Cross-scale simulation method and system based on the whole process of device radiation defect evolution

Abstract

The present invention provides a cross-scale simulation method and system based on the whole process of device radiation defect evolution, the method comprising: defining particle parameters and material parameters, and inputting them into the simulation system; establishing a molecular dynamics model based on space incident particles according to the basic principles of molecular dynamics, performing numerical simulation of the motion of space incident particles based on molecular dynamics, and obtaining the motion trajectory, speed, energy and distribution information of space incident particles in space devices at different times; using statistical physics methods to count the motion laws of space incident particles under different states, correlating the microscopic state of space incident particles with the macroscopic properties of materials, and performing cross-scale computational simulation. The present invention takes a single incident particle and material parameters as the research object, simulates the cross-scale defect evolution effect generated by the incident particle in the space device, and provides an important basis and idea for the material selection of semiconductor devices.

Description

Technical Field

[0001] The present invention relates to the field of simulation technology, and in particular to a cross-scale simulation method and system based on the entire process of device radiation defect evolution. Background Art

[0002] When energetic particles in the space environment are incident on the electronic system of a spacecraft, they deposit energy in an ionized or non-ionized manner, causing a variety of radiation effects in semiconductor devices, which seriously affect the reliability and life of the spacecraft. Most materials are caused by displacement defects after irradiation. The accumulation and evolution of these defects in structural materials changes the microstructure of the material, causing displacement damage, which in turn degrades the macroscopic physical properties of the material, such as shape, size, and mechanical properties, such as tension, compression, and shear, causing the material to fail, which may eventually lead to major safety accidents.

[0003] Material irradiation and testing are extremely costly and time-consuming, and sometimes even impossible due to the limited number of facilities, especially for some special energy particle irradiation. For irradiation experiments with different particles, post-irradiation analysis of irradiated samples and radioactive materials requires a long waiting time to reduce the radioactivity of the samples to background levels. The current experimental strategy is to simulate the reactor neutron irradiation environment with fission neutrons, high-energy protons or heavy ions. The results obtained in this way need to be extrapolated to real reactor irradiation conditions, but the accurate analysis of this process will cause many problems. In particular, it is difficult to distinguish the effects of particle type, particle energy, temperature, and cumulative damage, making it difficult to predict the overall damage rate.

[0004] The behavior of existing materials under new operating conditions, the need to develop new materials, and the increasing costs and challenges of material irradiation and testing have increasingly emphasized the importance of using computers to simulate the evolution of material microstructures under irradiation conditions. However, since the generation and evolution of displacement damage spans multiple time scales, it is difficult to simulate and there is still a lack of research in this area in the existing technology. Summary of the invention

[0005] The problem solved by the present invention is to provide a cross-scale simulation method for simulating the entire process of space irradiation defect evolution.

[0006] To solve the above problems, the present invention provides a cross-scale simulation method based on the entire process of device radiation defect evolution, comprising:

[0007] Calling different types of spatial incident particles from a particle database, and inputting parameters of the spatial incident particles into a simulation system for initialization; and calling different types of materials from a device material database, and inputting parameters of a simulation box corresponding to the material type into the simulation system;

[0008] According to the basic principles of molecular dynamics, a molecular dynamics model based on the spatial incident particles is established, a numerical simulation of the motion of the spatial incident particles based on molecular dynamics is performed, and the motion trajectory, velocity, energy and distribution information of the spatial incident particles in the spatial device at different times are obtained;

[0009] Statistical physics methods are used to calculate the motion laws of the spatial incident particles under different states, and the microscopic states of the spatial incident particles are associated with the macroscopic properties of the material to perform cross-scale computational simulation.

[0010] Preferably, the obtaining of the motion trajectory, velocity, energy and distribution information of the spatial incident particles at different times in the spatial device comprises:

[0011] The Verlet algorithm is used to calculate the actual position of the i-th atom at time t in the coordinate system of the spatial device, and the displacement and velocity of the spatial incident particle in the spatial device are obtained, thereby obtaining the motion trajectory, velocity, energy and distribution information of the spatial incident particle in the spatial device at different times.

[0012] Preferably, the method further comprises: when performing molecular dynamics simulation, adding fixed boundary conditions to the simulation box to simulate the physical phenomena produced by the specific interaction between the particles in the simulation box and the surrounding environment.

[0013] Preferably, the method further comprises: when performing molecular dynamics simulation, using a microcanonical ensemble and applying periodic boundary conditions only in the 1 / 2 direction.

[0014] Preferably, in the molecular dynamics simulation, the motion integration step is 1 fps and the total simulation time is 50 fps.

[0015] Preferably, the cross-scale computing simulation includes:

[0016] The OKMC method is used to simulate the short-time low physical damage level evolution of the spatial incident particles, so that the spatial incident particles are randomly or jump on a specified lattice to simulate the formation of defects in the spatial device;

[0017] Using the equivalent sphere method to identify the spatial incident particles corresponding to the defect;

[0018] The cluster dynamics method is used to evolve the high physical damage level of the spatially incident particles.

[0019] Preferably, the method of simulating the short-time low physical damage evolution of the spatial incident particles by using the OKMC method includes:

[0020] Set the initial time t=0;

[0021] Calculate the probability of local defect generation and give a list of local defect generations;

[0022] Calculate the total rate R, and obtain a uniform random number r in (0, 1] to select an event that may generate a defect;

[0023] Search for i by R×i - 1 < r×R < R×i to obtain the event of executing i, and execute event i;

[0024] Update the time , and accumulate the generation of defects along with the simulation time;

[0025] Determine the transition probability according to the Boltzmann constant, the absolute temperature in the system, and the Planck constant, and update the transition probability according to the temperature in the system.

[0026] Preferably, the method of using the equivalent sphere method to identify the space incident particles corresponding to the generation of defects includes: dividing the simulation box into several equivalent grids, checking whether the space incident particles are within a sphere with a specified radius centered on the ideal lattice position of the atoms in the space device. If the sphere centered on the ideal lattice position does not contain the atom, the lattice position corresponding to the atom is marked as a vacancy. If the displacement distance of the atom from its nearest ideal lattice position exceeds the radius of the sphere, the atom is marked as an interstitial atom.

[0027] Preferably, the particle database includes the corresponding relationships between the parameters of the space incident particles. The parameters of the space incident particles include the particle type and element type of the space incident particles; the device material database includes the relationship between the material type and the parameters of the simulation box. The parameters of the simulation box include the size, shape, and material of the simulation box.

[0028] The advantages of the present invention compared with the prior art are as follows:

[0029] The present invention studies the evolution and accumulation of irradiation defects by using the molecular dynamics method, conducts cross-scale computational simulations on the generation of defects in aerospace device materials by incident particles with different parameters (temperature, incident energy, and relative atomic mass), and the research object transitions from the particle scale to the mesoscopic system or even macroscopic system that includes the long-time evolution of defects, thereby improving the prediction efficiency and providing important basis and ideas for the material selection of semiconductor devices.

[0030] The present invention also provides a cross-scale simulation system based on the whole process of device radiation defect evolution, including:

[0031] An assignment module is used to call different types of spatial incident particles from a particle database, and input the parameters of the spatial incident particles into a simulation system for initialization; it is also used to call different types of materials from a device material database, and input the parameters of a simulation box corresponding to the material type into the simulation system;

[0032] A modeling and simulation module is used to establish a molecular dynamics model based on the spatial incident particles according to the basic principles of molecular dynamics, perform numerical simulation of the motion of the spatial incident particles based on molecular dynamics, and obtain the motion trajectory, velocity, energy and distribution information of the spatial incident particles in the spatial device at different times;

[0033] The computational simulation module is used to use statistical physics methods to statistically analyze the motion laws of the spatial incident particles under different states, associate the microscopic states of the spatial incident particles with the macroscopic properties of the material, and perform cross-scale computational simulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a flow chart of a cross-scale simulation method based on the whole process of device radiation defect evolution in an embodiment of the present invention;

[0035] Figure 2 It is a simulation calculation flow chart based on OKMC in an embodiment of the present invention. DETAILED DESCRIPTION

[0036] In the process of high-energy incident particles irradiating space devices, the incident particles collide with the lattice atoms of the device material to produce primary dislocated atoms. The primary dislocated atoms continue to interact with the surrounding atoms to cause cascade collisions, thereby causing defects in the device material. The generation of defects in the device and the aggregation and evolution of defects in the material cause changes in the microstructure of the material, and further lead to the degradation of the macroscopic properties of the material and the formation of various physical phenomena, such as radiation swelling, radiation hardening, radiation embrittlement, and decreased thermal conductivity.

[0037] It can be seen that the entire process of radiation damage generation roughly includes the process of generating primary displaced atoms in the device after the incident particles irradiate the device, the process of primary displaced atoms shifting cascade, and the process of defect generation and evolution. The present invention is aimed at the entire process of evolution of radiation defects in space devices. Since the generation and evolution of displacement damage spans multiple time scales, in order to fully understand the generation of single-particle displacement damage current and its annealing behavior, it is necessary to adopt a combination of multiple methods for research. The present invention combines existing experimental data and theoretical prediction models to obtain an understanding of material behavior under irradiation conditions without experimental data, and explains in detail the mechanisms and processes observed in the experiment, thereby accelerating material development, selection and service confirmation.

[0038] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0039] The present invention uses a single incident particle and material parameters as research objects to simulate the cross-scale defect evolution effect produced by the incident particle in the space device. Figure 1 As shown, a cross-scale simulation method based on the entire process of device radiation defect evolution is provided, including:

[0040] Different types of spatial incident particles are called from the particle database, and the parameters of the spatial incident particles are input into the simulation system for initialization; different types of materials are called from the device material database, and the parameters of the simulation box corresponding to the material type are input into the simulation system; According to the basic principles of molecular dynamics, a molecular dynamics model based on spatial incident particles (hereinafter also referred to as incident particles, particles) is established, and numerical simulation of the motion of spatial incident particles based on molecular dynamics is performed, and the motion trajectory, velocity, energy and distribution information of spatial incident particles at different times in the spatial device are obtained;

[0041] Statistical physics methods are used to calculate the motion laws of spatial incident particles under different states, and the microscopic state of spatial incident particles and the macroscopic properties of materials are correlated to perform cross-scale computational simulation.

[0042] The particle database includes the corresponding relationship between various parameters of the spatial incident particles, and the parameters of the spatial incident particles include the particle type (such as atom, proton, Chinese character, etc.), element type, etc. of the spatial incident particles;

[0043] Among them, the device material database includes the relationship between the material type and the parameters of the simulation box. The parameters of the simulation box include the size, shape (such as square, round and other self-built structural shapes), material, etc. of the simulation box.

[0044] For example, the damage caused by a certain type of incident particle irradiating a space device of a certain material is simulated. For example, a certain type of particle is input into the particle database, the parameters related to the particle are called out, and the particle parameters are input into the simulation system to simulate the incident particles; for example, if the device is a rectangular block made of aluminum and has a certain length, width and height, the parameters of the simulation box corresponding to the device material (such as the material, size and shape of the simulation box) are called from the device material database and assigned to the simulation system to simulate the device material. The parameters of the incident particles and the parameters of the simulation box corresponding to the device material parameters are input into the simulation system for modeling.

[0045] This embodiment uses molecular dynamics methods to study the evolution and accumulation of irradiation defects, and conducts cross-scale computational simulation of the generation of defects in aerospace device materials by incident particles with different parameters (temperature, incident energy and relative atomic mass). The research object transitions from the particle scale to the mesoscopic system or even the macroscopic system that includes the long-term evolution of defects, thereby improving the prediction efficiency and providing important basis and ideas for the material selection of semiconductor devices.

[0046] In some of the embodiments, in the particle irradiation simulation, different boundary conditions are first added to the simulation box, which represent the specific interaction between the atoms in the simulation box and the surrounding environment and allow the simulation of different phenomena. Because the incident particles act in a box within a certain range, the numerical simulation of the motion of spatial incident particles based on molecular dynamics in this embodiment includes: adding fixed boundary conditions in the molecular dynamics simulation to simulate the physical phenomena caused by the specific interaction between the particles in the box and the surrounding environment. In the molecular dynamics simulation, the microcanonical ensemble NVE is used, and periodic boundary conditions are applied only in the 1 / 2 direction. The motion integration step size of the molecular dynamics simulation is 1fps, and the total simulation time is 50fps.

[0047] In some of the embodiments, obtaining the motion trajectory, velocity, energy and distribution information of spatial incident particles in a spatial device at different times specifically includes: using the Verlet algorithm to calculate the actual position of the i-th atom at time t in the spatial device coordinate system, and obtaining the displacement and velocity of the spatial incident particles in the spatial device of calibration materials and structures, thereby obtaining the motion trajectory, velocity, energy and distribution of the spatial incident particles in the device structure at different times.

[0048] Furthermore, after calculating the position distribution, velocity and acceleration of the spatial incident particles, the statistical physics method is used to statistically calculate the motion law of the spatial incident particles under different states, and the microscopic state of the spatial incident particles is associated with the macroscopic properties of the material to achieve cross-scale computational simulation. In this embodiment, primary knockout atoms (PKA) of three energy points are selected as spatial incident particles, which are PKAs with energies of 2 keV, 5 keV and 10 keV, respectively. The corresponding simulation system sizes are 28a0× 28a0 × 25a0, 40a0 × 40a0 × 28a0, and 50a0 × 50a0 × 40a0, respectively. The Langevin heat bath is used to control the temperature of the outermost atoms in the simulation system except the upper surface, and the canonical ensemble is used in the internal cascade collision zone.

[0049] In some of these embodiments, the cross-scale computational simulation includes: First, use OKMC (Object Kinetic Monte Carlo) to simulate the short-time low-dpa (physical damage degree) evolution of spatially incident particles, then use the equivalent sphere method to identify the spatially incident particles corresponding to the generated defects, and then use the cluster dynamics method to perform the high-dpa defect evolution of the spatially incident particles.

[0050] Specifically, use OKMC to simulate the short-time low-dpa evolution of spatially incident particles, making the particles jump randomly or on specific lattices to simulate the formation of defects in the device. Calculate the probability of the jump by inputting the particle transition rate, then set a random direction in the lattice for the jump, and finally output the jump target positions of the particles at different times. Figure 2 For the process based on OKMC simulation, first initialize and input parameters such as incident particles (element type, mass, etc.) and materials (materials and sizes of the simulation box) by calling the database, then select the process of activating the particle action, use random direction transition or specified direction transition, as time increases, use OKMC to update the positions of the particles in the device, and finally complete the evolution process through the set time length, that is, end the process when the evolution time reaches the set time.

[0051] The specific process is as follows:

[0052] 1) Set the initial time t = 0;

[0053] 2) Calculate the probability of local defect generation and give a list of local defect generation;

[0054] 3) Calculate the total rate R and obtain a uniform random number r in (0, 1] to select an event that may generate a defect;

[0055] 4) Search for i by R×i - 1 < r×R < R×i to obtain the event to execute i, and execute event i (update the current state);

[0056] 5) Update the time , and accumulate the defect generation along with the simulation time;

[0057] 6) Determine the transition probability according to the Boltzmann constant, the absolute temperature in the system, and the Planck constant, and update the transition probability according to the temperature in the system.

[0058] Among them, after simulating the evolution of particle movement, in order to identify which spatial incident particles cause defects, the equivalent sphere method is used to check whether the spatial incident particles are within a sphere of a specified radius centered on the ideal lattice position of the atom. The so-called atom refers to the atom in the device, and the so-called ideal lattice position refers to the position where the atom should be when it is not affected by the spatial incident particles and its position changes, that is, the lattice position where it should be before deviation occurs. If the sphere centered on the ideal lattice position does not contain an atom, the lattice position is marked as a vacancy. Furthermore, if the atom is displaced from its nearest ideal lattice position by more than the radius of the sphere, the atom is marked as an interstitial atom. On this basis, cluster dynamics is used to carry out long-term, multi-particle, high-dpa defect evolution calculations.

[0059] Although the disclosure is disclosed as above, the protection scope of the disclosure is not limited thereto. Those skilled in the art may make various changes and modifications without departing from the spirit and scope of the disclosure, and these changes and modifications will fall within the protection scope of the present invention.

Claims

1. A cross-scale simulation method based on the entire process of device radiation defect evolution, characterized in that: Including: Call different types of space incident particles from the particle database, input the parameters of the space incident particles into the simulation system for initialization; and call different types of materials from the device material database, and input the parameters of the simulation box corresponding to the material type into the simulation system; According to the basic principles of molecular dynamics, establish a molecular dynamics model based on the space incident particles, conduct numerical simulation of the motion of the space incident particles based on molecular dynamics, and obtain the motion trajectories, velocities, energies, and distribution information in the space device of the space incident particles at different times; Adopt statistical physics methods to statistically analyze the motion laws of the space incident particles in different states, correlate the microscopic states of the space incident particles with the macroscopic properties of the materials, and conduct cross-scale computational simulations; The cross-scale computational simulation includes: Adopt the OKMC method to simulate the short-time low physical damage degree evolution of the space incident particles, and make the space incident particles jump randomly or on specified lattices to simulate the formation of defects in the space device; Adopt the equivalent sphere method to identify the space incident particles corresponding to the generated defects; Adopt the cluster dynamics method to conduct the high physical damage degree evolution of the space incident particles.

2. The cross-scale simulation method based on the whole process of device radiation defect evolution according to claim 1 is characterized in that: The obtaining of the motion trajectories, velocities, energies, and distribution information in the space device of the space incident particles at different times includes: Adopt the verlet algorithm to calculate the actual position of the i-th atom at time t in the space device coordinate system, obtain the displacement and velocity of the space incident particles in the space device, so as to obtain the motion trajectories, velocities, energies, and distribution information in the space device of the space incident particles at different times.

3. The cross-scale simulation method based on the whole process of device radiation defect evolution according to claim 2 is characterized in that: Also including: When conducting molecular dynamics simulation, add fixed boundary conditions to the simulation box to simulate the physical phenomena generated by the specific interaction between the particles in the simulation box and the surrounding environment.

4. The cross-scale simulation method based on the whole process of device radiation defect evolution according to claim 3 is characterized in that: Also including: When conducting molecular dynamics simulation, adopt the microcanonical ensemble and apply periodic boundary conditions only in the 1 / 2 direction.

5. The cross-scale simulation method based on the whole process of device radiation defect evolution according to claim 4 is characterized in that: In the molecular dynamics simulation, the motion integration step size is 1 fps, and the total simulation duration is 50 fps.

6. The cross-scale simulation method based on the whole process of device radiation defect evolution according to claim 1 is characterized in that: The adopting of the OKMC method to simulate the short-time low physical damage degree evolution of the space incident particles includes: Set the initial time t = 0; Calculate the probability of local defect generation and give a local defect generation list; Calculate the total rate R, and obtain a uniform random number r in (0, 1] to select an event that may generate a defect; Search for i through R×i - 1 < r×R < R×i to obtain the event to execute i, and execute event i; Update time , the accumulation of defects occurs along with the simulation time; Determine the transition probability according to the Boltzmann constant, the absolute temperature in the system, and the Planck constant, and update the transition probability according to the temperature in the system.

7. The cross-scale simulation method based on the whole process of device radiation defect evolution according to claim 1 is characterized in that: The adopting of the equivalent sphere method to identify the space incident particles corresponding to the generated defects includes: The simulation box is divided into several equivalent grids to check whether the spatial incident particle is within a sphere of a specified radius centered on the ideal lattice position of the atom in the spatial device. If the atom is not contained in the sphere centered on the ideal lattice position, the lattice position corresponding to the atom is marked as a vacancy. If the displacement distance of the atom from its nearest ideal lattice position exceeds the radius of the sphere, the atom is marked as an interstitial atom.

8. The cross-scale simulation method based on the whole process of device radiation defect evolution according to claim 1 is characterized in that: The particle database includes the corresponding relationship between the various parameters of the spatial incident particles, and the parameters of the spatial incident particles include the particle type and element type of the spatial incident particles; the device material database includes the relationship between the material type and the parameters of the simulation box, and the parameters of the simulation box include the size, shape and material of the simulation box.

9. A cross-scale simulation system based on the entire process of device radiation defect evolution, characterized in that: include: An assignment module is used to call different types of spatial incident particles from a particle database, and input the parameters of the spatial incident particles into a simulation system for initialization; it is also used to call different types of materials from a device material database, and input the parameters of a simulation box corresponding to the material type into the simulation system; A modeling and simulation module is used to establish a molecular dynamics model based on the spatial incident particles according to the basic principles of molecular dynamics, perform numerical simulation of the motion of the spatial incident particles based on molecular dynamics, and obtain the motion trajectory, velocity, energy and distribution information of the spatial incident particles in the spatial device at different times; A computational simulation module, for using a statistical physics method to calculate the motion laws of the spatial incident particles under different states, to correlate the microscopic state of the spatial incident particles with the macroscopic properties of the material, and to perform cross-scale computational simulation; The cross-scale computing simulation includes: The OKMC method is used to simulate the short-time low physical damage level evolution of the spatial incident particles, so that the spatial incident particles are randomly or jump on a specified lattice to simulate the formation of defects in the spatial device; Using the equivalent sphere method to identify the spatial incident particles corresponding to the defect; The cluster dynamics method is used to evolve the high physical damage level of the spatially incident particles.

Images