CN-121983150-A - Element segregation grain boundary model construction method

CN121983150ACN 121983150 ACN121983150 ACN 121983150ACN-121983150-A

Abstract

The invention discloses a construction method of a grain boundary model of element segregation, which relates to the technical field of microstructure analysis of metal materials, and the technical scheme is as follows: according to the method, firstly, the internal energy of a grain boundary is calculated and is used for distinguishing the chemical difference between a grain boundary effect and a material, then all atoms in a grain boundary model are traversed to replace atoms, then the partial energy of solute atoms is calculated, and finally, a partial energy configuration model is constructed and is used for subsequent study of structural stability, mechanical response or diffusion behavior. According to the method, the site which is prone to be occupied by the solute is judged through the partial aggregation energy, so that the rapid construction of the element partial aggregation model is realized.

Inventors

ZHANG LIANG
LIAO ZHUOJING
Chen Cuifan
HUANG XIAOXU

Assignees

重庆大学

Dates

Publication Date: 20260505
Application Date: 20251225

Claims (6)

1. A construction method of a grain boundary model of element segregation is characterized by comprising the following steps: S1, preparing in the early stage; S2, calculating internal energy of a grain boundary; S3, atom replacement; s4, calculating partial energy accumulation of solute atoms; s5, constructing a visualization and segregation model.
2. The method for constructing a grain boundary model with element segregation according to claim 1, wherein the step of preparing the S1 is as follows: (1) A grain boundary model, namely constructing and pre-relaxing a stable grain boundary structure; (2) Potential function, selecting potential function capable of correctly describing interatomic interaction of the alloy system and configuring parameter and pair_coeff mapping; (3) Writing a method and a command for calculating partial energy aggregation, and defining a variable target_id equivalent { target_id } to receive an externally-transmitted TARGET atom ID; (4) Python script is responsible for traversing all atom IDs, calling LAMMPS running in file, and collecting and summarizing the partial energy gathering result of each position.
3. The method for constructing a grain boundary model with element segregation according to claim 1, wherein the specific steps of calculating the grain boundary internal energy in S2 are as follows: Performing primary energy minimization on the undoped grain boundary model to obtain system initial energy v_E_initial; meanwhile, before formally calculating the grain boundary partial energy concentration, calculating the replacement energy of solute atoms in ideal bulk crystals ) And uses it as a reference value E 0 for distinguishing the chemical difference between the grain boundary effect and the material itself.
4. The method for constructing a grain boundary model for element segregation according to claim 1, wherein the specific steps of the atom replacement in S3 are as follows: locating the site, namely transmitting the atom ID in the cycle to the target_id of the in file; type substitution, namely temporarily changing the atom type of the site into a solute type; Energy minimization, namely, energy minimization is carried out again on the replaced structure to obtain v_E_modified; The partial energy accumulation calculation, namely calculating the partial energy accumulation of the site according to a formula delta E seg =(v_E_modified - v_E_initial) - v_E 0 , wherein a negative value indicates that the site is more beneficial to solute partial accumulation; Recording the result, namely additionally writing the id and delta E seg into segr _entries. And (4) cycling, namely restoring the atom type to the original type and continuing to the next atom ID.
5. The method for constructing a grain boundary model for element segregation according to claim 1, wherein the specific steps of calculating the solute atom segregation energy in S4 are as follows: after the circulation is finished, a partial energy gathering list corresponding to all atoms one by one is obtained; Python merges atom ID, type, coordinate with Δe seg to generate LAMMPS dump file with segr _energy field.
6. The method for constructing a grain boundary model for element segregation according to claim 1, wherein the specific steps of the visualization and the construction of the segregation model in S5 are as follows: Visualization, namely coloring the dump file according to color codes in OVITO and other software to obtain a spatial distribution map of grain boundary bias energy, wherein the lower the color is, the more easily the bias energy is, and the lower the color is, the more negative the color is; and (3) constructing a model, namely selecting a plurality of lowest energy sites to place solutes according to a distribution result, and constructing a partial aggregation model for subsequent study of structural stability, mechanical response or diffusion behavior.

Description

Element segregation grain boundary model construction method Technical Field The invention relates to the technical field of microstructure analysis of metal materials, in particular to a method for constructing a grain boundary model of element segregation. Background The grain boundary segregation plays an important role in controlling the strength and the thermal stability of the nano metal material. The research shows that the grain boundary stability is improved through the segregation of the proper alloying element Mo on the grain boundary, so that the strength of the nano metal Ni-Mo is greatly regulated and controlled. In addition, the segregation of La element in the grain boundary is utilized to stabilize fine nano grains in thermodynamics and dynamics, so that the thermal stability of 304L austenitic steel is remarkably improved. Grain boundary segregation utilizes chemical bonding energy between added solutes and grain boundaries to reduce excess grain boundary energy, bringing the nanostructure closer to equilibrium. Grain boundaries generally have the property of being structurally disordered and have a wide variety of site types or local atomic environments other than the bulk lattice environment, which determines the nature of the grain boundaries and their interactions with solute atoms. Thus, since the advent of molecular dynamics simulation, researchers have largely explored solute bias behavior at the grain boundaries at the atomic level to understand the relationships between their structure, composition, and properties. The construction of an element segregation boundary model is a primary consideration in researching grain boundary segregation. At present, the model is mainly constructed by several methods such as random substitution, monte Carlo simulation and the like. Random substitution atoms are a static configuration sampling method, i.e., within a given grain boundary model, the lattice sites are randomly selected for atomic substitution according to a predetermined probability (typically based on average concentration or simplifying assumptions). This method can rapidly generate a large number of samples of the biased configuration with statistical significance (average concentration) for structural analysis or as input to other calculations and is easy to implement in existing codes. However, the random substitution method generates a snapshot set of instantaneous atomic arrangement, and the generated configuration does not represent any dynamic process or equilibrium state, but only random realization of average concentration. Monte Carlo simulation is one of the most commonly used methods at present, which is to seek microscopic states of system energy minimization by randomly attempting atomic migration/substitution based on thermodynamic probability (Boltzmann factor). The simulated evolution reaches an equilibrium or metastable biased configuration. The method is definitely based on statistical mechanics and can simulate the process of approaching balance. The resulting configuration reflects a temperature, concentration dependent bias equilibrium state and is now widely used to simulate bias at various interfaces. However, monte carlo simulation requires a large number of iterative steps (especially low temperature, high barrier systems) to reach statistical equilibrium, which is time consuming. The convergence of the result is greatly influenced by the number of steps, the movement strategy, the temperature setting and the like, and the result needs to be carefully checked. And the potential functions currently used for monte carlo simulations are limited to embedded potential functions. A large number of atomic simulation results show that the segregation site of the solute at the grain boundary is closely related to the segregation energy, and the segregation energy is used as a core parameter to construct a solute grain boundary segregation model, so that the solute grain boundary segregation model is a direct and efficient modeling strategy from the thermodynamic driving force perspective. This approach has significant advantages and unique value over random permutation such as monte carlo simulation or simplifications. The advantage of constructing a bias model using bias energy is that the bias energy defines the total energy change of the system caused by the migration of solute atoms from the interior of the lattice to the grain boundary region. The greater the negative value, the stronger the driving force and the higher the equilibrium bias concentration. The complex atomic scale interactions and entropy effects (although simplified processing) are concentrated into one core thermodynamic quantity, making the modeling logic clear—atoms more prone to aggregation where energy is low. And is computationally extremely efficient without the need to perform millions of random attempts and accept/reject criteria to explore the configuration space to reach equilibriu