Evaluation method and system of molecular dynamics potential function based on material structure simulation

Abstract

The present application relates to a method for evaluating a molecular dynamics potential function based on material structure simulation. The method comprises: S1, receiving potential function and structural information; S2, obtaining a primitive cell structure according to the structural information, and analyzing the number of elastic coefficients, point defects and surface structures; S3, generating a supercell structure according to the primitive cell structure; S4, generating a job input file according to the potential function information, and submitting a calculation job to obtain the properties of the supercell structure after relaxation; S5, constructing at least two deformed substructures using the supercell structure, submitting a calculation job to obtain a relaxation energy value, and calculating the elastic coefficient; S6, constructing a corresponding point defect structure using the supercell structure, submitting a calculation job to obtain a relaxation energy value, and calculating the point defect formation energy; S7, constructing a corresponding surface structure using the supercell structure, submitting a calculation job to obtain a relaxation energy value, and calculating the surface formation energy. The present application can use any material structure to quickly and effectively evaluate any potential function.

Description

Technical Field

[0001] The present application relates to molecular dynamics simulation technology, and more specifically, to a method and system for evaluating molecular dynamics potential function based on material structure simulation. Background Art

[0002] As one of the important simulation methods in materials science, molecular dynamics (MD) plays an irreplaceable role in simulating material phenomena at the atomic scale. The simulation of radiation damage, defect evolution, phase change, fracture and other phenomena often requires more than one million steps of simulation on hundreds of thousands or even hundreds of millions of atoms. At such a small scale, a large number of simulation details of these phenomena are difficult to verify directly in experiments, and these details are precisely important information to reveal the mechanisms of various material phenomena and deepen the theoretical understanding of materials science. Therefore, how to ensure the accuracy of these simulation details becomes crucial.

[0003] Molecular dynamics simulations rely on potential functions to determine the interactions between atoms during simulations. However, the performance of potential functions is generally only mentioned by their authors when publishing literature, and they often only mention the favorable aspects and downplay the aspects with poor performance. This makes researchers lack a comprehensive understanding and knowledge of a potential function they use, which brings certain uncertainties to the simulation results. In order to effectively evaluate the potential function, it is necessary to use the potential function to simulate some basic physical properties of the relevant structure, and then compare these properties with experimental values. These physical properties to be evaluated should include but are not limited to the cohesive energy, elastic modulus, point defect formation energy, surface formation energy, melting point, thermal expansion coefficient, interface energy, etc. of the structure. There are many types of crystal potential functions published in the current scientific research community, covering substances such as elements, compounds, and metal alloys. Their crystal structures also mainly include cubic, tetragonal, orthorhombic, hexagonal and other crystal systems. Each type of substance and crystal system has a different number of physical quantities that need to be tested, and the structures and simulation parameters required for each physical quantity are different. The workload, complexity and tediousness of the evaluation simulation are relatively large. Therefore, the industry needs a universal, comprehensive and automated algorithm process to handle the evaluation of potential functions. Summary of the invention

[0004] The technical problem to be solved by the present application is to provide a universal, automatable method for evaluating molecular dynamics potential functions based on material structure simulation, which can evaluate any potential function based on any material structure, in response to the above-mentioned defects of the prior art.

[0005] The technical solution adopted by the present application to solve the technical problem is: to propose a method for evaluating the molecular dynamics potential function based on material structure simulation, comprising the following steps:

[0006] S1, receiving the potential function information input by the user and the structure name of the material structure to be simulated;

[0007] S2. For each material structure to be simulated, obtain a primitive cell structure according to the structure name, and analyze the elastic coefficient, point defects and number of surface structures related to the primitive cell structure;

[0008] S3, generating a supercell structure according to the primitive cell structure, wherein the supercell structure is an extension of the primitive cell structure in three main axis directions in three-dimensional space;

[0009] S4, generating a job input file for performing relaxation calculation simulation on the supercell structure according to the potential function information input by the user, and generating a script to submit the calculation job, and obtaining the relaxation energy value and structure size of the supercell structure after relaxation from the output information of the job;

[0010] S5. For each elastic coefficient analyzed in step S2, the supercell structure is deformed in the deformation direction to construct at least two deformed substructures, and a calculation job is submitted for each deformed substructure in the same manner as in step S4 to obtain relaxation energy values ​​of the at least two deformed substructures, and the relaxation energy values ​​of the at least two deformed substructures and the relaxation energy values ​​of the supercell structure obtained in step S4 are fitted to an energy-volume quadratic curve with deformation as a variable, and the elastic coefficient of the material structure is calculated using the curve according to the following formula:

[0011]

[0012] Wherein, E is the relaxation energy value, V is the volume, and V0 is the volume of the relaxed structure after the supercell structure obtained in step S4 is relaxed;

[0013] S6. According to the point defect information analyzed in step S2, a corresponding point defect structure is constructed using the supercell structure, and a calculation job is submitted for each point defect structure in the same manner as in step S4 to obtain a relaxation energy value of each point defect structure, and the relaxation energy value of the obtained point defect structure and the relaxation energy value of the supercell structure obtained in step S4 are calculated according to the following formula to obtain the point defect formation energy of the material structure:

[0014] E def =E n±m -(nE 0 ±mμ i )

[0015] Among them, E def is the point defect formation energy, E n±m is the structural energy including m defect atoms, E0 is the average relaxation energy per atom of the supercell structure obtained in step S4, n is the number of atoms in the supercell structure, m is the number of atoms in the defect itself, and μ i is the atomic chemical potential in the perfect structure of the i-th element, with a minus sign for vacancy defects and a plus sign for interstitial defects;

[0016] S7, according to the surface structure information analyzed in step S2, the corresponding surface structure is constructed using the supercell structure, and a calculation job is submitted for each surface structure in the same manner as in step S4 to obtain the relaxation energy value of each surface structure, and the relaxation energy value of the obtained surface structure and the relaxation energy value of the supercell structure obtained in step S4 are calculated according to the following formula to obtain the surface formation energy of the material structure:

[0017]

[0018] Among them, E (ijk) is the surface formation energy of the (ijk) surface, is the energy of the surface structure of n atoms, E 0 is the average relaxation energy per atom of the supercell structure obtained in step S4, A (ijk) is the surface area of ​​the (ijk) surface.

[0019] According to an embodiment of the method for evaluating the molecular dynamics potential function based on material structure simulation described in the present application, the potential function information received in step S1 includes a potential function name, a parameter value or a parameter file.

[0020] According to one embodiment of the method for evaluating the molecular dynamics potential function based on material structure simulation described in the present application, obtaining the primitive cell structure according to the structure name in step S2 includes: obtaining the primitive cell structure of the material structure from a general structure database of materials science according to the structure name.

[0021] According to one embodiment of the method for evaluating the molecular dynamics potential function based on material structure simulation described in the present application, the step S3 further includes: when the length of any side of the primitive cell structure is less than twice the cutoff distance of the potential function, expanding each direction of the supercell structure to an integer multiple of the size of the primitive cell structure until exceeding twice the cutoff distance.

[0022] According to one embodiment of the method for evaluating the molecular dynamics potential function based on material structure simulation described in the present application, step S3 further comprises: expanding the size of the supercell structure in each direction to at least

[0023] According to one embodiment of the method for evaluating the molecular dynamics potential function based on material structure simulation described in the present application, step S4 further includes: submitting the job to a high-performance computing server or a supercomputing system, and after the job is submitted, calling the molecular dynamics program in the system to perform the simulation work.

[0024] According to one embodiment of the method for evaluating the molecular dynamics potential function based on material structure simulation described in the present application, step S5 further comprises:

[0025] S51, for each elastic coefficient, using the supercell structure to perform 1%, 0.5%, -0.5%, -0.1% stretching or shearing deformation in the deformation direction, respectively, to construct four deformed substructures;

[0026] S52, submitting calculation jobs for the four deformed substructures respectively in the same manner as in step S4 to obtain relaxation energy values ​​of the four deformed substructures;

[0027] S53, fitting the relaxation energy values ​​of the four deformed substructures and the relaxation energy value of the supercell structure obtained in step S4 to an energy-volume quadratic curve with deformation as a variable, and calculating the corresponding elastic coefficient by fitting the curve;

[0028] S54, repeating steps S51 to S53 until all elastic coefficients analyzed in step S2 are calculated.

[0029] According to an embodiment of the method for evaluating the molecular dynamics potential function based on material structure simulation described in the present application, the point defects of the unit cell structure analyzed in step S2 include vacancy defects and gap defects; the step S6 of constructing the corresponding point defect structure using the supercell structure further includes: for vacancy defects, a vacancy defect structure is constructed for each position of each element in the unit cell structure; for gap defects, a gap defect structure is constructed for each element in each larger space in the unit cell structure, wherein the larger space is defined as a space inside the structure that can accommodate a space with a radius greater than The space in which a sphere is located that does not include any atoms.

[0030] According to one embodiment of the method for evaluating the molecular dynamics potential function based on material structure simulation described in the present application, the surface structure of the unit cell structure analyzed in step S2 includes a small index interface; the length, width, thickness and vacuum layer of the surface structure constructed in step S7 are all more than twice the cutoff distance of the potential function used.

[0031] The present application also proposes a system, including a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, the method for evaluating the molecular dynamics potential function based on material structure simulation as described above is implemented.

[0032] The present application also proposes a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the method for evaluating the molecular dynamics potential function based on material structure simulation as described above.

[0033] The implementation of the method for evaluating the molecular dynamics potential function based on material structure simulation of the present application has the following beneficial effects: (1) The method for evaluating the molecular dynamics potential function based on material structure simulation of the present application is a general method that can flexibly evaluate the molecular dynamics potential function. The method can use any material structure to evaluate any potential function, which has a significant promoting effect on the effective evaluation and screening of potential functions; (2) The method can be run in a pipelined high-throughput manner on a supercomputer or high-performance computing platform with minimal human intervention, thereby realizing rapid batch evaluation and screening of potential functions, greatly reducing the preliminary work of scientific researchers before conducting material simulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The present application will be further described below with reference to the accompanying drawings and embodiments, in which:

[0035] Figure 1 is a flow chart of a method for evaluating a molecular dynamics potential function based on material structure simulation according to one embodiment of the present application;

[0036] Figure 2 is a detailed flow chart of a method for evaluating a molecular dynamics potential function based on material structure simulation according to another embodiment of the present application;

[0037] Figure 3 It is a schematic diagram of the primitive cell structure of the α-U structure of uranium metal;

[0038] Figure 4 It is a schematic diagram of the supercell structure of the expanded α-U structure;

[0039] Figure 5 is the C for calculating α-U 11 Fitting curve of elastic coefficient;

[0040] Figure 6 is a schematic diagram of the vacancy defect structure of α-U;

[0041] Figure 7a It is a schematic diagram of an interstitial defect structure of α-U;

[0042] Figure 7bIt is a schematic diagram of another interstitial defect structure of α-U;

[0043] Figure 8 is a schematic diagram of the (001) surface structure of α-U;

[0044] Fig. 9 It is a schematic diagram of the primitive cell structure of the γ-U structure of uranium metal;

[0045] Fig.10 It is a schematic diagram of the supercell structure of the expanded γ-U structure;

[0046] Fig.11 is the C for calculating γ-U 11 Fitting curve of elastic coefficient;

[0047] Fig.12 is a schematic diagram of the vacancy defect structure of γ-U;

[0048] Fig.13 is a schematic diagram of the interstitial defect structure of γ-U;

[0049] Fig.14a is a schematic diagram of the (100) surface structure of γ-U;

[0050] Fig.14b is a schematic diagram of the (110) surface structure of γ-U;

[0051] Fig.14c is a schematic diagram of the (111) surface structure of γ-U;

[0052] Fig.15 It is a schematic diagram of the structure of a system according to an embodiment of the present application. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. In addition, the embodiments in the present application and the features in the embodiments may be combined with each other in the absence of conflict.

[0054] The present application proposes a method for evaluating the molecular dynamics potential function based on material structure simulation, which can automatically use the potential function to simulate multiple physical properties of a specified structure after knowing the minimum potential function and structural information, thereby achieving the purpose of comprehensive and rapid evaluation of the potential function. Figure 1 FIG. 1 is a flow chart of a method 100 for evaluating a molecular dynamics potential function based on material structure simulation according to an embodiment of the present application. Figure 1 As shown, the molecular dynamics potential function evaluation method 100 based on material structure simulation includes the following steps:

[0055] Step S1, receiving potential function information input by a user and the structure name of a material structure to be simulated, wherein the potential function information input by the user at least includes a potential function name, a parameter value or a parameter file.

[0056] Step S2, for each material structure to be simulated, obtain the primitive cell structure according to the structure name of the material structure, and analyze the elastic coefficient, point defects and the number of surface structures related to the primitive cell structure.

[0057] Step S3, generating a supercell structure according to the primitive cell structure, wherein the supercell structure is an extension of the primitive cell structure in three main axis directions in three-dimensional space.

[0058] Step S4, generating a job input file for relaxation calculation simulation for the supercell structure according to the potential function information input by the user, and generating a script to submit the calculation job, and obtaining the relaxation energy value and structure size of the supercell structure after relaxation from the output information of the job.

[0059] Step S5, for each elastic coefficient analyzed in step S2, deform the supercell structure in the deformation direction to construct at least two deformed substructures, and submit a calculation job for each deformed substructure in the same manner as in step S4 to obtain relaxation energy values ​​of the at least two deformed substructures, and fit the relaxation energy values ​​of the at least two deformed substructures and the relaxation energy values ​​of the supercell structure obtained in step S4 to an energy-volume quadratic curve with deformation as a variable, and use the curve to calculate the elastic coefficient of the material structure according to the following formula (1):

[0060]

[0061] Wherein, E is the relaxation energy value, V is the volume, and V0 is the volume of the relaxed structure after the supercell structure obtained in step S4 is relaxed.

[0062] Step S6, according to the point defect information analyzed in step S2, the corresponding point defect structure is constructed using the supercell structure, and a calculation job is submitted for each point defect structure in the same manner as in step S4 to obtain the relaxation energy value of each point defect structure, and the relaxation energy value of the obtained point defect structure and the relaxation energy value of the supercell structure obtained in step S4 are calculated according to the following formula (2) to obtain the point defect formation energy of the material structure:

[0063] E def =E n±m -(nE 0 ±mμ i ) (2)

[0064] Among them, E def is the point defect formation energy, E n±m is the structural energy including m defect atoms, E 0 is the average relaxation energy per atom of the supercell structure obtained in step S4, n is the number of atoms in the supercell structure, m is the number of atoms in the defect itself, and μ i is the atomic chemical potential in the perfect structure of the i-th element. The vacancy defect is calculated with a minus sign, and the interstitial defect is calculated with a plus sign.

[0065] Step S7, according to the surface structure information analyzed in step S2, the corresponding surface structure is constructed using the supercell structure, and a calculation job is submitted for each surface structure in the same manner as in step S4 to obtain the relaxation energy value of each surface structure, and the relaxation energy value of the obtained surface structure and the relaxation energy value of the supercell structure obtained in step S4 are calculated according to the following formula (3) to obtain the surface formation energy of the material structure:

[0066]

[0067] Among them, E (ijk) is the surface formation energy of the (ijk) surface, is the energy of the surface structure of n atoms, E 0 is the average relaxation energy per atom of the supercell structure obtained in step S4, A (ijk) is the surface area of ​​the (ijk) surface.

[0068] According to the molecular dynamics potential function evaluation method 100 based on material structure simulation according to the above embodiment of the present application, any potential function can be evaluated using any structure, which has a significant promoting effect on the effective evaluation and batch rapid screening of potential functions.

[0069] Figure 2 The flowchart of the method 200 for evaluating the molecular dynamics potential function based on material structure simulation according to another specific embodiment of the present application is further shown. Figure 2 As shown, the method 200 for evaluating the molecular dynamics potential function based on material structure simulation starts at step S201 .

[0070] Then in step S202, the potential function and structure information input by the user are received. The potential function information input by the user includes the potential function name, parameter value or parameter file, etc. The structure information input by the user includes at least the structure names of all structures to be simulated.

[0071] Then the method 200 simulates all input structures one by one, and in step S203 determines whether all structures have been simulated. If so, the process enters step S222 and ends. If not, it enters step S204.

[0072] In step S204, the ground state unit cell structure is obtained according to the structural information input by the user. The unit cell structure information includes the unit cell side length, angle and position of all atoms. In specific implementation, the unit cell structure of the material structure can be obtained from the general structure database of materials science according to the structure name input by the user. Commonly used structure databases include but are not limited to ICSD (https: / / icsd.products.fiz-karlsruhe.de), COD (http: / / www.crystallography.net / cod / ), AMCSD (http: / / rruff.geo.arizona.edu / AMS / amcsd.php), etc.

[0073] Then, in step S205, the elastic coefficients, point defects and surface structures related to the structure are analyzed according to the unit cell structure. The number of elastic coefficients of the unit cell structure is determined by the crystal structure of the unit cell according to general knowledge of materials science. For example, for a cubic structure, only C 11 , C 12 and C 44 ; For the hexagonal structure, add C to the cubic structure 13 and C 33 ; For the tetragonal structure, add C to the hexagonal structure 66 ; For the orthorhombic structure, add C on the basis of the tetragonal structure 22 , C 23 and C 55 For other complex crystal systems, corresponding adjustments are made according to materials science. For the most general structure, the elastic coefficients include up to 21 (C 11 , C 12 , ..., C 16 , C 22 , C 23 , ..., C 66 ), that is, all C ij And remove C ij =C ji The point defects of the unit cell structure include vacancy defects and interstitial defects. The surface structure of the unit cell structure includes the commonly used small index interface, the Miller index does not exceed 3, such as (100), (111), (310) surface.

[0074] Then, in step S206, a supercell structure is generated based on the primitive cell structure. The supercell structure is the expansion of the primitive cell structure in three main axis directions in three-dimensional space. If the length of any side of the primitive cell structure is less than twice the cutoff distance of the potential function, the supercell structure needs to be expanded to an integer multiple of the size of the primitive cell structure in the direction of the side until it exceeds twice the cutoff distance of the potential function. Preferably, the size of the supercell structure in each direction is expanded to at least

[0075] Then in step S207, a job input file is generated according to the potential function information input by the user to perform relaxation calculation simulation on the supercell structure, that is, energy minimization simulation without restricting structural changes.

[0076] Then in step S208, the corresponding job submission script is written according to the high-performance computing server or supercomputing system, and the molecular dynamics program of the system is called to simulate the ground state structure after the job is submitted. Preferably, the simulation program uses LAMMPS (Large-scale Atomic / Molecular Massively Parallel Simulator). The program was developed by Sandia National Laboratories in the United States and is mainly used for some calculations and simulations related to molecular dynamics. LAMMPS can support atomic and molecular systems of millions of levels, including gaseous, liquid or solid phases, various ensembles, and provides support for a variety of potential functions. All calculations are performed using the LAMMPS program, and the structure file, potential function parameters and input files are all in a LAMMPS compatible format. Specifically, the structural relaxation was simulated using the NPT (isobaric isothermal) method, using the fix npt command in LAMMPS, at 300 K and 0 bar for 500 000 steps, and then at 0.001 K and 0 bar for 200 000 steps, with a time step of 1 femtosecond per step.

[0077] Then in step S209, after the job is completed, the energy value and structure size of the supercell structure after relaxation are obtained from the output information of the job.

[0078] Then, the method 200 calculates each elastic coefficient analyzed in step S205 using the properties of the ground state structure obtained in step S209. In step S210, it is determined whether all elastic coefficients have been calculated. If so, the process proceeds to step S214, and if not, the process proceeds to step S211.

[0079] In step S211, for the calculation of each elastic coefficient, the supercell structure is used to perform 1%, 0.5%, -0.5%, -0.1% stretching or shearing deformation in the deformation direction to construct four deformation substructures;

[0080] Then in step S212, the LAMMPS calculation job is submitted for the four deformed substructures in the same manner and setting as in step S208 to obtain the relaxation energy values ​​of the four deformed substructures. Specifically, the NVT (isothermal isovolumetric) method can be used to simulate each deformed substructure, using the fix nvt command in the LAMMPS program, simulating 200,000 steps at a temperature of 300K, and then simulating 200,000 steps at a temperature of 0.001K, with a time step of 1 femtosecond per step.

[0081] Then in step S213, the relaxation energy values ​​of the four deformed substructures and the relaxation energy value of the supercell structure obtained in step S209 are fitted to an energy-volume quadratic curve with deformation as a variable, and the elastic coefficient is calculated using the curve according to the above formula (1).

[0082] Then, the process returns to step S210 and the aforementioned steps S211 to S213 are repeated until all elastic coefficients are calculated.

[0083] Then, the method 200 calculates the point defect formation energy for each point defect analyzed in step S205 using the properties of the ground state structure obtained in step S209. In step S214, it is determined whether all point defects have been calculated. If so, the process proceeds to step S218, and if not, proceeds to step S215.

[0084] In step S215, for each point defect, a corresponding point defect structure is constructed using the supercell structure. Point defects include vacancy defects and gap defects. For vacancy defects, a vacancy defect structure should be constructed for each position of each element in the primitive cell structure; for gap defects, a gap defect structure should be constructed for each element in each larger space in the primitive cell structure. A larger space can be defined as a structure that can contain a radius greater than If the sphere does not include any atoms, the space where the sphere is located can be considered as a larger space, and the atoms of the interstitial defects should be placed at the center of the sphere.

[0085] Then in step S216, a LAMMPS calculation job is submitted for each point defect structure in the same manner and setting as in step S208 to obtain the relaxation energy value of each point defect structure. Specifically, the point defect structure is simulated using the NVT (isothermal isovolumetric) method, using the fix nvt command in the LAMMPS program, simulating 500,000 steps at a temperature of 300K, and then simulating 500,000 steps at a temperature of 0.001K, with a time step of 1 femtosecond per step.

[0086] Then in step S217, the relaxation energy value of the point defect structure obtained and the relaxation energy value of the supercell structure obtained in step S209 are calculated according to the aforementioned formula (2) to obtain the point defect formation energy.

[0087] Then return to step S214 and repeat the above steps S215 to S217 until all point defects are calculated.

[0088] Then, the method 200 calculates the surface formation energy for each surface structure analyzed in step S205 using the properties of the ground state structure obtained in step S209. In step S218, it is determined whether all surface formation energies have been calculated. If so, the process returns to step S203, and if not, it proceeds to step S219.

[0089] In step S219, the corresponding surface structure is constructed using the supercell structure. As mentioned above, the surface structure of the original cell structure includes commonly used small index interfaces, with a Miller index of no more than 3, such as (100), (111), and (310) surfaces. The length, width, thickness, and vacuum layer of the surface structure constructed using the supercell structure must all exceed twice the cutoff distance of the potential function used. Preferably, the length and width are at least The thickness and vacuum layer are at least

[0090] Then in step S220, a LAMMPS calculation job is submitted for each surface structure in the same manner and setting as in step S208 to obtain the relaxation energy value of each surface structure. Specifically, the surface structure is simulated using the NVT (isothermal isovolumetric) method, using the fix nvt command in the LAMMPS program, simulating 500,000 steps at a temperature of 300K, and then simulating 500,000 steps at a temperature of 0.001K, with a time step of 1 femtosecond per step.

[0091] Then in step S221, the relaxation energy value of the obtained surface structure and the relaxation energy value of the supercell structure obtained in step S209 are calculated according to the above formula (3) to obtain the surface formation energy.

[0092] Then return to step S218, repeat the above steps S219 to S221 until all surface formation energies are calculated, and return to step S203.

[0093] According to the above-mentioned embodiment of the present application, the molecular dynamics potential function evaluation method 200 based on material structure simulation can be operated in a pipelined high-throughput manner on a supercomputer or high-performance computing platform with minimal human intervention, thereby realizing rapid batch evaluation and screening of potential functions, greatly reducing the preliminary work of scientific researchers before conducting material simulation.

[0094] The COMB potential function of uranium metal will be used as an example to evaluate its simulation accuracy for α-U and γ-U structures.

[0095] Step 301: Obtain the COMB potential function file and parameters, as well as the α-U and γ-U structure information to be simulated from user input.

[0096] Step 302: First, simulate the α-U structure. The ground state primitive cell structure of the α-U structure (see Figure 3 The space group of the ) is Cmcm, and the length of the three sides is After considering spatial symmetry, the structure has 1 atomic station with coordinates (0, 0.10, 0.25).

[0097] Step 303: The α-U structure is an orthorhombic system with 9 elastic coefficients; there is 1 atomic station, so 1 vacancy defect can be constructed with coordinates (0, 0.10, 0.25); there are 2 large gaps in the structure, so 2 interstitial defects can be constructed with coordinates (0.5, 0.30, 0.38) and (0, 0.41, 0.45).

[0098] Step 304: Based on the primitive cell structure information of step 302, a 11×6×7 supercell structure is constructed, with a total number of atoms of 1848 and three side lengths of (See Figure 4 shown).

[0099] Step 305: Set the input file according to the LAMMPS format based on the input potential function information, including but not limited to parameters such as units, pair_style, pair_coeff, thermo_style, timestep, dump, fix, run, etc.

[0100] units metal

[0101] pair_style comb3 polar_off

[0102] pair_coeff **UO

[0103] thermo_stylecustom step elapsed temp etotal

[0104] thermo_modify norm yes

[0105] timestep 0.001

[0106] dump datalmp all custom 100000datalmp.*id type qxyz fix 1all npttemp 300 300 0.01 tri 0 0 0.1

[0107] run 500000

[0108] Step 306: Generate a job submission script run.sh in the supercomputer system:

[0109] #! / bin / bash

[0110] #SBATCH-J job

[0111] #SBATCH-N 1

[0112] #SBATCH-n 32

[0113] mpirun-np 32lmp-in input>output

[0114] And use the command sbatch run.sh to submit the job.

[0115] Step 307: The total energy of the structure obtained in the job output information is -9798.1eV, and the average energy of the atoms is -5.302eV. In the last output structure of the job, the length of the three sides is The total volume is Then the side length of the primitive cell structure of the ground state is:

[0116]

[0117] Step 308: From the analysis in step 303, it can be seen that α-U has a common C 11 , C 12 , C 13 , C 22 , C 23 , C 33 , C 44 , C 55 , C66 etc. 9 elastic coefficients. 11 , stretch the supercell structure in the x direction by 1%, 0.5%, -0.5%, -1%, and construct side lengths of 30.82, 30.97, and 31.29. The side lengths of the deformed substructures in the y and z directions are kept at 35.05 and The four deformed substructures are submitted to the job for calculation in the same manner as step 306 using the same input settings as step 305 .

[0118] Step 309: After the calculation in step 308 is completed, the relaxation energy values ​​are respectively obtained as -9795.21, -9797.37, -9797.37, and -9795.21 eV.

[0119] Step 310: Fit the relaxation energy values ​​of the four deformed substructures and the original supercell structure in step 309 together to obtain a quadratic energy-volume curve (see Figure 5 As shown), the fitting formula is E = 2.095 × 10 -5 V 2 -1.557V+19127, and then use formula (1) to calculate

[0120]

[0121] Step 311: Stretch the supercell structure in the y direction by 1%, 0.5%, -0.5%, and -1%, construct four deformed substructures, and calculate the energy-volume curve to obtain C 22 =139GPa. Similarly, constructing deformation substructures in the z, xy, xz, yz directions, we can get C 33 , C 12 , C 13 , C 23 By constructing shear deformation substructures in the x-plane in the y and z directions and in the y-plane in the z direction, we can obtain C 44 , C 55 , C 66 The coefficients are shown in Table 1.

[0122] Table 1: Elastic coefficients of α-U structures (unit: GPa).

[0123] C11 C12 C13 C22 C23 249 61 43 139 123 C33 C44 C55 C66 214 41 18 26

[0124] Step 312: Based on the structural information in step 303, construct a vacancy defect structure with the structural side length of step 307 (see Figure 6 ) and 2 interstitial defect structures (see Figure 7a and Figure 7b). Using the same input settings as step 305, the three point defect structures are submitted to the job for calculation in the same manner as step 306.

[0125] Step 313: After the calculation in step 312 is completed, the relaxation energy values ​​are respectively obtained as -9790.9, -9800.8, and -9800.8 eV.

[0126] Step 314: Calculate the relaxation energy value of the point defect structure in step 313 according to formula (2). Therefore, the formation energy of the vacancy defect is:

[0127] E V =-9790.9-1847(-5.302)=1.9eV

[0128] The formation energy of the gap defect is:

[0129] E I =-9800.8-1849(-5.302)=2.6eV

[0130] Step 315: Based on the structural information in step 303, a (001) surface structure is constructed with the side length of the structure in step 307, and the length and width are thickness Vacuum layer like Figure 8 shown.

[0131] Step 316: Using the same input settings as step 305, the surface structure in step 315 is submitted to the job for calculation in the same manner as step 306, and after the job is completed, the relaxation energy value -9790.9 eV is extracted.

[0132] Step 317: Calculate the relaxation energy value of the point defect structure in step 316 according to formula (3). Therefore, the formation energy of the (001) surface is:

[0133]

[0134] Step 318: After calculating the performance of the α-U structure, start calculating the γ-U structure. Fig. 9 The lengths of the three sides of The atomic coordinates are (0,0,0) and (0.5,0.5,0.5). The structure is a cubic crystal system with 3 elastic coefficients, and can construct 1 vacancy defect, 1 interstitial defect and 3 surface structures.

[0135] Step 319: Based on the size of the primitive cell structure, a 9×9×9 supercell structure is constructed with a total number of atoms of 1458 and a side length of (See Fig.10shown).

[0136] Step 320: Use the potential function file set in step 305 to submit the job in the supercomputer system in the manner of step 306.

[0137] Step 321: The total energy of the ground state γ-U is -7718.65 eV and the average atomic energy is -5.294 eV in the output information. In the output structure, the length of the three sides is The total volume is Then the actual side length of the primitive cell structure is:

[0138] Step 322: γ-U structure requires calculation of C 11 , C 12 , C 44 And other three elastic coefficients. For C 11 , stretch the supercell structure in the x direction by 1%, 0.5%, -0.5%, -1%, and construct side lengths of 30.44, 30.60, 30.90, The side lengths of the deformed substructures in the y and z directions are kept at The four deformed substructures are submitted to the job for calculation in the same manner as step 306 using the same input settings as step 305 .

[0139] Step 323: After the calculation in step 322 is completed, the relaxation energy values ​​are obtained as -7717.13, -7718.27, -7718.27, and -7717.13 eV respectively.

[0140] Step 324: Fit the relaxation energy values ​​of the four deformed substructures and the original supercell structure in step 323 together to obtain a quadratic energy-volume curve (see Fig.11 As shown), the fitting formula is E = 1.795 × 10 -5 V 2 -1.0438V+7455.4, and then use formula (1) to calculate

[0141]

[0142] Step 325: Construct a stretching deformation substructure in the xy direction and a shearing deformation substructure in the y direction on the x plane, and obtain C 12 and C 44 coefficient.

[0143] Table 2: Elastic coefficients of γ-U structures (unit: GPa).

[0144] <![CDATA[C 11 ]]> <![CDATA[C 12 ]]> <![CDATA[C 44 ]]> 167 104 50

[0145] Step 326: Based on the structural information in step 318, construct a vacancy defect structure with the side length of the structure in step 321 (see Fig.12 shown) and a gap defect structure (see Fig.13 The two point defect structures are submitted to the job for calculation in the same manner as step 306 using the same input settings as step 305.

[0146] Step 327: After the calculation in step 326 is completed, the relaxation energy values ​​are obtained as -7712.0 and -7721.8 eV respectively.

[0147] Step 328: Substitute the relaxation energy value in step 327 into formula (2) to obtain the vacancy defect formation energy:

[0148] E V =-7712.0-(1458-1)(-5.294)=1.4eV

[0149] And the interstitial defect formation energy:

[0150] E I =-7721.8-(1458+1)(-5.294)=2.1eV

[0151] Step 329: According to step 318, γ-U has three surface structures. Construct a (100) surface with a length and width of 30.75×30.75 and a thickness of Contains 1458 atoms (see Fig.14a As shown); construct a (110) surface with a length and width of 33.83×30.75 and a thickness of Contains 1764 atoms (see Fig.14b As shown); construct a (111) surface with a length and width of 33.48×33.83 and a thickness of Contains 1848 atoms (see Fig.14c All surfaces have a vacuum layer of

[0152] Step 330: Using the same input settings as step 305, the surface structure in step 329 is submitted to the job for calculation in the same manner as step 306, and after the job is completed, the relaxation energy values ​​of -7403.0, -9072.0, and -9425.2 are extracted.

[0153] Step 331: Substituting the relaxation energy value in step 330 into formula (3), the surface formation energies of the (100) surface, the (110) surface, and the (111) surface are obtained as follows:

[0154]

[0155] At this point, all steps are completed.

[0156] Based on the evaluation method of the molecular dynamics potential function based on material structure simulation in the above-mentioned embodiment, the present application also proposes a system 400. Fig.15 As shown, the system 400 includes a processor 41 and a memory 42, and the processor 41 and the memory 42 are in communication connection. The memory 42 stores a computer program, and when the computer program is executed by the processor 41, the processor 41 implements the method for evaluating the molecular dynamics potential function based on material structure simulation of the aforementioned embodiment of the present application.

[0157] The present application also proposes a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the method for evaluating the molecular dynamics potential function based on material structure simulation of the aforementioned embodiment of the present application.

[0158] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A method for evaluating molecular dynamics potential function based on material structure simulation, characterized in that: The steps include: S1, receiving the potential function information input by the user and the structure name of the material structure to be simulated; S2. For each material structure to be simulated, obtain a primitive cell structure according to the structure name, and analyze the elastic coefficient, point defects and number of surface structures related to the primitive cell structure; S3, generating a supercell structure according to the primitive cell structure, wherein the supercell structure is an extension of the primitive cell structure in three main axis directions in three-dimensional space; S4, generating a job input file for performing relaxation calculation simulation on the supercell structure according to the potential function information input by the user, and generating a script to submit the calculation job, and obtaining the relaxation energy value and structure size of the supercell structure after relaxation from the output information of the job; S5. For each elastic coefficient analyzed in step S2, the supercell structure is deformed in the deformation direction to construct at least two deformed substructures, and a calculation job is submitted for each deformed substructure in the same manner as in step S4 to obtain relaxation energy values ​​of the at least two deformed substructures, and the relaxation energy values ​​of the at least two deformed substructures and the relaxation energy values ​​of the supercell structure obtained in step S4 are fitted to an energy-volume quadratic curve with deformation as a variable, and the elastic coefficient of the material structure is calculated using the curve according to the following formula: Wherein, E is the relaxation energy value, V is the volume, and V0 is the volume of the relaxed structure after the supercell structure obtained in step S4 is relaxed; S6. According to the point defect information analyzed in step S2, a corresponding point defect structure is constructed using the supercell structure, and a calculation job is submitted for each point defect structure in the same manner as in step S4 to obtain a relaxation energy value of each point defect structure, and the relaxation energy value of the obtained point defect structure and the relaxation energy value of the supercell structure obtained in step S4 are calculated according to the following formula to obtain the point defect formation energy of the material structure: E def =E n±m -(nE 0 ±mμ i ) Among them, E def is the point defect formation energy, E n±m is the structural energy including m defect atoms, E 0 is the average relaxation energy per atom of the supercell structure obtained in step S4, n is the number of atoms in the supercell structure, m is the number of atoms in the defect itself, and μ i is the atomic chemical potential in the perfect structure of the i-th element, with a minus sign for vacancy defects and a plus sign for interstitial defects; S7, according to the surface structure information analyzed in step S2, the corresponding surface structure is constructed using the supercell structure, and a calculation job is submitted for each surface structure in the same manner as in step S4 to obtain the relaxation energy value of each surface structure, and the relaxation energy value of the obtained surface structure and the relaxation energy value of the supercell structure obtained in step S4 are calculated according to the following formula to obtain the surface formation energy of the material structure: Among them, E (ijk) is the surface formation energy of the (ijk) surface, is the energy of the surface structure of n atoms, E 0 is the average relaxation energy per atom of the supercell structure obtained in step S4, A (ijk) is the surface area of ​​the (ijk) surface.

2. The method for evaluating the molecular dynamics potential function based on material structure simulation according to claim 1, characterized in that: The potential function information received in step S1 includes the potential function name, parameter value or parameter file.

3. The method for evaluating the molecular dynamics potential function based on material structure simulation according to claim 1, characterized in that: Obtaining the primitive cell structure according to the structure name in step S2 includes: obtaining the primitive cell structure of the material structure from a general structure database of materials science according to the structure name.

4. The method for evaluating the molecular dynamics potential function based on material structure simulation according to claim 1, characterized in that: The step S3 further comprises: when the length of any side of the primitive cell structure is less than twice the cutoff distance of the potential function, expanding the supercell structure in each direction to an integer multiple of the size of the primitive cell structure until it exceeds twice the cutoff distance.

5. The method for evaluating the molecular dynamics potential function based on material structure simulation according to claim 4, characterized in that: The step S3 further comprises: expanding the size of the supercell structure in each direction to at least 6. The method for evaluating the molecular dynamics potential function based on material structure simulation according to claim 1, characterized in that: The step S4 further includes: submitting the job to a high performance computing server or a super computing system, and after the job is submitted, calling a molecular dynamics program in the system to perform simulation work.

7. The method for evaluating the molecular dynamics potential function based on material structure simulation according to claim 1, characterized in that: The step S5 further comprises: S51, for each elastic coefficient, using the supercell structure to perform 1%, 0.5%, -0.5%, -0.1% stretching or shearing deformation in the deformation direction, respectively, to construct four deformed substructures; S52, submitting calculation jobs for the four deformed substructures respectively in the same manner as in step S4 to obtain relaxation energy values ​​of the four deformed substructures; S53, fitting the relaxation energy values ​​of the four deformed substructures and the relaxation energy value of the supercell structure obtained in step S4 to an energy-volume quadratic curve with deformation as a variable, and calculating the corresponding elastic coefficient by fitting the curve; S54, repeating steps S51 to S53 until all elastic coefficients analyzed in step S2 are calculated.

8. The method for evaluating the molecular dynamics potential function based on material structure simulation according to claim 1, characterized in that: The point defects of the primitive cell structure analyzed in step S2 include vacancy defects and gap defects; the step S6 of constructing the corresponding point defect structure using the supercell structure further includes: for vacancy defects, constructing a vacancy defect structure for each position of each element in the primitive cell structure; for gap defects, constructing a gap defect structure for each element in each larger space in the primitive cell structure, wherein the larger space is defined as a space in which a radius greater than The space in which a sphere is located that does not include any atoms.

9. The method for evaluating the molecular dynamics potential function based on material structure simulation according to claim 1, characterized in that: The surface structure of the unit cell structure analyzed in step S2 includes a small index interface; the length, width, thickness and vacuum layer of the surface structure constructed in step S7 are all more than twice the cutoff distance of the potential function used.

10. A system, characterized in that: The method comprises a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, the method for evaluating the molecular dynamics potential function based on material structure simulation according to any one of claims 1 to 9 is implemented.

11. A computer-readable storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a processor, the method for evaluating the molecular dynamics potential function based on material structure simulation according to any one of claims 1 to 9 is implemented.

Images