A method for simulating light intensity distribution in thick resist photolithography process

Abstract

The present invention discloses a method for simulating the light intensity distribution of thick-resist photolithography processes. The method comprises the following steps: constructing a corresponding model in FDTD based on actual contact / proximity photolithography conditions, applying a plane wave with controllable incident angle, polarization direction, and wavelength; processing the dispersive medium using a recursive convolution method; adding a convolutional perfectly matched layer (CPML) boundary to the simulation region; performing a loop calculation to update the electric and magnetic fields at each Yee grid point within each time step until a steady state is reached; and obtaining the light intensity distribution within the photoresist using a peak detection method based on the distribution of the electromagnetic field within the grid. This method addresses the current lack of high-precision simulation models for thick-resist photolithography processes, offers high flexibility, and is suitable for simulating complex photolithography scenarios.

Description

Technical Field

[0001] The invention relates to a light intensity distribution simulation method for a thick-resist photolithography process, and belongs to the field of MEMS processing simulation. Background Art

[0002] In the MEMS (Micro-Electro-Mechanical System) field, UV (Ultraviolet) thick resist lithography has become the mainstream process for fabricating high-aspect ratio structures. UV thick resist lithography is a complex, multi-parameter process whose accuracy depends on key setup parameters, the material properties of the photoresist, and the thickness of the photoresist. With the decrease in feature size and the increase in pattern complexity, the accuracy and precision of the thick resist microstructure during the lithography process are attracting increasing attention.

[0003] Numerical simulation can accurately predict the final photoresist topography during the photolithography process and study the effects of different process parameters, such as exposure conditions, mask geometry, and incident angle. This replaces time-consuming and expensive photolithography experiments, optimizes process parameters, and improves understanding of photolithography and process design. A typical photolithography simulation process for thick resist photolithography includes aerial imaging, exposure simulation, post-bake simulation, and development simulation. The aerial imaging step, which determines the intensity distribution within the photoresist, has a significant impact on the final developed profile. Aerial imaging is primarily based on two types of models: scalar diffraction theory and rigorous electromagnetic field theory. With the decrease in feature size and the increase in pattern complexity, methods based on scalar diffraction theory are no longer applicable. Therefore, it is of great significance to employ a method based on rigorous electromagnetic field theory, FDTD (Finite Difference Time Domain), to improve the aerial imaging model and thus obtain a more accurate intensity distribution within the photoresist. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to propose a light intensity distribution simulation method for thick resist photolithography process based on dispersive medium FDTD in response to the above-mentioned prior art, so as to simulate the light intensity distribution inside the photoresist with higher precision.

[0005] The present invention adopts the following technical solutions to solve the above technical problems:

[0006] The present invention provides a method for simulating light intensity distribution in a thick-resist UV lithography process, comprising the following steps:

[0007] Step 1: Build a corresponding lithography simulation model based on the actual lithography situation, discretize the spatial domain of the electric and magnetic fields using the Yee grid, and apply a plane wave with controllable incident angle, polarization direction, and wavelength;

[0008] Step 2: Obtain the material parameters of each Yee lattice point according to the lithography simulation model, use the recursive convolution method to process different types of dispersion materials, and calculate the corresponding calculation coefficients;

[0009] Step 3: Add a convolutional perfectly matched layer (CPML) boundary to the simulation area and calculate the calculation coefficient of the CPML boundary area.

[0010] Step 4: Based on steps 2 and 3, a set of modified finite-difference time-domain FDTD electromagnetic field update equations are obtained. The electric field and magnetic field of each Yee grid point are updated in each time step through cyclic calculation until a steady state is reached.

[0011] Step 5: According to the distribution of the electromagnetic field in the grid, the peak detection method is used to obtain the light intensity distribution inside the photoresist.

[0012] Furthermore, the light intensity distribution simulation method proposed in the present invention, in step 1, constructs a lithography simulation model based on the actual situation of proximity or contact lithography, simulates the mask plate, air gap, photoresist and silicon wafer materials in the proximity or contact lithography process, uses a matrix to store corresponding information, and uses a time-harmonic field source to simulate the basic light source used in the lithography process, which can control the incident angle, polarization direction and wavelength.

[0013] Furthermore, in the light intensity distribution simulation method proposed in the present invention, in step 2, a recursive convolution method is used to improve the FDTD method to process dispersive materials. The recursive convolution method first transforms the polarizability function of the medium from the frequency domain to the time domain, then expresses the time domain expression as an exponential function, and then calculates the convolution in the time domain constitutive equation of the medium in a recursive iterative manner, thereby obtaining the improved electric field intensity update equation in FDTD.

[0014] Furthermore, the light intensity distribution simulation method proposed in the present invention uses a recursive convolution method to express the electric displacement vector as the convolution between the electric field intensity and the frequency-dependent electric susceptibility. The relationship between the electric displacement vector and the electric field intensity is:

[0015]

[0016] Where D is the electric displacement vector, E is the electric field strength, t is the time, ε0 is the dielectric constant in vacuum, χ(τ) is the electric susceptibility, and ε ∞ is the relative dielectric constant when the frequency ω→∞;

[0017] Discretize time using Yee's grid and get:

[0018]

[0019] Where n is the field value at the nth time step. Substituting the above formula into Maxwell's equations can handle dispersive media;

[0020] Different materials are suitable for different dispersion medium models. The materials used in the lithography process can be simulated by the Debye medium model and the Lorentz medium model:

[0021] The polarizability of the Debye medium is expressed as:

[0022]

[0023] Where, ε s is the dielectric constant at zero frequency; t0 is the relaxation time of the medium; j is the imaginary unit, ω is the frequency;

[0024] The polarizability of a Lorentz medium is expressed as:

[0025]

[0026] Where ω0 is the resonant frequency of the Lorentz medium; δ is the attenuation coefficient.

[0027] Furthermore, in the light intensity distribution simulation method proposed in the present invention, in step 3, a convolution perfect matching layer CPML boundary is added to the simulation area, and a convolution perfect matching layer CPML is added to the boundary area of the calculation area. The passive Maxwell equations in the CPML boundary area are corrected to:

[0028]

[0029] Where H is the magnetic field intensity vector, E is the electric field intensity, μ is the magnetic permeability, ε is the dielectric constant, j is the imaginary unit, ω is the frequency; i, j, k represent the unit vectors in the three directions; s x =s(x),s y =s(y),s z =s(z) is the coordinate scaling factor.

[0030] Furthermore, in the light intensity distribution simulation method proposed in the present invention, the coordinate scaling factor is divided into the electric field scaling factor s ei and magnetic field stretching factor s mi , which follows:

[0031]

[0032] κ ei , κ mi , α ei , α mi The PML parameter that complies with the causal law is taken as κ ei ≥1, κ mi ≥1,α ei ≥0,α mi≥0,σ max is the maximum value of the conductivity distribution; ρ is the distance between the inner region and the PML boundary to the field component; d is the thickness of the PML layer; n pml is the order of the polynomial distribution; μ0 and ε0 are the vacuum permeability and vacuum permittivity, respectively.

[0033] Furthermore, in the light intensity distribution simulation method proposed in the present invention, in step 4, a set of modified FDTD electromagnetic field update equations is obtained, including an update equation group for the non-PML region and an update equation group for the CPML boundary region; wherein, the electric field equation group also applies the recursive convolution method to process the dispersive medium, making the FDTD algorithm suitable for lithography simulation, and obtaining a new electric field update equation; wherein,

[0034] In the non-PML region, the updated equation for the new x-component electric field is:

[0035]

[0036] in,

[0037]

[0038] Among them, I n (i, j, k) is a recursive term, and its recursive expression is shown in Equation (8), where Δt is the sampling time step, ε0 is the dielectric constant in vacuum, σ is the conductivity; χ(τ) is the electrical susceptibility; ε ∞ is the relative dielectric constant when ω→∞; C rec is a recursive constant;

[0039] The updated equations for the electric field in the y and z direction components have the same form as that in the x direction;

[0040] Among them, the update equation group of the CPML boundary area, the new electric field update equation is:

[0041]

[0042] in,

[0043]

[0044] Among them, I n Same as formula (8); σ ey , κ ey , α ey is σ in formula (6) ei , κ ei , α ei , i=y; The same logic applies;

[0045] In the update equations of the CPML boundary region, the magnetic field and electric field update equations for the y and z direction components are the same as those for the x direction.

[0046] Furthermore, in the light intensity distribution simulation method proposed by the present invention, in step 5, the light intensity distribution inside the photoresist is obtained by using the peak detection method, specifically obtaining the amplitude of each point in the Yee grid, and recording the field components F at the current and previous two moments. n 、F n-1 、F n-2 , when:

[0047] (F n-1 -F n )(F n-1 -F n-2 )<0 (11)

[0048] The average of the positive and negative peak values is taken as the peak value of the field, and then the peak value of the electromagnetic field is converted into light intensity through calculation, so as to obtain the light intensity distribution of each grid point, so as to better connect with the next module of the lithography simulation model and obtain a more accurate final result.

[0049] The present invention adopts the above technical means and has the following beneficial effects compared with the prior art:

[0050] (1) The present invention adopts a method based on rigorous electromagnetic field theory - FDTD, to establish a lithography spatial image forming model, simulate the propagation process of ultraviolet light in the lithography system, and strictly solve the target field at the mask. All diffraction, refraction, interference, absorption and polarization effects are calculated in the near field of the mask without the need for approximation.

[0051] (2) The present invention adopts the recursive convolution method to process the dispersive medium in the photolithography process, making the FDTD algorithm more suitable for the photolithography simulation model.

[0052] (3) The present invention adopts the CPML boundary truncation grid to better absorb electromagnetic waves from the calculation area.

[0053] In summary, the present invention offers high precision and flexibility, facilitates parallel computation, and is relatively adaptable to various challenges in photolithography simulation. Comparing the simulation results with experimental results reveals that the present invention can accurately simulate the light intensity distribution within the photoresist layer during thick-film UV lithography. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 This is a schematic diagram of the thick resist lithography simulation model based on FDTD.

[0055] Figure 2 It is the light intensity distribution curve diagram and corresponding light intensity contour diagram of different photoresist depths when vertically incident on SU8 thick resist. DETAILED DESCRIPTION

[0056] The present invention will be further explained below with reference to the accompanying drawings and specific examples.

[0057] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those skilled in the art in the art to which the present invention belongs. It should also be understood that terms such as those defined in common dictionaries should be understood to have meanings consistent with their meanings in the context of the prior art and, unless defined as such, will not be interpreted in an idealized or overly formal sense.

[0058] The present invention proposes a method for simulating light intensity distribution in a thick resist lithography process based on a dispersive medium finite-difference time-domain method, the specific steps of which include:

[0059] Step 1: Build a corresponding lithography simulation model based on the actual lithography situation, discretize the spatial domain of the electric and magnetic fields using an equally spaced staggered grid (Yee grid), and apply a plane wave with controllable incident angle, polarization direction, and wavelength;

[0060] Step 2. Obtain the material parameters of each Yee grid point based on the lithography simulation model, use the recursive convolution method to process different materials, and calculate the corresponding calculation coefficients; use the recursive convolution method to improve FDTD to process dispersive materials to ensure the stability of the FDTD algorithm; metal materials used in the lithography process, such as chromium, have negative dielectric constants under ultraviolet light, and the traditional FDTD algorithm becomes unstable; the recursive convolution method first transforms the polarizability function of the medium from the frequency domain to the time domain, and then expresses the time domain expression as an exponential function, and then calculates the convolution in the time domain constitutive equation of the medium in a recursive iterative manner, thereby obtaining the improved electric field intensity update equation in FDTD.

[0061] Step 3: Add CPML (convolutional perfectly matched layer) boundary conditions to the calculation area and calculate the calculation coefficients of the CPML boundary area;

[0062] Step 4: Set the time step number according to the actual lithography model, perform a loop calculation, and update the electric and magnetic fields of each Yee grid point in each time step until a steady state is reached.

[0063] Step 5: Based on the distribution of the electromagnetic field in the grid, the peak detection method is used to obtain the light intensity distribution inside the photoresist.

[0064] This specific example simulates the SU8 thick photoresist lithography process, and SU8 3000 is selected as the thick photoresist in the simulation.

[0065] As a specific embodiment of the present invention, in step 1, the present invention constructs a corresponding lithography simulation model in the dispersive medium FDTD according to the actual lithography situation, such as Figure 1 The figure shows a schematic diagram of a lithography simulation model based on FDTD. The outermost simulation area is the CPML boundary region, and the entire simulation area (including the boundary region) is divided into Yee grids. The lithography model is constructed based on actual conditions, and the material parameters and calculation parameters of different materials are recorded in the Yee grid.

[0066] like Figure 1 As shown, area a represents glass, area b represents chrome, area c represents air representing the mask aperture, and area d represents the photoresist. Incident light enters the simulation region at a set angle of incidence, polarization direction, and wavelength. Within the simulation calculation region, it propagates, reflects, diffracts, refracts, interferes, and is absorbed. It is completely absorbed at the boundary region, ultimately achieving stability.

[0067] As a specific embodiment of the present invention, in step 2, the electric displacement vector is expressed as the convolution between the electric field strength and the frequency-dependent electric susceptibility. The relationship between the electric displacement vector and the electric field strength is:

[0068]

[0069] Where D is the electric displacement vector, E is the electric field intensity, t is the time, τ is the integral symbol, ε0 is the dielectric constant in vacuum; χ(τ) is the electric susceptibility; ε ∞ is the relative dielectric constant when ω→∞.

[0070] Using Yee's grid to discretize time, we get:

[0071]

[0072] Where n is the field value at the nth time step. Substituting the above equation into Maxwell's equations can handle dispersive media.

[0073] The polarizability of the Debye medium can be written as:

[0074]

[0075] Where, ε s is the dielectric constant at zero frequency; t0 is the relaxation time of the medium, j is the imaginary unit, and ω is the frequency.

[0076] The polarizability of a Lorentz medium can be written as:

[0077]

[0078] Where ω0 is the resonant frequency of the Lorentz medium; δ is the attenuation coefficient.

[0079] In step 3, a boundary region, CPML (Convolution Perfect Matching Layer), is added to the calculation region. The passive Maxwell equations in the CPML boundary region are corrected as follows:

[0080]

[0081] Where H is the magnetic field intensity vector, E is the electric field intensity, μ is the magnetic permeability, ε is the dielectric constant; i, j, k represent the unit vectors in three directions; s x =s(x),s y =s(y),s z =s(z) is the coordinate expansion factor, which can be divided into the electric field expansion factor s ei and magnetic field stretching factor s mi , which follows:

[0082]

[0083] κ ei , κ mi , α ei , α mi The PML parameter that complies with the causal law is taken as κ ei ≥1, κ mi ≥1,α ei ≥0,α mi ≥0,σ max is the maximum value of the conductivity distribution; ρ is the distance between the inner region and the PML boundary to the field component; d is the thickness of the PML layer; n pml is the order of the polynomial distribution; μ0 and ε0 are the vacuum permeability and vacuum permittivity, respectively.

[0084] Specifically, step 4 includes the following steps:

[0085] Step 4-1: Apply the FDTD algorithm to process Maxwell's equations and derive the magnetic field update equations for the non-CPML region. Taking the x-direction component as an example (the other directions are the same), the magnetic field update equations are:

[0086]

[0087] in,

[0088]

[0089] Among them, Δt is the sampling time step, dx, dy, dz represent the grid size in each direction; μ0 is the vacuum permeability, μ r is the relative magnetic permeability, σ m is the magnetic permeability; represents the magnetic field intensity component in the x direction at the sampling time (n+0.5)Δt, represents the electric field intensity component in the z direction at sampling time nΔt, Represents the electric field intensity component in the y direction at the sampling time nΔt.

[0090] Step 4-2: Apply the CPML region parameters obtained in step 3 to derive the magnetic field update equations for the CPML region. Update the magnetic field components of the CPML region at time (n+0.5)Δt, using H x Taking the field component (the other directions are the same) as an example, its CPML area is y n 、y p 、z n 、z p region, the update equation is:

[0091]

[0092] in,

[0093]

[0094] Among them, σ my , κ my , α my are all σ in formula (6) mi , κ mi , α mi , i=y; The same logic applies.

[0095] Step 4-3: Apply the FDTD algorithm to process the Maxwell equations, and combine it with the recursive convolution method described in step 2 to derive the electric field update equations for the non-CPML region. At time (n+1)Δt, update the electric field components in the non-CPML region, using E x Taking the field component as an example (the other directions have the same form), the update equation is:

[0096]

[0097] in,

[0098]

[0099] Among them, I n (i, j, k) is a recursive term, and its recursive expression is shown in Equation (12), where Δt is the sampling time step, ε0 is the dielectric constant in vacuum, σ is the conductivity; χ(τ) is the electrical susceptibility; ε ∞ is the relative dielectric constant when ω→∞; C rec is a recursive constant.

[0100] Step 4-4: Apply the CPML region parameters obtained in step 3 and combine them with the recursive convolution method in step 2 to derive the electric field update equations of the CPML region. At time (n+1)Δt, update the electric field components of the CPML region, using E x Taking the field component as an example (the other directions are the same), its CPML area is y n 、y p 、z n 、z p region, the update equation is:

[0101]

[0102] in,

[0103]

[0104] Among them, I n Same as formula (12); σ ey , κ ey , α ey is σ in formula (6) ei , κ ei , α ei , i = y; ; Δt is the sampling time step, ε0 is the dielectric constant in vacuum; The same logic applies.

[0105] Step 4-5: Calculate repeatedly and continuously update the electric and magnetic fields at each grid point until they are stable.

[0106] In step 5, the peak detection method is used to obtain the amplitude of each point in the Yee grid, and the field components F at the current and previous two moments are recorded. n 、F n-1 、F n-2 , when:

[0107] (F n-1 -F n )(F n-1 -F n-2 )<0 (15)

[0108] The average of the positive and negative peak values is taken as the peak value of the field to obtain the light intensity distribution.

[0109] According to the above steps, the light intensity distribution simulation method of thick resist lithography process based on dispersive medium finite-difference time-domain method proposed by the present invention has the following main creative features:

[0110] First, based on the FDTD lithography model using the recursive convolution method, a new set of FDTD modified electromagnetic field update equations is obtained with CPML as the boundary.

[0111] 2. Use the peak detection method to convert the instantaneous value of the field intensity into light intensity.

[0112] This invention differs from existing methods by using CPML boundaries in the lithography model. In this model, the FDTD equations are modified using recursive convolution to accommodate the dispersive dielectric chromium in the mask. Furthermore, the present invention uses CPML as the boundary condition for the calculation region, resulting in a new electromagnetic field update equation that better absorbs electromagnetic waves emitted by the calculation region, thereby minimizing the impact of the boundary region on the simulation and achieving a more accurate light intensity distribution.

[0113] The present invention converts the instantaneous field intensity values obtained by the FDTD method into average values. The incident light in the present invention is a plane wave. The peak detection method is used to convert the instantaneous electromagnetic field values that vary with time during the calculation process into peak values, thereby obtaining an average value and the light intensity distribution at each grid point. This also facilitates docking with the next module of the lithography simulation model, resulting in a more accurate final result.

[0114] Figure 2 The light intensity distribution curve and corresponding light intensity contour map at different photoresist depths when vertically incident on SU8 thick photoresist. SU-83000 was selected as the simulated photoresist, and the I-line wavelength of 365nm ultraviolet light was used as the incident light. The incident light intensity was set to 2.6mW / cm 2 Under this condition, the refractive index of SU-8 3000 is 1.55 and the extinction coefficient is 0.0005.

[0115] Figure 2 (a) is a light intensity distribution curve at different photoresist depths under vertical incidence, and the curves represent the photoresist intensity distribution at 2μm, 10μm and 20μm from top to bottom respectively. Figure 2 (b) is the corresponding light intensity contour map. This example proves that the simulation results of the present invention are basically consistent with the experimental results. The present invention can accurately simulate the light intensity distribution inside the photoresist in the thick photoresist UV lithography process.

[0116] 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 principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A method for simulating light intensity distribution in a thick-resist UV lithography process, characterized in that: The steps include: Step 1: Build a corresponding lithography simulation model based on the actual lithography situation, discretize the spatial domain of the electric and magnetic fields using the Yee grid, and apply a plane wave with controllable incident angle, polarization direction, and wavelength; Step 2: Obtain the material parameters of each Yee lattice point according to the lithography simulation model, use the recursive convolution method to process different types of dispersion materials, and calculate the corresponding calculation coefficients; Step 3: Add a convolutional perfectly matched layer (CPML) boundary to the simulation area and calculate the calculation coefficient of the CPML boundary area. Step 4: Based on steps 2 and 3, a set of modified finite-difference time-domain FDTD electromagnetic field update equations are obtained. The electric field and magnetic field of each Yee grid point are updated in each time step through cyclic calculation until a steady state is reached. Step 5: According to the distribution of the electromagnetic field in the grid, the peak detection method is used to obtain the light intensity distribution inside the photoresist.

2. The light intensity distribution simulation method according to claim 1, characterized in that: In step 1, a lithography simulation model is constructed based on the actual situation of proximity or contact lithography to simulate the mask, air gap, photoresist and silicon wafer materials in the proximity or contact lithography process. The corresponding information is stored in a matrix. At the same time, a time-harmonic field source is used to simulate the basic light source used in the lithography process, which can control the incident angle, polarization direction and wavelength.

3. The light intensity distribution simulation method according to claim 1, characterized in that: In step 2, the recursive convolution method is used to improve the FDTD method to deal with dispersive materials. The recursive convolution method first transforms the polarizability function of the medium from the frequency domain to the time domain, then expresses the time domain expression as an exponential function, and then calculates the convolution in the time domain constitutive equation of the medium in a recursive iterative manner, thereby obtaining the improved electric field intensity update equation in FDTD.

4. The light intensity distribution simulation method according to claim 3, characterized in that: The recursive convolution method is used to express the electric displacement vector as the convolution between the electric field strength and the frequency-dependent electric susceptibility. The relationship between the electric displacement vector and the electric field strength is: Where D is the electric displacement vector, E is the electric field strength, t is the time, ε0 is the dielectric constant in vacuum, χ(τ) is the electric susceptibility, and ε ∞ is the relative dielectric constant when the frequency ω→∞; Discretize time using Yee's grid and get: Where n is the field value at the nth time step. Substituting the above formula into Maxwell's equations can handle dispersive media; Different materials are suitable for different dispersion medium models. The materials used in the lithography process can be simulated by the Debye medium model and the Lorentz medium model: The polarizability of the Debye medium is expressed as: Where, ε s is the dielectric constant at zero frequency; t0 is the relaxation time of the medium; j is the imaginary unit, ω is the frequency; The polarizability of a Lorentz medium is expressed as: Where ω0 is the resonant frequency of the Lorentz medium; δ is the attenuation coefficient.

5. The light intensity distribution simulation method according to claim 1, characterized in that: Step 3: Add a convolutional perfectly matched layer (CPML) boundary to the simulation area and a convolutional perfectly matched layer (CPML) boundary to the calculation area. In the CPML boundary area, the passive Maxwell equations are corrected as follows: Where H is the magnetic field intensity vector, E is the electric field intensity, μ is the magnetic permeability, ε is the dielectric constant, j is the imaginary unit, ω is the frequency; i, j, k represent the unit vectors in the three directions; s x =s(x),s y =s(y),s z =s(z) is the coordinate scaling factor.

6. The light intensity distribution simulation method according to claim 5, characterized in that: The coordinate expansion factor is divided into the electric field expansion factor s ei and magnetic field stretching factor s mi , which follows: κ ei , κ mi , α ei , α mi The PML parameter that complies with the causal law is taken as κ ei ≥1, κ mi ≥1,α ei ≥0,α mi ≥0,σ max is the maximum value of the conductivity distribution; ρ is the distance between the inner region and the PML boundary to the field component; d is the thickness of the PML layer; n pml is the order of the polynomial distribution; μ0 and ε0 are the vacuum permeability and vacuum permittivity, respectively.

7. The light intensity distribution simulation method according to claim 5, characterized in that: In step 4, a set of modified FDTD electromagnetic field update equations are obtained, including the update equations for the non-PML region and the update equations for the CPML boundary region; the electric field equations also apply the recursive convolution method to deal with dispersive media, making the FDTD algorithm suitable for lithography simulation, and obtaining a new electric field update equation; In the non-PML region, the updated equation for the new x-component electric field is: in, Among them, I n (i, j, k) is a recursive term, and its recursive expression is shown in Equation (8), where Δt is the sampling time step, ε0 is the dielectric constant in vacuum, σ is the conductivity; χ(τ) is the electrical susceptibility; ε ∞ is the relative dielectric constant when ω→∞; C rec is a recursive constant; The updated equations for the electric field in the y and z direction components have the same form as that in the x direction; Among them, the update equation group of the CPML boundary area, the new electric field update equation is: in, Among them, I n Same as formula (8); σ ey , κ ey , α ey is σ in formula (6) ei , κ ei , α ei ,i=y; The same logic applies; In the update equations of the CPML boundary region, the magnetic field and electric field update equations for the y and z direction components are the same as those for the x direction.

8. The light intensity distribution simulation method according to claim 1, characterized in that: In step 5, the peak detection method is used to obtain the light intensity distribution inside the photoresist, specifically to obtain the amplitude of each point in the Yee grid, and record the field components F at the current and previous two moments. n 、F n-1 、F n-2 , when: (F n-1 -F n )(F n-1 -F n-2 )<0 (11) The average of the positive and negative peak values is taken as the peak value of the field, and then the peak value of the electromagnetic field is converted into light intensity through calculation, so as to obtain the light intensity distribution of each grid point, so as to better connect with the next module of the lithography simulation model and obtain a more accurate final result.

Images