Molecular dynamics simulation method for cu interconnect dielectric materials

Abstract

The application discloses a kind of molecular dynamics simulation methods of Cu interconnect dielectric material.It is established interconnect dielectric material SiCOH molecular dynamics model using melting-cooling method according to molecular dynamics theory, the feasibility of model is verified using structure and dielectric property, on this basis, the optimal performance combination of interconnect dielectric material is obtained by calculating electric, thermal, mechanical properties and fracture mechanics properties with adjacent interconnect structure, and the result is fed back to adjust to guide material design.The application establishes the simulation model of dielectric material SiCOH performance by molecular dynamics simulation, obtains the optimal performance parameter combination for packaging structure and product, reduces the cost of test and production trial and error, so as to guide the design of high-performance, high-reliability very large scale integrated circuit post-process interconnect material and structure.

Description

TECHNICAL FIELD

[0001] The method belongs to the technical field of semiconductor integrated circuit manufacturing, and particularly relates to a molecular dynamics simulation method for dielectric material of an interconnection structure in a back-end process. BACKGROUND

[0002] With the increasing device density and the decreasing feature line width of ultra-large scale integrated circuits, the interconnection resistance coupling increases, signal crosstalk and delay and power consumption increase, and other problems, and the interconnection structure and material become the bottleneck of improving the performance of integrated circuits. With the shrinking of chip process, the number of wires per unit area is increasing, the spacing of the wires is decreasing, the parasitic capacitance between the wires is becoming significant, and the crosstalk and delay increase problems caused by the parasitic capacitance are more prominent. To solve this problem, low dielectric constant (low-κ) and ultralow dielectric constant (ultralow-κ) materials replace silicon dioxide (SiO2) and become the dielectric of the interconnection structure in the back-end process of chip manufacturing. For a multi-layer copper wiring of a double damascene structure, a plurality of process steps such as deposition of dielectric material, photolithography, reactive ion etching, copper metal plating, and chemical mechanical polishing are experienced, which puts higher requirements on the reliability of the low-κ material. To prepare a high-performance SiCOH material, it is necessary to continuously test to find a comprehensive optimal solution that meets the dielectric constant and related performance. There are usually two methods for preparing SiCOH: spin coating and chemical vapor deposition. Among them, the porous SiCOH deposited by plasma-enhanced chemical vapor deposition technology deposits the matrix material and the organic polymer (porogen) at the same time. Then, the organic polymer material is decomposed by solidification and the like, so as to convert the originally occupied position into micropores / nanopores, and the remaining matrix is reorganized and strengthened to form the final nanoporous film.

[0003] A large number of experiments on a macroscopic level can only obtain a limited isolated data set, and the time cost and economic cost required are high, and it is difficult to comprehensively consider various test conditions to optimize the whole process of SiCOH manufacturing. The generation of the network model of SiCOH is mainly based on the micro level, so it is difficult to meet the specific parameters of the test requirements only by testing, characterization and testing on a macroscopic level. The existing molecular scale calculation model needs to define the initial topological structure and use a specific relaxation method, which may be due to the difficulty in capturing complex bond formation and recombination by using traditional molecular dynamics force fields. Systematic micro-simulation and performance calculation of SiCOH have important significance and research value for the development and use of interconnection dielectric materials of ultra-large scale integrated circuits. SUMMARY

[0004] The application aims at solving the contradiction between reducing dielectric constant and improving performance of medium material in the post-process of ultra-large scale integrated circuit Cu / low-k interconnection, and proposes a molecular dynamics simulation method of medium SiCOH in the post-process.

[0005] The application is implemented as follows:

[0006] The application is implemented as follows:

[0007] S1: Collect and organize various effective information involved in the simulation process;

[0008] S2: Based on the molecular dynamics theory, a melting-cooling method is used to establish a molecular dynamics model of the interconnection medium material SiCOH;

[0009] S3: Calculate the dielectric properties and structural properties to verify the rationality of the grid topology structure of the established molecular dynamics model of the interconnection medium material SiCOH;

[0010] S4: Based on the verification of the rationality of the model network topology structure in S3, the thermal, mechanical, fracture mechanics and interconnection structure composite material mechanics properties are calculated to comprehensively evaluate the performance of the model;

[0011] S5: The performance calculation results of S3 and S4 are comprehensively analyzed to obtain the optimal performance combination of the interconnection medium material.

[0012] Further, the effective information in step S1 includes: element ratio and structure information of the interconnection medium material SiCOH, especially atomic ratio, doping elements, chemical bond polarization rate, pore and pore distribution information; molecular dynamics modeling method, interatomic interaction potential function theory, and theoretical basis and solving method of various performance calculations.

[0013] Further, in step S2, the melting-cooling method is used to establish the molecular dynamics model of the interconnection medium material, and the ReaxFF potential function or machine learning potential function is used to simulate the interaction between atoms, and the elements are parameterized.

[0014] Further, the specific process of step S2 is as follows:

[0015] According to the S1 investigation content, the atomic proportion of multiple groups of materials is determined and randomly placed in a cubic simulation box with periodic boundary conditions; the conjugate gradient method is used to relax the model, and the model is melted at 3000K temperature using isothermal-isobaric ensemble and canonical ensemble respectively; a reasonable cooling rate is used to linearly decrease from 3000K to room temperature; finally, the equilibrium model of multiple SiCOH is obtained by balancing under isothermal-isobaric ensemble and canonical ensemble respectively.

[0016] Further, the step S3 comprises the following sub-steps:

[0017] S3a verifies the rationality of the network topology structure of the established SiCOH molecular dynamics model by calculating the structure performance, wherein the structure performance is quantitatively evaluated by dielectric constant, density, porosity and pore distribution, radial distribution function, bond length, bond angle parameter;

[0018] S3b verifies the rationality of the network topology structure of the established SiCOH molecular dynamics model by calculating the dielectric performance, calculates the relationship between the dielectric constant and the frequency of electrons and ions, and obtains the static dielectric constant of the material.

[0019] Further, the step S4 comprises the following sub-steps:

[0020] S4a calculates the thermodynamic performance of the dielectric material SiCOH, obtains the changes of density, side length, volume and energy with actual temperature according to statistical mechanics, and obtains the thermodynamic parameters of glass transition temperature, thermal expansion coefficient and specific heat by using the equilibrium model of the molecular dynamics model of S2 at different temperatures, and obtains the thermal conductivity by using the temperature gradient field and heat flux according to the non-equilibrium molecular dynamics method based on Fourier heat conduction law;

[0021] S4b calculates the mechanical performance of the dielectric material SiCOH, obtains the atomic configuration, stress-strain response, stress evolution and pore evolution by uniaxial tensile test and nanoindentation test, and calculates the Young's modulus and hardness mechanical parameters;

[0022] S4c calculates the fracture mechanics performance of the dielectric material SiCOH, simulates the crack propagation process of the system with pre-made crack by using I-type uniaxial tensile test, obtains the atomic configuration, stress distribution, deformation, damage evolution and fracture morphology when completely broken, and calculates the fracture toughness and energy release rate block material fracture mechanics parameters, and the same method can be used to calculate the fracture mechanics performance under II and III type fracture loading mode.

[0023] S4d calculates the interfacial fracture mechanics behavior of the composite material with the dielectric material in the interconnection structure, in the BEOL back-end interconnection structure, the BEOL structure adjacent to the interconnection dielectric layer usually includes a copper diffusion barrier layer, an etching stop layer, and a sealing layer, a bimaterial interface model is established, relaxation is carried out by using the melting-cooling method, the loading mode in S4c simulates the molecular configuration, stress distribution, damage evolution, and fracture morphology when the composite material is completely fractured, and the interfacial fracture toughness mechanical parameters are calculated;

[0024] S4e establishes the quantitative relationship between the element ratio, temperature, porosity, doped atoms, impurities, and the thermal, mechanical, fracture, and composite material fracture performance of the dielectric material.

[0025] Further, the step S5 is as follows:

[0026] The performance calculation results of S3 and S4 are comprehensively analyzed to obtain the optimal performance combination of the interconnection dielectric material, and the results are fed back to adjust to establish a more optimal molecular dynamics model, thereby guiding the design of high-performance and high-reliability BEOL interconnection materials and structures.

[0027] Compared with the prior art, the present application has the following advantages:

[0028] The present application seeks the optimal performance combination of the dielectric material SiCOH in the Cu interconnection, establishes a molecular dynamics model without defining the initial topological structure and using a specific relaxation method by the molecular dynamics simulation method, and verifies the network topological structure of the model through the dielectric performance and the structure performance, and on this basis, the thermal, mechanical, and fracture performances of the composite material are calculated to guide the design of high-reliability dielectric materials and structures. BRIEF DESCRIPTION OF DRAWINGS

[0029] Figure 1 A flowchart of a molecular dynamics simulation method for the electrical, thermal, and mechanical performance of a Cu / low-κ interconnection dielectric material is provided for the embodiments of the present application;

[0030] Figure 2 A system framework structure diagram of a molecular dynamics simulation method for the electrical, thermal, and mechanical performance of a Cu / low-κ interconnection dielectric material is provided for the embodiments of the present application. DETAILED DESCRIPTION

[0031] The technical solutions in the embodiments of the present application will be described in detail below with reference to the contents of the present application and the drawings, and combined with examples. Obviously, the following described cases are only part of the examples to which the present application is applicable, and are not all the embodiments. Based on the related ideas and methods of the present application, other embodiments obtained by those of ordinary skill in the art without making creative improvements and labor on the inventive ideas themselves belong to the protection scope of the present patent.

[0032] Figure 1 For the flow chart of the present application, the core of the present application is the establishment of the molecular dynamics model of the dielectric material SiCOH and the performance calculation, and the optimal performance combination is sought to guide the design of high-reliability dielectric materials and structures. Figure 2 For the system framework structure diagram, the simulation results and the description parameters of different performances are specifically described, and the specific way of the input parameters required by each step and the output parameter transmission obtained is embodied in the diagram.

[0033] The present application will be further described in detail below in combination with the drawings and specific embodiments:

[0034] S1: Collect and organize various effective information involved in the simulation process;

[0035] S2: Based on the molecular dynamics theory, the melting-cooling method is used to establish the molecular dynamics model of the interconnection dielectric material SiCOH;

[0036] S3: Calculate the dielectric performance and structural performance to verify the rationality of the grid topology structure of the established molecular dynamics model of the interconnection dielectric material SiCOH;

[0037] S4: On the basis of verifying the rationality of the model network topology structure in S3, the thermal, mechanical, fracture mechanics and interconnection structure composite material mechanical properties are calculated to comprehensively evaluate the performance of the model;

[0038] S5: The performance calculation results of S3 and S4 are comprehensively analyzed to obtain the optimal performance combination of the interconnection dielectric material.

[0039] In the above embodiment, the effective information in step S1 includes: information such as element ratio and structure of the interconnection dielectric material SiCOH, especially atomic ratio, doping elements, chemical bond polarization rate, porosity and porosity distribution; molecular dynamics modeling method, interatomic potential function theory, and theoretical basis and solving method of various performance calculations.

[0040] In step S2, the dielectric model contains Si, C, O and H four elements, first of all, according to the research content of S1, the atomic number of Si, C, O and H four elements in the model is determined, the ReaxFF potential function or the machine learning potential function is used to simulate the interaction between atoms, the four elements are parameterized, all atoms are randomly placed in a cubic simulation box with periodic boundary conditions, and the minimum distance between each atom is ensured At this time, the simulation starts at a temperature and pressure level where the system dynamics is unstable, and there is no bond formation. Then the melting-cooling method is used to relax the model. The initial model of the dielectric material may also include doped atoms and impurity elements that may be introduced by other processes.

[0041] The potential function describing the interatomic interactions of the SiCOH model can be parameterized using ReaxFF potential function or machine learning potential function (such as DeepMD potential function) for the elements contained in the model. Both potential functions can simulate complex chemical reactions and bond formation / breaking, and are suitable for simulating material behavior under sustained load.

[0042] The ReaxFF reaction potential function is realized by training the empirical force field with quantum mechanical energy and structural information, which contains bond order information depending on atomic distance and is updated constantly in each molecular dynamics or energy minimization iteration, allowing continuous change. This potential function can simulate complex chemical reactions and bond formation / breaking, and is suitable for simulating material behavior under sustained load.

[0043] The potential function obtained by using machine learning (such as artificial neural network method) is specially developed for SiCOH material system. The deep learning method used is based on neural network fitting of first-principles data, and the potential function obtained is used to represent the many-body potential energy and solve molecular dynamics. This deep learning method can convert the data provided by the user into a deep potential model without human intervention, and can be used in subsequent molecular dynamics performance calculation process. It can improve the calculation speed of molecular dynamics by several orders of magnitude while maintaining the accuracy and accuracy of quantum mechanics.

[0044] In a further preferred embodiment, the element ratio of the medium model is an important design variable that can be changed, and can also include doping elements that can improve the dielectric constant performance, or other process-introduced impurity elements.

[0045] Further, the specific steps of the melting-cooling method used in the establishment of the molecular dynamics model are as follows: first, use the conjugate gradient method to relax the model; melt the model in the isothermal-isobaric (NPT) ensemble at 3000K and zero pressure, and in the canonical ensemble (NVT) at the same temperature; use a cooling rate of 2.5K / ps, and linearly reduce the model temperature from 3000K to 300K; obtain the amorphous structure at 300K, and relax it in the NPT and NVT ensembles respectively, and the system temperature and energy reach the equilibrium state. Finally, the SiCOH model quenched and relaxed at room temperature and 0bar is obtained. The melting temperature, cooling rate, simulation ensemble, and simulation time can be adjusted according to whether the system reaches equilibrium and whether the performance is optimal. This simulated annealing method not only does not require an input crystal structure representing the final topology or a well-defined initial structure, but also avoids the undesirable artificial constraints that may be imposed on the structure due to the relaxation of the initial structure representing the final network topology.

[0046] In a further preferred embodiment, step S3 specifically comprises calculation of dielectric properties and structural properties of the dielectric material to verify the reasonability of the network topology of the established SiCOH molecular dynamics model. The properties are quantitatively evaluated by parameters such as dielectric constant, density, porosity and pore distribution, radial distribution function, bond length, bond angle, etc.

[0047] In a further preferred embodiment, using the equilibrium model established in S2, the relationship between the dielectric function and frequency of the electrons and ions is calculated respectively using the first principle method after the structure reaches equilibrium, and then the dielectric constant of the electrons and ions is approximated to zero frequency and weighted respectively to obtain the static dielectric constant of the material. Generally, the dielectric constant of Low-κ material should satisfy κ < 3. The dielectric constant of dielectric material should be as small as possible to meet the interconnection requirements of the continuously reduced device size.

[0048] The equilibrium model established in S2 is analyzed for structural properties: the partial and overall radial distribution functions and the structure factor are calculated respectively to evaluate the short- and long-range order in the structure; the bond length and bond angle are used to analyze the local structural changes, and the bond lengths of Si-O, Si-C, C-H, etc. and the bond angles of Si-O-Si, Si-C-Si, etc. are calculated to evaluate the cage / network structure of the system; and the dihedral angle of Si-O-Si-O and the coordination number of Si can be used to evaluate the overall structure of the system. For example, the change in the Si-O-Si angle can be used to evaluate the bond distortion borne by the local tetrahedral unit to analyze the cage / network structure of the material. Since the introduction of pores has a huge impact on the dielectric properties of the material, it is crucial to simulate the porosity and pore distribution of the structure and use them as important design parameters to improve the model.

[0049] In a further preferred embodiment, the porosity and pore distribution of the S2 model are calculated using the identification method of quasi-molecular density isosurface. The porosity is usually between 10% and 60%, and the pore size is usually < 10 nm.

[0050] In a further preferred embodiment, the calculation results of dielectric properties and structural properties are compared with the research results of S1 on dielectric materials to verify the reasonability of the model network, and then other key material properties that affect reliability can be calculated.

[0051] In a further preferred embodiment, in step S4, the thermal, mechanical, fracture mechanical, and interconnection structure composite material mechanical properties are calculated. The following sub-steps are included:

[0052] S4a calculates the thermal properties, including glass transition temperature, thermal expansion coefficient, specific heat, thermal conductivity. The equilibrium model established in S2 is heated to 900K, and the NPT ensemble is used to equilibrate at every temperature, which is increased by 20K each time. The actual temperature and density of the system are sampled and averaged. The actual temperature and density of the system are plotted and linearly fitted, and the intersection is the glass transition temperature.

[0053] Further, the equilibrium model established in S2 is heated to 900K, and the NPT ensemble is used to equilibrate at every temperature, which is increased by 100K each time. The actual temperature, edge length, volume, and energy of the system are sampled and averaged. The edge length, volume, and energy of the system are plotted and polynomially fitted, and the linear / volumetric thermal expansion coefficient and specific heat can be obtained.

[0054] Further, the equilibrium model established in S2 is directly extended using the non-equilibrium molecular dynamics method based on Fourier's heat law. The equilibrium model established in S2 is extended along the z direction, and the simulation frame along the heat flow direction is divided into N equal-width plates. A heat source and a cold source are set in the model, and a certain heat flow is applied to the system. After the simulation reaches equilibrium, the average temperature of each plate is calculated, and the temperature-position relationship is plotted. The temperatures on the left and right sides of the heat source are linearly fitted, and the slope is the temperature gradient. According to Fourier's heat law, the thermal conductivity is calculated.

[0055] In the process of calculating the thermal conductivity, the system can also be equilibrated by applying a specific temperature to the heat source and cold source plates, and the heat flux of the system is calculated. According to Fourier's heat law, the thermal conductivity is calculated.

[0056] S4b calculates the mechanical properties, including atomic configuration, stress-strain response, stress evolution, and pore evolution, by uniaxial tension or nanoindentation. The Young's modulus, hardness, and other mechanical parameters are calculated.

[0057] S4c calculates the fracture mechanics properties of the medium material. The equilibrium model of S2 is subjected to uniaxial tension test, and the engineering strain is estimated using the length change of the simulation frame, and the engineering stress is estimated using the virial theorem. The stress-strain curve during the tension process is plotted, and the elastic modulus is calculated according to the slope of the elastic stage. The Poisson's ratio is calculated according to the transverse / longitudinal tensile strain relationship. At the same time, the atomic configuration, stress distribution (including principal stress, von Mises stress, etc.), and evolution of pore defects at different tension times can be obtained.

[0058] Further, the equilibrium model of S2 is subjected to nanoindentation test, using diamond indenter to load and unload the substrate of dielectric material, to obtain the indenter displacement and load curve, and the evolution of atomic configuration, stress distribution and defects such as pores with the depth of indenter, and to calculate the Young's modulus, hardness and other mechanical parameters.

[0059] S4d calculates the interfacial fracture mechanics behavior of the composite material with dielectric material in the interconnection structure. Type I uniaxial tensile test is used to simulate the crack propagation process of the bulk material with pre-existing crack. Specifically, the molecular configuration center of the equilibrium model of S2 is added to the initial crack of the rectangle, constant strain is applied to the rigid thin layer on both sides, the direction of constant strain is perpendicular to the initial crack, and uniaxial tensile test is carried out until the crack completely expands. The molecular configuration, stress distribution and damage evolution process describe the dynamic crack propagation process, and the fracture morphology describes the completely cracked morphology. According to the energy theory of fracture mechanism and the thermodynamic integral in the crack propagation process, the critical strain energy release rate is derived. The same method can be used to calculate the fracture mechanics performance under type II and type III fracture loading modes.

[0060] Further, the interfacial fracture mechanics behavior of the composite material with dielectric material in the interconnection structure is calculated. In the BEOL back-end interconnection structure, the materials adjacent to the interconnection dielectric material commonly include Si, SiO2, SiCN, SiC, α-Si3N4, β-Si3N4, etc. A double-material interfacial molecular structure model is established, the dielectric material and the adjacent material are placed in equilibrium, and the initial separation distance between the two layers is set to Relaxation is carried out by conjugate gradient method, and then relaxation is carried out by melting-cooling method as in S2, and the obtained structure has an interfacial transition zone, and a pre-crack is prepared in the middle of the interfacial blending zone. Type I uniaxial tensile test is used to simulate the dynamic crack propagation process, and the molecular configuration, stress distribution, damage evolution and fracture morphology at complete fracture of the whole crack propagation process are calculated, and the interfacial fracture toughness and other mechanical parameters are calculated.

[0061] Further, the evolution of pore defects of the system in step S4 under temperature and force load is calculated by the analysis method of pore in step S3.

[0062] When calculating the thermal, mechanical, fracture and composite fracture properties of the dielectric material in steps S3 and S4, it is necessary to minimize the interference of simulation conditions such as periodic boundary conditions, strain rate, size, etc. on the results.

[0063] When calculating the thermal, mechanical, fracture and composite fracture properties of the dielectric material in steps S3 and S4, it is necessary to establish the quantitative relationship between factors such as element ratio, temperature, porosity, doped atoms, impurities and the thermal, mechanical, fracture and composite fracture properties of the dielectric material, so as to facilitate seeking the best performance combination that meets the manufacturing and use requirements.

[0064] Further, the doping atoms considered in the calculation of the performance of the interconnection structure in steps S3 and S4 include doping elements F and the like that can form chemical bonds with lower polarizability, or impurity elements Na, N, S, K and the like introduced by other processes.

[0065] In a further preferred embodiment, the results of the performance calculation in steps S3 and S4 are comprehensively analyzed in step S5 to obtain the optimal performance combination of the interconnection dielectric material, on the basis of meeting the dielectric constant requirement, the stronger the heat transfer performance, the mechanical performance and the interface adhesion performance are the better. Specifically, on the basis of the dielectric constant of the interconnection dielectric material < 3.0, the Young's modulus > 2 Gpa, the interface energy release rate > 0.5 J / m 2 The results are fed back to adjust to establish a better molecular dynamics model, thereby guiding the BEOL interconnection material and structure with high performance and high reliability. It should be understood that the above description of the preferred embodiments is more detailed, and should not be considered as a limitation on the scope of patent protection of the present application. Those skilled in the art can make substitutions or modifications without departing from the scope of the claims of the present application, which are all within the scope of protection of the present application. The scope of protection of the present application should be subject to the appended claims.

Claims

1. A method of molecular dynamics simulation of Cu interconnect dielectric material, for performing molecular dynamics simulation, characterized in that, The method comprises the following steps: S1: collecting and organizing various effective information involved in the simulation process; the effective information includes: element ratio and structure information of the interconnection dielectric material SiCOH, atomic ratio, doping element, chemical bond polarization rate, pore and pore distribution information; molecular dynamics modeling method, theory of interatomic interaction potential function, and theoretical basis and solving method of various performance calculation; S2: based on the molecular dynamics theory, a molecular dynamics model of the interconnection dielectric material SiCOH is established by using the melting-cooling method; S3: the dielectric performance and the structure performance are calculated to verify the rationality of the grid topology structure of the established molecular dynamics model of the interconnection dielectric material SiCOH; including the following sub-steps: S3a: the rationality of the network topology structure of the established SiCOH molecular dynamics model is verified by calculating the structure performance; S3b: the rationality of the network topology structure of the established SiCOH molecular dynamics model is verified by calculating the dielectric performance, the relationship between the dielectric constant of electrons and ions and the frequency is calculated, and the static dielectric constant of the material is obtained; S4: on the basis of verifying the rationality of the model network topology structure in S3, the thermal, mechanical, fracture mechanics and interconnection structure composite material mechanical properties are calculated to comprehensively evaluate the performance of the model; including the following sub-steps: S4a: the thermodynamic performance of the dielectric material SiCOH is calculated, according to statistical mechanics, the equilibrium model of the molecular dynamics model in S2 at different temperatures is used to obtain the changes of density, side length, volume and energy with actual temperature, and the thermodynamic parameters of glass transition temperature, thermal expansion coefficient and specific heat are obtained, according to the non-equilibrium molecular dynamics method based on Fourier heat conduction law, the thermal conductivity is obtained by using temperature gradient field and heat flux; S4b: the mechanical properties of the dielectric material SiCOH are calculated, the atomic configuration, stress-strain response, stress evolution and pore evolution are obtained through uniaxial tensile test and nanoindentation test, and the Young's modulus and hardness mechanical parameters are calculated; S4c: the fracture mechanics performance of the dielectric material SiCOH is calculated, the system crack propagation process with a prefabricated crack is simulated by using I-type uniaxial tensile test, the atomic configuration, stress distribution, deformation, damage evolution and fracture morphology at complete fracture are obtained, and the fracture toughness and energy release rate block material fracture mechanics parameters are calculated, and the same method can be used to solve the fracture mechanics performance of II and III type fracture loading modes; S4d: the interface fracture mechanics behavior of the composite material composed of the dielectric material in the interconnection structure is calculated, in the BEOL rear interconnection structure, the BEOL structure adjacent to the interconnection dielectric layer usually includes copper diffusion barrier layer, etching stop layer and sealing layer, a double-material interface model is established, relaxation is carried out by using the melting-cooling method, the molecular configuration, stress distribution, damage evolution and fracture morphology at complete fracture of the composite material are simulated by using the loading mode in S4c, and the interface fracture toughness mechanical parameter is calculated; S4e: the quantitative relationship between the element ratio, temperature, porosity, doping atom, impurity and the thermal, mechanical, fracture and composite material fracture performance of the dielectric material is established. S5: Perform comprehensive analysis on the performance calculation results of S3 and S4 to obtain the optimal performance combination of the interconnection medium material.

2. The method of claim 1, wherein the Cu interconnect dielectric material is simulated by molecular dynamics simulation. In the step S2, a molecular dynamics model of the interconnection medium material is established using a melting-cooling method, a ReaxFF potential function or a machine learning potential function is used to simulate the interaction between atoms, and the elements are parameterized.

3. The method of claim 2, wherein the Cu interconnect dielectric material is simulated by molecular dynamics simulation. The specific process of the step S2 is as follows: According to the research content of S1, determine the atomic ratio of multiple groups of materials and randomly place them in a cubic simulation box with periodic boundary conditions; use the conjugate gradient method to relax the model, melt the model at 3000K temperature using the isothermal-isobaric ensemble and the canonical ensemble respectively; use a reasonable cooling rate to linearly decrease from 3000K to room temperature; finally, balance under the isothermal-isobaric ensemble and the canonical ensemble respectively, and finally obtain multiple equilibrium models of SiCOH.

4. The method of claim 1, wherein, The structural performance in the step S3a is quantitatively evaluated by dielectric constant, density, porosity and pore distribution, radial distribution function, bond length, bond angle parameters.

5. The method of claim 1, wherein, The steps of the step S5 are as follows: Perform comprehensive analysis on the performance calculation results of S3 and S4 to obtain the optimal performance combination of the interconnection medium material, and feedback adjustment to establish a more optimal molecular dynamics model to guide the design of high-performance and high-reliability BEOL interconnection materials and structures.

Images