the corresponding directions, and R3 and R4 are random numbers between 0 and 1.

[0033] Furthermore, the initial velocity of the charged particles is randomly sampled from the Maxwell distribution.

[0034] Furthermore, in step 3, the charges of the charged particles need to be distributed to the grid points of the corresponding grid, and all charged particles are traversed. For each charged particle, there is:

[0035]

[0036] Among them, r p ,z p are the radial and axial positions of the pth charged particle, dr and dz are the grid lengths in the corresponding directions, i and j are the integer values ​​of di and dj, respectively. The four grid points corresponding to the particle are (i, j), (i+1, j), (i, j+1) and (i+1, j+1), hr represents the distance of the charged particle from the grid point (i, j) in the r direction, and hz represents the distance of the charged particle from the grid point (i, j) in the z direction.

[0037] The charges of charged particles in the grid are distributed to the corresponding grid points through the following weight function:

[0038]

[0039] Among them, chg(i,j), chg(i+1,j), chg(i,j+1), chg(i+1,j+1) represent the accumulated charges at the grid points (i,j), (i+1,j), (i,j+1) and (i+1,j+1), and their initial values ​​are zero.

[0040] Furthermore, in step 4, the electric field at the location of the charged particle is obtained by summing the electric fields at the four corresponding grid points (i, j), (i+1, j), (i, j+1) and (i+1, j+1) around it through the weight function:

[0041]

[0042] Among them, E p represents the electric field at the location of the pth charged particle, E(i,j), E(i+1,j), E(i,j+1), (i+1,j+1) represent the electric field at the grid points (i,j), (i+1,j), (i,j+1) and (i+1,j+1).

[0043] Furthermore, the method of step 5 is as follows:

[0044] (1) Combined with the neutral gas concentration ρ obtained in the neutral particle simulation model, the probability of collision between neutral particles and charged particles is calculated according to the following formula:

[0045] P1=1-exp(-ρv′σ T Δt / m) (15)

[0046] Among them, P1 represents the collision probability, v′ represents the velocity difference between the charged particle and the neutral particle, σ T represents the collision cross-sectional area, Δt represents the unit time step, and m is the mass of the neutral particle;

[0047] (2) Generate a random number R5, compare the size of the random number R5 and the collision probability P. If R5 < P1, it represents a collision;

[0048] (3) If a collision occurs, calculate and update the velocity of the charged particle after the collision.

[0049] Beneficial effects: By adopting the above technical means, the present invention has the following beneficial effects compared with the prior art:

[0050] (1) The present invention adopts a method based on the first principle - PIC. By tracking and calculating the motion of a large number of charged particles, the macroscopic material properties and motion laws of capacitively coupled discharge plasmas are obtained, which is equivalent to directly solving the Boltzmann equation. This method has fewer assumptions and constraints, can fully consider the motion properties of plasmas, and the simulation results are closer to the actual physical process.

[0051] (2) The coordinates of the present invention are in the form of two-dimensional axisymmetric cylindrical coordinates. Calculating three-dimensional results in two-dimensional form reduces the amount of calculation.

[0052] (4) The present invention uses the DSMC method to calculate the concentration distribution of neutral gas in the chamber, which can further improve the accuracy of the PIC method.

[0053] (5) The present invention has high precision and high flexibility, is easy to parallel compute, and is suitable for plasma discharge simulations of different neutral gases. By comparing the simulation results with the experimental results, it can be concluded that the present invention can accurately simulate the distributions of particle energy, velocity, angle, etc. during capacitively coupled plasma discharge. Description of the Drawings

[0054] Figure 1 is a schematic diagram of a plasma simulation model based on the PIC - MCC method.

[0055] Figure 2 is the comparison result of the particle angle distribution of capacitively coupled plasma discharge of the present invention and the open - source software XOOPIC. Detailed Embodiments

[0056] The following further explains the present invention with reference to the drawings.

[0057] Such as Figure 1The figure shows a schematic diagram of the plasma simulation model based on the PIC-MCC method. The dashed line represents the simulation area, the solid line represents the plasma area inside the chamber, and the area outside the chamber represents the air domain. The outer wall of the chamber is grounded, that is, the potential of the Poisson equation boundary condition is always zero. The solid line below represents the sample base, and an RF voltage is applied to the base to stimulate neutral gas discharge by capacitive coupling. The discharge gas flows out through the gas outlet next to the base.

[0058] The present invention provides a numerical simulation method for capacitively coupled plasma discharge, the method comprising the following steps:

[0059] Step 1: construct a corresponding neutral particle simulation model for the flowing neutral gas in the chamber, and establish a grid model for the chamber and perform grid division to simulate the movement of neutral particles and obtain flow field information;

[0060] Step 2: construct a charged particle simulation model for the charged particles generated by the capacitive coupled discharge of the neutral gas, re-grid the chamber, and initialize the charged particles with initial velocities to be evenly distributed in the grid space;

[0061] Step 3: At the current moment, the charge of the charged particles in each grid is distributed to the grid points around the grid through the weight function, and a driving voltage is applied to the bottom boundary of the chamber model to calculate the electric field distribution on the grid points;

[0062] Step 4: The electric field on the grid point is added to the charged particles in the corresponding grid through the weight function, so as to promote the movement of the charged particles in unit time and update the speed and position information of the charged particles;

[0063] Step 5: traverse all charged particles, calculate the collision probability between charged particles and neutral particles according to the neutral gas concentration distribution extracted from the flow field information, generate a random number to determine whether a collision occurs, and if a collision occurs, update the velocity information of the charged particles after the collision;

[0064] Step 6: Repeat steps 2 to 5, update the electric field of the grid points and the velocity and position information of the charged particles in each unit time, until the concentration of the charged particles remains within a preset concentration range within a preset time, and obtain the property distribution of the particles.

[0065] As a specific embodiment of the present invention, the properties of the particle include one or more of particle energy, velocity, and an angle between the particle velocity direction and the vertical direction.

[0066] As a specific embodiment of the present invention, the grid division method in step 1 is the DSMC method and / or the grid division method in step 2 is the PIC method.

[0067] As a specific embodiment of the present invention, in step 1, the flow field information is obtained by statistically weighting the neutral particle information in the current grid:

[0068]

[0069]

[0070]

[0071] Among them, ρ is the neutral gas concentration, V is the volume of the grid where the neutral particles are located, m is the mass of the neutral particles, and n is the number of neutral particles in the grid; is the flow velocity, is the velocity of the pth particle; P is the pressure in the grid, and σ(v) represents the mean square deviation of the velocities of all neutral particles in the grid.

[0072] As a specific embodiment of the present invention, in step 1, when the DSMC method is used to establish a grid model for the chamber, the size of the grid is the mean free path of the neutral particles. And the grid considers different boundary conditions. For the solid boundary, the diffuse reflection model is used to describe the influence of the chamber wall on the neutral gas flow. The neutral particles moving to the solid boundary are given a new random velocity:

[0073]

[0074] V t =V n ·cosA (5)

[0075] V s =V n ·sinA (6)

[0076]

[0077] Among them, V n is the velocity component of the neutral particle in the normal direction of the boundary, V t , V s is the velocity component on the parallel boundary, k B is the Boltzmann constant, T wall is the set temperature of the solid boundary, R1 and R2 are random numbers ranging from 0 to 1;

[0078] For the inlet and outlet boundaries, the neutral particle inlet and outlet flow field is considered to satisfy the Maxwell rate distribution:

[0079]

[0080]

[0081] Among them, N in is the inlet and outlet neutral gas flow rate, n g is the neutral gas concentration at the inlet and outlet, A1 is the inlet and outlet area, erf is the error function, θ is the angle between the inlet and outlet normal vector and the velocity direction, is the velocity sampling amplitude of the flow field at the boundary, T in is the set temperature of the inlet and outlet boundaries.

[0082] As a specific embodiment of the present invention, in step 2, when the chamber is re-meshed, the length of each grid does not exceed the Debye length of the charged particle, and based on the axial symmetry of the cylindrical chamber, the coordinates in the grid are all in the form of cylindrical coordinates, and the entire simulation area is a two-dimensional axisymmetric RZ plane. At the beginning of the simulation, the charged particles are randomly and evenly distributed in the grid space:

[0083]

[0084] z=R4·LZ (11)

[0085] Among them, r and z are the random positions of the charged particles in the radial and axial directions, LR and LZ are the maximum lengths in the corresponding directions, and R3 and R4 are random numbers between 0 and 1.

[0086] As a specific embodiment of the present invention, the initial velocity of the charged particle is generated by random sampling from the Maxwell distribution.

[0087] As a specific embodiment of the present invention, in step 3, the charges of the charged particles need to be distributed to the grid points of the corresponding grid, and all the charged particles are traversed. For each charged particle, there is:

[0088]

[0089] Among them, r p ,z p are the radial and axial positions of the pth charged particle, dr and dz are the grid lengths in the corresponding directions, i and j are the integer values ​​of di and dj, respectively. The four grid points corresponding to the particle are (i, j), (i+1, j), (i, j+1) and (i+1, j+1), hr represents the distance of the charged particle from the grid point (i, j) in the r direction, and hz represents the distance of the charged particle from the grid point (i, j) in the z direction.

[0090] Considering the radial non-uniformity of the grid volume in cylindrical coordinates, the charges of charged particles in the grid are distributed to the corresponding grid points through the following weight function:

[0091]

[0092] Among them, chg(i,j) represents the accumulated charge amount at the lattice point (i,j), and its initial value is zero.

[0093] As a specific embodiment of the present invention, in step 4, the electric field at the position of the charged particle is obtained by accumulating the electric fields at the corresponding four surrounding lattice points (i,j), (i+1,j), (i,j+1), and (i+1,j+1) through a weight function:

[0094]

[0095] Among them, E p represents the electric field at the position of the p-th charged particle, and E(i,j) represents the electric field at the lattice point (i,j).

[0096] As a specific embodiment of the present invention, in step 5, it includes the following steps:

[0097] (1) Combining the neutral gas concentration ρ obtained from the neutral particle simulation model, calculate the probability of collision between neutral particles and charged particles according to the following formula:

[0098] P1 = 1 - exp(-ρv′σ T Δt / m) (15)

[0099] Among them, P1 represents the collision probability, v′ represents the velocity difference between the charged particle and the neutral particle, σ T represents the collision cross-sectional area, Δt represents the unit time step, and m is the mass of the neutral particle;

[0100] (2) Generate a random number R5, compare the size of the random number R5 and the collision probability P1. If R5 < P1, it means a collision occurs;

[0101] (3) If a collision occurs, calculate and update the velocity of the charged particle after the collision.

[0102] Figure 2 is the output of the ion angular distribution obtained by (a) the method used in the present invention and (b) running the open-source code XOOPIC. The initial electron and ion densities are both 5.12e14 m-3, the time step is 1.77e-11 s, the Ar concentration is 9.64e21 m-3, the electron temperature is 2.59 eV, the ion temperature is 0.0259 eV, the grid size is 1.98e-4, and the Debye length is 5.29e-4 m. The present invention compares the simulation results with the simulation results of the open-source code XOOPIC widely used for comparison and verification to confirm the model accuracy. It is found through verification that the simulation results are relatively consistent with the open-source code and can be used for numerical simulation of capacitive discharge plasmas.

[0103] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A numerical simulation method for neutral gas capacitively coupled plasma discharge, It is characterized in that The steps include: Step 1: construct a corresponding neutral particle simulation model for the flowing neutral gas in the chamber, establish a grid model for the chamber and perform grid division to simulate the movement of neutral particles and obtain flow field information; Step 2: construct a charged particle simulation model for the charged particles generated by the capacitive coupled discharge of the neutral gas, re-grid the chamber, and initialize the charged particles with initial velocities to be evenly distributed in the grid space; Step 3: At the current moment, the charge of the charged particles in each grid is distributed to the grid points around the grid through the weight function, and a driving voltage is applied to the bottom boundary of the chamber model to calculate the electric field distribution on the grid points; Step 4: The electric field on the grid point is added to the charged particles in the corresponding grid through the weight function, so as to promote the movement of the charged particles in unit time and update the speed and position information of the charged particles; Step 5: traverse all charged particles to determine whether charged particles collide with neutral particles. If a collision occurs, update the velocity information of the charged particles after the collision. Step 6: Repeat steps 2 to 5, update the electric field of the grid points and the speed and position information of the charged particles in each unit time until a steady state is reached, and obtain the property distribution of the particles; In step 3, the charges of the charged particles need to be distributed to the grid points of the corresponding grid, and all charged particles are traversed. For each charged particle, there is: Among them, r p ,z p are the radial and axial positions of the pth charged particle, dr and dz are the grid lengths in the corresponding directions, i and j are the integer values ​​of di and dj, respectively. The four grid points corresponding to the particle are (i, j), (i+1, j), (i, j+1) and (i+1, j+1), hr represents the distance of the charged particle from the grid point (i, j) in the r direction, and hz represents the distance of the charged particle from the grid point (i, j) in the z direction. The charges of charged particles in the grid are distributed to the corresponding grid points through the following weight function: Among them, chg(i,j), chg(i+1,j), chg(i,j+1), chg(i+1,j+1) represent the accumulated charges on the grid points (i,j), (i+1,j), (i,j+1) and (i+1,j+1), and their initial values ​​are zero; In step 4, the electric field at the location of the charged particle is obtained by summing the electric fields at the four corresponding grid points (i, j), (i+1, j), (i, j+1) and (i+1, j+1) around it through the weight function: Among them, E p represents the electric field at the location of the pth charged particle, E(i,j), E(i+1,j), E(i,j+1), (i+1,j+1) represent the electric field at the grid points (i,j), (i+1,j), (i,j+1) and (i+1,j+1).

2. A method for numerically simulating neutral gas capacitively coupled plasma discharge according to claim 1, It is characterized in that The properties of the particles include one or more of particle energy, velocity, and an angle between the particle velocity direction and the vertical direction.

3. A numerical simulation method for neutral gas capacitively coupled plasma discharge according to claim 1 or 2, It is characterized in that The meshing method in step 1 is the DSMC method and / or the meshing method in step 2 is the PIC method.

4. The method for numerical simulation of neutral gas capacitively coupled plasma discharge according to claim 1, It is characterized in that In step 1, the flow field information is obtained by statistically weighting the neutral particle information in the current grid: Among them, ρ is the neutral gas concentration, V is the volume of the grid where the neutral particles are located, m is the mass of the neutral particles, and n is the number of neutral particles in the grid; is the flow velocity, is the velocity of the pth particle; P is the pressure in the grid, and σ(v) represents the mean square deviation of the velocities of all neutral particles in the grid.

5. The method for numerical simulation of neutral gas capacitively coupled plasma discharge according to claim 4, It is characterized in that In step 1, when the DSMC method is used to establish a grid model for the chamber, the size of the grid is the mean free path of the neutral particles. And the grid considers different boundary conditions. For the solid boundary, the diffuse reflection model is used to describe the influence of the chamber wall on the neutral gas flow. The neutral particles moving to the solid boundary are given a new random velocity: In t =V n ·cosA (5) V s =V n ·sinA (6) Among them, V n is the velocity component of the neutral particle in the normal direction of the boundary, V t , V s is the velocity component on the parallel boundary, k B is the Boltzmann constant, T wall is the set temperature of the solid boundary, R 1 ,R 2 A random number between 0 and 1; For the inlet and outlet boundaries, the neutral particle inlet and outlet flow field is considered to satisfy the Maxwell rate distribution: Among them, N in is the inlet and outlet neutral gas flow rate, n g is the neutral gas concentration at the inlet and outlet, A1 is the inlet and outlet area, erf is the error function, θ is the angle between the inlet and outlet normal vector and the velocity direction, is the velocity sampling amplitude of the flow field at the boundary, T in is the set temperature of the inlet and outlet boundaries.

6. The method for numerical simulation of neutral gas capacitively coupled plasma discharge according to claim 1, It is characterized in that In step 2, when the chamber is re-meshed, the length of each grid does not exceed the Debye length of the charged particle, and based on the axial symmetry of the cylindrical chamber, the coordinates in the grid are all in cylindrical coordinate form, and the entire simulation area is a two-dimensional axisymmetric RZ plane. At the beginning of the simulation, the charged particles are randomly and evenly distributed in the grid space: z=R 4 ·LZ (11) Among them, r and z are the random positions of the charged particles in the radial and axial directions, LR and LZ are the maximum lengths in the corresponding directions, and R3 and R4 are random numbers between 0 and 1.

7. A numerical simulation method for neutral gas capacitively coupled plasma discharge according to claim 1 or 6, It is characterized in that The initial velocities of the charged particles are randomly sampled from a Maxwell distribution.

8. The method for numerical simulation of neutral gas capacitively coupled plasma discharge according to claim 1, It is characterized in that The method for step 5 is as follows: (1) Combined with the neutral gas concentration ρ obtained in the neutral particle simulation model, the probability of collision between neutral particles and charged particles is calculated according to the following formula: P1=1-exp(-ρv′σ T Δt / m) (15) Among them, P1 represents the collision probability, v′ represents the speed difference between charged particles and neutral particles, σ T represents the collision cross-sectional area, Δt represents the unit time step, and m is the mass of the neutral particle; (2) Generate a random number R 5 , compare the random number R 5 with the collision probability P1. If R 5 < P1, it means a collision has occurred. R 5 is a random number taken from between 0 and 1; (3) If a collision occurs, the velocity of the charged particle after the collision is calculated and updated.